raim-08 prepages to index

National Conference on Recent Advances in Innovative Materials

(RAIM-08)

www.excelpublish.com

Proceedings of the

National Conference on Recent Advances in Innovative Materials

(RAIM-08)

February 16-17, 2008

Editors

Dr. Subhash Chand Dr. A.S. Singha Dr. Kuldeep Kumar Sharma Dr. S.K.S. Parashar

Organised by

Department of Applied Sciences & Humanities, National Institute of Technology, Hamirpur – 177 005 (HP) India.

EXCEL INDIA PUBLISHERS

First Impression: 2008 © NIT, Hamirpur (HP) ISBN: 978-81-903047-7-1 No part of this publication may be reproduced or transmitted in any form by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the copyright owners. Published by EXCEL INDIA PUBLISHERS

61/28, Dalpat Singh Building, Pratik Market, Munirka, New Delhi-110067 Tel: +91-11-2671 1755/ 2755/ 5755 Fax: +91-11-2671 6755 E-mail: [email protected] Website: www.excelpublish.com Typeset and Printed by

Imprimerie 310 Gyandeep Complex, Munirka, New Delhi-110067 Tel.: +91-11-41668012, 20900870, (M) 9811123581

Preface

It gives us immense pleasure to bring out the proceedings of the National Conference on Recent Advances in Innovative Materials (RAIM-08) being held at National Institute of Technology, Hamirpur (HP), on February 16-17, 2008. Materials have always played a remarkable role in all stages of development of human civilization. This proceeding will highlight the important advances recently made in the area of materials which exhibit excellent properties and will find applications in various devices and systems.

The organizers of RAIM-08 have tried to arrange special invited lectures on various aspects of material fabrication, characterization and their applications. We take this opportunity to thank all the authors for contributing their papers to the conference which will benefit the readers in enhancing their knowledge. About 74 papers on various aspects of materials were received by us. All the papers went through a peer review process. About 20 papers were selected for oral presentations and 54 for poster presentations. The views expressed in the papers are exclusively that of the authors and editors are not responsible for any type of conflict or contradiction.

We are thankful to all the members of technical programme committee for their

efforts in putting together an excellent technical programme. Organising such a conference always need motivation and blessings of distinguished personalities. We are grateful to Dr. R.L. Chauhan, Honorable Chairman BOG, for his blessings and worthy Director, Professor I.K. Bhat for his constant encouragement, motivation and active contribution in the planning of this conference. We are also thankful to Dr. Piar Chand, Head, Department of Applied Sciences & Humanities for his consistent guidance and other authorities of NIT Hamirpur for making facilities available for the conference. We thank Dr. Arvind Kumar and Dr. Rajesh Kumar for their help. We also appreciate the untiring efforts by Mr. Manish Taunk and Mr. Atul Kapil and other research scholars at NIT Hamirpur.

We acknowledge D.R.D.O. New Delhi, Government of India for providing the

financial support.

Subhash Chand

Amar Singh Singha

Kuldeep Kumar Sharma

Sujay Kumar Singh Parashar

Contents

Preface v

1. Properties of Multiferroic BiMn2-xTixO5 (0 ≤ x ≤ 0.3)

D.K. Shukla, S. Mollah, R. Kumar, F. Singh and V. Ganeshan

1

2. Preparation and Characterization of Ni-Zn Spinel Ferrites A. Sharma, P. Mathur, A. Thakur and M. Singh

6

3. Preparation and Characterization of Strontium Ferrite by the Sol-Gel Method Shivendra Kumar Jaiswal and Jitendra Kumar

11

4. Study of Ni Substituted Mn-Zn Nanoferrite by Citrate Precursor Method Preeti Mathur, Ajay Sharma, Naveen Sharma, Atul Thakur and M. Singh

15

5. Characterization of Soft Ni-Zn Spinel Ferrites N. Sharma, Tanuj Sharma, P. Mathur, A. Thakur and M. Singh

19

6. Ferroelectric Relaxor Behaviour in Pb0.9Ba0.1(Fe0.5Ta0.5)O3 Alo Dutta, Chandrahas Bharti and T.P. Sinha

24

7. Effect of Annealing on Electrical Properties of Nanocrystalline CdSe Thin Films. Jeewan Sharma, G.S.S. Saini, N. Goyal and S.K. Tripathi

28

8. Self-Assembled and Ordered Templates by Anodic Oxidation of Aluminium S.K. Yadav, R. Gupta and K.N. Rai

33

9. Structural and Electrical Characterization of Lanthanum Substituted Barium Titanate Thin Films Pramod Singh Dobal, Anju Dobal and R.S. Katiyara

38

10. Synthesis of CdO Nanoparticles by Sol-Gel Technique Rajeev Kumar, R.K.Bedi, and Iqbal Singh

44

11. Deposition and Characterization of Nanocrystalline Thin Films of CdS from Chemical Route Jasim M. Abbas, Charita Mehta, G.S.S. Saini and S.K. Tripathi

47

viii Contents

12. Dielectric Spectroscopy of Silver Nanoparticles Embedded in Soda Glass Suman Bahniwal, Annu Sharma, Sanjeev Aggarwal, S.K. Deshpande and P.S. Goyal

51

13. AC Conductivity in a-Ge-Se-Ag Glasses Gurinder Singh, N. Goyal, G.S.S. Saini and S.K. Tripathi

53

14. Photoconductive Measurements of Thermally Deposited a-Ge20Se80-xInx Thin Films Ishu Sharma, Pankaj Sharma, P.B. Barman and S.K. Tripathi

57

15. Synthesis of MgB2 Superconductor with Different Method Kiran Singh, Rajneesh Mohan, N.K. Gaur and R.K. Singh

61

16. Synthesis and Characterization of Polyaniline Thin Films Atul Kapil, Inderjeet Kaur and Subhash Chand

64

17. Fabrication and Characterization of Indium Tin Oxide-PTCDA-Aluminium/ Silver Junctions Manish Taunk and Subhash Chand

67

18. Effect of HgO and AgO Addition on the Electrical Properties of YBa2Cu3O7-δ. Rajneesh Mohan, Kiran Singh, Nupinderjeet Kaur, S. Bhattacharya, N.K. Gaur and R.K. Singh

70

19. Dielectric Relaxation Studies of N, N-dimethyl Formamide in the Benzene Solution from Microwave Absorption Data Vimal Sharma and Nagesh Thakur

73

20. Dielectric Relaxation of Tetramethylurea in Benzene and Carbon Tetrachloride Solutions From Microwave Absorption Data Rajesh Kumar and Nagesh Thakur

77

21. Dielectric Response of Nitrogen Ion Implanted CR-39 Polymer Nidhi, Tanu Sharma, Sanjeev Aggarwal, Annu Sharma , S. Kumar, S.K. Deshpande and D. Kanjilal

82

22. Dielectric Relaxation of N, N-Dimethylacetamide in Benzene Solution from Microwave Absorption Data Raman Kumar and V.S. Rangra

85

23. Mixed Electronic-ionic Conductivity in Copper Phosphate Glasses Doped with Sodium Oxide P.S. Tarsikka and B. Singh

89

Contents ix

24. Properties of Alumina Particulate Reinforced Aluminum Alloy Produced by Stir Casting Ravindra Mamgain, H.S. Bains, A. Manna and S.N. Basu

92

25. Study of Polypyrrole Polymer Films for Sensing and Microwave Properties D.C. Tiwari, Rishi Sharma and Vikas Sen

96

26. Synthesization and Characterization of Co-doped ZnO Diluted Magnetic Semiconductor A.P. Singh, R. Kumar, P. Thakur, K.H. Chae, Basavaraj Angadi and W.K. Choi

100

27. Growth of Highly Oriented AgInSe2 Films for Solar Cell Applications Dinesh Pathak , R.K. Bedi and Davinder Kaur

104

28. Electrical Properties of Amorphous GaxSe1-x Thin Films Falah Ibrahim Mustafa, N. Goyal and S.K. Tripathi

107

29. Limiting Behaviour of AC Conductivity in Nanoferroelectric PZT based Ceramics S.K.S. Parashar, Kajal Parashar, R.N.P. Choudhary and B.S. Murty

110

30.

Effect of Annealing on the Optical Constants of (Ge20Se80)90Ag10 Thin Films Akshay Kumar, F.I. Mustafa, Shikha Gupta and S.K. Tripathi

114

31. Effect of UV-irradiation on Chemically Deposited Nanocrystalline CdSe Films Charita Mehta, Jasim M. Abbas, G.S.S. Saini, S.K. Tripathi

117

32. Optical and Structural Investigations of TiO2 Films Deposited on Transparent Substrates by Sol-Gel Technique Sudhir Kumar Sharma, M. Vishwas, K. Narasimha Rao, S. Mohan and K.V. A. Gowda

122

33. Effect of Annealing Temperature on the Optical Properties of (Pb0.8Sr0.2)TiO3 Thin Films K.C. Verma, K. Singh, Vinay Gupta and N.S. Negi

126

34. Non-Linear Optical Property of CdS Nanomaterials Synthesized by Chemical Route C.S. Tiwary, P. Kumbhakar, A.K. Mitra and R.N. Roy

129

x Contents

35. Optical Band Gap and Refractive Index of Vacuum Evaporated a-Ge10Se90-xTex (x = 0, 20) Thin Films Pankaj Sharma, Ishu Sharma and S.C. Katyal

134

36. Light Induced Changes in Se-Te Thin Film Vineet Sharma and Anup Thakur

138

37. Photo Bleaching in a-Ga50Se50 Thin Films Shikha Gupta, F.I. Mustafa, Akshay Kumar, N. Goyal, G.S.S. Saini and S.K. Tripathi

141

38. Preparation of Nanoparticles by Laser Ablation in Solution: Influence of the Laser Wave Length and the Surrounding Liquid Environment on Particle Size R. Sarkar, P. Kumbhakar and A.K. Mitra

145

39. The Synthesis of Copper Oxide Nanoparticles by Nitrate-Citrate Gel Method Iqbal Singh, R.K. Bedi and Rajeev Kumar

149

40. Corrosion Resistant Properties of Surface Tolerant Coatings Shyamjeet Yadav and D.D.N. Singh

153

41. Apparent Molar Volumes and Viscosities of Amino Acids in Aqueous Sodium Nitrate Solutions at 298.15 K Tarlok S. Banipal, Vaneet Dhir and Parampaul K. Banipal

159

42. Effect of Diluents on the Thermal Behaviour of Vinyl Ester resins Bharti Gaur

162

43. Evaluation of Crystallinity of Graft Copolymers of Flax with Binary Vinyl Monomers Susheel Kalia, B.S. Kaith and A.S. Singha

167

44. Synthesis and Structural Studies of N-(2-oxo-3-oxa-4(Phenyl)-Butanyl Benzene Sulphonamide and Related Compound as Potential Juvenile Hormone Analogue Pamita Awasthi, Shilpa Dogra and R.K Mahajan

170

45. Gelatin Grafted Polypropylene: Kinetics of Biodegradation Inderjeet Kaur and N. Deepika Khanna

176

46. Crystal and Molecular Structure of 1-(4-Chlorophenoxy) 3, 3-Dimethyl-1-4 (1, 2, 4-Triazole-1-Y-1)-2-Butanone Anjana Chauhan, Rachana Tiwari and R.K. Tiwari

182

Contents xi

47. Influence of Brass as a Filler on Load-Speed Sensitivity of Polymer Based Friction Composites Mukesh Kumar and Jayashree Bijwe

187

48. Reinforcement of Graft Copolymers of Binary Vinyl Mixtures of Methyl Meth Acrylate (MMA) onto Saccharum Cilliare Fiber with Urea-formaldehyde Matrix and Evaluation of their Mechanical Properties A.S. Singha and Anjali Shama

194

49. Application of Waste Bio-mass in p-r-f- based Composites and Study of their Properties A.S. Singha and Ashwarya Jyoti Khanna

200

50. Characterization and Salt-resistant Study of Gum Arabic and Methacrylic Acid Based Hydrogel B.S. Kaith and Shabnam Ranjta

204

51. Evaluation of Mechanical Properties of Grewia Optiva Fiber Reinforced Polymer Composites A.S. Singha and Vijay Kumar Thakur

208

52. Synthesis and Evaluation of Green Composites Based on Hibiscus Sabdariffa Reinforced Thermosetting Resin A.S. Singha and Vijay Kumar Thakur

212

53.

Phonon Properties of Intermetallic Compound AuIn2 M.M. Sinha

216

54. Stress Analysis of a Laminated composite Plate with Central Hole Under in-Plane Static Loading Using ANSYS – A Case Study M.P. Nagarkar, R.N. Zaware and R.R. Navthar

219

55. Low Temperature Specific Heat of Codoped Mg1-x(AlLi)xB2 Nupinderjeet Kaur, Rajneesh Mohan, N.K. Gaur and R.K. Singh

223

56. Restricted Diffusion in Nano-Channels K. Tankeshwar and Sunita Srivastava

227

57. Mechanical Strength and Phase Transformation of AlN P.S. Bisht, Virshali Joshi and U.P. Verma

230

58. Novel Feature of Quantum Transport through Ultra-Thin Quantum Film Santanu K. Maiti

235

xii Contents

59. Electronic Band Structure of Isolated Zigzag Single Wall Carbon Nanotubes Vikas Thakur, P.S. Bisht, U.P. Verma and P. Raja Ram

240

60. Modeling of Band-gap Using Strain Theory for Silicon V.K. Lamba, Ankur Gupta and Munish Verma

243

61. Pair Potentials of Expanded Fluid Rubidium Using Ab-initio Pseudopotentials Avneesh Sharma, Nitu Sharma and Raman Sharma

246

62. Modelling and Analysis of Interconnect on Wafer Munish Verma , S.S. Gill and V.K. Lamba

249

63. Effect of Germanium Content on Physical Properties of [Se80Te20]100-X GeX (X= 0,2,4,6) Mainika and Nagesh Thakur

253

64. High Pressure Phase Transition of Zinc-Blende Copper Chloride Deoshree Baghmara, Sadhna Singh, D.C. Gupta and N.K.Gaur

256

65. Stability of Na Clusters Inside C240 Molecule Harkiran Kaur, K. Ranjan and Keya Dharamvir

260

66. Phase Transition in ASe(A= Ca, Eu and Th) D.C. Gupta and K.C. Singh, Subhra Kulshrestha and Sonia Mehra

265

67. Parametric Analysis and Optimization of Cutting Parameters for Turning Operations Based on Taguchi Method Amar Patnaik , S.K. Pradhan and Maheshwar Dwivedy

268

68. Study of Formation of Diamond-Like Carbon Coatings and Theoretical Calculations of Stiffness of the Films Avnish Narula

273

69. A Study on Acoustic-Diffusive Wave Propagation Phenomenon in Semiconductor Materials in Contact with Fluid J.N. Sharma, Indu Sharma and Subhash Chand

278

70. Plane Harmonic Waves in Generalized Thermoelastic Materials with Voids D. Kaur and J.N. Sharma

283

71. Free Vibrations in a Cylindrical Panel of Heat Conducting Viscoelastic Material. P.K. Sharma, V. Walia and J.N. Sharma

288

Contents xiii

72. Transient Waves Due to Continuous Thermal Loads in Thermoviscoelastic Materials Rattan Chand, Nisha Sharma and J.N. Sharma

292

73. Forced Vibrations of Solid Thick Plate of Thermoviscoelastic Material P.K. Sharma, K.K. Sharma, V. Walia and A. Kumar

297

74. Thermoelastic Wave Propagation in Circumferential Direction of Transversely Isotropic Spherical Curved Plates Nivedita Sharma and J.N. Sharma

301

Author Index 305

Properties of Multiferroic BiMn2-xTixO5 (0 ≤ x ≤ 0.3)

D. K. Shukla1,*, S. Mollah1, R. Kumar2, F. Singh2 and V. Ganeshan3

1Department of Physics, Aligarh Muslim University, Aligarh 202002, India. 2Inter University Accelerator Centre, Aruna Asaf Ali Marg, New Delhi 110 067, India.

3UGC-DAE Consortium for Scientific Research, University Campus, Indore 452 017, India. E-mail: *[email protected]

Abstract

Polycrystalline single phase BiMn2-xTixO5 (0 ≤ x ≤ 0.3) has been prepared using standard solid state

reaction technique and characterized by x-ray diffraction (XRD), scanning electron microscopy (SEM), Raman spectroscopy and specific heat measurement techniques. XRD data shows that BiMn2-xTixO5 is orthorhombic (s.g. Pbam) up to x = 0.3, further SEM study indicates the decrease in porosity and grain size with Ti substitution. The change in phonon modes due to Ti substitution has been investigated from unpolarized Raman spectra. A low frequency Ag mode is found to be less asymmetrical with Ti substitution and is discussed elaborately. The specific heat anomaly associated with antiferromagnetic transition is suppressed with increasing Ti content and is consistent with the literature. Low temperature specific heat (LTSH) fitting shows the increasing magnetic contribution to the specific heat with Ti substitution. 1. Introduction

Magnetism arises due to spin nature of electron, whereas charge nature of electron turns out ferroelectricity. These two unrelated phenomena can co-exist together in certain materials in same phase and are known as multiferroics. Though the research on multiferroics was started in early 1960’s [1,2], a great upsurge has come since last six-seven years, due to the technological importance of these materials and are being extensively studied. [3-6]. The simultaneous occurrence of ferromagnetism/ antiferromagnetism (FM/AFM) and ferroelectricity (FE) and coupling between these two order parameters lead to the emergence of new storage media, which enable the electrically reading/writing of the magnetic memory and vice versa, yielding more degrees of freedom from device application point of view [4, 5].

Multiferroic materials RMn2O5 (R3+ Mn3+ Mn4+ O5

2-; where R is rare-earth, Y or Bi), order antiferromagnetically followed by a ferroelectric transition around ~ 40 K, has created a lot of interest in recent years [6-9]. In the context of reason behind co-existence of these unusual properties has been classified depending upon their ground state spiral magnetic structure and interactions [10, 11]. RMn2O5 system show mixed valence Mn-sites (Mn3+/Mn4+), having orthorhombic symmetry described by the space group Pbam (No. 55) and Z = 4. Detail about the crystal structure of BiMn2O5 designed (Fig. 1) for a unit cell structure of pure BiMn2O5 using refined

parameters from Rietveld refinement is presented elsewhere [12]. In orthorhombic RMn2O5, the spins of the Mn4+ and Mn3+ ions and R3+ are coupled together via the predominantly AFM superexchange (SE) interactions giving rise to a complex magnetic phase diagram [13]. The common features for all RMn2O5 are having a transition to a high-temperature Neel phase with two component incommensurate (IC) magnetic modulation characterized by a wave vector q = (qx, 0, qz) at TN ≈ 39 K and followed by a lock-in transition into a commensurate (CM) phase with q = (0.5, 0, 0.25) at TC ≈ 35 K. Any type of substitution at Mn site will lead to substantial change in its magnetic and dielectric properties because of change in magnetic coupling among Mn atoms. Doping of Ti4+ (d0) ion decreases the Mn-Mn interaction through oxygen and the Mn-Mn distance increases due to the formation of Mn4+-O-Ti4+ and Mn4+-O-Mn4+ chains, which may affect the ferroelectric and magnetic properties of BiMn2O5 compound.

Although a considerable number of reports have been cumulated on the structural, magnetic and dielectric properties of RM2O5, there are scarce data on the thermodynamic behaviour of these materials. Structural anomalies during transitions [7, 9 ] have been investigated in order to understand the nature of phonons participating in observations. In particular, neither theoretical nor experimental results on the magnons are available so far. In our earlier study [12] with the Ti substitution, we have observed the dilution of

2 Recent Advances in Innovative Materials

antiferromagnetism (TN ~ 39 K) and appearance of new weak magnetic anomaly at ~ 86 K and dielectric anomalies at ~120 K. In this work, we present the heat capacity measurement and Raman spectra of Ti substituted BiMn2O5 along with its structural characterization. Attempt has been made to correlate the effect of Ti substitution in the phonons and magnons through the Raman data and magnetic contribution in heat capacity measurement using fundamental specific heat model. 2. Experimental

Polycrystalline bulk BiMn2-xTixO5 (0 ≤ x ≤ 0.3) multiferroic samples have been synthesized using conventional solid-state reaction technique. The stochiometric amounts of Bi2O3, MnO and TiO2 powders of 99.99% purity were mixed thoroughly and precalcinated for 12 h at 800 °C. These were again ground and calcinated at 820 °C for 24 h. Finally, the samples were ground to fine powder, pressed into pellet forms and sintered at 840 °C for 24 h. This heat-treatment procedure was repeated for three times to get the better homogeneity in the samples. Powder x-ray diffraction measurements were performed using Bruker D8 X-ray Diffractometer with Cu Kα radiation at room temperature. Morphological study using scanning electron microscope (SEM), (JEOL-JSM-6360 with EDS unit) has been performed. Raman measurements were performed using micro-Raman setup from Renishaw having Ar ion laser excitation 514 nm at room temperature. To avoid the heating effect, laser beam was focused at very low power (< 2 mW, 20 X objective). The specific heat measurements have been performed using a commercial setup (14 Tesla PPMS, Quantum Design). The sample platform used for measurements above 2K is temperature as well as field calibrated and the addenda contributions are appropriately subtracted from the measured total specific heat. The typical error in the sample specific heat is estimated to be < 0.5%. 3. Results and discussion

X-ray diffraction of BiMn2-xTixO5 (0 ≤ x ≤ 0.3) at room temperature have been analyzed using Rietveld refinement [13] technique and refined plots are presented in Fig. 2.

Complete feature of XRD results are illustrated elsewhere [12] with tables of refined parameters. Refined XRD pattern for x = 0.0, 0.15

and 0.30 samples in the form of observed, calculated and difference (Fig. 2) illustrate a good agreement between observed and calculated profiles.

Fig. 1. Schematic crystal structure of BiMn2O5 and magnetic structure in ab plane along the c axis. octahedra in cryan represents Mn4+O6 geometry and pyramid in red shows Mn3+O5 arrangemnt. Green and white balls represent Bi and O atoms respectively.

20 30 40 50 60 70 80

x = 0.30(c)

2θ (degree)

x = 0.15(b)

Inte

nsity

(Arb

. Uni

ts)

Experimental Calculated Difference Bragg's position

BiMn2-xTixO5

x = 0(a)

Fig. 2. X-ray diffractions pattern of BiMn2-xTixO5 materials with (a) x = 0, (b) x = 0.15 and (c) x = 0.3.

O4 O4

b

c a

AFM

FM

FM AFM

Properties of Multiferroic BiMn2-xTixO5 (0 ≤ x ≤ 0.3) 3

(a) X = 0

(b) X = 0.15

(c) X = 0.30

Fig. 3. SEM micrographs of BiMn2-xTixO5 materials with (a) x = 0, (b) x = 0.15 and (c) x = 0.3.

From the XRD pattern, we can conclude that the samples with Ti concentration up to x = 0.3 are in single phase having orthorhombic structure with space group Pbam.

Further SEM micrographs have also been taken in order to study the surface morphology of the prepared samples. Figure 3 shows the SEM micrographs of BiMn2-xTixO5 materials for (a) x = 0, (b) x = 0.15 and (c) x = 0.3. SEM clearly shows that increasing content of Ti in BiMn2O5 decreases the porosity and grain size.

200 300 400 500 600 700

BiMn2-xTixO5

x = 0.30

x = 0.15

x = 0

Inte

nsity

(arb

. uni

ts)

Raman Shift (cm-1)

Fig. 4. Unpolarized Raman spectra of BiMn2-

xTixO5 materials for x = 0, 0.15 and 0.3, collected at room temperature.

Raman spectra were taken at room temperature. The site symmetry analysis [15] of the Pbam structure of BiMn2O5 yields a total 96 Г-point phonon modes, out of these 48 phonon modes are Raman-active (ГRaman = 13Ag + 13B1g + 11B2g + 11B3g). The Ag modes are expected to appear in the parallel xx, yy and zz scattering configuration and should not be seen in the crossed xy, xz and yz configurations. The B1g, B2g and B3g modes are expected respectively in xy, xz and yz configurations. Raman tensors for these modes can be represented in following form.

⎥⎥⎥

⎦

⎤

⎢⎢⎢

⎣

⎡=

cb

aAg

000000

,

⎥⎥⎥

⎦

⎤

⎢⎢⎢

⎣

⎡=

0000000

1 dd

B g ,

⎥⎥⎥

⎦

⎤

⎢⎢⎢

⎣

⎡=

00000

00

2

e

eB g and

⎥⎥⎥

⎦

⎤

⎢⎢⎢

⎣

⎡=

0000

000

3

ffB g

Table 1. Summery of fitting (with Eq. (1)) results to the data in Fig. 6. The units of different quantities are γ (J/mol K2), β(J/mol K4), B(J/mol K3) and ΘD (K). BiMn2-xTixO5 γ β B ΘD x = 0 0 0.0012 0.0049 235 x = 0.15 0 0.00025 0.0208 396 x = 0.30 0 10-6 0.0259 2494

Figure 4 shows the unpolarized Raman spectra of BiMn2-xTixO5 materials for (a) x = 0, (b) x = 0.15 and (c) x = 0.3 colleted at room

temperature. For all these three samples 19 Raman peaks were identified at 185, 195, 200, 235, 245, 280, 300, 325, 345, 360, 405, 445, 480,


510, 545, 565, 595, 610 and 655 cm-1 for the spectra collected between 100 to 800 cm-1. These Raman modes are in agreement to the earlier report by Garcia Flores et al [7] for BiMn2O5. An assignment of the Raman lines to definite modes is not possible at present. However, Raman mode frequencies in ionic materials, such as transition metal oxides are determined by the mass, charge and bond lengths of the participating atoms as well by the type of the atomic motions (stretching, bending or rotational). On the basis of mass and bond length consideration, it is reasonable to accept that the lines at > 300 cm-1 correspond to modes due to stretching and bending vibrations of lighter oxygen atoms. Whereas those at lower frequencies involve motion of heavier Mn and Bi atoms and involvement of Ti atoms in case of doped samples. Lines at lower frequencies are extremely sharp, indicating weak anharmonicity. In particular, with the Ti substitution the maximum effect is observed on the modes at ~ 200 cm-1. These modes corresponds to Ag modes as observed by Mihailova et al [9] in their polarized Raman study of RMn2O5 system. In our earlier study [12] through detail analysis of XRD and x-ray absorption spectroscopy (XAS) data we have confirmed that Ti is substituted at Mn4+ i.e. at octahedra site. For x = 0.15 small mass difference between Ti4+ and Mn4+ in octahedra environment has resulted the decrease in peak intensity and further substitution by x = 0.30 shows fusing feature of the peak at 195 and 200 cm-1. This may be the resemblance of removal of asymmetrical vibration at 195 and 200 cm-1. The manner of change in the Raman modes is in coherence with the dilution of antiferromagnetism due to increase of Ti content. Further the heat capacity measurement, we present here, also shows the diminishing behaviour of peak at its antiferromagnetic transition temperature (TN ~ 39 K) due to Ti substitution (see Fig. 5).

The temperature dependent specific heat measurement at zero magnetic field is shown in Fig. 5. A sharp anomaly at around antiferromagnetic ordering temperature TN ~ 39 K is in good agreement with report on magnetization measurement by us [12] and Munoz et al, [8]. With the Ti substitution, this anomaly diminishes in the same fashion as the magnetization data.

0 50 100 150 200

0

40

80

120

0

40

80

120

0

40

80

120

C (J

/mol

K)

T (K)

BiMn2-xTixO5

(c) x = 0.30

(b) x = 0.15

(a) x = 0

Fig. 5. Temperature dependence of the specific heat at zero magnetic field of BiMn2-xTixO5 materials for (a) x = 0, (b) x = 0.15 and (c) x = 0.3.

0 2 4 6 8 10 12 14 16

0123456701234567012345678

C = γT + βT3+ BT2BiMn2-xTixO5(a) x = 0β = 0.00122 (0.00012)B = 0.00494 (0.00158)

C (J

/mol

K)

T (K)

(b) x = 0.15β = 0.00025(0.00004)B = 0.02085 (0.00054)

(c) x = 0.30β = 0.000001 (0.0)B = 0.02592(0.00019)

Fig. 6. Low temperature (2-16 K) specific heat data fitted using the relation C = γT + βT3 + BTδ

for BiMn2-xTixO5 materials with (a) x = 0, (b) x = 0.15 and (c) x = 0.3. The low temperature specific heat (LTSH) data has been fitted by the expression

C = γT + βT3 + BTδ .…(1)

The linear term is associated with the charge – carrier contribution; the βT3 term corresponds to the lattice contribution and the last term BTδ (δ = 3/2 for ferromagnetic contribution and 2 for 2D antiferromagnetic) is related to the spin-wave excitations. First we fit the data using all three terms in Eq. (1) resulting in negative value of


parameters. By considering the data from 2 K to 16 K, the best fit is obtained if the linear term is neglected and δ = 2. The fitted parameters for lattice contribution, β (in J/mol K4) and magnon contribution, B (in J/mol K3) and Debye temperature ΘD is given in Table 1. The Debye temperature ΘD is calculated from β through the expression ΘD = (12π4nR / 5β), R being the gas constant and n the number of atoms per formula unit. For pure BiMn2O5 value of β is very less than B clearly indicating dominating contribution from antiferromagnetic ordering of Mn3+/Mn4+

atoms and values are in good agreement to that obtained by Munoz et al [8]. It is observed that with the Ti substitution by x = 0.15 the lattice contribution is further decreased and magnetic contribution is increased substantially. For x = 0.30, value of lattice contribution approaches almost to zero (10-6) and magnetic contribution is further increased i.e. at x = 0.3 specific heat is showing almost T2 dependence. Again it is to be noted that enhanced thermal excitation of spin wave corresponds to reduced magnetic dimensionality. Noticeably this observation shows the increased effect of net magnetic moment even if the AFM ordering diminishes with the Ti substitution. In our magnetic measurement also we have observed increasing net magnetic moment (μB /f.u.) with the Ti substitution. This is only possible if Mn3+/Mn4+

ratio increases i.e. Ti is substituted at Mn4+ site in antiferromagnetically ordered chain of Mn3+ - O - Mn4+. This clearly indicates that only Mn4+ ions are replaced by Ti substitution. 4. Conclusions Crystal structure, surface morphology, Raman spectra at room temperature and LTSH of BiMn2-xTixO5 (0 ≤ x ≤ 0.3) has been studied. From our data (1) Ti content up to x = 0.3 is accommodated in single phase orthorhombic structure having space group Pbam. (2) Ti doping decreases the porosity and grain size. (3) Change in phonon mode at lower frequencies is observed indicating the removal of asymmetrical vibration of Mn4+O6 octahedra due to Ti substitution. (4) Anomaly at TN ~ 39 K in specific heat vs temperature measurement decreases with Ti content, supports the weakening of antiferromagnetic ordering. However, (5) LTSH data fitted using C = γT + βT3 + BTδ relation shows dominating magnetic contribution to the specific heat at low temperature with Ti substitution and for x = 0.30 lattice contribution is

negligible. Debye temperature calculated for x = 0.30 is unexpectedly high. Acknowledgements

D.K.S. is thankful to the Inter-University Accelerator Centre (IUAC), New Delhi, India and CSIR, New Delhi for providing financial support. Department of Science and Technology (DST), Government of India, is acknowledged for funding the 14 Tesla-PPMS used in specific heat measurement. References [1] I. E. Dzyaloshinskii, Sov. Phys.-JETP 10

(1959) 628. [2] D. N. Astrov, Sov. Phys-JETP 11 (1960) 708. [3] M. Gajek, M. Bibes , S. Fusil, K.

Bouzehouane, J. Fontcuberta, A. Barthélémy, A. Fert, Nature Materials 6 (2007) 296.

[4] W. Erenstein, N. D. Mathur, J. F. Scott, Nature 442 (2006) 17.

[5] R. Ramesh, N. A. Spaldin, Nature Materials 6 (2007) 21.

[6] N. Hur, S. Park, P. A. Sharma, J. Ahn, S. Guha, S. W. Cheong, Nature (London) 429 (2004) 392.

[7] A. F. García-Flores, E. Granado, H. Martinho, R. R. Urbano, C. Rettori, E. I. Golovenchits, V. A. Sanina, S. B Oseroff, S. Park, S. W. Cheong, Phys. Rev. B 73 (2006) 104411.

[8] A. Munoz, J. A. Alonso, M.T. Casais, M. J. Martinez-Lope, J. L. Martinez, M. T. Fernandez-Diaz, Phys. Rev. B 65 (2002) 144423.

[9] B. Mihailova, M. M. Gospodinov, B. Güttler, F. Yen, A. P. Litvinchuk, M. N. Iliev, Phys. Rev. B 71 (2005) 172301.

[10] T. Kimura, Annu. Rev. Mater. Res. 37 (2007) 387.

[11] S. W. Cheong, M. Mostovoy, Nature Materials 6 (2007) 13.

[12] D. K. Shukla, S. Mollah, R. Kumar, P. Thakur, K. H. Chae, A.Banerjee, W.K.Choi, Communicated in J. Phys: Cond. Mater. (2007).

[13] G. R. Blake, L. C. Chapon, P. G. Radaelli, S. Park, N. Hur, S. W. Cheong, J. Rodríguez-Carvajal, Phys. Rev. B 71 (2005) 214402.

[14] J. Rodriguez-Carvajal, Physica B 192 (1993) 55; ibid J. Recent Developments of the Program FULLPROF, in Commission on Powder Diffraction (IUCr) Newsletter 26 (2001) 12.

[15] D. L. Rousseau et al., J. Raman Spectrosc. 10 (1981) 253.

Preparation and Characterization of Ni-Zn Spinel Ferrites

A. Sharma, P. Mathur, A. Thakur and M. Singh

Material Science Laboratory, Department of Physics, H. P. University, Summer Hill, Shimla -5, India. E-mail: [email protected]

Abstract

In the present study, preparation and characterization of Ni-Zn soft ferrites with formula NixZn1-

xFe2O4, where x = 0.1, 0.2 and 0.3, by co-precipitation method has been reported. Microstructural properties studied with the help of X-ray diffraction and scanning electron microscopy shows that the samples have almost uniform sized crystallites and have single phase spinel structure. The properties like initial permeability, permeability loss, dielectric constant, dielectric loss are studied as function of frequency. The d.c. resistivity of the samples is also studied. Variation of dc resistivity with composition, x, of Ni-Zn sample shows that pure zinc ferrite has more resistivity than that of nickel substituted zinc ferrite. The dispersion in magnetic and electrical properties has been discussed by using various models and theories.

1. Introduction

Soft ferrites such as Mn-Zn and Ni-Zn have a wide range of applications in electronic components such as inductors, wide band transformers, antenna cores, HF transformers [1-5]. Spinel Ni-Zn ferrites are particularly suitable for read-write heads used in high speed digital tapes or discs [6]. The electrical and magnetic properties of soft ferrites have been studied by many workers [7-11]. The usefulness of the ferrite is strongly influenced by the physical and chemical properties of the materials and depends upon many factors including the processing technique used to synthesize them. In order to get good quality ferrite with reproducible stoichiometric composition and desired microstructure, a recently reported co-precipitation method has been used to synthesize ferrite. In the present communication, we have studied the properties like initial permeability, permeability loss, dielectric constant, dielectric loss and d.c. resistivity of NixZn1-xFe2O4 ferrites with x = 0.0, 0.2 and 0.4. 2. Experimental details

Ni-Zn ferrites of composition Nix Zn1-xFe2O4 with x = 0.0, 0.2 and 0.4 were prepared by co-precipitation method [12-13]. The materials used were nickel chloride (98% Merck, India), zinc chloride (96% Merck, India), iron (III) chloride (98% Merck, India) and sodium hydroxide (96% Merck, India). One molar solution of these materials was made with distilled water. 70ml sodium hydroxide was taken from one molar solution in approximately 1760ml of distilled

water to have the concentration of 0.37 mol/litre and heated to boiling. Nickel chloride, zinc chloride and iron (III) chloride were taken from their molar solution in accurate stoichiometric proportions. These solutions were poured as quickly as possible into boiling solution of NaOH under vigorous stirring produced by the glass mechanical stirrer (~ 500 r.p.m.). Mixing is very important otherwise segregation of phases can take place. After co-precipitation, pH is set 12.5-13. Reaction vessel is covered with plastic cover to diminish evaporation of the solution. Reaction is continued for 30-40 minutes at temperature 90-100°C under vigorous stirring. Reaction vessel is cooled to ambient temperature and particles precipitate. Total volume is reduced to ~500ml by aspiration of supernatant. Suspension is centrifuged in 4 beakers at 7000 r.p.m. for 10 minutes. Precipitate is washed with distilled water of ~900ml and centrifuged once more in 4 beakers at 7000 r.p.m. for 10 minutes. The residue is dried and was calcinated at 5000 in a box type furnace for 15 hours at the rate of 200°C /h to obtain a ferrite powder. This powder was mixed with 2% P.V.A. binder and pressed into pellets of 1.50cm diameter and 0.20 cm thickness under pressure of 5 tons (1 ton = 1.016 x 103 kg) and rings of 1.5 cm outer diameter, 1.0 cm inner diameter and 0.20 cm thickness under a pressure of 3 tons. These samples were sintered in air at 1000°C at a heating rate of 250°C/h for 30 minutes and were subsequently cooled to room temperature by switching off the power supply of the furnace. The pellets were coated with silver paste to provide electrical contacts and the rings were wound with about 50 turns of 30 SWG enameled copper wire to form torroids.

Preparation and Characterization of Ni-Zn Spinel Ferrites 7

X-ray diffraction measurements were taken on a Rigaku Geiger Flex 3 kW diffractometer using CuKα source. Scanning electron micrograph (SEM) was recorded by using Cambridge Stereo Scan 360 scanning electron microscope. Dielectric constant, dielectric loss, initial permeability and loss factor were measured by using Agilent Technologies 4285A Precision LCR Meter upto 30MHz. Resistivity as a function of temperature was measured by using Two-Probe method. 3. Results and Discussion

Fig.1 shows the X-ray diffraction pattern of Ni0.2Zn0.8Fe2O4 ferrite. The figure shows the position and relative intensities of the various X-ray diffraction peaks. The diffraction pattern showed characteristic lines of spinel structure ferrite and no extra lines were observed, which indicate that all the samples had single phase spinel structure.

20 30 40 50 60 70 80150

200

250

300

350

400

450

500

550

Fig.1 X-ray diffraction pattern of Ni0.2Zn0.8Fe2O4 sintered at 10000C

533

440

511

420

400

311

220

111

coun

ts

2θ

The average grain particle size has been evaluated from (FWHM) i.e. full width at half maximum of the reflection of maximum intensity in the XRD pattern using Scherrer’s formula [14]

d = θ

λcosBK

……………………….(1)

where B2 = 22SM BB − , d is the particle size in

A0, K is the shape factor (taken as 0.9), λ is the X- ray wavelength (1.54 A0), BM & BS are measured peak broadening and instrumental broadening in radian respectively. The calculated average particle sizes were found to be around 1µm.

Fig. 2 shows the scanning electron micrograph of Ni0.2Zn0.8Fe2O4 ferrite sample.

Fig. 2. Scanning electron micrograph of Ni0.2Zn0.8Fe2O4 sintered at 10000C

The micrograph indicates that the sample has almost uniform sized crystallites with uniform grain growth. The samples prepared by the conventional method [15] and other non-conventional methods [16] had grain sizes larger than those observed in the present work. Hence, co-precipitation method produces ferrites with uniform grain growth & uniform size crystallites.

The observed variation of initial permeability, μi, as a function of frequency of the applied field in the frequency range from 75 kHz to 30 MHz is shown in Fig. 3 for the NixZn1-

xFe2O4 (x = 0.0, 0.2 and 0.4) ferrite sample. It is observed that initial permeability, μi, decreases with increasing frequency. The permeability decreases sharply in the lower frequency range. The initial permeability, μi, which is flux induced in the material at very low fields arises as a result of reversible movement of domain walls [17]. Greater the number of domain walls, higher is the permeability.

100 1000 10000-1000

0

1000

2000

3000

4000

5000

6000

7000

8000

Fig.3 Variation of initial permeability with log frequency for NixZn1-xFe2O4 with x = 0.0, 0.2 and 0.4

μ ι

log frequency(kHz)

ZnFe2O4 Ni0.2Zn0.8Fe2O4 Ni0.4Zn0.6Fe2O4


The variation of initial permeability, μi, with frequency can be understood on the basis of Globus model [18-20]. It was shown that for relaxation character

(μi –1)2 fr = constant………………………..(2) where μi is the static initial permeability and fr is the relaxation frequency. These workers also showed that the transformation of magnetic spectra from relaxation character to resonance character changes Eq.2 to

(μi -1)1/2 fr = constant…………………….(3) It follows from the above mentioned

equation that the dispersion frequency is expected to be lower for specimen of higher permeability. This is due to fact that for materials of lower permeability, the demagnetizing fields, which appear during wall movement, result in increasing the restoring force, thereby increasing the relaxation frequency. Also, the high anisotropy of low permeability ferrites is known to increase the intrinsic restoring force of the domain walls. An increase in the restoring force due to the wall discontinuity changes the relaxation character to the resonance character. For the ferrite with x = 0.0, the resonance peak is found to be at lower frequency as its permeability is high and for the sample with x = 0.4, the peak is not observed, which may be due to the reason that the resonance frequency may lie beyond the range of our work.

Variation of permeability loss factor, tanδµ, with frequency is shown in Fig. 4 for NixZn1-

xFe2O4 ferrite sample with x = 0.0, 0.2, 0.4. The loss is decreasing initially with increase in frequency of applied ac field and then it starts increasing at high frequency. This is the normal behaviour and also observed by other workers [12-13].

100 1000 10000

0.0

0.2

0.4

0.6

0.8

1.0

Fig.4 Variation of permeability loss with log frequency for NixZn1-xFe2O4 with x = 0.0, 0.2 and 0.4

tanδ

μ

log frequency (kHz)


The variation of dielectric constant, ε, for Nix Zn1-xFe2O4 ferrites was studied as a function of frequency. Frequency was varied from 75 kHz to 30 MHz. The results for these measurements are show in Fig 5.

100 1000 100000

1000

2000

3000

4000

5000

6000

7000

8000

9000

Fig.5 Variation of dielectric constant with log frequency for NixZn1-xFe2O4 with x = 0.0, 0.2 and 0.4

diel

ectri

c co

nsta

nt

log frequency(kHz)

ZnFe2o4 Ni

0.2Zn0

0.8Fe

2O

4 Ni0.4Zn00.6Fe2O4

It is observed that for each sample, the dielectric constant decreases slowly upto 100 kHz and is nearly constant above 100 kHz. After 10MHz, dielectric constant shows an increase with the increase in frequency. The variation of dielectric constant reveals that the dispersion due to Maxwell-Wagner [21-22] interfacial polarization is in agreement with Koops phenomenological theory [23]. Rezlescue et al [24] have correlated dielectric polarization and Verwey type of conduction mechanism between Fe2+ ↔ Fe3+, Ni2+↔ Ni3+. This correlation gives local displacement of electrons in the direction of applied field, which induces polarization in ferrites.

Increase in dielectric constant with frequency can be explained as, Iwauchi25 has pointed out that there is a strong correlation between conduction mechanism and the dielectric behavior of ferrites. The conduction in ferrites is considered as due to hopping of electrons between Fe2+ ↔ Fe3+. As such, when hopping frequency is equal to that of frequency of externally applied electrical field, a maximum of dielectric constant, ε, may be observed.

The initial slow decrease in the dielectric constant as predicted by Koops model has been observed in the present study. Workers [23-24] have studied the variation of the dielectric constant with frequency at low frequencies. The physical reason for dispersion in the dielectric constant can be understood on the basis of hopping of electrons between Fe2+↔ Fe3+ and

Preparation and Characterization of Ni-Zn Spinel Ferrites 9

Ni2+↔ Ni3+ pairs of ions. The applied electric field displaces the electrons slightly from their equilibrium positions, thus producing polarization.

Variation of dielectric loss factor, tanδε, with frequency are shown in Fig. 6 for Nix Zn1-xFe2O4 (x = 0.0, 0.2, 0.4) ferrite samples.

100 1000 10000

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Fig.6 Variation of dielecric loss with log frequency for NixZn1-xFe2O4 with x = 0.0, 0.2 and 0.4

tan

δ ε

log frequency(kHz)


Frequency is varied from 75 kHz to 30 MHz. It is observed that for each sample of Ni-Zn ferrite, the loss factor decreases slowly upto 100 KHZ and is nearly constant above 100 kHz. After 10MHz, slight increase in tanδε with an increase in frequency is observed, which can be explained on the basis of Koops phenomenological model [23]. A quantitative description of this increase in tanδε can be given by Iwauchi [25]. The conduction in ferrites is considered as due to hopping of electrons between Fe2+ and Fe3+. As such, when hopping frequency is equal to that of frequency of externally applied electric field, a maximum of loss tangent may be observed.

The value of d.c. resistivity is found to be 18.67 x 106 ohm-cm, 4.96 x 106 ohm-cm and 12.03 x 106 ohm-cm for the concentration x = 0.0, 0.2 and 0.4 respectively. It shows that Ni-Zn ferrites show high resistivity, which makes these ferrites useful for high frequency applications where eddy current losses become appreciable. It is found that resistivity of zinc ferrite is higher than that of nickel substituted zinc ferrites. Resistivity in Ni-Zn ferrite decreases up to x = 0.2 and then increases for x = 0.4. Since resistivity decreases with addition of Ni2+, conduction is prevalent in Ni-Zn ferrites. The conduction is due to the hole transfer from Ni3+ to Ni2+ ions [26].

Ni2+ + Fe3+ ↔ Ni3+ + Fe2+………(4) Hopping of electrons from Fe2+ to Fe3+ ions

is also responsible for conduction in Ni-Zn ferrites. The mechanism which appears to be responsible for conduction in the present compositions is Verwey hopping mechanism. The conduction in the Ni-Zn ferrites may be attributed to the presence of very small quantities of Fe2+ and Ni3+ ions along with a large number of Fe3+ and Ni2+ ions. Since all these ions have preference for B sites, the conduction can take place through Verwey hopping mechanism. The Fe2+ is formed due to partial reduction of Fe3+ ions during sintering. A complete reoxidation of these Fe2+ ions does not take place, particularly in dense samples when a reasonably fast cooling rate is maintained. The Ni3+ ions on the other hand are formed during cooling of ferrites. Thus electron hopping can take place mainly between Fe2+ and Fe3+ ions in the oxygen deficient regions and between Ni2+ and Ni3+ ions in the oxygen rich regions [27]. Besides the electron hopping between the similar ions, the presence of nickel at the octahedral sites may also cause conduction given by Eq. 5. The increase in d c resistivity at high concentration can be explained on the basis of inverse proportionality between dielectric constant and resistivity [28].

We are successful in synthesizing high permeability ferrites in our laboratory by co-precipitation method, which could be used from low frequency to high frequency applications, data storage devices, inductors, cores of transformers etc. The synthesized ferrites have almost uniform sized crystallites, which are confirmed by scanning electron micrograph. References [1] G. Kumar, M. Kanthwal, B. S. Chauhan and

M. Singh, Ind J Pure & Appl. Phys, 44 (2006) 930.

[2] Lal M, Sharma K D & Singh M, J Pure & Appl Phys, 43 (2005) 291.

[3] Shukla S J, Jadhav K M & Bichile G K, J Pure & Appl Phys, 39 (2001) 226.

[4] Mathur P, Thakur A & Singh M, J Mater Sci (2007) in press.

[5] Singh A K, Goel T C, Mandiratta R G, Thakur O P & Prakash C, J Appl Phys, 92 (2002)3874.

[6] Patil KC, Manoharan S S & Gajapathy, “Hand book of ceramic and composites” Marcel Decker, New York 1 (1990) 469.

[7] Daniel J M & Rosebwaig A, Can J Phys 4, 48 (1970) 381.


[8] Srivastava C M, Bull Mater Sci16, (1984) 7.

[9] Naik A B & Powar J I, Ind J Pure & Appl. Phys, 23 (1985) 436.

[10] Van uitert LG, “Proceeding of the I R E”, 44 (1956) 1294.

[11] Koops C G, Phys Rev 83 (1956) 647. [12] Thakur A, Mathur P & Singh M, Z

Phys Chem, 221(2007) 1-9. [13] Thakur A, Mathur P and Singh M, J

Phys Chem Solids, 68 (2007) 378. [14] Cullity B D, “Elements of X-ray

diffraction (Addision-Wesley, Reeding, M A, 1978).

[15] Bermejo E, Mercier T I C & Quartan M, J Amm Ceram Soci, 78 (2) (1995) 365.

[16] Jhonson D W, Amm Ceram Bull, 53 (2) (1974) 163.

[17] Globus A, Acad Sci, 255 (1962) 1709-1711.

[18] Geieraltowski J and Globus A, IEEE Trans Magn, 13 (1977). 1359.

[19] Globus A, Proc J Phys Colloq, 38 C-1 (1977).

[20] Globus A, Duplex P, Guyot M, IEEE Trans Magn, 7 (1971) 617.

[21] Maxwell J C, “Electricity and Magnetism” Oxford University Press London Vol. 3 (1958) p. 328.

[22] Wagner K W, Anand Der Phys, 210 (1913) 817.

[23] Koops C G, Phys Rev, 83 (1951) 121. [24] Rezlescue N & Rczlescu E, Phys Status

solidi (a), 23 (1974) 575. [25] Iwauchi K, Jap J Appl Phys, 10 (1971)

1520. [26] Van uitert LG, J Chem Phys, 24 (1956)

306. [27] Dietzmann G, Krotzsch M & Wolf S,

Phys Stat Sol, 2 (1962) 1762. [28] Thakur A & Singh M, Ceram Inter, 29

(2003) 506.

Preparation and Characterization of SrFeO3-δ by Sol-Gel Method

Shivendra Kumar Jaiswal and Jitendra Kumar Materials Science Programme, Indian Institute of Technology- Kanpur, 208016, India

E-mail: [email protected]

Abstract

Strontium ferrite (SrFeO3-δ) powder with single perovskite-type cubic phase and lattice parameter 3.858±0.002 Å has been synthesized by decomposition of oxalate (obtained by a sol-gel process using strontium nitrate, iron nitrate and oxalic acid as precursors and ethanol and water as solvents) at 800oC. The compound is shown to exhibit hysteresis loop at room temperature with high coercivity (~ 3748 Oe) and remanance magnetization (0.6016 emu/g). The ferromagnetic to paramagnetic transition is not abrupt and corresponds to superparamagnetic behavior with Curie temperature of (TC) 589oC. The oxygen desorption is accompanied by reduction of Fe3+ to Fe2+ ions and occurs at 64 and 650 oC from the surface of crystallites and the bulk, respectively. 1 Introduction

Perovskite-type ferrites (AFeO3, A – rare earth or alkaline-earth metals) have attracted immense interest in the recent past because of their potential application as gas sensor, cathode material for solid oxide fuel cell, oxygen permeable membrane, etc.[1-9]. Among them, strontium ferrite (SrFeO3-δ) based system is identified as most promising material. Of several synthesis techniques, viz., EDTA-citric acid complexing [10], solid state sintering [11], co-precipitation method [12] etc., sol-gel method is quite popular now because it offers several advantages, e.g., homogeneous mixing at atomic/molecular level of constituents, better stoichiometry control, high purity, low temperature processing, cost effectiveness, etc. The objective of this paper is to discuss synthesis of a single phase of strontium ferrite through oxalate based sol-gel route and study its thermal stability, morphology, oxygen desorption and magnetic properties. 2 Experimental

Strontium ferrite was synthesized by sol gel technique using appropriate amount of Sr(NO3)2, Fe(NO3)3 and oxalic acid as precursors and ethanol as a solvent. Due to low solubility of Sr(NO3)2 in ethanol, distilled water was added drop wise to achieve complete dissolution. The nitrate salt solutions were mixed together and then oxalic acid solution added to form a gel. Subsequently, the gel was subjected to digestion for 4 h and drying at 150 oC for 24 h. To ascertain the steps involved in the formation of strontium ferrite, thermogravimetric analysis of the dried sol gel

product was carried out by raising its temperature at a rate of 3oC per minute up to 850 oC. Accordingly, the dried sol gel product was calcined at 800 oC for 10 h and then ground and sieved through a 240 mesh to obtain SrFeO3-δ fine powder for further study. While a Rich Seifert X-ray diffractometer (model ISO Debye flux 2002) was used to identify the phase(s), a vibrating sample magnetometer (Princeton VSM model-150) was employed to ascertain magnetic parameters, like magnetization (M), coercive field (Hc) and ferromagnetic transition temperature (Tc). Also, oxygen temperature programmed desorption (O2-TPD) set up ( Micromeritics model 2705 TPD) was utilized for studying desorption behavior of SrFeO3-δ by raising its temperature at a rate of 5oC/minute from ~25oC to 950oC with helium (flow rate ~ 20 ml/minute) as a carrier gas, an oxygen sensor and a built in automatic data acquisition system. 3 Results and discussion 3.1 Thermogravimetric (TG) analysis

Fig.1 shows the (%) weight vs temperature plot of the dried sol gel product. It displays a number of steps for weight change, which can be visualized better in –(dW/dT) vs T plot (see, e.g., figure 1). Table 1 lists various steps, evolution of species with temperature ranges and respective percentage weight loss. Above 730 oC, the product became was quite stable and amounted to about 45% of the initial value. Therefore, the dried sol gel product was calcined at a higher temperature of 800oC for 10h to obtain a stable compound for further study.

12 Recent Advances in Innovative Materials Table 1. Percentage weight loss, temperature range, temperature for maximum rate loss and respective attributes

Fig.1. TGA plot and corresponding –(dW/dT) versus T curve of the sol gel product. 3.2 Phase evaluation

X –ray diffraction pattern of the stable compound obtained by calcining the dried sol gel product at 800oC for 10 h is shown in the fig. 2. It corresponds to a single cubic phase having lattice parameter a = 3.858±0.002 Å and matches well with perovskite- type structure of SrFeO3-δ; known ‘a’ value being ~ 3.869 Å for δ =0 [16, 17]. In this, the strontium and oxygen ions assume corner and face centred positions displaying a cubic-close packed arrangement (with some anion vacancies, of course), whereas, iron ions occupy the octahedral site at the cube centre. This result in conjunction with the TG analysis indicates formation of stable SrFeO3-δ compound at as low as ~ 730oC. In contrast, synthesis of strontium ferrite by EDTA-citric acid complexing method [10], solid state sintering [11] and co-precipitation [12] requires

Fig. 2. XRD pattern showing single perovskite-type cubic phase of SrFeO3-δ powder. relatively higher temperatures i.e. ~900, ~1200, ~1100 oC, respectively. 3.3 Oxygen desorption

The oxygen- temperature programmed desorption profile of SrFeO3-δ is shown in fig. 3. It indicates two major peaks around 64 and 650 oC which may be attributed to desorption of oxygen from the surface and the bulk, respectively. The corresponding areas under the peak in arbitrary units are 2.06 x107 and 8.48 x 107, respectively. The later being four times more and so indicates substantial desorption of oxygen around 650 oC. The XRD pattern of the sample after O2 – TPD (OTPD) experiments continue to show the perovskite structure of SrFeO3-δ without any decomposition (fig. 4). Obviously, the partial loss of oxygen due to desorption does not effect the structural framework of SrFeO3-δ up to 900 oC at least.

%Weight loss Temperature range (oC)

Temperature for maximum rate loss (oC)

Attributes

~10 ~15 ~ 5 ~15 ~11

120-210 210-290 290-350 350-425 610-710

170 246 323 409 685

Water [13] Crystallizing water[13] Residual nitrates and organic matter [14] CO2 [15] CO [15]

100 200 300 400 500 600 700 80040

50

60

70

80

90

100

-0.4

-0.3

-0.2

-0.1

0.0

0.1

% W

eigh

t(w)

Temperature (0C)

- (dw

/dT)

20 40 60 80 100

321

22231

0220

211

200

111

110

Inte

nsity

(Arb

itrar

y)

2θ (in degrees)

Preparation and Characterization of SrFeO3-δ by Sol-Gel Method 13

0 2 0 0 4 0 0 6 0 0 8 0 0 1 0 0 04 .0 x 1 0 5

4 .2 x 1 0 5

4 .4 x 1 0 5

4 .6 x 1 0 5

4 .8 x 1 0 5

Rec

orde

r Res

pons

e

T e m p e r a tu r e ( o C )

20 40 60 80 100

31022

0

111

200 21

1

222 32

1

110

Inte

nsity

(a.u

.)

2θ (in degree)

Fig. 3. O2- temperature programmed desorption signal vs temperature plot of SrFeO3-δ Figure 5 shows the scanning electron micrographs of SrFeO3-δ before and after O2-TPD process. Clearly, the morphology appears to be similar but particles split into smaller sizes following oxygen desorption.

The oxygen ions diffuse from bulk to surface (via anion vacancies present) and recombine for desorption to occur as molecule by releasing their electrons to the lattice. These electrons, in turn, create defect centres and /or change the oxidation state of cations (e.g., Fe3+ to Fe2+). In the later case, concentration of Fe2+ ions should increase with the progress of desorption. The preliminary Mossbauer studies of SrFeO3-δ show emergence of Fe2+ ions after the O2-TPD process. Thus, the substantial oxygen desorption observed at ~ 650oC in SrFeO3-δ system appears to be accompanied by Fe3+ to Fe2+ conversion. These findings is consistent with the reported reduction of cations from high valance to lower valance state in the Fe- containing perovskites in the temperature range 300-600oC [10] and in La1-xSrxCoO3(0≤x≤1) at 500 and 820oC [18] with desorption of oxygen.

Fig. 4. XRD pattern of powder after oxygen desorption indicating retention of perovskite-type cubic phase of SrFeO3- δ.

Fig.5. Scanning electron micrograph of (a) SrFeO3-

δ and (b) SrFeO3-δ after oxygen desorption. 3.4 Magnetic measurements

Fig. 6 shows the magnetization verses field plot for SrFeO3-δ compound at room temperature. Clearly, the hysteresis loop is not saturating and shows high coercive field and remanance. It corresponds to magnetization of 2.81 emu/g (or 0.08 μB per iron atom) at 1.7 Tesla, the remanent magnetization ~ 0.6016 emu/g, coercivity ~ 3748 Oe, and the area of the loop 8996 Oe-emu/g. The strontium hexaferrite prepared by combustion route is reported to have magnetization of 62 emu/g at 1.5 Tesla and coercivity value ~ 1950 Oe [14]. However, high value of coercivity (~ 2000 Oe) has been observed in strontium ferrite, prepared by co- precipitation and sintering at 1200oC, and attributed to the submicron particle size [19]. These results suggest that the magnetic parameters of strontium ferrite depend largely on the synthesis route itself. The products obtained in the present investigation indeed correspond to large magnetization and coercivity values possibly due to small particle size, morphology, and high magnetocrystalline anisotropy [14, 20]. The absence of saturation is perhaps caused by distribution of particles and/or the field ~ 9-10 kOe

a)

b)


is insufficient to align all the spins. Fig. 7 shows the variation of magnetization with temperature (T) of SrFeO3-δ at a fixed magnetic field of 500 Oe. Accordingly, the magnetization remains constant up to ~ 200oC and then decreases at various rates depending upon the temperature interval and eventually becomes zero at 589 oC. This behavior also indicates the presence of particles of various sizes with different Curie temperatures (TC) and /or superparamagnetic characteristics

Fig. 6. Room temperature hysteresis loop of strontium ferrite powder depicting non-saturation behavior and high coercivity.

Fig. 7. Variation of magnetization with temperature of SrFeO3-δ compound under an external magnetic field of 500 Oe; the inflection points are indicated by arrows 4. Conclusions

The perovskite- type SrFeO3-δ compound with single phase (cubic, a ~ 3.858±0.002 Å.) can be synthesized by decomposition of oxalate obtained by sol-gel process using strontium nitrate, ferric

nitrate and oxalic acid as precursors, and ethanol and water as solvents at 800oC. It desorbs substantial amount of oxygen at 64 and 650 oC from the surface and bulk, respectively. The process is accompanied by reduction of Fe3+ to Fe2+. Also, SrFeO3-δ exhibits ferromagnetic ordering with Curie temperature (TC) of ~ 589oC and high coercivity (~3748 Oe) and remanent magnetization (0.6016 emu/g). The small particle size, morphology of crystallites and high magnetocrystalline anisotropy seem to be responsible for the magnetic behavior. References [1] B.C.H. Steele, Mater. Sci. Eng. B13 (1992)79. [2] Y. S. Lin, Y. Zeng, J. Catal. 164 (1996) 220. [3] K.R. Kendall, C. Navas, J.K. Thomas, H.-C.

zur Loye, Solid State Ionics 82 (1995)215. [4] E. Maguire, B. Gharbaga, F.M.B. Marques,

J.A. Labrincha, Solid State Ionics127 (2000) 329-335.

[5] A. Cherrak, R. Habaut, Y. Barbaux, G. Mairesse, Catal. Lett. 15 (1992) 377.

[6] J. Barrault, C. Grosser, M. Dion, et.al., Catal. Lett. 16 (1992) 203.

[7] R. Di Cosimo, J.D. Burrington, R.K. Grasselli, J. Catal. 102 (1986) 234.

[8] A. Lofberg, S.Boujmini, E.Capoen, M.C. Steil, C. Pirovano, R.N. Vannier, G. Mairesse, E.Bordes-Richard, Catal. Today 91-92 (2004)79.

[9] J.C. Boivin, C. Pirovano, G.Nowogrocki, G. Mairesse, Ph.. Labrune, G. Lagrange, Solid State Ionics 113-115 (1998) 630.

[10] Zongping Shao, Weishen Yang, You Cong, J. of mem. sci. 172 (2000) 177.

[11] Hui. Lu, Jianhua Tong, Z. Deng, Y. Cong, W. Yang, Matr. Res. Bull. 41 (2006) 683.

[12] C.O.Augustin, L.John Berchmans, R.K. Selvan, Mater. Lett. 58 (2004).

[13] Dong Hwang Chen, Yuh-Yuh Chen, Mat. Res. Bull. 37 (2002) 801.

[14] Yen-Pei Fu, Cheng- Hsiung Lin, Ko-Ying Pan, J. of alloys and compound 349 (2003) 228.

[15] S.K. Singh, J. Kumar, J. of Physics and Chemistry of Solids 67 (2006) 1687-1691

[16] Harry L. Yakel, Jr. Acta Cryst. (1955) 8, 394. [17] JCPDF files (data number 34-0638) [18] N. Yamazoe, Y. Teraoka, T. Seiyama, Chem.

Lett. (1981) 1767. [19] S. Kulkarni. J Shrotri, S. K. Date, J. Mat.

Sci.,24 (1989)3739-3744. [20] A. Ataie, S H Manesh, J. Eur. Ceram. Soc. 21

(2001) 1951.

-20000 -15000 -10000 -5000 0 5000 10000 15000 20000-3.5

-3.0

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Mag

netiz

atio

n(em

u/g)

Applied field(Oe)

0 100 200 300 400 500 6000.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Mag

netiz

atio

n (e

mu/

g)

Temperature(oC)

Study of Ni Substituted Mn-Zn Nanoferrite by Citrate Precursor Method

Preeti Mathur, Ajay Sharma, Naveen Sharma, Atul Thakur and M. Singh

Department of Physics, Himachal Pradesh University, Summer Hill, Shimla, 171005, India, E-mail: [email protected]

Abstract

In the present work, Mn0.4Ni0.2Zn0.4Fe2O4 nano-ferrite has been prepared by citrate-precursor method.

The effect of oxidizing agent on dielectric behaviour is investigated. The dc resistivity of this ferrite has been improved by using citrate precursor method as compared to conventional ceramic method. Further improvement in the d.c. resistivity has been observed by adding H2O2 (hydrogen peroxide), which acts as a strong oxidizing agent. We have shown by means of X-ray diffraction that the resulting ferrite is made up of nanocrystallites and the average size of these nanocrystallites, calculated by Scherrer’s formula, also depends on the polarizer. The average size of these nanocrystallites is found to be approximately 70 nm. The size measurement is also confirmed from scanning electron microscopy. Both the results are found to be in good agreement. The decrease in the dielectric constant and dielectric loss factor by the addition of oxidizing agent is attributed to the inverse proportionality of resistivity and dielectric constant. Possible mechanisms contributing to these processes have been discussed 1. Introduction

Processing of ferrites has gained tremendous importance in recent times to meet the high performance demands on ferrites in keeping with fast emerging technologies. The quality of ferrite powder has strong influence on performance of final device. Soft ferrites are among the most widely used magnetic materials having low cost, high performance in high frequency applications [1-3]. These ferrites are stable, easily manufactured and possess environmental stability [4]. Soft ferrite properties arise from interaction between metallic ions occupying particular position relative to the oxygen ions in its spinel structure. The spinel Mn-ni-Zn ferrites have been used in various components for application in high frequency range. The usefulness of the ferrites is strongly influenced by the physical and chemical properties of the materials and depends upon many factors including the processing method used to synthesize them. In order to get good quality ferrites with reproducible stoichiometric composition and desired microstructure, citrate precursor method [5] has been used to synthesize ferrites in the present work. Citrate precursor method offers a synthesis route for the production of ferrites, which are uniform and non-aggregated [6]. We have found that the subtle changes in the synthesis chemistry profoundly affect the structure and chemistry of resulting ferrite particles..

2. Experimental Details

Mn-Ni-Zn ferrite of composition Mn0.4Ni0.2Zn0.4Fe2O4 was prepared by citrate precursor method. The materials used were Manganese chloride (98% Merck, India), Zinc chloride (96% Merck, India), iron (III) chloride (98% Merck, India) and sodium hydroxide (96% Merck, India). The experimental details have been described by authors elsewhere [5]. Here two samples were prepared. In one sample, 20 ml of H2O2 (hydrogen peroxide) is added and the other sample is prepared without H2O2. The powder so obtained was mixed with 2% P.V.A. binder and pressed into pellets of 1.50cm diameter and 0.20 mm thickness under pressure of 5 tons (1 ton = 1.016 x 103 kg). These samples were sintered in air at 900, 1000, 1100, 1200 and 1300°C at a heating rate of 250°C /h and were subsequently cooled. The pellets were coated with silver paste to provide electrical contacts. X-ray diffraction (XRD) measurements were taken on a Rigaku Geiger Flex 3 kW diffractometer using CuKα source. Dielectric constant and loss factor were measured upto 30 MHz by using Agilent Technologies 4285A Precision LCR Meter (upto30MHz). The dc resistivity was measured by using Two-Probe method.

3. Results and Discussion The X-ray diffraction patterns of all the

samples show single phase spinel structure and


only reacted constituents were present in the samples. XRD pattern of the ferrite samples sintered at 1000°C with H2O2 and without H2O2 is shown in Fig.1. The observed diffraction lines were found to correspond to those of standard pattern of Manganese ferrite with no extra lines, indicating thereby that the samples have single

30 40 50 60

(b)

(a)

440

511/333

422400

311

220

Cou

nts

2θ

Fig. 1. XRD of Mn0.4Ni0.2Zn0.4Fe2O4 ferrite (a) with H2O2 (b) without H2O2

phase spinel structure and no unreacted constituents were present in these samples. Lattice constant ‘a’ for the samples with H2O2 and without H2O2 was calculated by using Bragg’s law and was found to be 8.318°A and 8.329°A respectively. The particle size of the samples has been estimated from the broadening of the X-ray diffraction peaks using the Scherrer equation [7] for Lorentzian peak d = 0.9 λ (w - w1) cosθ where, d is the grain diameter, w and w1 are the half intensity width of the relevant diffraction peak and the instrumental broadening respectively, λ is the X-ray wavelength and θ is the angle of diffraction. The average particle size was found to be ~65 nm with H2O2 and ~75 nm without H2O2. The small particle size with H2O2 can be explained as it oxidizes Fe+2 to Fe+3, which have smaller size.

Fig. 2. (a) SEM of fractured surface of Mn0.4Ni0.2Zn0.4Fe2O4 ferrite with H2O2

Fig. 2. (b) SEM of fractured surface of Mn0.4Ni0.2Zn0.4Fe2O4 ferrite without H2O2

Fig.3 shows the variation of dc resistivity with sintering temperature. It is observed that the dc resistivity obtained in case of citrate precursor method (~ 106ohm.cm) is greater by at least two orders of magnitude as compared to the conventionally (~ 104 ohm.cm) normal prepared Mn0.4Ni0.2Zn0.4Fe2O4 ferrites [5].

Fig. 3. Variation of DC resistivity with sintering tempering temperature

The higher value of dc resistivity is due to

homogeneity and smaller grain size (~70 nm to 80 nm). The smaller grains contain greater number of grain boundaries. The grain boundary

Study of Ni Substituted Mn-Zn Nanoferrite by Citrate Precursor Method 17

is a region of mismatch between the energy states of the adjacent grains and hence acts as a barrier to the flow of the electrons and therefore decreases the conductivity. In conventional ceramic method, the sintering temperature being higher, the grains are relatively bigger. Further, the extensive ball-milling required in the conventional method may result in loss of few grains of the material and give the introduction of impurities. This would result in non-stoichiometric and non-homogeneous sample. In citrate precursor method, these possibilities being negligible, higher resistivity of the prepared samples is therefore expected.

It is found that the dc resistivity increases upto 1000°C; further increase in sintering temperature however, reduces the resistivity. The decrease in resistivity at higher sintering temperature is expected because of the increase in grain size. Also, at higher sintering temperature, the probability of production of ions in more than one valence state is high, means hopping is large and resistivity is lower. The dc resistivity is further improved by the addition of H2O2 (polarizer) as shown in Fig.3. H2O2 converts Fe+2 ions into Fe+3 ions; thereby blocking the conduction between Fe+2 and Fe+3 ions, which is the reason for increase of dc resistivity [8]. In the sample, Fe+2 ions are easily polarizable, reduction in the number of Fe+2 ions results in the increase of their dc resistivity. By addition of H2O2, the charge/size ratio increases and hydration of the Fe+3 ions by water molecules also increases. This results in an increase in the size of Fe+3 ions and decrease of ionic mobility of Fe+3 ions. Everett et al.[9] have shown the oxidizing nature of H2O2 in ferrites.

The variation of dielectric constant of Mn0.4Ni0.2Zn0.4Fe2O4 ferrite prepared with and without H2O2 is shown in Fig.4.

Fig. 4. Variation of dielectric constant with log frequency at 10000 C

The values of dielectric constant are quite low and are about 103 times lower than those obtained for the samples prepared by the conventional ceramic method [10-11]. The low value of dielectric constant is attributed to homogeneity, better symmetry, uniform and small grains. By an application of electric field, the electrons reach the grain boundary through hopping. Since the resistance of the grain boundary is high, electrons pile up at grain boundaries and produce polarization. The low dielectric constant indicates lower interfacial polarization because of high resistivity of the sample, which is due to the absence of appreciable number of Fe+2 ions. As expected, the dielectric constant of Mn-Ni-Zn ferrite with H2O2 is less as compared to without H2O2, because of inverse proportionality between resistivity and dielectric constant.

The variation of dielectric loss factor of Mn0.4Ni0.2Zn0.4Fe2O4 ferrite prepared with and without H2O2 is shown in fig.5.

Fig. 5. Variation of dielectric loss with log frequency at 10000 C

The value of loss tangent obtained in the present work is in the range of 10-2-10-3 at frequencies from a few kHz to about MHz. These values of loss tangent are about two orders of magnitude lower than those obtained in the ferrites prepared by conventional ceramic method [10-12]. The dielectric loss arises, if the polarization lags behind the applied altering field and is caused by the presence of impurities and structural inhomogeneities. The low dielectric loss obtained in the present work is therefore attributed to more structurally perfect and homogeneous ferrites processed by citrate precursor method. The decrease in dielectric loss tangent with increase in frequency is in accordance with the Koop’s Phenomenological


model [11]. The dielectric loss factor of Mn-Zn ferrite with H2O2 is also less as compared to without H2O2.

4. Conclusions

The results indicate definite improvement in the electric properties like dc resistivity and low dielectric loss factor; thus the approach of polarizer H2O2. The particle size of the ferrite is also reduced by addition of H2O2. The optimum sintering schedule is proved to be useful in the development of core materials for application in high frequency. The ferrite materials offer inexpensive alternative in many new applications of modern technologies.

References [1] M. Singh, S.P. Sud, Mater. Sci. Eng. B 83

(2000) p.180. [2] M. Singh and S. P. Sud, Mod. Phys. Lett. 14

(2000) pp.531-537.

[3] M. I. Rosales, E. Amano, M. P. Cuautle, R. Valenzuela, Mater Sci Eng B, 49 (1997) p.221

[4] S.F.Wang, Y.R. Wang, C.K.Y. Thomas, P.J.Wang & C.A.Lu, J. Magn. Magn. Mater. 217 (2000) p.35.

[5] A. Thakur, M. Singh, Ceramic international 29 (2003) pp.505-511.

[6] M.P. Pileni, J. Phys. Chem. 97 (1993) p.6961.

[7] B. D. Cullity, Elements of X-ray Diffraction (Addison Wesley Reading, MA. 1978).

[8] E.J.W.Verwey, J.H.D. Boer, Rec. Trav. Chem. Phys.55 (1936) p.531.

[9] E.C.Everett, C.J.O.Vincent, G.Harris, J. Appl. Phys., 85 (1999) p.5175.

[10] L.Radhapiyari, S.Phanjoubam, H.N.K. Sharma & C.Prakash, Mater. Lett. 44 (2000) p.65.

[11] C.G.Koops, Phys.Rev. 83 (1951) p.121. [12]M.A.Haiti, J.Mag.Mag.Mater.192 (1999)

p.305.

Characterization of Soft Ni-Zn Spinel Ferrites

N. Sharma, Tanuj Sharma, P. Mathur, A. Thakur and M. Singh Department of Physics, Himachal Pradesh. University, Summer Hill, Shimla, 171005, India.


Abstract

Nanoparticles of spinel ferrite NixZn1-xFe2O4 were prepared by co-precipitation method. The ferrite samples were pre-sintered at 2000 C at the rate of 2000 C/hr for 15 hours. Finally, one set of samples was sintered at 4000 C and the other set at 5000C at the rate of 2000 C/hr for another 15 hours. The spinel structure was confirmed by X-ray diffraction (XRD) pattern. The particle size was calculated by using Debye-Scherrer’s equation for Lorentzian peak. The calculated average particle size lies between 19-53 nm. The particle size was also investigated by using scanning electron microscopy (SEM). Both the results were found to be in good agreement. The resistivity data was taken by two probe method and resistivity, activation energy and porosity were calculated. There was an increase in the resistivity and activation energy when the ferrite samples were sintered at a higher temperature. Possible models, theories and mechanisms contributing to these processes have been discussed. 1. Introduction

The soft ferrites have gained technological importance by virtue of their high resistivity and negligible eddy current losses [1-6]. These ferrites have a wide range of applications from low frequency to microwave frequencies on account of their low cost and high efficiency [7-10]. The importance of soft ferrites, particularly Mn-Zn ferrites, in low frequency inductors, antenna rods, wide-band and pulse transformers are well known [11-14] and their uses in these fields have been continuously increasing for several decades. For high frequency electronic devices such as electromagnetic wave absorbers or inductor devices, extrinsic parameters play an important role. Therefore, the trend of research is shifted towards the study of the properties of the ferrites depending upon the extrinsic parameters such as porosity, grain size, sintering temperature etc. Rikukawa [15] established the semi-empirical relationship between apparent permeability, porosity and average grain size. Numerous chemical routes, such as reverse micelle synthesis [16], citrate-precursor [17], combustion [18], high energy ball milling [19] and sol-gel [20] have been used to synthesize the nanoparticles of ferrites. Recently, co-precipitation [21] has been extensively employed to obtain the nanoparticles in large amounts. The co-precipitation method ensures an ease of preparation and control of composition, particle size, purity, homogeneity and uniform grain growth. The inter-particle spacing and surface to volume ratio in particles play a predominant role

in influencing the material properties. In nanoparticles, a large fraction of atoms resides at the surface as compared to the bulk particles which causes significant changes in the structural, electrical and magnetic properties. We have found out an increase in resistivity and activation energy when the ferrites were sintered at higher temperature i.e. 5000C. The nanocrystallites were formed at as low as 4000C temperature if the duration of sintering is increased to 15 hours. 2. Experimental details

Ni-Zn ferrite Ni0.2Zn0.8Fe2O4 of composition was prepared by the co-precipitation method. The materials used were nickel chloride (98% Merck, India), zinc chloride (96% Merck, India), iron (III) chloride (98% Merck, India) and sodium hydroxide (96% Merck, India). One molar solution of these materials was made with distilled water. 70 mL of sodium hydroxide was taken from one molar solution in approximately 1760 mL of distilled water to have the concentration of 0.37 mol/L and heated to boiling. Nickel chloride, zinc chloride and iron (III) chloride were taken from their molar solution in accurate stoichiometric proportions. These solutions were poured as quickly as possible into the boiling solution of NaOH under vigorous stirring produced by the glass mechanical stirrer (~ 500 r.p.m.). Mixing is very important otherwise segregation of phases can take place. After co-precipitation, pH is set 12.5-13. Reaction vessel is covered with plastic cover


to diminish evaporation of the solution. Reaction is continued for 30-40 minutes at a temperature of 90-100°C under vigorous stirring. Reaction vessel is cooled to an ambient temperature and particles precipitate. Total volume is reduced to ~500 mL by aspiration of supernatant. Suspension is centrifuged in 4 beakers at 7000 r.p.m. for 10 minutes. Precipitate is washed with distilled water of ~900 mL and centrifuged once more in 4 beakers at 7000 r.p.m. for 10 minutes. The residue is heated at 400C till it gets dried and thereafter, calcinated in a box type furnace at 2000 C for 15 hours at the rate of 200° C /h to obtain a ferrite powder. This powder was mixed with 2% P.V.A. binder and pressed into pellets of 1.50 cm diameter and 0.20 cm thickness under a pressure of 5 tons (1 ton = 1.016 x 103 kg). These samples were sintered in air at 500°C at a heating rate of 200° C/h and were subsequently cooled to the room temperature by switching off the power supply. The pellets were coated with the silver paste to provide electrical contacts. The X-ray diffraction (XRD) measurements were taken on a Rigaku Geiger Flex 3 kW diffractometer using CuKα source. Scanning electron micrograph (SEM) was recorded using Cambridge Stereo Scan 360 scanning electron micrograph. The resistivity measurements were made by using Two-Probe method. Curie temperature was measured by using the gravity separation method. 3. Results and discussion

The XRD patterns of the series Ni0.2Zn0.8Fe2O4 is studied. The observed diffraction lines are found to correspond to those of the standard pattern of the manganese ferrite with no extra lines, indicating thereby that the samples have a single-phase spinel structure and no unreacted constituents are present in these samples. The typical diffraction patterns, when sintered at 4000C and 5000C are shown in Fig.1(a) and Fig.1(b) respectively .

We have used the 311 reflection line in XRD patterns for obtaining the average particle size with the help of the Debye – Scherrer’s equation [22]:

t =BB θ

λcos

9.0; …………………………..(1)

where, B= ( ) 2/122SM BB − ,

t is the thickness (diameter) of the particle, λ is the X-ray wavelength (1.54 A0), BM and BS are the measured peak broadening and the instrumental broadening in radian respectively and θB is the Bragg angle of the reflection.

Fig. 2. (a) Scanning Electron Micrograph of fractured surface of Ni0.4Zn0.6Fe2O4 ferrite sintered at 4000C

Fig. 2. .(b) Scanning Electron Micrograph of fractured surface of Ni0.4Zn0.6Fe2O4 ferrite sintered at 5000C

Characterization of Soft Ni-Zn Spinel Ferrites 21

The calculated particle size for the sample sintered at 5000 C lies between 19-53 nm. The particle size is also investigated by using scanning electron microscopy. Fig. 2(a) and Fig. 2(b) shows SEM at x = 0.4. For the samples sintered at 5000 C, both the results were found to be in good agreement because of their homogeneous and uniform structure. But, for the samples sintered at 4000 C, the results do not follow a systematic pattern, as the samples are having inhomogeneous and non-uniform structures. Lattice constant ‘a’ calculated by using Bragg’s law and average particle size of each sample is shown in Table I. Since the ionic radius of Zn2+

(0.74 A0) is smaller than that of Ni2+ (0.83 A0), an increase in the value of the lattice constant and grain size with an increase in x, as observed in the present work, is expected. Table 1. Lattice constant, Grain size and Saturation magnetization at 4000C and 5000 C

Lattice constant

‘a’

(A0)

Grain size

‘d’

(nm)

NixZn1-

xFe2O4

with x

4000 C 5000 C 4000 C 5000 C

0.0 8.481 8.471 56.65 19.00

0.1 8.482 8.478 55.00 22.50

0.2 8.435 8.487 43.50 27.30

0.3 8.533 8.494 89.00 33.70

0.4 8.525 8.499 95.00 37.50

0.5 8.543 8.509 98.50 40.00

0.6 8.499 8.518 63.00 43.35

0.7 8.540 8.522 99.40 48.00

0.8 8.499 8.529 83.00 50.10

0.9 8.521 8.534 92.50 51.80

1.0 8.544 8.539 97.50 53.00

It is found that the temperature dependence of the resistivity obeys the relation: ρ = ρo exp(Eρ/kT)………...(2) where, Eρ is the energy of activation [23], which is the energy needed to release an electron from an ion for a jump to the neighbouring ion, so giving rise to the electrical conductivity. Upon a jump, a displacement of the ions occurs in the neighbourhood of the electron in the question.

The activation energy, Eρ, was calculated from the slopes by using the relation: Eρ = 0.198 × 10-3 × d(log ρ)/ d(1/T) eV/K…. (3) The plot of log ρ versus 1/T x 103 shows linearity with a sharp change in the slope at the Curie temperature for all the compositions. The activation energies calculated from these slopes and observed from gravity the separation method are given in Table II. Both the values are found to be in good agreement. It can be seen that the activation energies obtained at 4000C have lower values than those obtained at 5000 C for all the compositions. For a better comparison, the graph of log ρ ×103 versus 1000/T at two sintering temperatures for the composition Ni0.6Zn0.4Fe2O4 is plotted in Fig.3.

The straight line II for the sample sintered at 4000C lies far below the line I which is for the sample sintered at 5000C. The change occurring at the slopes of the lines I and II can be clearly seen. However, the values of the Curie temperature remain constant. This kind of behaviour is observed for all the compositions. The observed or experimental density ρobs of the samples is calculated by using the Archimedes principle. For this purpose, each sample is taken in the form of a pellet which is cleaned, dried and weighed in air. Then the same sample is immersed in distilled water to determine the loss of weight. This standard method is described by Smit and Wijn [11]

The porosity of each sample is calculated by using the following relation

Porosity = Th

obsTh

ρρρ )( −

….……...(4)

Table II indicates that the porosity is reduced at a sintering temperature of 5000 C as compared to 4000 C for all the samples and the resistivities are considerably increased or the conductivities are much decreased when the sintering temperature is high.


Table 2. Sintering temperature, Porosity, Resistivity, Activation energy and Curie temperature at 4000C and 5000C

Composition Sintering

temperature (0C)

Porosity Resistivity ×106Ω cm

Activation energy

(eV)

Curie temperature (K)

Obs. Cal. 400 0.248 4.98 0.1978 NiFe2O4 500 0.216 63.8 0.5468

Nil Nil

400 0.243 11.75 0.2052 Ni0.1Zn0.9Fe2O4 500 0.227 141.33 0.4287

311 311

400 0.258 83.6 0.3845 Ni0.2Zn0.8Fe2O4 500 0.246 472 0.4372

348 348

400 0.282 92.0 0.3926 Ni0.3Zn0.7Fe2O4 500 0.255 176.21 0.4284

387 388

400 0.297 3.76 0.3117 Ni0.4Zn0.6Fe2O4 500 0.264 330 0.3435

476 475

400 0.309 121.78 0.3672 Ni0.5Zn0.5Fe2O4 500 0.269 237.66 0.4473

492 493

400 0.342 2.17 0.4028 Mn0.6Zn0.4Fe2O4 500 0.277 287 0.5874

507 508

400 0.342 251 0.3572 Ni0.7Zn03Fe2O4 500 0.281 256 0.5391

564 564

400 0.348 212 0.4852 Ni0.8Zn0.2Fe2O4 500 0.286 242 0.6832

583 582

400 0.353 209 0.4273 Ni0.9Zn0.1Fe2O4 500 0.288 231 0.6628

612 610

400 0.356 195 0.3967 NiFe2O4 500 0.289 218 0.7139

639 638

The observed results can be explained on the basis of microstructural changes brought about by the sintering conditions. The pores may or may not be filled by air, but these pores invariably introduce the insulating or the impeding paths to the electrons. In other words, the pores offer extrinsic contribution to the activation energy. Table II shows that for the intermediate compositions, NixZn1-xFe2O4 with x = 0.4 and 0.6, the reduction in resistivity is of the order of 10-2

, when sintered at 5000C. According to Verwey and deBoer [24], the

conduction in ferrites takes place by hopping via

activation of the states involving cations changing valence as: Fe2+ Fe3+ ………….(5) and vice-versa. The presence of Ni ion on the octahedral sites favors the conduction mechanism as proposed by Van Uitert [25], viz. Ni2+ + Fe3+ Ni3+ + Fe2+ ..(6) which, explains the predominant conduction mechanism of Ni-Zn system under investigation. According to Mössbauer study carried by Daniel and Rosencwaig [26], the cation distribution of Ni-Zn ferrites is given as: (Znx

2+ Fe1-x3+)[Ni1-x

2+Fe1-x3+O4

2-]………..(7) As zinc is a nonmagnetic element and ZnFe2O4 is a normal ferrite, Zn2+ ions occupy the tetrahedral site, while in NiFe2O4, a

Characterization of Soft Ni-Zn Spinel Ferrites 23

predominantly inverse ferrite, Ni ions occupy the octahedral B-site. In the samples sintered at a higher temperature, Ni3+ may also be present along with Ni2+ and hopping of holes from Ni3+ to Ni2+ is also probable according to the mechanism given in Eq.6. 4. Conclusion

We are successful in synthesizing high density Mn-Zn nano-sized ferrites in our laboratory by the co-precipitation method. Although spinel structure of the nano-ferrites is achieved at 4000C, but to obtain a homogeneous and uniform structure, we have optimized sintering temperature of 5000C. Further, from the long series of NixZn1-xFe2O4 ferrites, Ni0.4Zn0.6Fe2O4 sample sintered at 5000C is optimized to get the better results because its resistivity is comparatively higher than that of the other samples. Moreover, its uniform and homogeneous structure makes it suitable for memory storage device, read-write head, low and high frequency applications. References [1] Atul Thakur, Preeti Mathur and M. Singh, Z.

Phys. Chem. 221(2007) 837-845 [2] Preeti Mathur, Atul Thakur and M. Singh, J.

Mater. Sci. 42, No.19 (2007) 8189-8192. [3] A. Thakur, P. Mathur and M. Singh, J. Phys.

Chem. Solids 68 (2007) 378. [4] Preeti Mathur, Atul Thakur and M. Singh ,

Modern Physics Letters B Vol.21, (2007) 1425-1430.

[5] Atul Thakur and M. Singh Z. Phys. Chem. 221 (2007) 887-895

[6] M. Singh and S.P. Sud, Mod. Phys. Lett. B 14 (2000) 531.

[7] A.Verma, T.C. Goel and R. G. Mendiratta, Mater. Sci. Techol. 16 (2000)712.

[8] B. S. Chauhan, R. Kumar, K.M. Jadhav and M Singh, J. Magn. Magn. Mater. 71 (2004) 283.

[9] S. F. Wang, Y. R. Wang, C. K. Y. Thomas, P. J. Wang and C. A. Lu, J. Magn. Magn. Mater. 217 (2000) 35.

[10] M. Singh, Mod. Phys. Lett. B 20 (2006) 1163.

[11] J. Smit and H. P. J. Wijn, Ferrites, (Philips Technical Library, Eindhovan, The Netherlands), 1959.

[12] B. S. Chauhan, R. Kumar, K. M. Jadhav and M. Singh, J. Magn. Magn. Mater. 71 (2004) 283.

[13] Y. Purushotam, M. Singh, S. P. Sud and P. V. Reddy, Int. J. Mod. Phys B 12 (1998) 2247.

[14] A. Thakur, P. Mathur and M. Singh, Z. Phys. Chem. 221 (2007) 1-9.

[15] H. Rikukawa , IEEE. Trans. Magn. (USA)18 (1982) 1535.

[16] C.N. Chinnasamy, A.Narayanasamy , N. Ponpandian, K. Chattopadhyay, H. Gueralt and J.M.Greneche, J. Phys. Condens. Mater.12 (2000) 7795.

[17] A. Thakur and M. Singh, Ceramic International 29 (2003)505.

[18] A. M. Abdeen, J. Magn. Magn. Mater.185 (1998) 199.

[19] S. K. Sharma, S. N. Dolia, R. Kumar, M. Knobel, V. V. S. Kumar and M. Singh, J. Pure & Appl. Phys. 44 (2006) 711.

[20] B. Rolling, A. Happe, K. Funke and M.D.Ingram, Phys. Rev. Lett.78 (1997) 2160.

[21] A. Thakur, P. Mathur and M. Singh, J. Phy. Chem. Solids 68 (2007) 378.

[22] B. D. Cullity, Elements of X-ray Diffraction (Addison Wesley Reading, MA. 1978).

[23] E. J. W. Verwey, P.W. Haaijman, F. C. Romeyn and G. W. Van Oosterhout, Philips Res. Rep. 5 (1950) 173-187.

[24] E .J. W. Verwey and J.H. de Boer, Recl. Trav. Chim. Pays-Bas. (Nitherlands) 55 (1936) 531.

[25] L.G.Van Uitert, J. Chem. Phys. (USA) 24(1956) 306.

[26] J. M. Daniel and A. Rosencwaig , Can. J. Phys. (Canada),48 (1970) 381.

Ferroelectric Relaxor Behaviour in Pb0.9Ba0.1(Fe0.5Ta0.5)O3

Alo Dutta, Chandrahas Bharti1 and T.P. Sinha Department of Physics, Bose Institute, 93/1 A. P. C. Road, Kolkata -700009, India

1University Department of Physics, T. M. Bhagalpur University, Bhagalpur-812007, India Email: [email protected]

Abstract

The relaxor ferroelectric lead barium iron tantalate, Pb0.9Ba0.1(Fe0.5Ta0.5 )O3 (PBFT) is synthesized by Coulombite precursor method. Scanning electron micrograph of the sample taken at room temperature shows the average grain size ≈2.5 µm. The frequency dependence of dielectric response is measured in a temperature range from 118 K to 303 K. The temperature dependence of permittivity (ε′) shows broad maxima at various frequencies. The frequency dependence of the permittivity maximum temperature (Tm) has been modeled using Vogel-Fulcher relation. Cole-Cole plot confirms the polydispersive nature of relaxation time.

1. Introduction

Relaxors have found numerous technological applications due to their excellent dielectric, electromechanical, electro-optical and some other interesting properties. These materials exhibit a non Debye dielectric dispersion in a large frequency range around and below the temperature Tm for which the dielectric constant assumes its maximum value. Such characteristics are of great interest for their practical applications in ferroelectric related devices. Lead based complex perovskite Pb(B′xB′′1-x)O3 is a prominent group among the relaxor compounds where two kinds of cations with different valencies, ionic radii and electro negativities occupy perovskite B sites. Perovskite based Pb(B′1/2B′′1/2)O3 and Pb(B′1/3B′′2/3)O3 relaxor ferroelectrics (RFE) such as Pb(Sc1/2Nb1/2)O3 (PSN) and Pb(Mg1/3Nb2/3)O3 (PMN), are technologically important transducer or actuator materials with extraordinary dielectric and electromechanical properties [1]. Unlike a normal ferroelectric (FE), the dielectric constant (ε′) of a RFE exhibits a wide peak, over a broad temperature range, with strong frequency dispersion, which clearly indicates relaxation processes at multiple time scales [2-6].

The volatilization of toxic PbO during high temperature sintering not only causes environmental pollution but also generate instability of composition and electrical properties of the products [7]. Also products containing Pb based gadgets are not recyclable. Taking these aspects into account, search of environment-friendly low lead or lead free compounds having their either comparable or

superior electrical properties for such applications are the main trends of research now-a-days [7-10].

In this paper, we have tried to reduce the amount of lead and studied the relaxor behaviour of modified Pb(Fe0.5Ta0.5)O3, lead iron tantalate (PFT) with barium doping at A-site i.e., lead barium iron tantalate, Pb0.9Ba0.1(Fe0.5Ta0.5)O3 (PBFT). An analysis of real and imaginary parts of the dielectric permittivity with frequency has been performed at various temperatures. The relaxor behaviour of PBFT is obtained by monitoring the variation of its dielectric permittivity with temperature from 118 K to 303 K in the frequency range 0.1 KHz–1 MHz. 2. Experimental

The PBFT sample is synthesized by Coulombite precursor technique. The mixture of Fe2O3 and Ta2O5 taken in stoichiometric ratio is calcined at 1200 oC for 7 hrs. The resulting mixture is mixed with proper amount of PbO and BaCO3 and finally calcined at 1100 oC for 7 hrs to get PBFT. The scanning electron micrograph of the sample is taken to get the proper compactness of the material. The average grain size is found to be ∼2.5 µm. The sintered pellet of the sample is polished and silver electroded and connected to an LCR meter (Hioki) for dielectric measurement. The frequency dependence of dielectric constant and loss tangent were obtained in the frequency range from 0.1 KHz–1 MHz and in the temperature range from 118 K to 303 K. All the dielectric data were collected while heating at a rate of 0.5 oC min-1. The temperature was controlled by a programmable oven.

Ferroelectric Relaxor Behaviour in Pb0.9Ba0.1(Fe0.5Ta0.5)O3 25

3. Results and discussion

Figure 1 shows the scanning electron micrograph of the sample taken at room temperature by FEI Quanta 200 SEM. The average grain size is found to be ≈2.5 µm. Figure 2 shows the temperature dependence of real (ε′) and imaginary (ε′′) parts of dielectric constant for PBFT at various frequencies. There is a broad peak at around 173 K in the ε′-T plot. With increasing frequency the Tm increases, while the magnitude of the peak value (the value of ε′ at Tm) decreases. There is a strong dielectric dispersion in the frequency region around and below Tm in the ε′-T curve. This shows that PBFT is a relaxor ferroelectric [15]. Fig. 3 shows the inverse of ε′ as a function of temperature at 175 KHz and its fit to the experimental data by Curie-Weiss law. A deviation from Curie-Weiss law starting at Tdev can be seen clearly. The parameter ΔTm, which is often used to show the degree of deviation from the Curie-Weiss law, is defined as ΔTm = Tdev - Tm.

Fig. 1. Scanning electron micrograph of PBFT at room temperature. The Tdev as determined from the Curie-Weiss fit is found to be 195 K, and hence ΔTm is found to be 21 K at 175 KHz. A modified Curie-Weiss law has been proposed by Uchino and Nomura, [11] to describe the diffuseness of the ε′ at and around Tm. It is given as (1/ε′-1/ ε′m) = (T - Tm)γ/C1 , (1)

where ε′ is the dielectric constant at temperature T. γ and C1 are modified constant. The above characterization is done on the basis of Curie-Weiss law and the value of empirical parameters like ΔTm, γ (=1.6) and ΔTrelax = Tm(100KHz)- Tm(100Hz) (=5K) suggest that the permittivity of PBFT follows Curie-Weiss law only at temperatures much higher than Tm.

1200

125 250

75

150

40KHz 58KHz 83KHz 121KHz 175KHz

T (K)

ε'

ε" 40KHz 58KHz 83KHz 121KHz 175KHz

Fig. 2. Temperature variation of dielectric permittivity (ε′ andε′′) of PBFT at various frequencies.

150 200

0.8

1.0

1.2

Tm

175 KHz

TCW

Tdev

1 / ε

'

T (K)

Fig. 3. The inverse dielectric constant (1/ε′) as a function of temperature at 175 kHz for PBFT. The symbol represents experimental data points and the solid line shows fitting of the Curie-Weiss law.


11 12 13

5.67

5.76

103 /

T m

ln ν

Fig. 4. Frequency dependence of Tm for PBFT. The symbols indicate experimental data points and the solid line is the fit for Vogel-Fulcher relationship.

The frequency dependence of Tm is shown in Fig. 4. This frequency dependence of Tm can be modeled using the Vogel-Fulcher relation given by [12,13]

υ = υ 0 exp [-)T ( m fBk

EaΤ−

] … (2)

where υ0 is the pre-exponential factor, Ea is the activation energy and Tf is the freezing temperature. Tf is regarded as the temperature where the dynamic reorientation of dipolar cluster polarization can no longer be thermally activated. The solid line in Fig.4 is the curve fitted to the data using expression (2). The values derived from the curve give activation energy of 0.098 eV, a pre-exponential factor of 1.9 × 1015

Hz and a static freezing temperature of 130 K. This value of Tf is very reasonable, as it is below the temperature where, the maximum occurs. The close agreement of the data with the Vogel-Fulcher relationship confirms the relaxor behaviour in PBFT. In Fig.5, we have plotted ε″ against ε′ usually known as Cole-Cole plot [14] at temperatures 303 K.

2000 4000 6000

1000

2000

3000

4000

303 K

ε"

ε'

Fig. 5. Complex Argand plane plot between ε″ and ε′ at 303 K for PBFT

The plot shows a circular arc with end points on the axis of real and centre below this axis. The complex dielectric constant in such situations is known to be described by the empirical relation [5]

αωτεε

εεεε −∞

∞ +−

+=′′−′= 1*

)(1 ii s

,

where εs and ε∞ are the low-and high–frequency values of ε′, α is a measure of the distribution of relaxation times, τ=ω-1 and ω=2πυ. The parameter α is determined from the location of the centre of the Cole-Cole circle and is found to be 0.32, Thus the relaxation process differs from monodispersive Debye process for which α=0. All these analyses suggest that PBFT is a relaxor ferroelectric. 4. Conclusions

Lead barium iron tantalate, Pb0.9Ba0.1(Fe0.5Ta0.5 )O3 (PBFT) is synthesized by Coulombite precursor method. Scanning electron micrograph of the sample is taken at room temperature. The average grain size is found to be ∼2.5 µm The field dependence of dielectric response is measured in a frequency range 0.1 KHz-1 MHz and in a temperature range from 118 K to 303 K. The temperature dependence of permittivity (ε′) shows broad maxima at various

Ferroelectric Relaxor Behaviour in Pb0.9Ba0.1(Fe0.5Ta0.5)O3 27

frequencies. The frequency dependence of the permittivity maximum temperature (Tm) has been modeled using Vogel-Fulcher relation. Cole-Cole plot confirms the polydispersive nature of relaxation time. References [1] L. E. Cross, Ferroelectrics 76 (1987) 241.

(b) L. E. Cross, Ferroelectrics 151 (1994) 305.

[2] G. A. Samara, Phys. Rev. Lett. 77 (1996) 314.

[3] C. A. Randall and A. S. Bhalla Jpn. J. Appl. Phys. 29 (1990) 327.

[4] R. Blinc, V.V Laguta, and B Zalar Phys. Rev. Lett. 91 (2001) 247601.

[5] S. Saha and T. P. Sinha J. Phys.Condens. Matter 14 (2002) 249.

[6] B. P. Burton, E. Cockayane and U. V. Waghmare Phys. Rev. B 72 (2005) 064113.

[7] Y. Li, W. Chen, Q. Xu, J. Zhou, X. Gu, S. Fang, Mater. Chem. Phys. 94 (2005) 328

[8] Q. Xu, S. Chen, W. Chen, S. Wu, J. Zhou,H. Sun, Y. Li, Mater. Chem. Phys. 90 (2005) 111

[9] J. R. Gomah-Pettry, A. N. Salak, P. Marchet, V. M. Ferreira, J. P. Mercurio, Phys. Stat. Sol. 241 (2004) 1949

[10] Y. Hosono, K. Harada, Y. Yamashita, Jpn. J. Appl. Phys. 40 (2001) 5722

[11] K. Uchino and S. Nomura, Integr. Ferroelectr. 44 (1982) 55.

[12] H. Vogel, Phys. Z. 22 (1921) 645 [13] G. S. Fulcher, J. Am. Ceram. Soc. 8 (1925)

339. [14] K. S. Cole and R. H. Cole, J. chem. Phys. 9

(1941) 341.

Effect of Annealing on Electrical Properties of Nanocrystalline CdSe Thin Films

Jeewan Sharma, G.S.S. Saini, N. Goyal and S.K. Tripathi*

Centre of Advanced Study in Physics, Panjab University, Chandigarh-160014, India. E-mail: [email protected]; [email protected]

Abstract

The present paper reports the effect of annealing in vacuum on electrical properties of nanocrystalline CdSe (n-CdSe) thin films. Semiconducting Cd30Se70 was prepared by melt-quenching technique. Thin films of this material were prepared by inert gas condensation method in a conventional vacuum coating system on well degassed Corning 7059 glass substrates. These films were deposited in the presence of argon gas at partial pressure of 2×10-1 mbar. Film was then annealed at 473 K for one hour. Annealing was done in vacuum of ~ 2×10-3 mbar. The structural (XRD) and electrical properties were studied before and after annealing. 1. Introduction Study of materials in nanometer scale is an emerging area these days because in nanometer scale structures, finite size effects give rise to novel electronic, magnetic, optical and structural properties. The desires to identify, understand and exploit the size-dependent properties of materials at the nanometer scale motivate the study of nanometer scale crystals. For example, semiconductor nanomaterials exhibit strong size-dependent properties [1, 2] that may provide insight into the scaling limits of magnetic storage and microelectronics, key components in information technology. Thus, there is a tremendous scope to design new materials with unusual properties. The drive towards miniaturization of electronic components and integration to accommodate huge number of them in small volume has been there for decades [3].

Nanomaterials are potential candidates for next generation electronic devices and the circuits such as, transistors, resistors and capacitors, are reduced in size. By achieving a significant reduction in their size, the microprocessors, which contain these components, can run much faster, thereby enabling computations at far greater speeds. Also, the flat-panel displays constructed out of nanomaterials [4] possess much higher brightness and contrast than the conventional ones owing to their enhanced electrical and magnetic properties.

The electrical and magnetic properties of nanocrystalline materials will probably form the basis for their widespread and industrial applications. The unique combination of

chemical and mechanical properties with appropriate electrical conductivity could lead to the applications in the semiconductor and microelectronics industries as well as in battery technology.

In search of new semi conducting materials for efficient solar energy conversion through photo electrochemical solar cells, metal chalcogenides are increasingly studied. These materials possess the following criteria to make them potential candidates in photo electrochemical solar cells: (i) the band gap is between solar energy spectrum, making them capable of absorbing a major portion of solar energy, (ii) they are chemically and electrochemically stable in either acid or alkaline condition and (iii) the constituent elements are abundantly available and cheap. Among the many chalcogenide semiconductors, CdSe, ZnSe, SnSe, GeS and GeSe meet these conditions closely and are promising materials for energy conversion [5].

The properties of nanomaterials are critically dependent on the nature of preparation method and subsequent heat treatments like annealing in air, vacuum or different gaseous environments like H2, Ar, N2 etc. [6]. In the present study, nanocrystalline thin films of CdSe are deposited using inert gas condensation method using argon as carrier gas. Effect annealing in vacuum is studied on structural and electrical properties of these films.

2. Experimental

The semi conducting Cd30Se70 is prepared from its constituents (5N pure) by melt-quenching technique. Thin films of this material are prepared by inert gas condensation method on well degassed Corning 7059 glass substrates as described

Effect of Annealing on Electrical Properties of Nanocrystalline CdSe Thin Films 29

elsewhere [7]. The deposited film is then annealed at 473 K for one hour in a vacuum of ~ 2×10-3 mbar. Crystallographic study was carried out before and after annealing using a Phillips PW-1710 X-ray diffractometer with CuKα radiation in the 2θ range from 10o to 70o. The electrical measurements of these thin films were carried out in a specially designed metallic sample holder. All the electrical measurements are done under vacuum. 3. Results and discussion Fig. 1 (i) shows the XRD pattern of as-deposited and annealed films. The spectrum in Fig. 1 (i) (a) shows the diffraction spectrum of as-deposited thin film. There is a highest intensity reflection peak at 2θ = 25.3o <111>, with two another small intensity peaks at 2θ = 41.8 o <220> and 49.5o <311>. Some new peaks at 2θ = 24.1o <100>, 27.4o <101> and 45.9o <103> are observed after annealing. The intensity of peaks is increased on annealing the film at 473 K. The broad hump in the background [Fig. 1 (i) (a)] is due to the amorphous glass substrate and also due to some amorphous phase present in the n-CdSe thin film. The comparison of observed ‘d’ values with standard ‘d’ values [8, 9] confirms that the as-deposited film is having sphalerite cubic (zinc-blende type) nanocrystalline structure and the new peaks are due to hexagonal phase present in the annealed film.

10 20 30 40 50 60 70

1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.80.000.030.060.090.120.150.18

[311]C[112]H[103]H

[220]C[110]H

[101]H

2θ

[100]H

(a)

(b)

Inte

nsity

[111]C / [002]H (i) n-CdSe

(βco

sθ)/λ

(sinθ)/λ

(a)

(b)

(ii) n-CdSe

Fig. 1. (i) XRD pattern and (ii) size and strain analysis plots for (a) as-deposited and (b) annealed n-CdSe thin films.

Information of the strain and the particle size is obtained from the full width at half maximum (FWHMs) of the diffraction peaks [7]. Fig. 1 (ii) represents the plots of (βcosθ)/λ

vs. (sinθ)/λ for as-deposited and annealed films which are straight lines. The slopes of the plots give the amount of residual strain, which turns out to be 8.13×10-2 and 6.39×10-3 for as-deposited and annealed films respectively. The reciprocal of intercepts on the (βcosθ)/λ axis give the average particle size as ~ 3.9 nm and 20.8 nm respectively. The values of strain and particle size are given in Table 1. The increase in the particle size on annealing shows the improvement in the crystallinity of the film. The negative value of residual strain for as-deposited film indicates the compressive strain while positive value of residual strain for annealed film indicates the tensile strain in this film. If the film deposited is free from impurities, the compressive strain is generated at the film substrate interface, when the very small crystallites are bonded to substrate, due to surface tension. The tensile strain developed in annealed film may be due to the difference in thermal expansion coefficients of the substrate and deposited material. Table 1. Representation of strain and particle size for as-deposited and annealed n-CdSe thin films. Film Strain Particle size (nm)

As-deposited -8.13×10-2 3.9

Annealed 6.39×10-3 20.8

DC Conductivity Measurements: Fig. 2 shows the temperature dependence of dark conductivity for as-deposited and annealed n-CdSe thin films in the temperature range 253-370 K. The values of σd for as-deposited and annealed n-CdSe thin films are found to be

2.8 3.0 3.2 3.4 3.6 3.8-21

-20

-19

-18

-17

-16

-15

-14

-13

-12

-11

Lnσ d

(Ω-1

cm-1

)

1000/T (K-1)

(a)

(b)

n-CdSe

Fig. 2. Temperature dependence of dark conductivity for (a) as-deposited and (b) annealed n-CdSe thin films.


(4.62±0.02) ×10-8 Ω-1cm-1 and (7.23±0.02)×10-8 Ω−1cm-1 respectively at 298 K. The electrical conductivity shows typical Arrhenius type of activation

⎟⎠⎞

⎜⎝⎛ Δ−

=kT

Eod expσσ (1)

where ΔE is the activation energy for conduction and k is the Boltzmann’s constant. The plots of lnσd vs. 1000/T are straight lines in the measured temperature range. This implies that the conduction in as-deposited and annealed n-CdSe thin films is an activated process having single activation energy. The activation energies for dc conduction have been calculated from the slopes of lnσd vs. 1000/T curves and is found to increase from (0.31±0.01) eV to (0.58±0.01) eV on annealing. The values of σd and ΔEd for as-deposited and annealed n-CdSe thin films are given in Table 2. The value of σd and ΔEd increases on annealing n-CdSe thin film.

The increase in conductivity after annealing in these films is attributed to the improvement in crystallite size of films [10, 11]. The increase in electrical conductivity and activation energy after annealing may be due to the change in structural parameters, improvement in crystallite and grain size, decrease in inter-crystallite boundaries (grain boundary domains) and removal of some impurities (adsorbed and absorbed gases). Excess atoms of compound are also possible [12] due to a small change in stoichiometry after annealing.

Steady State Photoconductivity

Fig. 3 shows the temperature dependence of photoconductivity for as-deposited and annealed n-CdSe thin films. The values of photo-

2.8 3.0 3.2 3.4 3.6 3.8-17.0

-16.5

-16.0

-15.5

-15.0

-14.5

-14.0

-13.5

-13.0

-12.5

-12.0

Lnσ p

h(Ω

-1cm

-1)

1000/T (K-1)

(a)

(b) n-CdSe8450 Lux

Fig. 3. Temperature dependence of photo conductivity for (a) as-deposited and (b) annealed n-CdSe thin films.

conductivity are calculated to be (2.93±0.02) ×10-7 Ω−1cm-1 and (4.22±0.02) ×10-7 Ω−1cm-1 for as-deposited and annealed n-CdSe thin films respectively at 298 K. Clearly, the value of σph increases on annealing. The photo activation energy (ΔEph) has been calculated using the slopes of Fig. 3 and is found to increase from (0.17±0.01) eV to (0.30±0.01) eV on annealing (Table 2). The activation energies for photoconduction are much smaller than for the dark conduction. No maximum in the steady state photoconductivity with temperature has been observed in the measured temperature range. Photosensitivity of n-CdSe film decreases after annealing. Fig. 4 shows the intensity (F) dependence of σph for as-deposited and annealed n-CdSe thin films. It is clear from the figure that lnσph vs. lnF curves are straight lines indicating that σph follows a power law with intensity, i.e., σph ∝ Fγ. In as-deposited and annealed n-CdSe thin films, the values of γ are found to be in between 0.5 and 1.0. According to Rose [13], 0.5 < γ < 1.0, can not be understood by assuming a set of discrete trap levels but requires the existence of a continuous distribution of traps in the band gap. The values of γ lying between 0.5 and 1.0, indicates that continuous distribution of localized states exists in the mobility gap of the present material and the resulting recombination mechanism is bi-molecular [13], where the recombination rate of electrons is proportional to the number of holes.

3 4 5 6 7 8 9-20.5-20.0-19.5-19.0-18.5-18.0-17.5-17.0-16.5-16.0-15.5-15.0

lnσ p

h(Ω

-1cm

-1)

lnF(Lux)

(a) γ =0.51

(b) γ =0.58

n-CdSe

Fig. 4. Intensity dependence of photo conductivity for (a) as-deposited and (b) annealed n-CdSe thin films. Transient Photoconductivity

Fig. 5 shows the rise and decay curves of Iph for as-deposited and annealed n-CdSe thin films. Iph rises to a steady state value and no peak is observed. It is evident that decay of Iph is slow. During decay, the photocurrent does not reach zero

Effect of Annealing on Electrical Properties of Nanocrystalline CdSe Thin Films 31

for a long time after the incident light is switched off. A persistent photocurrent is observed in both cases. This type of photoconductive decay has also been reported in various other semiconductors [14, 15]. In the present case, the non-exponential decay of photoconductivity is observed. The values of τd at different times have been calculated using

11

−

⎥⎥⎦

⎤

⎢⎢⎣

⎡⎟⎟⎠

⎞⎜⎜⎝

⎛−=

dtdI

Iph

phdτ (2)

for as-deposited and annealed n-CdSe thin films

0.0

2.0x10-9

4.0x10-9

6.0x10-9

8.0x10-9

0 100 200 300 400 500 600

0.05.0x10-11

1.0x10-10

1.5x10-10

2.0x10-10

2.5x10-10

I ph

(A)

(b)

n-CdSe8450 Lux

Time (s)

(a)

Fig. 5. Decay of photocurrent at 298 K for (a) as-deposited and (b) annealed n-CdSe thin films. from the slopes (at different times) of decay curves of Fig. 5. The decay times observed for n-CdSe thin films are found to be time dependent. The value of σd increases with time, which confirms the non-exponential decay of photocurrent. Table 2. Representation of dark-, photo-conductivity, dark-, photo-activation energy, γ and carrier life time for as-deposited and annealed n-CdSe thin films.

Property As-deposited Annealed

σd (Ω-1 cm-1) (4.62±0.02)×10-8 (7.23±0.02)×10-8

σph(Ω-1 cm-1) (2.93±0.02)×10-7 (4.22±0.02)×10-7

ΔEd (eV) (0.31±0.01) (0.58±0.01)

ΔEph (eV) (0.17±0.01) (0.30±0.01)

γ 0.51 0.58

(lnτd)t=0(sec) 1.3 1.6

Fig. 6 shows the plots of lnτd vs. lnt for as-deposited and annealed n-CdSe thin films at a temperature 298 K and intensity 8450 Lux. The extrapolation of the curves at t = 0, gives the values of the carrier life time [16] and are found to be 1.3 and 1.6 seconds for as-deposited and annealed n-CdSe thin films respectively. Clearly, the carrier life time increases with annealing the film (increasing size). The straight lines in Fig. 6, obey a power law of the form t-N, with N = d(lnτd /lnt) and the values of N are 0.91 and 0.72 for as-deposited and annealed n-CdSe thin films respectively.

1.5 2.0 2.5 3.0 3.5 4.02.5

3.0

3.5

4.0

4.5

lnτ d

(s)

lnt (s)

(a) N=0.91

(b) N=0.72

n-CdSe8450 Lux

Fig. 6. Plot of lnτd vs lnt for (a) as-deposited and (b) annealed n-CdSe thin films.

A persistent current is found in both states of n-CdSe thin film. This persistent photocurrent may not be simply due to carriers trapped in the localized states [17]. So, for simplifying the analysis, the persistent photocurrent is subtracted from the measured photocurrent. In the present case, the slope of decay curves goes on decreasing continuously as the time of decay increases. This indicates that the traps are present at all the energies in the band gap of all samples. These traps have different time constants and hence giving non-exponential decay of photocurrent. The value of decay time constant (τd) increases with time, which confirms the non-exponential decay of photocurrent in all the samples. The carrier life time is found to increase on annealing. 4. Conclusions

As-deposited film is found to have zinc-blende type nanocrystalline structure while hexagonal character starts increasing when film is annealed in vacuum at 473 K for one hour. The particle size increases from 3.9 nm to 20.8 nm after annealing.


The strain is found to decrease from -8.13×10-2 to 6.39×10-3 after annealing. The strain is compressive in as-deposited films and tensile in annealed film. The value of σd and ΔEd increases as the particle size of n-CdSe increases. Photosensitivity increases after annealing. Steady state photoconductivity studies indicate that there is continuous distribution of localised states. Decay of photocurrent is slow which may be due to the presence of deeper localised states in this material. Acknowledgements

We acknowledge the financial support provided by the CSIR, New Delhi for the completion of this work.

References [1] K.K. Nanda, Chemical Physics Letters 419

(2006) 195. [2] C.C. Yang, Q. Jiang, Mater. Sci. Engin. B

131 (2006) 191. [3] M. Bangal, S. Ashtaputre, S. Marathe, A.

Ethiraj, N. Hebalkar, S.W. Gosavi, J. Urban, S.K. Kulkarni, Hyperfine Interactions 160

(2005) 81. [4] M.B. Yu, Rusli, S.F. Yoon, S.J. Xu, K.

Chew, J. Cui, J. Ahn, Q. Zhang, Thin Solid Films 377-378 (2000) 177.

[5] B. Subramanian, T. Mahalingama, C. Sanjeeviraja, M. Jayachandran, M. J. Chockalingam, Thin Solid Films 357 (1999) 119.

[6] R.B. Kale, C.D. Lokhande, Appl. Surf. Sci. 223 (2004) 343.

[7] J. Sharma, G.S.S. Saini, N. Goyal, S.K. Tripathi, J. Optoelect. Adv. Mat. 9 (2007) 3194.

[8] JCPDS Data File No. 8-459. [9] JCPDS Data File No. 19-191. [10] R.B. Kale, C.D. Lokhande, Appl. Surf. Sci.

223 (2004) 343. [11] S. Ghosh, A. Mukherjee, H. Kim, C. Lee,

Mater. Chem. Phys. 78 (2003) 726. [12] G.I. Rusu, M.E. Popa, G.G. Rusu, I. Salaoru,

Appl. Surf. Sci. 218 (2003) 222. [13] A. Rose, Concepts in Photoconductivity and

Allied Problems, Interscience, New York (1963).

[14] D.V. Harea, I.A. Vasilev, E.P. Colomeico, M.S. Iovu, J. Optoelect. Adv. Mater. 5 (2003) 1115.

[15] M.S. Iovu, S.D. Shutov, V.I. Arkhipov, G.J. Adriaenssens, J. Non-Cryst. Solids 299 (2002) 1008.

[16] D.P. Padiyan, A. Marikani, K.R. Murali, Cryst. Res. Technol. 35 (2000) 949.

[17] V. Sharma, A. Thakur, P.S. Chandel, N. Goyal, G.S.S. Saini, S.K. Tripathi, J.Optoelect. Adv. Mater. 5 (2003) 1243.

Self-Assembled and Ordered Templates by Anodic Oxidation of Aluminium

S. K. Yadav1, R. Gupta1,2 and K. N. Rai1

1Materials Science Programme, Indian Institute of Technology Kanpur, Kanpur - 208016, U. P. INDIA. 2 Department of Physics, Indian Institute of Technology Kanpur, Kanpur - 208016, U. P. INDIA.

E-mail: [email protected] , [email protected]

Abstract

Anodic oxidation of aluminium sheet has been investigated to produce ordered nanoporous templates for encasing Ni particles to catalyze ordered growth of carbon nanotubes. Anodization was carried out under potentiostatic conditions, using oxalic acid as an electrolyte. Different combinations of processing parameters (etchants, anodization time, temperature and number of anodization steps) were used in order to optimize the process. Anodized alumina samples were characterized with scanning electron microscopy. Disordered and hexagonally ordered nano-porous alumina structures were obtained in single step and double step anodization respectively, with commercial and highly pure aluminium foils. The average diameters of these pores ranged from 30 to100 nm and pore to pore spacing varying from 80-200nm using oxalic acid as electrolyte and by changing the electrochemical cell voltage. We have investigated pore formation process for different types of aluminium foils (purity and thickness) and observed variation in pore diameters and pore ordering as well. The average interpore separation Dip is found to be a linear function of anodizing voltage ’V’ represented by a equation of the type Dip = mV + C, with m = 2.2nm/V and C =16nm . These templates were used as starting materials for depositing Ni nanodots on different substrates for other applications.

1. Introduction

Thrust on synthesis of nanomaterials has generated interest on using porous alumina as templates for growth. Ordered porous anodic alumina (PAA), formed by anodic oxidation of aluminium has attracted much interest as a starting material for fabrication of nanoscale electronic and optoelectronic devices [1], magnetic nanodots and nanowires [2], quantum sized particles [3], as mask in fabrication of nano porous GaAs, InP and diamond like film for various applications [4,5]. PAA membranes possess many interesting physical properties. They exhibit relatively ordered hexagonal array of pores [6] and a narrow distribution of pore size and inter pore spacing. Pore diameter, ordering and pore to pore spacing strongly depends upon the working voltage, electrolyte, and film purity. A typical structure of anodized aluminium comprises of a porous layer (facing electrolyte) over an impervious (non porous) alumina layer, supported on anodizing aluminium sheet It is also possible to remove residual aluminium on the back side and the barrier layer to get through porous films.

In the present work different types of aluminium foils have been used for anodization. PAA films in single and double step anodization

have been produced. Different pore diameters, ordering and porosity are observed depending upon the anodizing voltage, electrolyte, film quality and etching time. We use these templates to deposit Ni nano-particles on different substrates. These ordered Ni nano-islands can be useful for the growth of carbon nanotubes and other nano-structures. 2. Experimental

Different types of aluminium foils 100 µm (99.5% purity), 130µm (>99.99%) and 150 µm thickness (commercial) were taken for the study. Each type of aluminium foil was cleaned in acetone and annealed at 5000C in ambient atmosphere for 5hrs in order to enlarge the grain size and to obtain strain free homogeneous microstructure over large area. Following the annealing, foils were degreased with acetone in an ultrasonic bath followed by rinsing in methanol and distilled water. Aluminium foils processed as above were electro-polished for 20 sec. in a solution of phosphoric acid: sulfuric acid: distilled water in the ratio of 2:2:4 (by weight) at 16V. The surface roughness after electro-polishing improved and a mirror like finishing of aluminum surface can be seen. Pretreatment of aluminium foil is


recommended though, electro-polishing is not essential in fabrication of ordered pore arrays. For anodizing aluminium foils were mounted in a home designed- electrochemical cell made out of Teflon. It essentially consisted of an anode holder, a cathode holder and a variable spacer to control the spacing between the two electrodes. Aluminum foils to be anodized were tightened at the anode holder, in between two Teflon plates with help of Teflon screws and the cathode (Al or graphite) was put in the cathode holder. A circular hole of 0.5cm2 area was made through which only one face of aluminum foil faced the electrolyte hence limiting the current flow in the cell. For all our experiments we fixed the distance between cathode and anode to 5 cm. A regulated dc power supply (Aplab, 120V-1.2A model-7231) was used for the preparation of PAA films. Experiments were carried out in 0.3M oxalic acid solution on different types of aluminum foils. To ensure uniformity in the growth conditions and avoid concentration gradients in the solution we stirred the solution using a magnetic stirrer at the rate of 100rev/minute. This also prevented the solution from heating locally. A double jacket beaker attached to a chiller was used to cool the electrolyte solution. We anodized 150µm (commercial) 100µm (99% pure) and 130µm (>99.99) thick foils in 0.3M oxalic acid at 100C solution temperature with different applied electrode potential (30-80V). Our cell design was such that aluminum foils oxidized from front face only. To remove remaining aluminum on the backside after oxidization, we used 10 wt% cupric chloride and saturated mercuric chloride solutions. Pore widening process was carried out using phosphoric acid. 3. Result and Discussion

Fig.1. shows a typical current-time curve during the oxidation of aluminum foil at applied voltage of 40V in 0.3M oxalic acid at 100C solution temperature. Initially high current is observed just after applying the voltage. A non porous oxide barrier layer forms quickly as the potential was applied to the electrode. Due to the fast growth of the insulating oxide layer, resistance of the electrode increases resulting in a sharp decrease in current. The local heating resulting from the resistance and electric field enhancement due to the formation of a thin oxide layer promotes local dissolution of barrier layer, resulting in pore nuclei formation [7,8] hence a slow increase in current.

0 100 200 300 400 500

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

Cur

rent

Den

sity

mA

/cm

2

Time(seconds)

Fig. 1. Current-time curves for anodization process at 40V

According to the previous reports [7,9] well-ordered nanopore arrays could be obtained from the aluminum foils which have been pretreated with high temperature annealing and electropolishing. However, our results suggest that the two-step anodization process and film qualities are much more significant factors to obtain well ordered pore arrays rather than the pretreatments. Fig.2 shows SEM image of PAA prepared by anodization of 150µm thick aluminium foil (commercial) for 30 hrs at 40V applied potential in 0.3M oxalic acid solution after 20 minute etching in H3PO4. Remaining aluminium on the backside is dissolved in 10wt% CuCl2 solution. We observe that the pores are randomly arranged having size (pore diameter) from 25 nm to 65 nm with average pore size of 49nm.

Fig. 2. Commercial aluminum foil anodized in 0.3M oxalic acid at 40V. As observed in Fig. 2. most of the pores are noncircular (except few). Average pore to pore

Self-Assembled and Ordered Templates by Anodic Oxidation of Aluminium 35

distance and pore density are found to be 98nm

and 2cm/ 10101.1× respectively. The SEM image of PAA fabricated from

100µm thick aluminum foil (99% pure) anodized in 0.3M oxalic acid solution at 40V for 3hrs is shown in Fig.3. For barrier layer dissolution and pore widening H3PO4 has been used .The pores are found to be almost circular with average pore diameter of 57 nm and relatively narrow size distribution. Partially hexagonal ordered pores can be seen (Fig.3). The average pore to pore distance and pore density was found to be 103nm and

2/cm9109.5× respectively.

Fig. 3.Aluminium foil (99%) anodized in 0.3M oxalic acid at 40V. Furthermore, a 20µm thick 99.99% pure aluminium foil is anodized under same conditions as discussed earlier for 5 hrs.

SEM image of the bottom surface is shown in Fig.4. Almost circular pores with hexagonal ordering were observed. Obtained values of pore diameter (Dp), pore to pore distance (Dip) and pore density are given in Table1. It is to be noted that the pore density observed for commercial aluminium foil is a bit larger than that of 100µm and 20 µm films because of wide distribution of pores. The smaller pore to pore distance also indicates a larger pore density.

Fig. 4.Pure aluminum foil (>99.99%) anodized in 0.3M oxalic acid at 40V. Table 1. Pore diameter (Dp), pore to pore distance (Dip) and pore density for different grades of material.

Purity Dp (nm)

Dip (nm)

Pore density N/ cm2

commercial 49 98 2/cm10101.1× 99.00% 57 103 2/cm9109.5× >99.99 53 105 29 /cm107.08×

In the next step commercial aluminum foils were anodized in 0.3M oxalic acid solution at 30, 40, 50, 60, and 80V respectively. The pore size, pore to pore distance and pore density for various applied voltage are summarized in Table2.We see that as applied potential increases both the pore size and pore-pore separation increases. Table 2. Pore size, pore to pore distance and pore density for different voltage.

V Dp (nm) Dip (nm) Pore density/cm2

30 37 78 101052.1 × 40 49 99 101009.1 × 50 58 123 91016.6 × 60 69 136 91016.8 × 80 103 203 91081.1 ×

A plot of pore diameter and pore to pore distance with applied voltage is shown in Fig.5. We see that both pore diameter and pore to pore distance vary almost linearly with applied voltage consistent with earlier reported work in the literature.


30 40 50 60 70 80

40

60

80

100

120

140

160

180

200

40

60

80

100

120

140

160

180

200

Inte

r-por

e D

ista

nce(

nm)

Pore

Dia

met

er(n

m)

Applied Voltage(V)

Fig. 5. Variation of pore diameter and inter-pore distance with applied voltage. We extrapolated both curves and found that at zero value of applied potential pore diameter is almost zero and pore to pore distance is almost 16nm which is close to the reported value of 18nm[10]. Further, we prepared PAA film in a two step anodization process. Pure aluminum foil (>99.99%) of 130µm thickness was electropolished in 2:2:4 H2SO4:H3PO4:H2O solution. The polished aluminium foil was anodized twice.

1ststep: Electro polished aluminum foil anodized in 0.3M oxalic acid solution for 1hrs at 40V at 100C. Formed aluminium oxide was dissolve in 6wt% H3PO4 for 1hrs at 600C.

2nd step: Second anodization was carried for 20hrs at same conditions and remaining aluminium at backside was removed in cupric chloride solution. The sample was broken into several pieces and etched in different conditions. Fig.6 shows SEM image of the bottom of porous anodic alumina without etching just after the aluminium dissolution. A regular arrangement of particles of spherical shape can be seen which indicates that the anodization takes place through the bottom of the pores. A few defects can also be observed. The average size of particles, inter-particle distance and particle density are calculated as 98nm, 110nm and 8.5x109/cm2 respectively. In the top view we observed open pores. This sample was etched for 30 minutes in 5wt% H3PO4 at room temperature and we still observe a structure similar to that given in Fig.6. A fresh solution was made of 5wt% phosphoric acid and this sample was etched for 15mn more at room temperature (Fig.7).

Fig.6. Bottom view of anodized aluminium oxide without etching after dissolving aluminium in CuCl2.

We observe open pores at the bottom as well

as at top. The average size of pores, inter-pores distance and pore density are calculated as 24nm, 113nm and 9.7x109/cm2 respectively. A hexagonal ordered arrangement of pores can be seen. Each pore has exactly six coordination number.

Fig.7. Bottom view after etching in 5wt%H3PO4 for 30minute and again etching for 15minute in fresh 5wt%H3PO4 solution at RT.

In the next step we proceeded to use the templates for growing ordered arrays of Ni nanostructures. These are very important both for the basic research and for their potential application in growth of carbon nanotubes, magnetic recording media, sensors and other devices. Nickel was electrodeposited on titanium coated silicon substrate through porous anodic alumina from an electrolyte containing 300g/l

Self-Assembled and Ordered Templates by Anodic Oxidation of Aluminium 37

NiSO4.7H2O, 45g/l NiCl2.H2O and 40g/l H3BO3 for 20 sec.

Fig. 8. Face view PAA after Ni deposition for 20 second. Fig.8 shows the SEM image of a PAA film electro-deposited with Ni for 20sec. As can be seen from the image, a large number of pores are empty suggesting a poor filling ratio with nearly equal number of pores filled with nickel and remaining open. Therefore, we increased the deposition time to 120 sec. with all other parameters kept same. The SEM photograph of the top surface is shown in Fig.9. As seen from the image Ni does not cover all the pores and a few pores are still open. However, the ratio of the number of open pores to the number of filled pores is very large signifying a large filling ratio.

Fig. 9. (a) Face view PAA after Ni deposition for 120 sec.

4. Conclusion

We have studied self–organization of two-dimensional pore arrays formation in different types of aluminum foils in anodic oxidation of aluminium. We find irregular pore array in commercial aluminum foil, circular and relatively ordered pore arrays in pure aluminum foil (99%) and circular and perfectly ordered pore arrays in highly pure aluminium foil (99.99%). We also observed the difference in average pore size, pore spacing and their distribution as a function of film purity. Pores of diameter ranging from 30-100nm are fabricated having inter pore distance in the range of 80-200nm by changing anodizing voltage in 0.3M oxalic acid. Ni nano dots have been electro-deposited on titanium coated silicon substrate through PAA as mask. Morphology of PAA films after nickel deposition has been examined. A very good filling ratio is found for 120sec. deposition of Ni. Further work is needed to explore the use of these templates for other interesting applications such as photonic band gap materials. Acknowledgements SKY thanks CSIR, India for financial support. Dr. K. N. RAI is an Emeritus Fellow of UGC India. References [1] H. Masuda, M. Yotsuya, M. Asano, and K.

Nishio, Appl. Phys. Lett., 78, 6, (2001). [2] M. Nakao,S. Oku, T. Tamamura, K. Yasui and

H. Masuda, Jpn. J.Appl. Phys. 38,1052, (1999).

[3] P. Hoyer, N. Baba, and H. Masuda, Appl. Phys. Lett. 66, 20,(1995).

[4] H. Masuda, Electrochemical and Solid-State Letters, 4 (11) ,G101, (2001).

[5] A. Packer, J. Phys. Chem. 62,1025, (1958). [6] Hideki Masuda, Masato Yotsuya, Mari Asano,

and Kazuyuki Nishio Appl. Phys. Lett., Vol. 78, No. 6, 5 February 2001

[7] J. H. Yuan, F.Y. He, D.C. Sun. and X.H. Xia Chem. Mater.2004, 16,1841-1844

[8] A.P. Li, F. Muller, A. Birner, K. Nielsch and U. Gosele, J. Appl. Phys. 84, 6023,(1998) .

[9] J. J. Schneider, N. Engstler, K.P. Budna, C. Teichert and S. Franzka, Eur. J. Inorg. Chem 2352. (2005).

[10] K.N. RAI and E. Ruckenstein , J. Catal. 40, 117 (1975)

Structural and Electrical Characterization of Lanthanum Substituted Barium Titanate Thin Films

Pramod Singh Dobal, Anju Dobala and R. S. Katiyarb

Department of Physics, VSSD College, Kanpur 208002 (UP), INDIA. a Department of Material Science & Physics, CSJM University, Kanpur 208024

b Department of Physics, University of Puerto Rico, San Juan PR 00931 USA E-mail: [email protected]

Abstract

Thin films of Ba1-xLaxTiO3 with different La contents (x = 0.0, 0.03, 0.05, and 0.10) were synthesized

on platinum substrates using sol–gel technique. The effect of trivalent La3+ substitution on structural, ferroelectric and dielectric properties of these films was investigated. Combined X-ray diffraction and Raman spectroscopic measurements on these compositions revealed a slight increase in tetragonal distortion of the unit cell with increasing La content. Accordingly, an increase in the tetragonal to cubic transition temperature TT/C was exhibited by the temperature dependent Raman spectroscopy in the range 70-5000 K. The temperature dependent dielectric measurements in these films showed broad and diffused dielectric maxima as well as their merging behavior with increasing La content. The remnant polarization for La substituted films, which decreased with increasing La content, was found to be much higher than that for undoped BaTiO3. 1. Introduction

Barium titanate (BaTiO3) is a well-known perovskite that has widespread applications owing to its high dielectric constant, piezoelectric, electrostrictive, and ferroelectric properties [1-3]. Many of the useful properties of BaTiO3 (BTO) can be induced and tailored by modifying its stoichiometry by incorporating appropriate dopants in its perovskite structure [4-6]. However, the effect of a specific dopant on the electrical properties of BaTiO3 depends on various parameters like the doping site, valence change of the dopant, charge compensating defects, possible defect pairs between the dopant ions and the charge compensating defects, etc. La3+ and Pr3+ usually substitute on Ba sites of BTO whereas other lanthanides Y3+, Tb3+, Gd3+ and Er3+ tend to have simultaneous substitution on both Ba and Ti sites. The substitution of Ba2+ by La3+ in BTO has been reported to have a switching from ‘donor-doping’ mechanism to an ‘ionic compensation’ [7]. In donor-doping mechanism A-site doping with La3+ can be represented by;

−

−++

++++⇒+

+⇒

eOOTiLaTiOOLaeLaBa

oTiBa 2216222 2232

32

It leads to the composition Ba1-xLaxTiO3 with [Ba+La]/ [Ti] = 1. On the other hand, an ionic compensation mechanism

+++ ⇒+ 342 41 LaTiBa leads to formula Ba1-xLaxTi1-x/4O3 with [Ba+La]/ [Ti] > 1. Considering the radii of La3+, Ba2+, Ti4+ ions ( rLa

3+ = 1.36 A0, rBa2+

=1.44 A0 and rTi4+ = 0.74A0), substitution of

Ba2+ by La3+ is expected to be more on Ba-sites of the BTO compared to that on Ti-sites. Random lattice distortion produced by substitution can lead to the formation of dipolar impurities and defects that have profound influence on the structural properties. In this work, La3+ has been incorporated on Ba2+ sites of the BaTiO3 perovskite lattice to result in Ba1-xLaxTiO3 (BLT) using sol-gel technique and a comprehensive study of substitutional effects on structural, ferroelectric, and dielectric properties has been presented . 2. Results and discussion

X-ray diffraction (XRD) patterns obtained from Ba1-xLaxTiO3 (BLT) films are shown in Fig.1(a) for La content x = 0.0, 0.03, 0.05, 0.10. The inset shows the splitting of (001) and (100) reflections. The diffraction pattern obtained from undoped BTO (x = 0.0) powder is also presented for comparison. The comparison of powder and film XRD patterns clearly indicates that the films have been entirely crystallized into the perovskite structures. Except the Pt substrate peaks at

Structural and Electrical Characterization of Lanthanum Substituted Barium 39

39

about 39.80 and 46.20 2θ values no other impurity/ constituent phases were detected in the films. The splitting of diffraction peaks in all BLT compositions confirmed the tetragonal structure of the films. The lattice parameters were calculated from the XRD patterns of the powder and thin film samples, which showed a marginal increase in the tetragonal distortion (c/a ratio) with increasing La content (Fig.1b). The similarity in the crystal structure of the undoped and the lanthanum-doped barium titanate is attributed to the low concentration of the dopant.

Fig. 1. (a) X-ray diffraction patterns of BLT films with La content. Inset shows 001 and 100 reflections; (b) the variation of tetragonal distortion with La content in BLT.

Compositional dependent Raman Scattering results from Ba1-xLaxTiO3 films are displayed in Fig.2(a). The spectra of x=0.0 compositions obtained from both powder and thin film contain one overdamped E1(TO) mode and other E1 and A1 modes. Accurate assignment of these peaks requires the phonons to be propagating along any principal axis. For oblique angles the phonon dispersion makes it difficult to isolate different E(TO) modes from the corresponding A1(TO) modes [8]. Therefore the mode assignment in polycrystalline films is done based on their frequencies. An anti-resonance effect at about

180cm-1 has been understood originating from the anharmonic coupling between the sharp A1(TO1) mode at about 180 cm-1 with broad 267 cm-1 A1(TO2) mode [8-11]. Third asymmetric A1(TO3) mode appears at about 525 cm-1. The E(TO) mode, which is associated with the tetragonal –cubic phase transition [12] was observed at about 306 cm-1. The A1(LO3) mode, which is at about 727 cm-1 in tetragonal BTO crystal for phonon propagating along c-axis transforms to the 715 cm-1 E(LO3) phonon propagating in the ab plane. The mode around 715 cm-1 in BTO is therefore assigned as E(LO3) mode. Because of the random grain orientations in the film, the directions of the phonon wave vectors are randomly distributed from one grain to another with respect to the crystallographic axes. As a result, the Raman lines resulting from the mode mixing and long-range electrostatic force effects are relatively broader than those found in BaTiO3 single crystal. The phonon frequencies can shift remarkably to lower/ higher values depending upon the tensile/compressive nature of the stresses.15-16 The observed frequencies of E(TO) modes in BaTiO3 films were found slightly lower compared to the powder samples indicating stresses in the films. Such stresses in thin films can be attributed to the differences in lattice parameters and thermal expansion coefficients of the film and substrate material.

Fig. 2. Raman spectra of BLT films at (a) 300K and (b) 70K.

The sharp A1(TO1) mode at 180 cm-1 corresponds to the vibration of the Ba ions against the TiO6 octahedra. Raman spectra with La substitution did not show any remarkable shift of the anti-resonance effect at


about 180 cm-1 due to the similar masses of La and Ba ions. However, the coupling between the sharp A1(TO1) and broad A1(TO2) modes weakens as the intensity of A1(TO2) mode decreases. All other Raman modes weaken and broaden with increasing La content. The breakdown of long-range order can allow off Brillioun zone center (k≠0) phonons and lead to a change in the spectral distribution of the Raman modes. The E(TO1) mode, which appeared as a small hump at about 40 cm-1 in the spectra of x=0.0, 0.03 and 0.05 compositions was downshifted and broadened in x=0.10 composition film spectrum. This phonon shift could be due to the enhanced two-dimensional compressive stresses, which result in tetragonal distortions in the films at higher La content.

To understand the phase transition behavior of BLT films, Raman spectra were recorded in 70-500 K temperature range. The depolarized Raman spectra of BLT films at 70K are shown in Fig. 2(b). The disappearance of overdamped E1(TO) mode as well as the strong appearance of E(TO3) mode at about 485 cm-1 in these compositions clearly reflect a different phase (believed to be rhombohedral) at 70K. The appearance of extra bands in the vicinity of 640 cm-1 is interpreted as caused by the distortions of Ti-O-Ti bonds in the above low temperature phase. Like the room temperature Raman spectra, the spectral features in all compositions remain same with increasing La content.

Fig. 3. Temperature variation of Raman modes in BTO film (x=0.00).

Temperature dependent Raman spectra from BTO film (x=0.00) are shown in Fig. 3. The spectra recorded in the rhombohedral phase (below 200K) could not be distinguished from those in the orthorhombic phase (200-290K). The disappearance of 306 cm-1 E(TO) mode suggests a tetragonal to cubic phase transition at about 410 K and two very broad bands centered at about 288 and 516 cm-1 characterize the cubic paraelectric phase, as reported for BaTiO3. A slight increase in the tetragonal to cubic transition temperature in thin film compared to the single crystal is likely to originate from the stresses in the films. The E(TO3) mode at 485 cm-1 loses its intensity with increasing temperature. The temperature evolution of this mode has been studied in the literature. The E(TO3) mode was observed arising only below 180K in rhombohedral mode of BTO crystal [13]. However, the same mode was observed in the infrared spectra of BTO from all the phases [14, 15]. Later studies reveal that the observation of this mode depends on the excitation geometries utilized [8, 16,17]. The appearance of E (TO3) mode at 485 cm-1 in the low temperature Raman spectra can be attributed to a low temperature phase. With increasing temperature the material rearranges itself in the new phase and this mode looses its intensity and finally disappears. The disappearance of the E(TO3) mode can therefore be attributed to a phase change. Since this mode disappears at 390K in BTO Raman spectra, which is the orthorhombic to tetragonal transition temperature of BTO, the disappearance temperature has been termed as TO/T. Similarly, the disappearance of 306 cm-1 E(TO) mode represents the tetragonal to cubic transition temperature (TT/C), which was observed at 410, 420, 430, 440 K for x = 0.00, 0.03, 0.05 and 0.10 compositions, respectively. The variation of transition temperatures with La content are plotted in Fig.4. The observed increase in the transition temperature is also obvious from the XRD observations showing marginal increase in the tetragonal distortion with increasing La contents in the BLT compositions.


41

Fig. 4. Variation of orthorhombic to tetragonal (TO/T) and tetragonal to cubic (TT/C) transition temperatures with La contents.

The dielectric response of the film was analyzed in terms of the dielectric constant (ε) and losses (tanδ) as function of temperatures and composition. Fig. 5(a) depicts the variation of room temperature measured dielectric constant and loss as a function of composition at 100 kHz frequency. The substitution of La for Ba substantially increases the dielectric constant.

An increase of dielectric constant with La doping has been reported in BLT ceramics [18].

Fig. 5. Variation of 100kHz dielectric constant (ε) and losses (tanδ) with La content x (a), dielectric constant (ε) with temperature for all x values (b).

compared to the ceramics, our BLT films show higher dielectric constant and decreased dielectric losses.

The phase transition behavior of BLT thin films was also characterized from the temperature dependent dielectric constant in the range 77-500K. The temperature variations of the dielectric constant of BLT films measured with 100 KHz are shown in Fig. 5(b) for the compositions studied. The undoped BaTiO3 (x = 0.0) exhibits dielectric maximum ε’max of ~ 543 at about 410K and a small maximum associated with the rhombohedral to orthorhombic phase transition temperature, TR/O at 218K. The orthorhombic to tetragonal transition TO/T, appeared as a small hump about 310 K. The observed transitions temperatures for BaTiO3 are comparable to those reported in the literature [11,18]. On increasing the La contents in the films, the transition temperatures TR/O, TO/T and TT/C could not be isolated due to the broadness and diffuse nature of the dielectric response. Temperature dependent dielectric measurements on BLT ceramics have clearly indicated a systematic increase in the tetragonal to cubic transition temperature TT/C with increasing La contents[18]. Therefore, the broad and diffuse dielectric response in BLT films can be understood as a result of two-dimensional stress in these films, which increases with La contents. As a result, a very broad dielectric response was observed for x= 0.10 BLT film. Such dielectric behavior is common in ferroelectric relaxor materials [19-21]. However, no relaxor behavior (frequency dependent dielectric constant) was evidenced from BLT films.

Figure 6 shows the typical ferroelectric

polarization – electric field (P-E) hysteresis loops of the BLT films. The observations of hysteresis loop suggested ferroelectric behavior and tetragonal structure of BLT films. However, the film with x=0.10 composition exhibited a slim hysteresis curve compared to lower La doped composition films.


Fig. 6. Ferroelectric polarization – electric field (P-E) hysteresis loops of the BLT films at 15V.

The values of remnant polarization (2Pr) and saturation polarization (2Ps) obtained at an applied voltage of 15V are given in Table I. Compared to undoped BTO film, La doped (x=0.03) films exhibited much larger 2Pr and 2Ps values (7.2 and 54µC/cm2 respectively), which were found decreasing with further increase in La contents. The Coercive field (Ec) in La substituted films was found larger compared to BTO film. The decrease in Pr and increase in Ec is believed to be due to increasing grain sizes observed in these films using atomic force microscopy. Table 1. Polarization – electric field (P-E) hysteresis parameters obtained from BLT films.

La content

2Pr, (µC/cm2)

2Ps (µC/cm2)

Ec (kV/Cm)

x = 0.00 0.01 0.1 11.4

x = 0.03 8.9 68.2 23.7

x = 0.05 5.60 36.54 40

x = 0.10 5.32 41.42 28

3. Conclusions

It is shown that the La3+ substitution in BaTiO3 slightly increases the unit cell tetragonal distortion Accordingly, an increase in orthorhombic to tetragonal (TO/T) as well as tetragonal to cubic (TT/C) transition temperature was observed using temperature dependent Raman spectroscopy in the range 77-500 K. The ferroelectric transition temperature in the films was found slightly higher compared to the powders primarily due to the stresses in the films. The temperature dependent dielectric measurements in La doped BTO films showed broad and diffused maxima making the estimation of transition temperature difficult. The broad dielectric maxima however, exhibited merging behavior with increasing La contents in the films. The Tc obtained from the dielectric constant variation with temperature was found greater than the Curie temperature T0 indicating a first order phase transition in the compositions studied. From ferroelectric measurements of these films the remnant polarization for the La substituted films, which decreases with increasing La contents, was found to be higher for x=0.03. Acknowledgements This work was supported in parts by the NSF-INT-0097018 and NASA-NCC3-1034 grants. References [1] G.A. Smolenski, Ferroelectric and

related materials, Gordan & Breach Science Publishers, London, 1993.

[2] F.John and G. Shirane, erroelectric crystals, Pergamon, New York, 1962.

[3] L. E. Cross, Ferroelectrics 76 (1987) 24. [4] P.S. Dobal, A. Dixit, R. S. Katiyar, Z.

Yu, R. Guo and A. S. Bhalla, J. Appl. Phys. 89 (2001) 8085.

[5] G. A. Samara, J. Phys.:Condens. Matter 15 (2003) R367.

[6] F.M. Pontes, J. Appl. Phys. 91 (2002) 5972.

[7] F.D.Morrison. D.C.Sinclair and A.R.West, J. Am. Ceram. Soc. 84 (2001) 474.

[8] A. Scalabrin, A.S. Chaves, D.S. Shim and S.P.S. Porto, Phys. Stat. Solidi (b) 79 (1977) 731.

[9] A. S. Chaves, R. S. Katiyar and S. P. S. Porto, Phys. Rev. B 10 (1974) 3522.


43

[10] J. A. Sanjarjo, R. S. Katiyar and S. P. S. Porto, Phys. Rev. B 22 (1980) 2396.

[11] P.S. Dobal, A. Dixit, R.S. Katiyar and A.S. Bhalla, Proceedings of SPIE, Vol. 4333, Ed. Christopher S. Lynch (2001) 111.

[12] B. D. Begg, K. S. Finnie and E. R. Vance, J. Am. Ceram. Soc. 79 (1996) 2666.

[13] C.H. Perry and D. B. Hall, Phys. Rev. Lett. 15 (1965) 700.

[14] J.T. Last, Phys. Rev. 105 (1957) 1740. [15] Y. Luspint, J.L. Servoin and F. Gervais, J.

Phys. C: Solid St. Phys. 13 (1980) 3761. [16] M. D. Domenico Jr, S. H. Wemple, and S. P.

S. Porto, Phys. Rev. 174 (1968) 522.

[17] A. Pinczuk, W. Taylor and E. Burstein, Solid State Commun. 5 (1967) 429.

[18] M. Aparna, T. Bhimasankaram, G. Prasad, and G.S. Kumar, Bull. Mater. Sci. 24 (2001) 497.

[19] I.G. Siny, R.S. Katiyar and A.S. Bhalla, Ferroelectric Review, 2 (2000) 51.

[20] E. Buixaderas, M. Savinov, M. Kempa, S. Veljko, S. Kamba, J. Petzelt, R. Pankrath and S Kapphan, J. Phys.: Condens. Matter 17 (2005) 653.

[21] A. Dixit, S.B. Majumder, P.S. Dobal, R.S. Katiyar and A.S. Bhalla, Thin Solid Film 447 (2004) 284.

Synthesis of CdO Nanoparticles by Sol-Gel Technique

Rajeev Kumar†, R.K.Bedi† and Iqbal Singh* †Material Science Laboratory, Department of Physics, Guru Nanak Dev University, Amritsar-143005.

*PG Department of Physics, Khalsa College, Amritsar-143005 (Punjab), INDIA E-mail: [email protected]

Abstract

As an n- type semiconductor, CdO is of interest for low voltage and short wavelength electro-optical deices such as light emitting dioxides and diode laser. Cadmium oxide powder was prepared by sol gel technique using cadmium nitrate and citric acid as starting material. The phase evolution of CdO gel and calcined at 500˚C for 4 hours was investigated by XRD technique. Scanning electron micrograph of powder samples, indicate randomly distributed CdO nano-rods. CdO nanoparticles of average crystal size 33 nm were successfully obtained by this technique. It has been observed from their respective XRD patterns that crystallinity of the calcined powder is comparatively better than that of as prepared gel sample. A peak corresponding to Cd crystalline phase is noticed along with CdO peaks. This indicate that there is an amount of unoxidized Cd grains exists with oxide. 1. Introduction

The wide band gap semiconductors like ZnO, CdO, SnO2 and In2O3 have distinctive properties and now widely used as transparent conducting oxide and sensors[1]. As an n- type semiconductor CdO is of interest for low voltage and short wavelength electro-optical devices such as light emitting dioxides and diode laser [2]. It is widely used in field emission flat panel display with the development of various optical devices. CdO adopts the centro symmetric rock salt structure .It exhibits interesting electronic and optical properties that have been studied in detail [3]. Several techniques have been employed to prepare CdO nano particles including thermal evaporation[4], memberane template method [5], micro emulsion method [6], spray pyrolysis [7], sputtering[8], activated, sol gel method[9], mechanochemical method [10] and metal organic chemical vapour deposition[11]. However, there have been no reports on the preparation of CdO nanoparticle by sol-gel auto combustion method. The sol-gel combustion synthesis has evolved as a standard technique for oxide nanoparticle fabrication such as used in the formation of Al2O3-ZrO2-Nb composite, nanocrystalline MgAl2O4, Fe2O3 and (ZnO)mIn2O3 [12-16].

Combustion synthesis has been one of the methods most commonly used to obtain powders with compositional uniformity. Combustion synthesis processes are characterized by high temperature, fast heating rates and short reaction

times. All these features make the sol-gel combustion technique an important and attractive method for the manufacturing of oxide materials. In the present study a simple sol gel combustion route is used for the preparation of cadmium oxide nanostructures. The as prepared powder and CdO calcined at 500˚C in air for 4 hours have been characterized by XRD and SEM. 2. Experimental

Nanocrystalline CdO powder was prepared using citrate/nitrate smoldering combustion method. Cadmium Nitrate (Cd(NO3)2.4H2O), hydrated Citric acid (C6H8O7.7H2O) were used as starting materials. The solution of (0.5M) cadmium nitrate was made by dissolving nitrate salt in deionised distilled water with appropriate dosage of citric acid. The prepared solution is heated at 80ºC for 30 hrs and then it turns to yellowish sticky gel. The gel was taken on pre heated hot plate, a combustion process took place accompanying with the evolution of brown fume. Finally, a brownish fluffy product was obtained. It was then grinded and calcined at 500˚C for 4 hours in muffle furnace in free supply of air to effectively remove residual nitrate and organic material.

Synthesis of CdO Nanoparticles by Sol-Gel Technique 45

Fig. 1. shows the preparation procedure of CdO nanopowder using citrate nitrate gel combustion synthesis 3. Measurements

Phase identification of as prepared and calcined powder were performed via X-ray diffraction analysis using nickel filtered CuKα (wavelength-1.5405 A˚) radiations in the range of 2θ ( 20-70˚) by Philips diffractrometer.

Fig. 2. XRD pattern of (a) as synthesized powder, (b) powder calcined at 500˚C for 4 hours Diffraction analysis using nickel filtered CuKα (wavelength-1.5405 A˚) radiations in the range 4. Results and discussion

Fig. 2 shows the XRD pattern of (a) as prepared powder and (b) the powder after calcinations at 500˚C. Fig. 2(b) reveals that crystallinity of the calcined powder is comparatively much better than that of as prepared sample. A peak corresponding to (102) plane of the Cd crystalline phase is noticed along with CdO peaks obtained due diffraction from plane (111), (200), (311) and (322) . This reveals that there is an amount of unoxidized Cd grains exists with cadmium oxide in as prepared powder. It has been observed that a peak corresponding to (102) reflection of Cd disappears and a CdO peak which is due to reflection from (220) plane appears for the calcined sample as shown in (fig 2(b)). This confirms the formation of CdO powder. The intensity of different peaks associated with CdO increases with thermal treatment.

Half peak widths and grain diameter for the preferential orientations of samples were calculated using Scherrer relation given below D = 0.9λ / βcosθ (1)

Where β is the half peak width as radian of the peak which has maximum intensity, D is the grain diameter, θ is Bragg’s angle and λ is the wavelength of light used. The average crystallite size appears to slightly increase from 32-33 nm after the calcination of CdO powder which may be attributed to crystallization of nanoparticles. This will result decrease in surface to volume

(0.5M)Cd(NO3)2.3H2O + (0.36M)C(OH)(COOH). (CH2.COOH)2.H2O + Deionised water

Slowly heat to remove all the inorganics at 80°C

Put on hot plate preheated to 200°C

XRD,SEM characterization

Decomposed gel calcined at 500˚C

2(a)

2(b)


ratio which leads to decrease in defect states causing better crystallinity of the product.

The surface morphology of CdO nanoparticles is shown in fig. 3(a-b) In general these micrograph indicate randomly distributed polycrystalline microstructure in the form of rods. It has been found that CdO calcined at 500˚C show crystallites in some region tangled together with rods of varying diameters and length [10,17]. 5. Conclusion

Nanostructured CdO powder has been successfully synthesized using a citrate-nitrate combustion process followed by a short-time calcination.

Fig. 3. SEM micrgraph of (a) as synthesized powder, (b) powder calcined at 300˚C for 4 hours

The calcined powder shows CdO particles in the form rods. The citrate-nitrate process appears to be an attractive method for fabrication of cost-effective nanostructured materials for device fabrication.

References

[1] Gopel W. & K. D. Schierbaum, Sens. Actuators, B 26, (1995), 1.

[2] Gulino, F. Castelli, P. Dapporto, P. Rossi & I. Fragala, Chem. Mater, 14, (2002), 704.

[3] Dou Y., R. G. Egdell, D. S. L. Law, N. M. Harrison & B. G. Searle, J. Phys. Condens. Mater., 10, (1998), 8447.

[4] A. A. Dakhel & F. Z. Henari, Cryst. Reas. Technol., 30, (2003), 979.

[5] Yang Y. W., C. H. Liang, G. Z. Wang & T. Gao, J. Mater. Sci. Lett., 20, (2001), 1687.

[6] Dong W. & C. Zhu, Opt. Mater., 22, (2003), 227.

[7] C. H. Bhosale, A. V. Kambale, A.V kokate, K. Y Rajpure, Mater. Sci. & Engg. B, 122, (2005) 67.

[8] T. K. Subramanyam, G. M. Rao, S. Uthanna, Mater. Chem. & Phys., 69, (2001), 133.

[9] J. Santos-Cruz, G. Torres-Delgado, R. Castanedo- Perez, S. Jimenez-Sandoval, O. Jimenez-Sandoval, C.I. Zuniga-Romero, J. Marquez Marin & O. Zelaya- Angel, Thin Solid Films, 493, (2005), 83.

[10] H. Yang, G. Qiu, X. Zhang, A. Tang, W. Yang, J. nano. Res., 6, (2004), 539.

[11] Z. Zhao, D. L. Morel, C. S. Ferekides, Thin Solid Films, 413,(2002), 203.

[12] K. C. Patil, S.T.Aruna, S. Ekambaram, Solid State & Mater. Sci. 2 (1997) 158.

[13] Roberto T, Munir ZA,, J. Am. Ceram. Soc. 82(1999)1985.

[14] Bhaduri S, Bhaduri SB, Prisbery KA, J. Mater. Res. 14(1999)3571.

[15] V.A Hiremath, A. Venkataraman, Bull. Mater. Sci 26(2003) 391.

[16] S Kikkawa, J. Am Ceram Soc 88 (2005) 308.

[17] P.K. Ghosh, S.Das and K.K. Chattopadhyay, J. Nano. Res., 7, (2005), 219.

3(a)

3(b)

Deposition and Characterization of Nanocrystalline Thin Film of CdS from Chemical Route

Jasim M. Abbas, Charita Mehta, G.S.S. Saini and S.K. Tripathi

Department of Physics, Centre of Advanced Studies in Physics, Panjab University, Chandigarh 160 014. E-mail: [email protected], [email protected]

Abstract

We have synthesis cadmium sulfide nano-crystalline (n-CdS) powder using cadmium acetate and

thiourea, trisodium citrate, from chemical route at different pH. The formation of CdS has been confirmed with help of infrared (IR) spectroscopy by observing bands corresponding to the multi phonon absorption. We have also observed the IR bands at 1403 and 1560 cm-1 due to symmetric and asymmetric stretching of COO of the capped sodium citrate at CdS nano-crystalline particles. The Size of crystal is determined as 3-5 nm with the help of X-ray diffraction. We also deposited the thin film of CdS on glass substrate from the solution using self- aggregation approach. The film thickness is measured as 366 nm with profile meter. The optical band gap is calculated as 2.6-2.75 eV. Its value is found to depend on pH. We also measured dark and photoconductivity at different temperatures. 1. Introduction

Synthesis of semiconductor nano-particles

or quantum dots has been at the focus of intense research due to their enormous application for optical and optoelectronic devices. In addition, their study provides an opportunity to observe the evolution of collective behavior of the bulk from the discrete nature of molecular properties [1,2]. In particular, II-VI and III-V compound semiconductors have been extensively investigated due to the their strong quantum confinement effects resulting in significant variation in their electrical and optical properties with size, non linear optical properties, reactive and selective photo-catalysis and enhancement in fluorescence efficiency [3,5]. This has resulted in the development of many viable routes for their synthesis in liquids and at solid interfaces [6-8]. II-VI semiconductor materials are synthesized generally by wet chemical methods. Thin films now occupy a prominent place in basic research and solid state technology. The use of thin film semiconductors has attracted much interest in an expanding variety of applications in various electronic and optoelectronic devices due to their low production costs. CdS is one of the most promising II–VI materials because of its wide range of application in various optoelectronic [9-10], piezo-electronic [11,12] and semiconducting devices. Thin films of CdS are of considerable interest for their efficient use in the fabrication of solar cells [13].

Nano-crystalline thin films are also polycrystalline in nature but with sizes of crystallites of the order of a few nanometers.

Many properties of nano-crystalline materials are found to deviate from those of coarse grained polycrystalline materials with the same average chemical composition. These deviations result from dimensionality of nanometer sized grains and numerous interfaces between adjacent crystallites [14-16]. A major research goal of recent years is to understand the size dependent properties of nano-crystalline materials and as a result considerable amount of work on synthesis and characterization of nano-crystalline thin films are being carried out [17-18]. Out of several methods for deposition of nano-crystalline thin films, chemical bath deposition (CBD) technique is suitable for the deposition of film on large area substrate.

In the present work, we report the preparation of n-CdS particles and thin films on glass substrate by CBD method. These have been characterized by infrared (IR), UV-Vis, XRD spectroscopic techniques and electrical conductivity measurements. 2. Experimental

In order to prepare adherent and uniform films, we carefully clean the surface of the substrates used within the deposition experiments. We have synthesized n-CdS powder at different pH (pH=10,12) in the presence of capping agent (tri-sodium citrate). Aqueous solution of CdS has been prepared by dissolving a nominal amount of cadmium-acetate, capping agent (tri-sodium citrate) and thiourea for sulfide ion source in 50 ml deionized water and the resultant mixture has been stirred at


500 1000 1500 2000 2500 3000 3500 4000.0

0.2

0.4

0.6

0.8

1.0

155413

92

652

Tran

smitt

ance

-1

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16

Wavenumber/cm

0.0260

0.0265

0.0270

0.0275

0.0280

0.0285

0.0290

(βco

sθ)/λ

elevated temperature. The solid phase is isolated by filtration and finally dried in hot bath, from the residue solution. This solid phase consists of CdS nano-crystals. Thin films have been deposited on cleaned glass, quartz substrates for the optical, electrical and structural measurements. Crystallographic study has been carried out using a Phillips PW-1610 X-ray diffractometer with CuKα radiation in the 2θ range from 10 to 70. The IR spectrum is determined on a Perkin-Elmer PE-Rx 1 FTIR spectrophotometer. The spectral resolution of the IR spectrophotometer was 1 cm-1 throughout the experiment. To study the optical properties of n-CdS thin films, the transmission spectra are recorded using a double beam UV/VIS/NIR spectrometer [Hitachi-330] in the transmission range 300-1000 nm for all samples. The electrical measurements of these thin films were carried out in a specially designed metallic sample holder. Planar geometry of the films (length ~ 1.0 cm; electrode gap ~ 8×10-2 cm) is used for the electrical measurements. Thick In electrodes are used for the electrical contacts. Thickness of the film is measured as ~ 465 nm with a profile meter.The elictrical conductivity is noted manually from a digital picoammeter (DPM-11Model).The accuracy in current measurements is typically 1 pA.

(sinθ)/λ

CdS


Fig. 1 shows the XRD pattern of n-CdS film deposited on the glass substrates (pH = 12). The spectrum in Fig. 1 shows the diffraction peaks at 2θ values of 26.4, 43.7 and 51.5 on the film deposited at pH = 12. The highest intensity of reflection peak is at 2θ = 26.4 [111], with two another small intensity peaks at 2θ =

Fig. 1. XRD spectrum of n-CdS. 43.7 [220] and 51.5 [311] indicating that [111] is the preferred direction. The intensity of these peaks increases as the pH value decreases

(figures not shown here). Information of the strain and the particle size are obtained from the full width at half maximum (FWHMs) of the diffraction peaks. The FWHMs (β) can be expressed as a linear combination of the contributions from the strain (ε) and particle size (L) through the following relation [19]

λ

θελ

θβ sin1cos+=

L (1)

Fig. 2. shows the plot of (βcosθ)/λ versus (sin θ)/λ for n-CdS film at (pH = 12) which is a straight line. The reciprocal of intercept on the (βcosθ)/λ axis gives the average particle size as ~ 4 nm. The particle size increases (4.6 nm to 5.4 nm) as the pH value decreases from 10 to 12. Fig. 2. Plot between (βcosθ)/λ versus (sin θ)/λ.

Further, the formation of n-CdS has been confirmed with help of IR spectroscopy by observing a band at 652 cm-1 corresponding to the multi phonon absorption of CdS. We have also observed IR bands at 1403 and 1560 cm-1

due to symmetric and asymmetric stretching of COO¯ of the capped sodium citrate at n-CdS particles as shown in Fig. 3.

Fig. 3. IR spectrum of n-CdS.

Fig. 4 shows the temperature depen-dence of dark conductivity (σd) and photo-conductivity (σph) of n-CdS thin films deposited at pH 12 and 80o C. Similar behavior have been found for the

Deposition and Characterization of Nanocrystalline Thin Film of CdS from Chemical Route 49

2.6 2.7 2.8 2.9 3.0 3.1 3.2 3.3-22

-21

-20

-19

-18

-17

-16

-15

-14

B

A

A. Photo 0.59 eVB. Dark 0.80 eVCdS pH 12, 80o

Ln

films deposited at different pH and temperatures. The electrical conductivity shows typical Arrhenius type of activation

σ

1000/T (K-1 )

⎟⎠⎞

⎜⎝⎛ Δ−

=kT

Eod expσσ (2)

0.00E+0002.00E+0094.00E+0096.00E+009

00E+0091.00E+0101.20E+0101.40E+0101.60E+0101.80E+010

where ΔE is the activation energy for dc conduction; k is the Boltzmann’s constant. Fig. 4. Temperature dependence of dark conductivity of n-CdS thin films.

The values of σd and σph, calculated using Eq. (2), are (4.1 ± 0.02)×10-8 Ω-1cm-1 and (1.6 ± 0.02)×10-7 Ω-1cm-1 for n-CdS film respectively. The value of σd increases as the particle size of n-CdS increases at different pH. The increase in electrical conductivity and decrease in the activation energies as the value of pH decreases may be due to the change in structural parameters, improvement in crystallite and grain size, decrease in inter-crystallite boundaries (grain boundary domains) and removal of some impurities (adsorbed and absorbed gases).

Fig. 5 shows the plot between (αhν)2 vs. hν for n-CdS films deposited at different pH (9 & 12). Form the intercept of the hν-axis, we have obtained the values of optical gap (Eg) which are 2.75 and 2.88 eV respectively. The increase in the optical gap from the bulk value (2.42 eV) can be explained on the basis of the quantum confinement effect. 4. Conclusions

The n-CdS powder as well as thin films have been deposited at different pH values of the deposition bath solution. The particle size has been calculated using XRD data and found to be 3-5 nm. IR data confirms the formation of CdS nano-particles.

1.5 2.0 2.5 3.0 3.5 4.0

8.

αhυ

)2

(hυ)

CdS 800C 12 9

(

Fig. 5. Plot of (αhν)2 vs. hν for n-CdS films deposited at different pH (9 & 12). The optical band gap has been calculated using optical data and it is found that the band gap increases from the bulk value which is due to the quantum confinement effect. The particle size decreases as the value of pH increases. Electrical conductivity measurements show that the value of dark conductivity increases and activation energy decreases as the particle size increases.

Acknowledgement: This work is financially supported by CSIR (Major Research Project), New Delhi. References [1] C.B. Murray, D.J. Norris, M.G. Bawendi, J.

Am.Chem. Soc. 115 (1993) 8706. [2] K. Murakoshi, H. Hosokawa, M. Saitoh, Y.

Wada, T. Sakata, H. Mori, M. Satoh, S. Yanagida, J. Chem. Soc. Faraday Trans. 94 (1998) 579.

[3] O. Enea, A. Bard, J. Phys. Chem. 90 (1986) 301.

[4] P. Lianous, J.K. Thomas, Chem. Phys. Lett. 125 (1986) 299.

[5] U. Resch, A. Eychmuller, M. Hasse, H. Weller, Langmuir 8 (1992) 2215.

[6] Y. Iyechika, G. Wigner, D. Jager, A. Witt, C. Klingshirn SPIE Opt. Comput. 88 (1988) 103.

[7] S.S. Dhumure, C.D. Lokhande, Indian J. Pure Appl. Phys. 31 (1993) 512.

[8] J.-H. Lee, J.-S. Yi, K.-J. Yang, J.-H. Park, R.- D. Oh, Thin Solid Films 344 (2003) 431.

[9] J. de Klerk, E.F. Kelly, Appl. Phys. Lett. 5 (1964) 2.

[10] D. Patidar, R. Sharma, N. Jain, T.P. Sharma, N.S. Saxena, Bull. Mat. Sci. 29 (2006) 21.

[11] T.L. Chu, S.S.Chu, J. Britt, C. Ferekids, C. Wang, C.Q. Wu, H.S. Ullal, IEEE Electron. Dev. Lett. EDL 13 (1992) 303.


[12] J. Britt, C. Ferekids, Appl. Phys. Lett. 62 (1993) 2851.

[16] S.N. Behera, S.N. Sahu, K.K. Nanda, Indian J. Phys. A74 (2000) 81.

[13] H. Tong, Y.-J. Zhu, Nanotechnology 17 (2006) 845.

[17] S.N. Shau. J. Mater. Sci. Mater. Electron. 6 (1995) 43.

[14] P. Nandakumar et al, Bull. Mater. Sci. 20 (1997) 579.

[18] S.N. Sahu. S. Chandra. Solar Cell 22 (1987) 163.

[15] M.L. Breen, J.T. Woodward IV, D.K. Schwartz, A.W. Apblett, Chem. Mater. 10 (1998) 710.

[19] S.B.Qadri, E.F. Skelton, D. Hsu, A.D. Dinsmore, J. Yang, H.F. Gray, B.R. Ratna, Phys. Rev. B 60 (1999) 9191.

Dielectric Spectroscopy of Silver Nanoparticles Embedded in Soda Glass

Suman Bahniwal*, Annu Sharma*1, Sanjeev Aggarwal*, S. K. Deshpande# and P. S. Goyal#

*Department of Physics, Kurukshetra University, Kurukshetra – 136 119, India # UGC-DAE Consortium for Scientific Research, Mumbai Centre, Bhabha Atomic Research Centre,

Mumbai 400 085, India 1Email: [email protected]

Abstract

Soda glass metal nanocomposite having Ag nanoparticles have been synthesized by the combined use

of ion exchange method and subsequent thermal annealing. The frequency response of dielectric constant ( 'ε ) and dielectric loss (tan δ) has been studied both in pristine soda glass and Ag nano particles embedded soda samples in the frequency range 1000 kHz to 100MHz. It has been observed that dielectric constant increases and corresponding dielectric loss decreases in case of Ag nanoparticles embedded soda glass samples as compare to pristine samples. The possible correlation between the modification of dielectric parameters and the changes induced due to embedded silver nanoparticles in soda glass has been discussed.

1. Introduction

The physics of metal nanoparticles embedded in dielectrics has attracted considerable attention in recent years because of their novel properties like quantum size effects and their potential applications in non-linear optics [1-3]. The non linear response of the glassy matrix is increased with the introduction of small amount of metal and such composites have immense potential applications in optoelectronic and optical switching devices [4-5]. Coulomb blockade is one of most important effects due to nanoparticles, which gives us a new idea about designing new nanocomposite as an insulating material [6-8]. Normally metal particles always act as conducing fillers in insulating matrix. According to percolation theory, metal particles can form the conduction network to change composite in to conductor when the content of metal particles reaches the percolation threshold. When metal nanoparticles are dispersed uniformly in insulator at certain content, the composite can be confirmed the inverse tendency, with improved insulating properties by the theory of coulomb blockade. In the present the dielectric properties were measured to check the effects of metal nanoparticles on the soda glass. 2. Experimental

Silver exchanged soda glass slides were prepared by dipping soda glass slides in a molten

salt bath of AgNO3 and NaNO3 mixture at a temperature of 380oC for few minutes. [9]. Then these silver exchanged glass slides were washed with distilled water to remove AgNO3 adhering to the sample surfaces and were subsequently annealed at 500oC and 600oC for 1 hour. All these experiments were carried out in air.

The dielectric measurements were carried out over a frequency range of 105Hz to 108Hz using impedance Gain-Phase Analyzer (Model HP 4194A) available at UGC-DAE Consortium for Scientific Research, Mumbai Centre, BARC, Mumbai. All measurements were carried out at room temperature in rotary pump vacuum. 3. Results and Discussions

Dielectric properties are a complex function of permittivity, conductivity, size, shape spatial arrangement of the constituents (the filler and the matrix) and the testing frequency. Fig 1 depicts the variation of dielectric constant ε׳ with angular frequency ω for soda glass and Ag nano particles embedded soda glass samples annealed at different temperatures. Fig. 2 presents the variation of dielectric loss tangent versus angular frequency for the same samples. It has been observed that the dielectric constant increases for silver nanoparticles embedded soda glass samples as compared to pure soda glass samples. This can be explained on the basis of coulomb blockade theory [10], according to which, when metal nanoparticles of appropriate quantities are added to a glassy matrix and are dispersed


evenly with proper space distribution, many tunneling knots are formed, which prevent electrons from moving directionally in a certain

106 107 108

0.8

1.0

1.7

Die

lect

ric c

onst

ant (ε, )

Angular frequency (Hz)

PURERT500600

Fig. 1. Variation of dielectric constant ε׳ with angular frequency ω for soda glass and Ag nano particles embedded soda glass samples.

104 105 106 107

0.2

0.4

0.6

0.8

1.0

Die

lect

ric lo

ss


PURERT500600

Fig. 2. Variation of dielectric loss (tan δ) with angular frequency ω for soda glass and Ag nano particles embedded soda glass samples. electric field, as a result, resistance of the glassy matrix increases. This phenomenon is observed only when coulomb blockade condition e2/2c>>KβT is satisfied. In case of Ag- soda glass nanocomposite, Ag nanoparticles are dispersed in soda glass, they act as many

tunneling knot, which prevent the charge carrier from moving freely in a glassy matrix. As a result the insulating properties of glassy matrix are enhanced and this leads to an increase in dielectric constant and decrease in dielectric loss with increasing angular frequency. However at a frequency of ~108 Hz the dielectric constant as well as dielectric loss merges.

4. Conclusion

It has been found that the dielectric constant increases and dielectric loss decreases due to silver nanoparticle formation in soda glass by ion exchange method. This has been explained on the basis of coulomb blockade theory. References [1]. M. Mensnaoui, M. Maazaz, C. Parent, B.

Tanguy and G. Le Flem, Eur. J. Solid State Inorg. Chem., 29 (1992) 1001.

[2]. S. E. Paje, J. Llopis, M. A. Villegas and J. M. Fernandez Navarro, Appl. Phys. A: Mater. Sco. Process., 63 (1996) 431.

[3]. J. W. Haus, N. Kalyaniwala, R. F. Haglund, Jr. L. Yang, J. E. Witing, J. Appl. Phys., 65 (1989) 1420.

[4]. C. Flytzanis, F. Hache, M.C. Klein, D. Ricard, and P. H. Roussignol, Prog. Opt., 29 (1991) 323.

[5]. A. Tamura, K. Higeta, and T. Ichinokawa, J. Phys. C, 15 (1982) 4975.

[6]. M. Ferrari, L. M. Gratton, A. Maddalena, M. Montagan, and C. Tosella, J. Non-Cryst. Solids, 191 (1995) 101.

[7]. K.A.Mateev, L.I.Glazman, and H.U.Baranger, Surface Sciecne, 361-362 (1996) 623.

[8]. Ghozhong Cao, Nanostructures and Nanomaterials, (2004) Imperial college press.

[9]. Suman Bahniwal, Annu Sharma, Sanjeev Aggarwal and Shyam Kumar, Indian Journal of Physics, (In press).

[10]. Qunqiang Feng, Zhimin Dang, Na Li and Xiaolong Cao, Materials Science and Engineering B, 99 (2003) 325-328.

AC Conductivity in a-Ge-Se-Ag Glasses

Gurinder Singh, N. Goyal, G.S.S. Saini and S.K. Tripathi Centre of Advanced Study in Physics, Panjab University, Chandigarh-160014, India

E-mail: [email protected]; [email protected]

Abstract

The present paper reports the effect of Ag doping (low ~ 4 at. % and high ~ 10 at. %) on the ac conductivity of a-Ge20Se80 glass. Frequency dependent ac conductivity of the samples over a frequency range ~ 100 Hz to 50 kHz has been taken in the temperature range ~ 268 K to 358 K. σac is proportional to ωs for undoped and doped samples. The value of frequency exponent (s) decreases with the increase in temperature. It has been observed that as the concentration of Ag increases in the Ge20Se80 system, the conductivity of the system increases. These results have explained on the basis that when the concentration of Ag increases, the clusters Ag2X connect themselves or the amorphous structure becomes homogeneous and the sample (at higher doping concentration of Ag i.e. ~ 10 at. %) turns into ionic conductor. Figure 2 shows the temperature dependence of σac for doped and undoped samples at frequency 10 kHz.

1. Introduction

Chalcogenide glasses have been recognized as promising materials for IR optical elements and transfer of information. They exhibit unique infrared transmission and electrical properties, which makes them potentially useful for applications such as memory devices, holographic recording media and solar energy conversion [1-3]. Studies of the electronic nature of amorphous material give information about its electrical behaviour and this may be related to structural properties. The disorder in atomic configuration is responsible for the existence of localized electronic states within the material. Because the charge carriers are localized, ac technique is often employed to probe their behaviour.

Silver as an additive in chalcogenide glasses has attracted widespread interest in glass science [4]. The interest stems in part from light-induced effects relevant to optical recording and information processing [4]. It also stems from the drastically increased electrical conductivity [5] of Ag doped chalcogenide glasses. Although several studies have been undertaken on Ag-doped chalcogenide glasses, especially on Ge-Se-Ag systems [6-8]. GeSeAg glassy system share many structure and transport characteristic with the (network former)-(network modifier) family of glasses. In this glass, Ge is termed the network forming cation, Ag the mobile cation and Se the anion [6]. The structure of this glass has been investigated by many authors [9-11] and a short range order due to GeSe4/2 tetrahedra was reported. However, the correlation of the Ag is

motive of controversy. The glass homogeneity generates dispute too. Some works propose that (Ge0.25Se0.75)0.75Ag0.25 glasses are homogeneous bulk glasses [9, 10]. However in a recent work, ternary (GezSe1-z)1-yAgy bulk glasses in the Se-rich region (z < 1/3) are shown to be intrinsically phase separated into an Ag2Se-rich glass and a residual GetSe1-t (t > z at y ≠ 0) with Ag acting as network modifier [12]. They observed bimodal glass transition temperatures. In contrast, Ge-rich glasses (z > 5/2) were reported as homogeneous, wherein Ag acts as a network former, replacing available Ge sites of the backbone to be threefold coordinated to Se.

The present paper reports the effect of Ag doping (low ~ 4 at. % and high ~ 10 at. %) on the ac conductivity (σac) of a-Ge20Se80 glass. The temperature and frequency dependent σac of (Ge20Se80)100-xAgx (x = 0, 4 and 10) glassy alloys have been studied. The results reported in the present study provide information about the effect of Ag alloying in Ge20Se80 glass. The modified Correlated Barrier Hopping model (CBH) [13, 14] has been used to explain the effect of Ag doping. 2. Experimental

(Ge20Se80)100-xAgx (x = 0, 4 and 10 at %) glassy materials have been prepared by taking the constituent elements (M/s Alfa Aesar, U.S.A.; purity ~ 99.999 %) in required atomic weight percentages. The materials were mixed together and sealed in quartz ampoules (length ~10 cm, diameter ~1cm) under a pressure of ~ 1.33 x 10-5 mbar. These ampoules were mounted in a furnace for heating. In the beginning, the


temperature of the furnace was raised slowly and kept constant at the melting temperature of each constituent element (1234 K for Ag, 490 K for Se and 1210 K for Ge) for about two hours each. The temperature was then raised and maintained at about 50 K more than the highest melting point of the constituents (1284 K) for about 24 hours. The ampoules were immediately cooled in liquid nitrogen, for the materials to go into the glassy state. The glassy materials, in the form of ingots, were obtained by breaking the ampoules. To verify the amorphous nature of these glasses, X-ray diffraction (XRD) study on all samples was done (Philips X-Ray Generator, Model: PW1729 along with a PW1710 Diffractometer). No peaks have been observed for the materials prepared by above-mentioned technique, thereby, confirming the amorphous nature of the prepared materials.

Compressed pellets (diameter: 0.687 cm; thickness: 1~2 mm) were prepared by grinding the bulk-ingots into fine powder and compressing the resulting powder in a die under a hydraulic press (pressure ~ 106 kg/m2). While making compressed pellets, one must ensure that the powder is compressed to maximum compaction so that there were no voids in the sample. A three terminal sample holder has been fabricated for the measurement of σac of the pellet shaped samples. Provision was made for measuring the vacuum inside the sample holder. Cylindrical brass jacket was used as outer cover to provide excellent electromagnetic shielding. A built in micrometer was used for the measurement of sample thickness. Temperature of the sample was noted down with the help of a thermocouple which was kept close to the sample. A low temperature bath (Julabo F-70 VC/K) was used for controlling the temperature of the samples. Vacuum pumping system (Model: VS-65D, H.H.V India) was used to achieve a vacuum, up to ~ 1.33 x 10-5 mbar inside the sample holder. A General Radio Bridge (Model: 1615-A) was used for the measurements of frequency dependent ac conductivity. The bridge consists of an audio oscillator (Model: 1311), a tuned amplifier (Model: 1232-A) and a null detector, which permits balance to a resolution of one part in a million. 3. Results and discussion

Figures 1 shows the frequency dependence of measured conductivity (σac) for the undoped Ge20Se80 sample at different temperatures.

Similar results have been obtained for the doped samples (Ge20Se80)96Ag4 and (Ge20Se80)90Ag10. The conductivity data presented in these figures can be fitted with the relation σac = Aωs, where‘s’ is frequency exponent and A is a constant. The decrease in slope with increasing temperature indicates that the value of ‘s’ decreases with increasing temperature.

Figure 2 shows the temperature dependence of σac of Ge20Se80, (Ge20Se80)96Ag4 and (Ge20Se80)90Ag10 glasses at frequency 10 kHz. Similar results have been obtained at other frequencies. From the figure, it is clear that the conductivity increases with temperature more in case of pure sample. However, in case of lower and higher doping of Ag (x = 4 and 10 at. %), the increase in conductivity is small with temperature. However, a large increase in conductivity has been observed at higher Ag (x = 10 at. %) concentration at all temperatures. Figure 3 shows the temperature dependence of the frequency exponent‘s’ for all the three samples. It is clear from the figure that the value of‘s’ decreases with the increase in temperature and with the increase in Ag concentration also. This decrease is more in case of higher Ag (10 at. %) doping concentrations as compared to the lower Ag concentration. The decrease in s is

Fig. 1. Frequency dependence of conductivity at different temperatures in a-Ge20Se80 alloy.

2.5 3.0 3.5 4.0 4.5 5.0

-11.5

-11.0

-10.5

-10.0

-9.5

-9.0

-8.5

Log f(Hz)

a-(Ge20Se80)

Log σ ac

(Ω-1cm

-1)

358K 338K 318K 288K 268K

AC Conductivity in a-Ge-Se-Ag Glasses 55

more at higher temperatures in all Ag concentrations.

Using a modified Correlated Barrier Hopping Model (CBH) [13, 14], the above results can be explained as follows: The total conductivity is the combined mechanism of these processes. These processes are bipolaron hopping between D+ and D- centers, single polaron hopping between D0 and D- centers and D0 and D+ centers. Here, W is equal to WM, which is slightly less than the band gap for the bipolaron hopping. However, it is equal to W1 and W2 for the two types of single polaron hopping mechanisms, which are substantially less than WM for bipolaron hopping. The smaller values of W1 and W2 for single polaron hopping means that the value of Rw is much greater for single polaron hopping as compared to bipolaron hopping. The structure of Ag-Ge-Se glasses has been investigated by X- ray diffraction [15], neutron diffraction [10] neutron inelastic and Raman scattering [16] and differential anomalous X-ray scattering (DAS) [9]. The structure of the glass corresponding to a stoichiometric composition GeSe2 was thoroughly investigated and was concluded to be formed by tetrahedral GeSe4 connected by edge and corner sharing [17, 18] to construct quasi-two-dimensional distorted layers, which is in good agreement with the outrigger-raft structure model [19]. The structure of GeSe3 glass is formed of similar GeSe4 units as in GeSe2 glass and excess Se-Se connections. The excess selenium creates short chain or screw type of (-Se-)n and also causes the depolymerization of the layered network of GeSe4 tetrahedra. Unusual changes in ionic conductivity of silver in the present study suggest a structural change around silver ions in the glasses. A sharp transition from semiconductor to superionic conductor in (Ge20Se80)100-xAgx glasses has been detected by the author. This is evident from fig. 2 where sharp increase in conductivity for x = 10 at. % has been reported. A percolation transition at x = 10 is supposed in [20]. This phenomenon of percolation [20] could be explained for the systems of GeSeAg glasses. Armand et al. [21] suggested that glasses GeXAg (X = S, Se) containing small amounts of Ag2X, have an inhomogeneous structure composed of Ag2X clusters embedded in a GeX2 network, which becomes homogeneous at higher concentration of Ag2X. Therefore, when the Ag content is low, the clusters do not connect themselves and the sample does not conduct Ag+ ions. Then, when the Ag content increases to x = 10 at. %, the clusters Ag2X connect themselves

Fig. 2. emperature dependence of ac conductivity at 10 kHz.

260 280 300 320 340 3600.75

0.80

0.85

0.90

0.95

1.00

1.05 a- (Ge20Se80)100-xAgx

S

T(K)

x = 0 x = 4 x = 10

Fig. 3 . Variation of frequency exponent‘s’ with temperature.

2.8 3.0 3.2 3.4 3.6 3.8-13

-12

-11

-10

-9

-8

-7

-6

-5

-4a- (Ge

20Se

80)100-x

Agx

Log σ ac

(Ω-1cm

-1)

1000/T(K-1)

x = 0 x = 4 x = 10


or the amorphous structure become homogeneous and the sample turn into ionic conductor. Mirandou et al. [22] have reported ionic conduction in the (Ge25Se75)100-xAgx glasses at about x = 10. Borisova et al. [23] studied the glass forming ability of the (Ge0.25Se0.75)100-xAgx ternary system by water quenching. They found that in samples with composition Ge25Se75, the vitreous behaviour could be retained by introducing a maximum Ag content of 30 at. %. When the concentration of silver is increased, the total electrical conductivity steeply increases from 10-13 Ω-1cm-1 to 10-5 Ω-1cm-1 at about x ~ 10 at. %. Kawasaki et al. [20] studied the electrical properties of Agx(GeSe3)1-x (0 ≤ x ≤ 0.571) glasses and investigated the impedance spectra and EMF (Electric Motive Force) of these samples. They found that as the concentration of silver is increased the total electrical conductivity increases from 10-14 S/cm to 10-3 S/cm at about x = 0.3. This rapid increase is chiefly due to an appearance of silver ion migration. 4. Conclusions

Electrical conductivity measurements have been done in undoped Ge20Se80 and doped with Ag (low and high concentrations) metal. Frequency dependent ac conductivity (ac) measurements at different temperatures show that the value of ac increases at low concentration as well as at high concentration of Ag. The value of‘s’ decreases at all concentrations at high temperatures. It is observed that the addition of Ag increases the conductivity of GeSeAg glassy system. This is because when the Ag content increase, the clusters Ag2X connect themselves or the amorphous structure becomes homogeneous and the sample turns into ionic conductor. Acknowledgements This work is financially supported by the C.S.I.R. (major project), New Delhi. References

[1] L. Calvez, H.L. Ma, J. Lucas, X.H. Zhang, Adv. Mat. 19 (2007) 129. [2] L. Calvez, P. Lucas, M. Roze, H.L. Ma, J. Lucas, X.H. Zhang, J. Appl. Phys. A 89 (2007) 183.

[3] A. Nemcsics, S. Kovacs, Z. Labadi, K.F. Hesse, M. Czank, P. Turmezei, S. Motrya, Solar Energy Materials and Solar Cells 89 (2005) 175-183. [4] H. Fritzsche, Philos. Mag. B 68 (1993)

561. [5] E. Bychkov, V. Tsegelnik, Yu Vlasov, A. Pradel, M. Ribes, J. Non-Cryst. Solids 208 (1996) 1. [6] M.A. Urena, M. Fontana,B. Arcondo, M.T. Clavaguera –Mora, J. Non-Cryst. Solids

320 (2003) 151. [7] Y. Wang et al. J. Phys.: Condens. Matter.

15 (2003) S1573. [8] R.J. Dejus et al. J. Non-Cryst. Solids 143 (1992) 162. [9] J.D. Westwood, P. Gergopoulos, D.H. Whitmore, J. Non-Cryst. Solids 107

(1988) 88. [10] R. Dejus, S. Susman, K. Volin, D.

Montague, D. Price, J. Non-Cryst. Solids 143 (1992) 162. [11] A. Piarristeguy, M. Mirandou, M.

Fontana, B. Arcondo, J. Non-Cryst. Solids 273 (2000) 30.

[12] M. Mitkova, Y. Wang, P. Boolchand, Phys. Rev. Lett. 83 (1999) 3848.

[13] K. Shimakawa, Phil. Mag. B 46 (1982) 123.

[14] K. Shimakawa, Phil. Mag. B 48 (1983) 778.

[15] L.C. Bourne, S.C. Rowland, A. Bienenstock, J. Phys. Colloque C4 (1981) 951.

[16] R. Dejus, D.J. Lepoire, S. Susman, K.Volin, D.L. Price, Phy. Rev. B-Cond. Mat. 44 (1991) 11705.

[17] C. Peyroutou, S. Peytavin, M. Ribes, H. Dexpert, J. Solid State Chem. 82 (1989) 70.

[18] U. Walter, D.L. Price, S. Susman, K.J. Volin, Phys. Rev. B 37 (1988) 4232. [19] N. Kumagai, J. Shirafuji, Y. Inuishi, J.

Phys. Soc. Jpn. 42 (1977) 1261. [20] M. Kawasaki, J. Kawamura, Y.

Nakamura, M. Aniya, Solid State Ionics 123 (1999) 259.

[21] P. Armand, A. Ibanez, H. Dexpert, D. Bittencourt, D. Raoux, E. Philippot, J. Phys. IV Colloque C22 (1992) 2. [22] M. Mirandou, M. Fontana, B. Arcondo, J. Mat. Process. Tech. 143 (2003) 420. [23] Z.U. Borisova, T.S. Rykova, E.Yu. Turkina, A.R. Tabolin, Izv. Akad. Nauk. SSR Neorg. Mater. 20 (1984) 1796.

Photoconductive Measurements of Thermally Deposited a-Ge20Se80-xInx Thin Films

Ishu Sharma1 , Pankaj Sharma1 , P. B. Barman1 and S. K. Tripathi2

1. Department of Physics, Jaypee University of Information Technology, Waknaghat, Solan, H.P. 173215 (India) 2. Department of Physics, Panjab University, Chandigarh, India


Abstract

Steady state and transient photoconductivity has been measured on Ge20Se80-xInx (x = 0, 5, 15) vacuum evaporated thin films. Study of temperature dependent dark conductivity and photoconductivity measurements in the temperature range 303-375 K, shows that the conduction in this glass is through an activated process having single activation energy. The activation energy value of photoconduction is smaller in comparison to activation energy in dark. The photosensitivity shows a maximum value at 15 at. % of In concentration. This is attributed to the decrease in the density of defect states of Ge-Se alloy with increase of In content. The results of intensity dependent steady state photoconductivity follow a power law with intensity, where the value of power lies between 0.5 and 1.0, suggesting bimolecular recombination. Rise and decay of photocurrent for different concentration of Indium shows that photocurrent rises monotonically to the steady state value and the decay of photocurrent is also very fast. An attempt has been made to explain the results on the basis of defects and density of states.

1. Introduction

Chalcogenide glasses have been given a great deal of attention mainly because of their unusual electrical and optical properties as well as different properties of glass formation. They are excellent and unique for ultrafast all-optical devices and having low phonon energy, high photosensitivity, easy fabrication and processing, good chemical durability. These glasses provide the key functions for frequency conversion, all optical switching (AOS) and are used as all-optical processing devices. Chalcogenide glass fibers transmit in IR region. These glasses are used for fabricating active devices, such as infrared transmitting optical fiber, optical amplifiers, infrared lasers, blue laser diodes and efficient femto second switches [1-3]. Common feature of these glasses is the presence of localized states in the mobility gap due to the absence of long range order as well as inherent defects. Photocurrent measurements have been widely used for understanding the defect states in these glassy chalcogenides. The study of the photoconductivity (PC) provides an understanding of the photo generations and transport of the free carriers.

Ge-Se chalcogenide glasses are very good glass-former and are widely studied system for its optical and electrical properties. The Ge20Se80 glassy alloy lies at the threshold of mode change having an average coordination number <r> = 2.4 [4]. The energy band

gap and lattice perfections play a major role in the preparation of devices for a particular wavelength, which can be modified by the addition of dopants [3,5]. The addition of Indium (In) to Ge20Se80 system in an effective way, control it’s electrical and optical properties which leads the system towards intermediate region. Moreover, Ge-Se-In system is of special interest because of the fact that it forms glasses over a wide domain of compositions [6] and the incorporation of In to Ge-Se alloy expands the glass forming area and creates compositional and configurational disorder in the system.

The photoconductive properties of chalcogenide

glasses depend on the structural configuration of the system and addition of metallic additive creates the structural disorder. So, authors have decided to workout for the photoconductive properties of Ge20Se80-xInx system (x = 0, 5, 15) in detail which is very important from the basic physics and application point of view. Present paper reports the steady state and transient photoconductivity measurements on thermally evaporated Ge20Se80-xInx thin films. Temperature dependent (303 – 375 K) steady state photoconductivity and intensity dependent (3-1035 Lux) photocurrent is measured for all compositions. Rise and decay of photocurrent is measured at room temperature at 1035 Lux intensity for system.


2. Experimental Procedure

Glassy alloys of Ge20Se80-xInx (x = 0, 5, 15) have been prepared by the quenching technique as described elsewhere [7]. Thin films of the glassy alloy have been prepared by vacuum evaporation technique, keeping the substrates at room temperature and a base pressure ~ 2 ×10−5 mbar. Amorphous nature of the bulk samples and thin films was confirmed by X-ray diffraction technique as no sharp peak was observed. The compositions of evaporated samples are measured by an electron microprobe analyzer (JEOL 8600 MX) on different spots (size ~ 2 µm) on the samples. Pre-deposited thick indium electrodes on well-degassed Corning 7059 glass substrates have been used for the electrical contacts. A planar geometry of the film (length ~ 2.4 cm; electrode gap ~ 4×10−2 cm) is used for the electrical measurements. The thickness of the film is ~ 2000 Å. The films have been kept in the deposition chamber in dark for 24 h before mounting in the metallic sample holder to attain thermodynamic equilibrium. The photoconductivity of the amorphous film has been studied by mounting it in a specially designed metallic sample holder where heat filtered white light (200W tungsten lamp) can be shone through a transparent quartz window. A vacuum of about 10−3 mbar is maintained throughout these measurements. The light intensity is measured using a digital Luxmeter (Testron, model TES-1332). The photocurrent (Iph) is obtained after subtracting the dark current (Id) from the current measured in the presence of light. For measurements of transient photoconductivity, light is shone on the thin film and the rise in photocurrent is noted manually from a digital picometer (DPM-111 Model). The accuracy in Iph measurements is typically 1 pA. The films are annealed at 375 K for 2 h in a vacuum of about 10−3 mbar, and the dark conductivity and photoconductivity measurements are carried out.


The temperature dependent dc dark conductivity

for glassy thin films of a-Ge20Se80-xInx is shown in Fig. 1. The plots of lnσd vs. 1000/T are found to be straight indicating that conduction is through an activated process having single activation energy in the temperature range 303–375 K. In most of the chalcogenide glasses, can, therefore, be expressed by the Arrhenius relation.

Fig. 1. Temperature dependence of dark conductivity of Ge20Se80-xInx (x = 0, 5, 15) thin films. The values of activation energy ΔE and σd at 303 K are calculated from slope of the curves of Fig. 1 and are inserted in table 1. it is found that with the increase of In content to Ge-Se alloy both ΔE and σd increases.

The values of ΔEph and σph are calculated from the slope of the curves of Fig. 2. It is clear from the figure that the photoconductivity is an activated process which is explained on the basis of bimolecular recombination and the activation energy for photoconduction is much smaller than dark conduction in all samples.

Fig. 2. Temperature dependence of photo conductivity of Ge20Se80-xInx (x = 0, 5, 15) thin films.

Intensity (F) dependence of steady state photoconductivity has also been studied at room temperature (303 K) to investigate the nature of recombination processes in thin films. The plots of ln σph vs. lnF are straight lines for all the compositions and shown in Fig. 3,

2 . 6 2 . 7 2 . 8 2 . 9 3 . 0 3 . 1 3 . 2 3 . 3 3 . 4

- 2 5

- 2 4

- 2 3

- 2 2

- 2 1

- 2 0

- 1 9

- 1 8

- 1 7

- 1 6

- 1 5

lnσ d (Ω

-1

cm-1

)

1 0 0 0 / T ( K - 1 )

G e 2 0 S e 8 0 G e

2 0S e

7 5I n

5 G e 2 0 S e 6 5 I n 1 5

2 . 6 2 . 7 2 . 8 2 . 9 3 . 0 3 . 1 3 . 2 3 . 3 3 . 4

- 2 3

- 2 2

- 2 1

- 2 0

- 1 9

- 1 8

- 1 7

- 1 6

- 1 5

- 1 4 G e2 0

S e8 0

G e 2 0 S e 7 5 I n 5 G e 2 0 S e 6 5 I n 1 5

lnσ ph

(Ω-1

cm-1

)

1 0 0 0 / T ( K - 1 )

Photoconductive Measurements of Thermally Deposited a-Ge20Se80-xInx Thin Films 59

Fig. 3. Intensity dependence of photo conductivity of Ge20Se80-xInx (x = 0, 5, 15) thin films at 303 K.

which indicate that the photoconductivity follows

a power law with intensity i.e. σph α Fγ For γ = 0.5, the recombination is predominantly bimolecular in nature,

γ = 1.0, the recombination is monomolecular in nature. For 0.5 ≤ γ ≤ 1.0, according to Rose [8], the value of γ can not be interpreted by assuming a set of discrete trap levels, but considering the existence of continuous distribution of trap levels in the band gap. In a- Ge20Se80-xInx thin films also, the value of γ , are found to be close to 0.5 for all compositions. This square root dependence of photocurrent on intensities indicates the existence of bimolecular recombination on illumination of thin films with light in which recombination rate of electron is proportional to number of holes.

Photosensitivity (σph/σd) is an important parameter to describe the usefulness of a particular material from optoelectronic device point of view in the photoconductivity measurements. The values of σph/σd at 303 K and at intensity 1035 Lux are calculated and show a maximum value at 15 at. % of In concentration, (Fig.4) indicating that the Ge-Se-In system becomes more rigid.

Fig. 4 Variation of photosensitivity vs. In content (x at. %).

σph/σd depends on the lifetime time of excess charge carriers which in turn depends upon the density of localized states in a particular material. The higher the density of states, the lower will be the lifetime, as these defect states may act as recombination centers in the presence of light. This might be due to the formation of intimate valence alternation pairs (IVAPs) under illumination. If the density of defect states is more, then the IVAPs will also be more [9]. Therefore, σph/σd at a particular temperature and intensity, will be maximum when the density of states will be minimum. This result is further confirmed by the minimum value of nσ at this particular concentration. The charge carrier concentration (nσ) for different samples is calculated using the equation [10]

⎟⎟⎠

⎞⎜⎜⎝

⎛ Δ−⎟⎠⎞

⎜⎝⎛=

TkE

hTmk

nB

B σσ

πexp

2 2

23

2 (2)

The calculated values of nσ are listed in table 1. It is clear that the value of nσ decreases with the increase of In concentration which indicates that the number of localized states decreases.

Rise and decay of photocurrent have been measured for all compositions at room temperature and 1035 Lux intensity (results not shown here). It is clear from the results that the photocurrent raises monotonically to the steady state value. When the light is switched off, the photocurrent decreases rapidly approximately to the value of the saturated dark current indicating less number of defects states in the system.

Table 1. Dark conductivity (σd), the activation energy for dc conduction (ΔEd), power exponent (γ) and charge carrier density (nσ) for a-G20Se80-xInx (x = 0, 5, 15) thin films

X dσ

(Ω-1cm-1) (303K)

dEΔ (eV)

γ (303K)

σn (cm-1)

0 1.4 x 10-11

0.59 0.51 4.8 x1015

5 2.8 x 10-11 0.66 0.57 2.4 x1014

15 9.8 x 10-10 0.70 0.50 6.9 x1013

1 2 3 4 5 6 7- 2 6 . 0

- 2 5 . 5

- 2 5 . 0

- 2 4 . 5

- 2 4 . 0

- 2 3 . 5

- 2 3 . 0

- 2 2 . 5

- 2 2 . 0

- 2 1 . 5

- 2 1 . 0

- 2 0 . 5

- 2 0 . 0

- 1 9 . 5

- 1 9 . 0

- 1 8 . 5

- 1 8 . 0

- 1 7 . 5

lnσ ph

(Ω-1

cm-1

)

l n F ( L u x )

G e 2 0 S e 8 0 G e 2 0 S e 7 5 I n 5 G e 2 0 S e 6 5 I n 1 5

0 5 1 0 1 50

1 0

2 0

3 0

4 0

5 0

6 0

7 0

σ ph/σ

d

x a t . %


4. Conclusion

The study of dark and photoconductivity of Ge20Se80-xInx (x = 0, 5, 15 at.%) thin films as a function of temperature reveals that the conduction is through an activated process with a single activation energy in the temperature range 303 to 375 K. The intensity dependence of photocurrent indicates the existence of bimolecular recombination in Ge20Se80-xInx thin films in which recombination rate of electrons is proportional to the number of holes. Charge carrier concentration decrease with In addition. Dark activation energy increases with addition of indium.

Photosensitivity also shows maximum value at 15

at. % of In addition, the transient photoconductivity measurements for investigating thin films at room temperature and 1035 Lux intensity shows that the photocurrent decays very fast indicating less number of defects states in the system.

References [1] V. Trnovcona, I. Furar and D. Lezal, J. Non-Cryst.

Solids 353 (2007) 1311. [2] M. Drechsler, B. K. Meyer, D. M. Hofmann, P.

Ruppert and D. Hommel, Appl. Phys. Lett. 71 (1997) 1116.

[3] N. Goyal, A. Zolanvari and S. K Tripathi, J. Mater. Sci.: Mater. Electr. 12 (2001) 523.

[4] M. Micoulaut and J. C. Phillips. Phys. Rev. B 67 (2003) 104204.

[5] E. Marquez, T. Wagner, J. M. Gonzalez-Leal, A. M. Bernal-Olive, R .Prieto-Aleon, R. Jimenez-Garay and P. J. S. Ewen, J. Non-Cryst. Solids 274 (2000) 62.

[6] Z. U. Borisova, Glassy Semiconductors, Plenum Press, New York, 1981.

[7] Ishu, S. K. Tripathi and P. B. Barman, J. Phys. D 40 (2007) 4460.

[8] A. Rose, Concepts in Photoconductivity and Allied Problems, Interscience, New York, 1963.

[9] K. Shimakawa, J. Non-Cryst. Solids 77 (1985) 1253.

[10] V. Sharma, A. Thakur, N. Goyal, G. S. S Saini and S.K. Tripathi, Semicond. Sci. Technol. 20 (2005) 103.

Synthesis of MgB2 Superconductor with Different Method

Kiran Singha, Rajneesh Mohana, N. K. Gaura and R. K. Singhb

a Department of Physics, Barkatullah University, Bhopal- 462026 (M.P.), India b Institute of Professional and Scientific Studies and Research, Chaudhary Devilal University, Sirsa -125055.


Abstract The polycrystalline samples of MgB2 superconductor have been synthesized by using in-situ solid state reaction method in two different ways: by using the Tantalum foil and stainless steel sheet. The prepared samples were characterized for phase purity and superconducting transition temperature (Tc) by using the X-ray diffraction and four-probe resistivity measurement setup, respectively. All the samples are crystalline into AlB2 hexagonal structure with space group p6/mmm. The resistivity measurement indicates that all the samples have the same Tc.

1. Introduction The recent discovery of superconductivity in MgB2 [1] with higher Tc 39 K has initiated much activity in experimental as well as theoretical studies. The study of MgB2 is quite interesting due to the various advantages over low temperature and high temperature superconductors. MgB2 has wide range of operating temperature between 20K – 30K. The raw materials used for the preparation of MgB2 superconductor is very cheap, easy to prepare, stable and can be cooled using electrical refrigerator, rather than messy cryocans, which is less complex and less expensive than those of operating at 4K. MgB2 has been prepared in various forms like, bulk, single crystals, thin films, wires and tapes. In MgB2 the critical current density (Jc) is mostly determined by flux pinning, rather than by the connections at grain boundaries as in high temperature superconductors (HTSC). Unlike HTSC, MgB2 has lower anisotropy, larger coherence length and transparency at the grain boundaries to the current flow. Several theoretical studies have revealed that MgB2 can be treated as a phonon-mediated superconductor with strong coupling [2, 3]. However, experimental techniques, like photoemission spectroscopy [4], tunneling spectroscopy [5], isotope effect measurement [6, 7] and inelastic neutron scattering measurement [8] have shown that MgB2 is a typical BCS superconductor with a strong electron-phonon (e-ph) interaction.

The enormous efforts have been directed towards improving the Jc by using different methods, such as chemical doping [9-14],

irradiation with heavy ions [15] and different methods of preparation [16,17], but doping is one of the best method to increase the Jc. The Jc of pure MgB2 is around 104A/cm2 at 20 K [18]. Generally, sealed quartz ampoules or vacuum furnace are used for the preparation of MgB2 because of the high reactivity of Mg with oxygen to form insulating MgO phase. Earlier we have reported the optimized parameters for the preparation of Cu doping [19].

In the present paper, we are reporting the preparation of MgB2 superconductor through in-situ solid state reaction method in two different ways (using Ta foil and stainless steel sheet). The Ta foil is very costly, so we tried to synthesis MgB2 by using SS sheet instead of Ta foil. The results related to phase purity and Tc are presented in the following sections. 2. Experimental

The polycrystalline samples of MgB2 have been synthesized by using the solid-state reaction method. The stoichiometric amount of Mg and B powders were thoroughly ground in an agate mortar for 2 hrs and compacted into the circular pellets of 12 mm diameter. The three pellets were wrapped in Tantalum foil (abbreviated A) and other three were put in a stainless steel sheet boat (self made) (abbreviated B), and heated in a quartz tube at 850°C for 2 hrs in the presence of high purity argon flow followed by the furnace cooling. These samples were characterized by X-ray diffractometer (Cu Kα radiation) in the range 20° ≤ 2θ ≤ 80° for identifying the phase formation. The low temperature dependence of the electrical resistivity was performed by the standard four-probe method from


20 K to 300 K by closed cycle refrigeration equipment (CTI Cryogenics). Scanning Electron Microscopy is used to study the surface morphology of the samples.


The XRD patterns of both A and B samples are shown in fig. 1. All the peaks are associated with the MgB2 phase (P6/mmm space group). It can be observed from fig. 1 that both A and B samples are monophasic with small amount of MgO at 2θ = 62.36, this peak was observed in various reports [9, 19-21]. It is evident from the XRD patterns (B) that there is an additional impurity peak of MgO around 2θ = 44.54. This impurity peak could be removed by taking some more precautions. The lattice parameters were calculated by least square method. The lattice parameters are in good agreement with those reported earlier [1], there is no difference in the lattice parameters for both (A) and (B) samples

The variation of normalized electrical resistivity in the temperature range 30 K to 300 K of A and B samples are shown in fig. 2. It can be seen that both the samples show the metal to superconducting transition. The normal state resistivity of sample (B) is slightly higher than that of (A). This might be due to the additional MgO phase observed in

sample (B). The residual resistivity ratio of both the samples remains the same i.e. ~3. It can be clearly seen from fig. 2 that there is no change in the Tc of both the samples till Tc = 38.50 K. Also the transition in both cases is very sharp (≤1K), which also supports the purity of the samples. The details of the lattice parameters and transition temperature are given in Table I. The SEM image of MgB2 prepared by using Ta foil is shown in fig. 3. It is clear from fig. 3 that the average grain size is around 1μm.

4. Conclusion We have synthesized MgB2 superconductor by using the Ta foil and SS sheet. Our results on the XRD and R-T measurements show that both the samples (A) and (B) has the same lattice parameters and Tc except an additional impurity peak of MgO observed in sample (B) which can be removed by taking precautions. Here, it is to be noted that the SS sheet is very economic as compared to Ta foil. Thus, one can prepare the MgB2 with low cost by using the SS sheet.

Lattice parameters Sample ‘a’ (Å) ‘c’(Å) ‘V’(Å3)

Tc (K)

ΔTc (K)

A 3..0837 3.523 29.012 38.50 1 B 3.0819 3.524 28.987 38.48 1

Fig. 1. XRD patterns of MgB2 superconductor using Ta foil and stainless steel sheet

(A)

(B)

Fig. 2. The variation of normalized resistivity with temperature for MgB2 prepared by using Ta foil and stainless steel sheet

Table 1. Lattice parameters and Tc of A and B samples

Synthesis of MgB2 Superconductor with Different Method 63

Acknowledgement The authors are thankful to University Grants Commission, New Delhi for providing the financial support to do this work. One of us, Kiran Singh is thankful to Council of Scientific and Industrial Research, New Delhi for providing the financial support in the form of Senior Research Fellowship (SRF). References [1] J. Nagamatsu, N. Nakagawa, T.

Muranaka, Y. Zenitani, J Akimistu, Nature 410 (2001) 63.

[2] J. Kortus, I. I Mazin, K. D. Belashchenko, V. P. Antropov, L. L. Boyer, Phys. Rev. Lett. 86 (2001) 4656.

[3] J. M. An, W. E. Pickett, Phys. Rev. Lett. 86 (2001) 4366.

[4] T. Takahashi, T. Sato, S. Souma, T. Muranaka, J.Akimitsu, Phys.Rev. Lett. 86 (2001) 4915.

[5] G. Rubio-Bollinger, H. Suderow, S. Vieira, Phys

Rev. Lett. 86 (2001) 5582. [6] S. L. Bud’ko, G. Lapertot, C. Petrovic, C. E. Cunningham, N. Anderson, P. C. Canfield, Phys. Rev. Lett. 86 (2001) 1877. [7] D. G. Hinks, H. Claus, J. D. Jorgensen, Nature 411 (2001) 457.

[8] R. Osborn, E. A. Goremychkin, A. I.

Kolesnikov, D. G. Hinks, Phys. Rev. Lett. 87 (2001) 017005. [9] V. P. S. Awana, M. Isobe. K. P. Singh, E.

T. Muromachi and H. Kishan, Supercond. Sci. Technol. 19 (2006) 551.

[10] S. X. Dou, S. Soltanian, J. Horvat, X. L. Wang, S. H. Zhou, M. Ionescu, H. K. Liu, P. Munroe, M. Tomsic, Appl. Phys. Lett. 81 (2002) 3419.

[11] J. Wang, Y. Bugoslavsky, A. Berenov, L. Cowey, A. D. Caplin, L. F. Cohen, J. L. MacManus-Driscoll, L. D. Cooley, X. Song, D. C. Larbalestier, Appl. Phys. Lett. 81 (2002) 2026.

[12] H. Kumakura, H. Kitauchi, A. Matsumoto, H. Hatakeyama, Appl. Phys. Lett. 84 (2004) 3669.

[13] M. D. Sumption, M. Bhatia, M. Rindfleisch, M. Tomsic, S. Soltanian, S. X. Dou, E. W. Collings, Appl. Phys. Lett. 86 (2005) 092507.

[14] Y. Ma, H. Kumakura, A. Matsumoto, H. Hatakeyama, K. Togano, Supercond. Sci. Technol. 16 (2003) 852.

[15] Y. Bugoslavsky, L. F. Cohen, G. K. Perkins, M. Polichetti, T. J. Tate, R. Gwilliam, A. D. Caplin, Nature 410 (2001) 561.

[16] A. Serquis, L. Civale, D. L. Hamaman, X. Z. Liao, J. Y. Coulter, Y. T. Zhu, M. Jaime, D. E. Peterson, F. M. Muller, Appl. Phys. Lett. 82 (2003) 2847.

[17] Y. Ma, A. Xu, X. Li, X. Zhang, S. Awaji, K. Watanabe Japan. J. Appl. Phys. 45 (2006) L493.

[18] P. C. Canfield, S. L. Budko, D. K. Finemore, Physica C 385 (2003) 1.

[19] K. Singh, R. Mohan, N. Kaur, N. K. Gaur, M. Dixit, V. Shelke, R. K. Singh, Physica C 450

(2006) 124. [20] R. G. Abhilash Kumar, K. Vinod, R.P.

Aloysius, U. Syamaprasad, Mater. Lett. 60, 3328 (2006).

[21] D. Wang, Y. Ma, Z. Yu, Z. Gao, X. Zhang, K. Watanabe, E. Mossang, Supercond. Sci. Technol. 20, 574 (2007).

[22] W. Du, H. Xu, H. Zhang, D. Xu, X. Wang, X. Hou, Y. Wu, F. Jiang, L. Qin, J. Crystal

Growth 289, 626 (2006).

Fig. 3. SEM image of MgB2 superconductor using Ta foil

Synthesis and Characterization of Polyaniline Thin Films

Atul Kapil, Inderjeet Kaur and Subhash Chand Department of Applied Sciences, National Institute of Technology, (Deemed University)

Hamirpur–177005 (HP), India E-mail: [email protected]

Abstract

Among the organic conducting polymers, Polyanilines (PANI) are the most important class of conjugated polymers for the possible use in commercial applications. Polyaniline systems were prepared by the oxidative polymerization of aniline with ammonium peroxydisulfate in the presence of acids. Both organic and inorganic acids have been included. The hydrochloric acid (HCl) and p-toluenesulfonic acid (p-TSA) were used as dopants during polymerization. One Polyaniline system with partial doping of HCl and p-TSA was synthesized. The syntheses of PANI by the in-situ doping with HCl resulted in a better conductivity. Prepared polymer was characterized by SEM, and TGA scan analysis. Electrical conductivity of these PANI systems at room temperature in the form of films and pellets was measured with four-probe method and was found in good agreement with reported ones. 1. Introduction Polyaniline (PANI) is one of the important classes of intrinsically conducting polymers. It has received increasing attention recently due to its attractive physical, chemical [1,2] and material properties [3], high conductivity, flexibility of synthesis. The wide range of electrical, electrochemical and optical properties of polyaniline along with its excellent stability makes it a useful electronic material for various applications. Some of the potential devices based on polyaniline are organic light emitting diodes [4], gas sensors, photovoltaic cells [5] and schottky devices [6,7]. It has been demonstrated that the Polyaniline (PANI) may be prepared in three different distinct oxidation states of the polymer as fully reduced lucoemeraldine base, the half oxidized emeraldine base and the fully oxidized pernigraniline base. Among these, emeraldine base is the most stable form of PANI in the lab atmosphere [8]. Depending upon the mode of preparation, polyaniline can be classified into salt and base forms. Polyaniline has good stability in air, even at elevated temperature [9]. The emeraldine salt form, which is partially protonated, can be synthesized by electrochemical or chemical oxidative polymerization of aniline in aqueous acidic media by a variety of oxidizing agents [10,11]. The oxidizing agent traditionally employed in the polymerization of aniline has been ammonium peroxydisulfate, i.e., (NH4)2S2O8 in aqueous HCl [12,13]. Common inorganic acids

and organic sulfonic acids are strong enough to produce favourable conditions for the polymerization of aniline. On the other hand, most carboxylic acids are not strong enough to provide sufficient acidity for efficient polymerization of aniline and, consequently, the conductivity of PANI prepared in their presence is reduced. Yet the preparation of PANI in the presence of various acids is of great interest, because other important properties, like the processability in general and the solubility in particular, are affected by the nature of the acid. The present paper deals with the preparation as well as characterization of polyaniline doped with HCl and p- toluenesulfonic acid. 2. Experimental 2. 1. Materials Aniline, ammonium persulfate (APS), p-toluene sulfonic acid (TSA), N,N-Dimethylformamide (DMF), 1-Methy-2-pyrrolidone (NMP) and acetone were analytical grade purchased from Alfa Aesar. Aniline and all other reagents were used as received.

2. 2. Preparations

PANI was synthesized by oxidative polymerization of aniline in hydrochloric acid by APS. In a typical procedure, 1.5 ml aniline was dissolved in 150 ml of hydrochloric acid and brought to the desired temperature (100C) with

Synthesis and Characterization of Polyaniline Thin Films 65

magnetic stirring. Afterwards, 50 ml of hydrochloric acid solution containing 289.5 g APS maintained at the same temperature was added drop wise to start the polymerization, and the reaction mixture was allowed to continue for 5hr after completion of the addition. The product was filtered and the precipitate cake was washed exhaustively with deionized water and ethanol. The cake was further treated with 5.0wt% ammonia water with stirring for 1.0 hr and then washed to neutral with de-ionized water. The dedoped product was then dried in vacuum (500 C) for 48 hr and then powdered in an agate mortar/pestle to obtain the EB form of PANI powder. The EB powders were mixed with p- TSA solution with stirring for 20 h and afterwards dried in vacuum (500C) to prepare p- TSA doped PANI (PANI-TSA). The solution of synthesized PANI in different solvents like N,N-Dimethylformamide (DMF) and 1-Methy-2-pyrrolidone (NMP) was stirred (1 hr) and spin cast on various substrates at 1500 rpm for 5 min (Apex SCU 05 spinner) or drop cast into each micro-well area of the substrates. The freshly cast films were dried at 500C for 24 hr under vacuum followed by cooling to room temperature for at least 3 hr. 2. 3. Measurements

Thermal stability measurements were made by TGA analysis. The observations of the sample morphology were carried out using a SEM scanning electron microscopy. Conductivity measurements were made on the films as well as on the compressed pellets of the powder using a conventional four-probe technique at 25 °C. About 100 mg of PANI powder was compressed into a pellet of 10 mm in diameter and 1 mm thick with a manual press. All the contacts for the conductivity measurement were made with silver paste. 3. Results and discussion The polymerization of aniline is a exothermic oxidation. A detailed discussion of the mechanism of PANI formation [14] has been reported earlier. An induction period is typical for the oxidation of aniline under strongly acidic condition. During this stage, the temperature of reaction mixture practically does not change. The colourless mixture becomes blue as oligomeric cation radicals are produced [15]. Temperature rises and a macroscopic polyaniline (PANI) precipitate is produced at this stage. The mixture becomes deep blue due to the presence of the

protonated pernigraniline [3,14]. The temperature reaches its maximum, and the mixture changes its colour as the blue pernigraniline is converted to the green protonated emeraldine in the final stage of polymerization [14]. Although the polymerization of aniline took less than 5 hours using APS as an oxidant, the conductivity of polyaniline had reached 0.74 S cm−1 at 25 °C (Fig. 3), which was close to that of reported ones [10].

0

0.5

1

1.5

2

2.5

0 0.2 0.4 0.6 0.8 1

Voltage(mV)

Cur

rent

(mA

)

Fig. 1. Current-voltage characteristics of PANI (HCl+TSA) with four probe method.

0 5 10 15

Voltage(mV)

Cur

rent

(mA

)

Fig. 2. Current-voltage characteristics of PANI (p-TSA) with four probe method.

0

0.5

1

1.5

2

2.5

0 0.2 0.4 0.6 0.8

Voltage(mV)

Cur

rent

(mA

)

Fig. 3. Current-voltage characteristics of PANI (HCl) with four probe method.

Polyaniline doped with p-toluenesulfonic

acid showed less conductivity as compare to the polyaniline doped with HCl due to the lower acidity of former with respect to later (Fig. 2) [10]. Partially doped polyaniline showed a comparable conductivity (Fig. 1).


Fig. 4. Scanning electron micrograph of the PANI doped with HCl. SEM of conducting PANI synthesized by chemical oxidative method is shown in figure 2. SEM taken on the powdered samples of the HCl doped PANI shows individual three-dimensional lumps [16]. Conducting polymers are very sensitive to the temperature. Due to the interaction between electron and sample, considerable amount of heat is generated which causes the development of small cracks during SEM recording. The contrast in the image is a result of differences in the scattering from different surface areas due to geometrical differences [17].

Fig. 5. TGA curves of PANI doped with HCl. Thermal degradation studies were performed using Shimadzu DT-30 at a linear heating rate of 100C/ min from room temperature to 10000C. The thermo gravimetric analysis (TGA) of the sample as shown in Fig. 1 indicates that the mass loss starts at 500C and continues up to 4500C and in the next stage the mass loss occurs rapidly upto the temperature 5500C. The initial mass loss may be due to the loss of water molecules and for the next stage due to oxidative degradation of polymer in air [18]. 4. Conclusion Polyaniline systems doped with HCl and p-Toluenesulfonic acid have been fabricated. They

appear to be particularly attractive with good conductivity. The polyaniline doped with HCl was shown to have the highest conductivity. Future study will be dedicated on evaluating other polymerization methods and additional protonic acids to improve the conductivity and solubility of polyaniline. References [1] J. Stejskal, R.G. Gilbert, Pure Appl. Chem.

74 (2002) 857. [2] A.G. MacDiarmid, Synth. Met. 84 (1997)

27. [3] A.G. MacDiarmid, Synth. Met. 125 (2002)

11. [4] H.L. Wang, A.G. MacDiarmid, Y.Z. Wang,

D.D. Gebler, A.J. Epstein, synth. Met. 78 (1996) 33.

[5] L. Ding, M. Jonforsen, L.S. Roman, M.R. Anderson, O. Inganas, Synthe. Met. 110 (2000) 133.

[6] S.S. Pandey, M.K. Ram, V.K. Srivastava, B.D. Malhotra, J. Appl. Polym. Sci. 65 (1997) 2745.

[7] H.K. Chandhari, D.S. Kelkar, J. Appl. Polym. Sci. 61 (1996) 561.

[8] Miroslava Trchova, P. Matejka, J. Brodinova, A. Kalendova, J. Prokes, J. Stejskal, Polym. Degrad. Stab. 91 (2006) 114.

[9] J. Prokes, J. Stejskal, Polym. Degrad. Stab. 86 (2004) 187.

[10] Xiaoxuan Li, Xingwei Li, Materials Letters 61 (2007) 2011.

[11] M. Trchova, I. Sedenkova, E. Tobolkova J. Stejskal, Polym. Degrad. Stab. 86 (2004) 179.

[12] I. Sedenkova, M. Trchova, Natalia V. Blinova, J. Stejskal, Thin Solid Films 515 (2006) 1640.

[13] P. Wang, K.L. Tan, E.T. Kang, K.G. Neoh, Appl. Surf. Sci. 193 (2002) 36.

[14] J.Stezskal, P. Kratochvil, A.D. Jenkins, Polymer 37 (1996) 367.

[15] J.Stezskal, P. Kratochvil, A.D. Jenkins, collect. Czech. Chem. Commun. 60 (1995) 1747.

[16] R. Singh, V. Arora, R.P. Tandon, S. Chandra, N. Kumar, A. Mansingh, Polymer 38 (1997) 4897.

[17] D. Kumar, R. Chandra, Ind. J. Engg. Mater. Sci. 8 (2001) 209.

[18]M. Ghosh, A.K. Meikap, S.K. Chattopadhyay, S. Chatterjee, J. Phys. Chem. Solids 62 (2001) 475.

Fabrication and Characterization of Indium Tin Oxide-PTCDA-Aluminium/Silver Junctions

Manish Taunk and Subhash Chand

Department of Applied Sciences, National Institute of Technology Hamirpur-177 005 (H.P.), INDIA. E-mail: [email protected], [email protected]

Abstract

Thin films of organic semiconductor 3,4,9,10-Perylene Tetra-Carboxylic Di-Anhydride (PTCDA) are

fabricated in sandwiched configuration between indium tin oxide (ITO) and metals (Al & Ag) and their current voltage characterization is investigated for molecular electronics. Band gap of PTCDA thin film is also determined from transmission spectrum. Electrical properties of metal/organic interfaces are largely dominated by chemistry. Metal-molecule chemical reactions are responsible for the creation of interface gap states. The possible reason for Ohmic nature of reactive metal Al contacts is attributed to carrier conduction through a high density of chemistry induced interface gap states, whereas the partial blocking nature of noble metal Ag contacts is due to the unreactive and abrupt interface with thin films of PTCDA. 1. Introduction

The field of organic electronics is continuously growing and in the past few decades it becomes the thrust area of current research. Investigations of material properties, device structures and characteristics show that a new class of electronic and photonic devices can be successively prepared from these complex structured organic molecular compounds. The low synthesis costs and relative easiness of handling makes this new class of materials attractive for the wide range of applications. Furthermore, the chemical compatibility of organic materials with plastics allows the low cost fabrication of flexible, unbreakable, and transparent displays [1]. However, the low mobility of organic semiconductors precludes the complete replacement of compound inorganic materials especially for high frequency applications. On the other hand, organic semiconductors can be used in hybrid organic/inorganic devices to tailor the electronic transport properties and performances of conventional devices [2-5].

Experimentally these organic molecular materials are characterized in two ways: first by the width of their forbidden gap which is of the order of (1.0-3.5)eV and are classified as semiconductors. Secondly due to their parallel growth and their planer spaces are very near which results in extensive overlapping of p-orbital. This overlapping result in delocalization of electron and high intrinsic conductivity is observed in these materials [6]. 3,4,9,10-Perylene Tetra-Carboxylic Di-Anhydride

(PTCDA) i.e. C24O6H8 is a crystalline molecular organic semiconductor and is of particular interest due to its excellent properties and electronic potential for the making of optoelectronic devices. It is truly a remarkable electronic material also because of its crystal structure. In thin films of PTCDA, molecules form monoclinic structure with two molecules per unit cell, (102) crystallographic planes parallel to the substrate surface with an intermolecular packing distance of 3.21Å [7, 8].

Fig. 1. Chemical structure of PTCDA molecule showing perylene core and anhydride end groups.

Novel devices based on organic

semiconductors such as Organic Light Emitting Diodes (LEDs) or Organic Field Effect Transistors (OFETs) require metallic contacts. These metallic contacts on molecular organic thin films play an important role of charge injection electrodes. Therefore, the interface between metals and organic semiconductors deserves special attention. Low work function metals are used for electron injection and metals with higher work function are used for hole injection with reference to the semiconductor work function.


It was observed that PTCDA self organizes spontaneously over large areas on amorphous substrates such as glass. This material is used by various researchers as a passivating layer on reactive semiconductor surfaces and also as a modifier of diode parameters in case of metal-semiconductor or Schottky contacts [9-12]. Earlier PTCDA is much debated for its conductivity type whether electron transport or hole transport material. The di-anhydride groups present in PTCDA are electron withdrawing, so it should be n-type semiconductor. Forrest’s group have done great work in this regard and shown that PTCDA is a p-type semiconductor. They also noticed that air exposure to this material changes its conductive properties and suggested atmospheric p-type doping [13].

2. Experimental

The device in our study consists of single layer of PTCDA sandwiched between Indium Tin oxide (ITO) coated glass and metals Al & Ag. ITO coated glass slides are used as the substrate for the deposition of thin films of PTCDA (procured form Alfa Aesar, Lancaster). Thickness of the ITO layer on glass substrate is 150 nm with resistivity (15-20) Ωcm-2. Substrates were degreased in acetone, methanol and de-ionized water (18.2MΩ) several times prior to deposition. We have deposited the PTCDA thin films on pre-cleaned ITO coated glass substrates by thermal evaporation technique inside a vacuum chamber under a base pressure of 10-6 mbar at the rate of 3-5Ǻ/sec. Thickness of the film is measured by digital thickness monitor provided in coating unit. Circular dots of high purity metals with diameter 1mm are deposited on PTCDA layer by same thermal evaporation technique.

Current Voltage response of an electronic circuit is one of the major characterization techniques in which molecular systems act as conducting element. Current voltage data of these contacts is taken with the help of silver paste and silver wires on ITO and metallic dots as two electrodes, and is measured by keithley 2400 source meter. All the measurements are done at room temperature. PTCDA thin film is also fabricated on pre-cleaned micro glass slides and their transmission spectrum is recorded in the visible region by spectrophotometer. Data acquisition from source meter is done by computer as controller using IEEE-488 interface in visual basic.

3. Results and Discussion

Transmission spectrum in visible region of 2500Ǻ thick PTCDA thin film is shown in fig. 2. The first trench edge in this spectrum is obtained at wavelength 560nm. This wavelength corresponds to 2.21eV energy and is equal to the optical band gap of PTCDA. It means at this energy electrons can be excited from Highest Occupied Molecular Orbital (HOMO) to Lowest Unoccupied Molecular Orbital (LUMO). The obtained value of band gap proves it as a semi-conducting material and agrees well with previously reported results for this material [14].

350 400 450 500 550 600 650 700 750 800

10

20

30

40

50

60

70

80

90

100

PTCDA Thin Film Transmission Spectrum

Tran

smis

sion

(%)

Wavelength (nm)

Fig. 2. Transmission spectrum of 2500Ǻ thick PTCDA thin film.

Schottky-Mott model is used to align position of vacuum and energy levels at metal/organic interfaces. Difference between metal work function and ionization potential (p-type) and electron affinity (n-type) of the particular semiconductor determines the interface barrier height. In this model, it is also assumed that no interface dipole is formed at the junction. Detailed studies on metal/organic semiconductor interfaces have demonstrated that these interfaces are more complex than metal/inorganic semiconductor interfaces where simple vacuum alignment theory serves the purpose [15]. Softness of Van Der Waals bonded organic molecular solids, lead to considerable chemical and structural deviation from ideal interfaces when contact is formed. The measured I-V curve of ITO-PTCDA-Al structure is shown in fig. 3. The current response taken by sweeping voltage in one direction from +2V to -2V leads to Ohmic behavior. If we see the contact formation between Al and PTCDA in the light of Schottky-Mott theory then it should be rectifying, but it is

Fabrication and Characterization of Indium Tin Oxide-PTCDA-Aluminium/Silver Junctions 69

Ohmic as shown by I-V plots. This is due to the high chemical reactivity of Al with PTCDA.

-2 -1 0 1 2

-3.0x10-2

-2.0x10-2

-1.0x10-2

0.0

1.0x10-2

2.0x10-2

3.0x10-2

Cur

rent

(A)

Voltage (V)

ITO-PTCDA-Al ITO-PTCDA-Ag

Fig. 3. Current-Voltage (I-V) characteristics of ITO-PTCDA-Al & Ag sandwiched structure.

Al and other metals like In, Ti, Sn react extensively with the oxygen containing anhydride group of PTCDA molecule, creating a large density of electronic states within the energy gap of PTCDA. These newly generated electronic states between LUMO and HOMO of organic semiconductors are known as interface gap states. All reactive metals result in an Ohmic contact, which is attributed to carrier conduction through these interface states [16-18].

Fig. 3. shows the measured I-V response of ITO-PTCDA-Ag contact structure in the same voltage range. This deviated linear curve gives us information that may be the interface of Ag/PTCDA is non reactive and abrupt. This is also justified by chemical reactivity scale of metals where Al is more reactive metal than silver (Ag).

Hirose et al deeply studied the chemical reactivity of various metal/PTCDA interfaces by more precise techniques like synchrotron radiation photoemission spectroscopy. They also investigated that In and Al diffuses deep in thin films of PTCDA and their penetration is found to be inversely related to their first ionization energy. While for noble metals Au and Ag no interface chemical reaction and diffusion is observed which results in a blocking carrier transport forming an abrupt junction with PTCDA [19-21].

This suggests us that classical Schottky-Mott model of contact formation is inapplicable in case of reactive metals. Thus it is concluded that, while dealing with reactive metal/organic interfaces factors related to chemical reaction, diffusion of metal and alignment of energy levels

must be taken into account. More detailed studies of these interfaces are in our future plans. References [1] V. Saxena, B. D. Malhotra, Curr. Appl.

Phys. 3 (2003) 293. [2] Reinhard Scholz, Marion Friedrich,

Georgeta Salvan, T. U. Kampen, D. R. T. Zahn, T. Frauenheim, J. Phys. Condens. Matter 15 (2003) S2647.

[3] D. R. T. Zahn, T. U. Kampen, Henry Mendez, Appl. Surf. Sci. 212-213 (2003) 423.

[4] G. Ginev, T. Riedl, R. Parashkov, W, Knowalsky, J. Phys. Condens. Matter 15 (2003) S2611.

[5] A. R. Vearey-Roberts, D. A. Evans, Appl. Phys. Lett. 86 (2005) 72105.

[6] P. E. Burrows, S. R. Forrest, Appl. Phys. Lett. 64 (1994) 2285.

[7] T. U. Kampen, G. Salvan, M. Friedrich, D. R. T. Zahn, Appl. Surf. Sci. 212-213 (2003) 501.

[8] D. R. T. Zahn, T. U. Kampen, A. Bekkali, I. Thurzo, A. Bolognesi, T. Ziller, P. Lugli, Appl. Surf. Sci. 234 (2004) 313.

[9] D. R. T. Zahn, T. U. Kampen, D. Tenne, R. Scholz, Appl. Surf. Sci. 175-175 (2001) 326.

[10] D. R. T. Zahn, T. U. Kampen, A. Das, S. Park, W. Hoyer, Appl. Surf. Sci. 234 (2004) 333.

[11] M. Eremtchenko, J. A. Schaefer, F. S. Tautz, Nature 425 (2003) 602.

[12] R. Temirov, S. Soubatch, A. Lucian, F. S. Tautz, nature 444 (2006) 350.

[13] S. R. Forrest, J. Phys. Condens. Matter 15 (2003) S2599.

[14] S. R. Forrest, M. L. Kaplan, P.H. Schmidt, J. Appl. Phys 55 (1984) 1492.

[15] T. U. Kampen, Appl. Phys. A 82 (2006) 457.

[16] Y. Hirose, A. Kahn, C.I Wu, V. Aristov, Appl. Surf. Sci. 113/114 (1997) 291.

[17] S. R. Forrest, Y. Hirose, A. Kahn, Appl. Phys. Lett. 64 (25) (1994) 3482.

[18] S. Park, T. U. Kampen, D. R. T. Zahn, T. Kachel, P. Bressler, W. Braun, Appl. Surf. Sci. 190 (2002) 376.

[19] Y. Hirose, A. Kahn, V. Aristov, Appl. Phys. Lett. 68 (2) (1996) 217.

[20] Y. Hirose, A. Kahn, V. Aristov, P. Soukiassian, V. Batovic, S. R. Forrest, Phys. Rev. B 54 (1996) 13748

[21] S. Picozzi, B. Delley, M. Elstner, Phys. Rev. B 68 (2003) 195309.

Effect of HgO and AgO Addition on the Electrical Properties of YBa2Cu3O7-

Rajneesh Mohana, Kiran Singha, Nupinderjeet Kaura, S. Bhattacharyab, N. K. Gaura and R. K. Singhc

a Department of Physics, Barkatullah University, Bhopal- 462026 (M.P.), India bTechnical Phyics and Prototype Engineering Division, Bhabha Atomic Research Centre Mumbai -400085.

c Institute of Professional and Scientific Studies & Research, C. D. University, Sirsa -125055 (HR), India E-mail : [email protected]

Abstract

We have investigated the effect of HgO and AgO on the electrical properties of YBa2Cu3O7-δ samples synthesized using solid-state reaction method. These samples were sintered at 950 °C in open atmosphere for 24 hrs. They were characterized through standard four probe technique for low temperature electrical resistivity measurement. Results of RT measurement show that with the addition of HgO and AgO the resistivity of YBCO superconductor decreases gradually. 1. Introduction

High temperature superconductors (HTS) and their parent compound remain ones of the most intriguing systems despite an impressive quantity of collected experimental material. After the first discovery of the superconducting ceramic system La-Ba-Cu-O [1] with critical transition temperature between 30–40 K, other families of copper-oxide based ceramics have been synthesized with higher critical temperatures. These oxides include the Y-Ba-Cu-O series (Tc ~ 90K) [2], the Bi-Sr-Ca-Cu-O series (Tc ~ 80 - 115K) [3], the Tl-Ba-Ca-Cu-O group (Tc ~ 85-125K [4] and Hg-Ba-Ca-Cu-O (Tc ~ 133 K) [5]. Y-Ba-Cu-O is the first superconductor with Tc > 77 K, remains the best studied ceramic superconductor, although other ceramic oxide systems based on Bi-Sr-Ca-Cu-O or Tl-Ba-Ca-Cu-O and Hg-Ba-Ca-Cu-O have been prepared and found to have somewhat higher Tc’s than Y-Ba-Cu-O. It is termed as “Y-123”. Y-123 is brittle and has low tensil strength, and can not be melt-processed into wires due to peritectic decomposition, which make its processing challenging. Y-123 is an oxygen deficient compound and required long annealing time in oxygen ambient. In this regard the study of metal superconductor composites is of immense importance to improve the physical and transport properties. There are many reports on the study of addition of metals oxides to the Y-123 [6-11].

The HgO has been considered as a potential material for doping oxygen because of its lower decomposition temperature (476˚C) and high oxygen ambient created during decomposition.

HgO decomposes into mercury, which escapes from the matrix leaving the crystal unaltered and oxygen, which provide an excellent ambient for the formation of a stoichiometric oxide compound [11-15] and might influence the grain boundary structure. This will also helpful in designing the suitable GBs for high current density applications.

Silver oxide is also a preferred additives for Y-123 since its addition does not deteriorate superconducting properties, but significantly increases the thermal and electrical conductivity as well as the mechanical strength of the fabricated composites. Furthermore its potential beneficial effect on increasing critical current densities has been widely documented. Thus, the presence of Ag in Y-123 may affect many technologically important steps in the application of HTSC’s of this family. Studies of Ag substitution effects by, e.g. Behera et al. [16] concluded that the silver could be doped in the grains rather than precipitated at the grains in the samples prepared at a relatively lower sintering temperature. Sintering at higher temperature results in diffusing of most of the Ag to grain boundaries causing the decrease of normal-state resistivity and increase of the critical current without a significant effect on the critical temperature. All these effects are distinctly influenced by the changes in microstructure and grain links in the samples.

It can also provide oxygen ambient as a result of its decomposition into Ag and O, which considerably improves many characteristics of the Y-123 superconductors. It enhances oxygen diffusion, Silver acts as a flux, its addition improves the grain growth in the sintering process and the texturing. The strength and fracture

Effect of HgO and AgO Addition on the Electrical Properties 71

toughness were observed to increase with Ag content in Y-123/Ag composites. Silver doping was also found to reduce the effect of uniaxial compression on the I-V curves [17]. It was assumed that silver, as a soft metal, would be capable of lowering local strains at grain boundaries, which are weak links in HTSC ceramics, as well as favor the creation of internal long-range strain fields in a sample. The study of the Y-123/Ag system was mostly performed in the framework of technical and largescale applications, involving the transport and mechanical [18] properties.

In order to improve our understanding of these phenomena, this paper addresses our findings in a study on the effect of additives like HgO and AgO on the electerical properties of Y-123 superconductor in the range of 0–30 wt.%. 2. Experimental Procedure

A master batch of polycrystalline samples of YBa2Cu3Oy have been prepared by standard solid-state reaction method from the appropriate amount of high purity powders of Y2O3, BaCO3 and CuO. The samples were calcined after thorough grinding at 900 °C, 910 °C and 920 °C for 24 hrs respectively with intermediate grinding. The resultant powder is devided into seven parts. HgO and AgO in the molar ratio of 0.1, 0.2 and 0.3 respectively was added to these seven different parts. These parts were then grinded thoroughly in separate pestle and mortes. Then each mixture of powder was pelletized into 12 mm circular pellets by applying 8 ton of pressure using a hydraulic press. These pellets were sintered at 950 °C for 24 hrs in open atmosphere. one rectangular bar was cut from the sintered pellets from each sample. On these bars four indium contacts were made.

Low temperature electrical transport measurement on these rectangular bars were performed using the standard four-probe method using a closed cycle helium cryostat (CTI, Cryogenics). 3. Results and discussion

The resistivity –temperature (ρ-T) behaviour of HgO and Ago added samples were shown in Fig 1. and 2 respectivily. The ρ-T behaviour near superconducting transition is shown in the inset of the respective figures.

It is clear from the figures 1 and 2 that with the addition of HgO and AgO the resistivity of Y-123 superconductor decreases gradually. This is due to the increase in the oxygen content provided as a result of decomposition of HgO and AgO. It has already been shown that HgO decomposes into Hg metal and atomic oxygen during heating [12-14]. The Hg escapes from the matrix leaving the crystal structure unaltered and atomic oxygen released during decomposition provides an excellent ambient for the formation of a stoichiometric Y-123 superconductors. On the other hand AgO decomposes into Ag and atomic oxygen. The Ag results in diffusing of most of the Ag to grain

boundaries causing the decrease of normal-state resistivity and connectivity across the different grains. In figure 3, the ratio of resistivity at 290 K (ρ290 K) and resistivity at 100 K (ρ100 K) of the HgO and AgO added samples is presented. It can be seen from fig.3 that with AgO addition the ratio ρ290K/ρ100K increases progressively with AgO conc. It means that Y-123 becomes more metallic in normal state, because of the diffusion of Ag atoms across the grains. In case of HgO added samples,

Fig. 2. ρ-T behaviour of AgO added samples. Inset ρ-T behviour near transition.

0 50 100 150 200 250 3000.0

0.5

1.0

1.5

2.0

2.5

3.0

84 86 88 90 92 940.0

0.5

1.0

1.5

2.0

ρ (m

Ω c

m)

T (K)

Pure Y-123 0.1 AgO added 0.2 AgO added 0.3 Ago added

Pure Y-123 0.1 AgO added 0.2 AgO added) 0.3 AgO added

ρ (m

Ω c

m)

T (K)

Fig. 1. ρ-T behaviour of HgO added samples. Inset ρ-T behviour near transition.

0 50 100 150 200 250 3000.0

0.5

1.0

1.5

2.0

2.5

3.0

78 80 82 84 86 88 90 920.0

0.5

1.0

1.5

2.0

ρ (m

Ω c

m)

T (K)

Pure Y-123 0.1 HgO added 0.2 HgO added 0.3 HgO added

T (K)

ρ (m

Ω c

m)

Pure Y-123 0.1 HgO added 0.2 HgO added 0.3 HgO added


the ratio of ρ290K/ρ100K decreases for 10% addition after that it increases linearly. It is because of negligibly small amount of oxygen go into the YBCO matrix at low HgO conc. As very increase the HgO conc. more oxygen will go into the sample matrix and thus increasing the metallic character. If we compare the both curves in fig. 3, we see that AgO added samples have more metallic character than HgO added ones. This is due to the reason that AgO provide both metallic Ag, which diffuse across the grains and O, but HgO provides only O, because Hg leaves the matrix on decomposition.

4. Conclusions

On the basis of an over all discussion, it may be concluded that the addition of HgO and AgO decreases the normal state resistivity of YBCO. AgO besides providing Oxygen ambient and also make Ag to diffuse across the Y-123 grains. Acknowledgements The authors are thankful to the University Grants Commission (UGC) for financial support. References [1] G. Bednorz and K.A. Müller, Z. Phys. B, 64

(1986) 189-197.

[2] M.K. Wu, J.R. Ashburn, C.J. Torng, P.H. Hor, R.L. Meng, L. Gao, Z.J. Huang, Y.Q. Wang and C.W. Chu, Phys. Rev. Lett. 58 (1987) 908.

[3] H. Maeda, Y. Tanaka, M. Fukutomi and T. Asano, 27 (1988) L209.

[4] Z.Z Sheng and A.M. Hermann. Proc. in World Congress on Superconductivity. Singapore: World Scientific, 1988

[5] Cantoni, M., A. Schilling, H.U. Nissen and H.R. Ott, Physica C, 215 (1993) 11

[6] T. Nishizaki, M. Maeda, T. Sato, N. Kobayashi, Physica C 426–431 (2005) 18.

[7] Y. Feng, A.K. Pradhan, Y. Zhao, Y. Wu, N. Koshizuka, L. Zhou, Supercond. Sci. Technol. 14 (2001) 224.

[8] S. Nariki, N. Sakai, M. Murakami, I. Hirabayashi, Physica C 439 (2006) 62.

[9] J.D. Riches, J.A. Alarco, J.C. Barry, Physica C 336 (2000) 43.

[10] F. Ben Azzouz, M. Zouaoui, K.D. Mani, M. Annabi, G. Van Tendeloo*, M. Ben Salem, Physica C 442 (2006) 13.

[11] A. Mellekh, M. Zouaoui, F. Ben Azzouz, M. Annabi, M. Ben Salem Sol. Stat. Commun. 140 (2006) 318

[12] R. Mohan, K. Singh, N. Kaur , S. Bhatacharya, N.K. Gaur, V. Shelke, S.K. Gupta and R.K. Singh, Sol. Stat. Commun. 141 (2007) 605.

[13] M. Dixit, S. Bhattacharya, R. Mohan, K. Singh, P.S.R. Krishna, V. Shelke, N.K. Gaur and R.K Singh, Pramana J. Phys. 63 (2004) 233.

[14] M. Dixit,V. Shelke, S. Bhattacharya, N.K. Gaur and R.K. Singh, J. Supercond. 553 (2002) 15.

[15] A. Pandey, Y.S. Reddy and R.G. Sharma, J. Mat. Sci. 32, (1997) 3701.

[16] D. Behera, N.C. Mishra, K. Patnaik. J. Superconduct 10 (1997) 27.

[17] T. S. Orlova, B. I. Smirnov, V. V. ShpeÏzman, et al., Sov. Phys. Solid State 32 (1990) 606.

[18] L.K. Markov, T.S. Orlova, N.N. Peschanskaya, B.I. Smirnov, Yu. P. Stepanov, and V.V. Shpeizman, Phys. of the Sol. Stat., 45 (2003) 1629

Fig. 3. The ratio of resistivity at 290 K (ρ290 K) and resistivity at 100 K (ρ100 K) of the HgO and AgO added samples.

Dielectric Relaxation Studies of N, N-Dimethylformamide in the Benzene Solution from Microwave Absorption Data.

Vimal Sharma† and Nagesh Thakur‡

Department of Physics, Paniapt Institute of textile & Engg., Samalkha, Panipat, India. Department of Physics, H. P. University, Shimla–171005, Himachal Pradesh, India.


Abstract

Dielectric constant (ε′) and dielectric loss (ε″) of dilute solutions N, N-dimethylformamide (DMF) in benzene have been measured at microwave frequency 9.883 GHz and at different temperatures (25, 30, 35 and 400C) using short-circuited waveguide method of Heston et al [1]. Standing microwave techniques and Gopala Krishna’s single frequency concentration variation method [2] have been used for above measurements. The measured values of dielectric constant (ε′) and dielectric loss (ε″) have been used to evaluate dipole moment (μ) and relaxation time (τ). Energy parameters (ΔHε, ΔFε and ΔSε) for the dielectric relaxation process of DMF in benzene have been calculated using Eyring’s rate equations [3]. Comparison has been made with the corresponding energy parameters for viscous flow process. It is found that the dielectric relaxation process can be treated as the rate process like the viscous flow process. Plot of log (τT) versus 103/T (Figure 1) for DMF is found to be linear, showing that the decay of relaxation time with temperature is exponential. This indicates that the dielectric relaxation process can be treated as a rate process. Variation of ε΄ and ε˝ with weight fraction of solute in benzene is found to be linear (Figures 2 & 3). This ensures the applicability of the Debye theory [4] and hence, that of Gopala Krishna’s method in the studied concentration range. The change in dipole moment of DMF with temperature indicates the existence of solute-solvent type of molecular association. The solute-solvent associations may be due to interaction of positive fractional charge at the sight of N-atom in DMF molecule and π-delocalized electron cloud in the benzene ring of benzene molecule (Fig. 4).

1. Introduction

N, N- Dimethylformamide (DMF) is an important non-aqueous solvent with dielectric constant ε΄ = 36.70 [1] and dipole-moment μ = 3.86D [2]. DMF is a widely used solvent for many recently developed synthetic procedures because of its powerful solvating properties and its chemical stability in the absence of acidic or basic catalysts [3]. DMF as pure solvent is certainly to some extent associated by means of non specific dipole-dipole interactions, and is of particular interest because any significant structural effects are absent due to the lack of hydrogen bonds; therefore it may work as an aprotic protophilic solvent of large dipole moment and high dielectric constant. The associative molecular nature of DMF motivated the authors to undertake the experimental work concerned with the dielectric relaxation process of DMF. For this standard microwave techniques and solution methods have been used.

When polar molecules are subjected to electromagnetic waves in microwave frequency

region, they absorb considerable microwave energy thereby perturbing dipole moment of molecules due to molecular rotations. The study of dielectric relaxation of polar liquids in non-polar solvents from the microwave absorption studies give valuable information about various types of the molecular associations present in the solutions as microwaves can detect weak molecular interactions [4-9]. The present investigation is concerned with the study of dielectric relaxation of DMF in benzene using standard microwave standing wave techniques and single frequency concentration variation method of Gopala Krishna [10]. Measurements have been made at different temperatures (25, 30, 35 and 400C). The various energy parameters like enthalpy of activation (ΔHε, free energy of activation (ΔFε and entropy of activation (ΔSε) for dielectric relaxation process and similar parameters (ΔHη ΔFεand ΔSε) for viscous flow process have been calculated and compared. It is found that the dielectric relaxation process is a rate process like that of the viscous flow process. Solute-solvent types of the molecular


associations have been proposed for DMF in benzene solution.

2. Experimental details

For the present investigation different pure samples of DMF and benzene have been procured from standard companies and used after fractional distillation. Benzene (GR Grade) from E Merck Ltd., Mumbai (India) was dried by refluxing over freshly cut sodium metal for 6-8 hours and then distilled through a long vertical fractionating column. DMF (AR grade) from Sisco Research laboratory Pvt. Ltd. Bombay (India) was dried over 4Å for 8-10 hours and then distilled through long vertical fractional column. The middle fraction of each distilled solutions were collected for use. The X-band microwave bench was used to measure wavelengths in the dielectric and the voltage standing wave ratio (VSWR). The dielectric constant (ε΄) and the dielectric loss (ε˝) of dilute solutions of DMF in benzene were calculated following microwave absorption technique of Heston et al. [11]. All the measurements were carried out at temperatures 25, 30, 35 and 400C and the temperature was thermostatically controlled within ±0.50C. Using the Gopala Krishna’s single frequency concentration variation method, dielectric relaxation time (τ) and dipole moment (μ) were calculated.

The viscosities and densities of the solutions were measured by Ubbelohde viscometer and sealable type of pycnometer, respectively.


The dielectric constant (ε΄) and the dielectric loss (ε˝) for DMF in the benzene solution have been calculated using the short-circuited waveguide method of Heston et al. This method is the highly accurate for the measurement of ε΄ and ε˝ of polar mixtures in dilute solutions of non-polar solvent at very low concentrations. The accuracy in measurements of ε΄ and ε˝ values was ±1% and ±3% respectively. Following equations have been used:

2

02

0⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=′

dc λλ

λλ

ε (1)

dnd

d

g

d

ρλλ

λλ

πε

202" ⎟⎟

⎠

⎞⎜⎜⎝

⎛= (2)

Here λ0, λc, λg and λd are the wavelengths of microwave in free space, the cut off wavelength, the waveguide wavelength and the wavelength in the waveguide filled with solution, respectively and ρ is the inverse of the voltage standing wave ratio. (dρ/dn) is the slope of the curve of ρ versus n. Here, n is the integer (n = 1, 2, 3…) such that, (nλd/2) represents the length of dielectric filled waveguide. The relaxation time (τ) and dipole moment (μ) of the molecular entities were calculated using single frequency concentration variation method of Gopala Kirshna, using the following equations.

⎟⎠⎞

⎜⎝⎛=

dXdY

cπλ

τ2

0 (3)

and

dWdX

dXdY

NdkTM

⎥⎥⎦

⎤

⎢⎢⎣

⎡⎟⎠⎞

⎜⎝⎛+=

22 1

49

πμ (4)

Variation of ε΄ and ε˝ with weight fraction of solute in benzene is found to be linear. This ensures the applicability of the Debye theory and hence, that of Gopala Krishna’s method in the studied concentration range. Measured dipole moment value of DMF is found to be very near to the literature value of the dipole moment. This shows that pure DMF exist in the monomer form in benzene solution.

Table 1 represents the values of weight fraction, dielectric constant, dielectric loss, dielectric relaxation time (τ) and dipole moment (μ) for DMF in benzene solution at different temperatures (25, 30, 35 and 400C). The variation of dipole moment with rise in temperature shows solute-solvent association while no change in dipole moment with temperature shows absence of solute-solvent association [12]. There is a small variation in the dipole moment of DMF in benzene solution with the rise in temperature. This could be explained on the basis of the solvent effects [13]. The small change in dipole moment with temperature may be due to the stretching of bond moments and change in bond angles. This change may also be due to the breaking of solute-solvent associations with rise in temperature. The solute-solvent associations may be due to interaction of positive fractional charge at the sight of N-atom in DMF molecule and π-delocalized electron cloud in the benzene ring of benzene molecule (Fig. 1).

Dielectric Relaxation Studies of N, N-Dimethylformamide in the Benzene Solution 75

Fig. 1. Solute-solvent molecular association of DMF molecule in benzene.

The relaxation time increases with the size

of the molecule and may be discussed in terms of the molecular shape and solvent microscopic viscosity. The decrease in the relaxation time with rise in temperature is observed in the present case. This decrease in relaxation time with the rise in temperature can be explained on the basis of Debye theory [14]. Thermal energy of the system increases with rise in temperature and decreases the relaxation time of the molecular entities. At higher temperature, due to more number of collisions rate of loss of energy increases hence reorientation of molecules becomes faster. Table 1. Dielectric relaxation time (τ) and dipole moment (μ) for DMF in benzene solution at different temperatures.

Temperature 0C

Wt. fraction of solute in benzene

τ X 10-12 sec

μ (D)

25

30

35

40

0.0023 0.0046 0.0073 0.0094

0.0023 0.0046 0.0073 0.0094

0.0023 0.0046 0.0073 0.0094

0.0023 0.0046 0.0073 0.0094

4.05

3.77

3.56

3.28

3.74

3.80

3.84

3.88

Plot of log (τT) versus 103/T for DMF is found to be linear, showing that the decay of relaxation time with temperature is exponential. This indicates that the dielectric relaxation process can be treated as a rate process. The slopes of the linear plots between log (τT) and log (η) with 103/T for DMF have been used to calculate various thermodynamic parameters for relaxation and viscous flow processes from Eyring’s equations [15]. These thermodynamic parameters are presented in table 2. The observations show that the free energy of activation (ΔFε) for the dielectric relaxation process is less than the free energy of activation (ΔFη) for viscous flow process. This may be explained on the basis that, the dielectric relaxation process involves the rotation of molecular entities whereas in the flow process, the rotational as well as the translational motion of the molecules are involved. Enthalpy of activation depends upon the local environment of the molecules. The enthalpy of activation (ΔHε) for dielectric relaxation process is found to be different from the enthalpy of activation (ΔHη) for viscous flow process. This difference shows that, dielectric relaxation process involves different types of bonding and breaking of bonding to different extents. The ratio of enthalpies of activation for relaxation and viscous flow processes (ΔHε ΔHη for DMF is more than 0.6. According to Krishnaji and Mansingh classification [16], polar liquids for which this ratio is greater than 0.55 do not show solid rotator phase but those for which it is less than 0.45 should show a solid rotator phase. Table 2 Enthalpies of activation (ΔHε, ΔHη in kJ. mole-1), free energies of activation (ΔFε, ΔFη in kJ. mole-1), and entropies of activation (ΔSε, ΔSη in kJ. mole-1 K-1) for DMF in benzene solutions at different temperatures.

Tem 0C

ΔHε ±0.17

ΔFε ±

0.14

ΔSε ±

0.31

ΔHη ΔFη ΔSη

25 30 35 40

6.76 6.76 6.76 6.76

8.27 8.30 8.32 8.35

-5.06 -5.06 -5.06 -5.06

11.02 11.02 11.02 11.02

12.20 12.23 12.26 12.32

-3.96 -3.99 -4.00 -4.14

The entropy of a system is the measure of

the orderly nature of the system. The positive value of the change in entropy (ΔSε) for activated process indicates the non-cooperative environment of the system and the activated state is unstable. If the environment of the system is

N CH3

CH3 C

O

δ + H


cooperative for the activated process, the change in entropy (ΔSε) becomes negative. In the present case, it is observed that the change in entropy of the dielectric relaxation process is negative indicating that the environment of the system is cooperative like that of the activated viscous flow state. References

1. G. Mamantov, in “Characterization of solutes in non-aqueous solvents” (Plenum Press, New York and London, 1976) p. 53.

2. M. H. Hutchinson and L. E. Sutton, J. Chem. Soc., (1958) 4382.

3. Vogel, in “Textbook of Practical Organic Chemistry” (Longman Singapore Publishers Pte Ltd., 1989) Edition Vth, p. 409.

4. N. Thakur & D. R. Sharma, Indian J. Pure & Appl. Phys., 38, (2000) 328.

5. A. D. Vyas & V. A. Rana, Indian J. Pure & Appl. Phys., 40, (2002) 69.

6. V. S. Rangra & D. R. Sharma, Indian J. Phys., 78B (1), (2004) 111.

7. A. Chaudhari, S. Ahire & S. C. Mehrotra, J. Mol. Liq., 94, (2001) 17.

8. .G. D. Rewar & D. Bhatnagar, Indian J. Pure & Appl. Phys., 40, (2002) 430.

9. .R. J. Sengwa, R. Chaudhary & S. C. Mehrotra, Molec. Phys., 99, (2001) 1805.

10. K. V. Gopala Krishna, Trans. Faraday Soc., 33, (1957) 767.

11. W. M. Heston (Jr), A. D. Franklin, E. L. Hennely & C. P. Smyth, J. Am. Chem. Soc., 72, (1950) 3443.

12. J. S. Dhull and D. R. Sharma, J. Phys. D Appl. Phys., 15, (1982) 2307.

13. Nora E. Hill, Worth E. Vaughan, A. H. Price & Mansel Davies, in “Dielectric properties and molecular behaviour” (Van Nostrand-Reinhold, London, 1969) p. 253.

14. P. Debye, in “Polar Molecules” (New York Chemical, Catalog, 1929).

15. H. Eyring, S. Glasstone & K. J. Laidler, in “Theory of rate process” (Mc Grow-Hill, New York, 1941) p. 541.

16. Krishnaji & Man Singh, J. Chem. Phys., 44, (1966) 1590.

Dielectric Relaxation of Tetramethylurea in Benzene and Carbon Tetrachloride Solutions from Microwave Absorption Data

Rajesh Kumar and Nagesh Thakur

Electronics Research Laboratory, Department of Physics, Himachal Pradesh University, Shimla 171 005. E-mail: [email protected]

Abstract

Using standard standing wave microwave techniques and following the single frequency (9.885 GHz) concentration variation method, the dielectric relaxation time (τ) and dipole moment (μ) of dilute solutions of tetramethylurea (TMU) in benzene and carbon tetrachloride (CCl4) solutions at different temperatures (25, 30, 35 and 400C) have been calculated. From these studies, monomer structure of TMU in both solutions has been inferred. The presence of solute-solvent molecular associations has been proposed. A comparison of energy parameters for the dielectric relaxation process and viscous flow process shows that dielectric relaxation process can be treated as a rate process just like the viscous flow process.

1. Introduction Tetramethylurea is an important non-aqueous solvent with dielectric constant (ε′ = 23.45±0.06) and dipole moment (μ = 3.37D) [1]. The combination of high dipole moment, moderate dielectric constant, low viscosity and low specific conductance, makes TMU a useful solvent for studying the behaviour of electrolytes in solution [1]. Dielectric relaxation studies of polar molecules in nonpolar solvent from microwave absorption data have been frequently attempted by a number of research workers [2-8]. Dielectric relaxation studies in the microwave region provide meaningful information about various types of molecular associations present in the solution [9]. This is because of the capability of microwaves to detect even weak molecular interactions. This paper is concerned with the dielectric relaxation studies of TMU in benzene and carbon tetrachloride (CCl4) solutions at different temperatures (25,30,35 and 40 oC) from microwave absorption measurements at 9.885 GHz. It is found that, TMU exist in monomer form in both benzene and carbon tetrachloride (CCl4) solutions. Solute-solvent molecular association is predicted. Energy parameters (ΔHε, ΔFε, ΔSε) for the dielectric relaxation process have been calculated and compared with the corresponding energy parameters for viscous flow process. It is found that dielectric relaxation process is a rate process like the viscous flow process.

2. Experimental details

Tetramethylurea (Fluka, Germany) was dried over anhydrous BaO for 48 hours and then

distilled through a long vertical fractionating column and the middle fraction was used with in a week to avoid decomposition. Benzene extra-pure AR (Sisco Research Laboratories Pvt. Ltd., Mumbai) was dried by refluxing over freshly cut sodium metal for 6-8 hours and then distilled through a long vertical fractionating column. The middle fraction of distilled benzene was collected for use. Carbon tetrachloride extra-pure AR (Sisco Research Laboratories Pvt. Ltd., Mumbai) was distilled through a long vertical fractionating column and the middle fraction was collected for use. The X-band microwave bench was used to measure the wavelength in the dielectric medium and voltage standing wave ratio (VSWR) using a short-circuiting plunger. Microwave techniques of Heston et al. [10] were used to calculate dielectric constant (ε′) and dielectric loss (ε″) of dilute solutions of TMU in benzene and carbon tetrachloride (CCl4) solutions at different temperatures. Circulation of thermostated water around the dielectric cell controlled the temperature of the solution. The temperature control of the thermostat was ± 0.05 oC. The viscosity and density of benzene at different temperatures were measured with Ubbelohde viscometer and a pycnometer respectively. Dielectric relaxation time (τ) and dipole moment (μ) of TMU in benzene and carbon tetrachloride (CCl4) solutions has been calculated following the single frequency concentration variational method of Gopala Krishna [11]. Eyring rate equations [12] have been used to calculate the energy parameters for dielectric relaxation process and viscous flow process.



The dielectric constant (ε′) and dielectric loss (ε″) of the dilute solutions of TMU in benzene and carbon tetrachloride (CCl4) solutions at different temperatures have been calculated by the method of Heston et al. [10]. Following equations have been used:

22

⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=′

d

o

c

o

λλ

λλ

ε (1)

and

⎟⎠⎞

⎜⎝⎛

⎟⎟⎠

⎞⎜⎜⎝

⎛=′′

dnd

d

g

d

o ρλλ

λλ

πε

22

(2)

where λo, λc, λg and λd are the free space wavelength, the cut-off wavelength, the waveguide wavelength and the wavelength in the waveguide filled with solution, respectively. ρ is the inverse of voltage standing wave ratio (VSWR) and dρ/dn is the slope of ρ versus n curve, where n = (1,2,3…). ε′ and ε″ were reproducible within ±0.5% and ±3.8%, respectively. The relaxation time (τ) and dipole moment (μ) as given in Table 1 have been calculated by following the Gopala Krishna’s single frequency concentration variation method [11].

Gopala Krishna’s method makes use of Debye’s theory of dielectric relaxation. Debye’s equation for complex permittivity for the dielectric medium as a function of frequency of the applied electric field [13] can be written as:

,1

19

421

21 2

1*

*

ωτμπ

εε

εε

jkTN

++

+−

=+−

∞

∞ (3)

where N1 is the number of polar molecules per unit volume and εεε ′′−′= j* is the complex

permittivity of the medium. ∞ε is the optical permittivity and ω is the angular frequency. Separating real and imaginary parts of both sides of eq. (3) yields

( ),

11

94

21

22

22

21

22

22

τωμπ

εε

εεεεε

+⋅+

+−

=′′++′−′′+′+′

∞

∞

kTN

( )

.19

423

22

21

22 τωωτμπ

εεε

+⋅=

′′++′′′

kTN

Putting

( ),

22

22

22

εεεεε

′′++′−′′+′+′

=X (4)

( ),

23

22 εεε

′′++′′′

=Y (5)

and

,21

+−

=∞

∞

εε

P

in the above equations, one gets

.ετYPX += (6)

The value of P in eq. (6) slightly varies over the range of concentrations of dilute solutions. But as the variation of ε′ and ε″ is far higher than the variation of P due to the change in concentration of dilute solutions at microwave frequencies, it could be treated as constant over the range of concentration variation. From the slope of curve Y versus X the value of relaxation time (τ) of polar molecules in non-polar solvents could be determined and it can be written as:

.2

0 ⎟⎠⎞

⎜⎝⎛=

dXdY

cπλ

τ (7)

For the determination of dipole moment (μ) eq. (6) can be written as

,12KWdPX += (8) with

( ),194

22

2

τωμπ+

=kTM

NK

and

,121 M

WNdN =

where N is the Avogadro number, M is the molecular weight of polar substance, W is the weight fraction and d12 is the density of solution. At low concentration, the variation of density of the solution with weight fraction W may be taken as linear and given by the relation:

( ),1012 Wdd α+= where d0 is the density of the solvent. In the limited experimental range of concentration variation, the graph between X and W can be taken as a straight line with its slope (dX/dW) as Kd0. From this the value of dipole moment may be calculated, using the following relation:

Dielectric Relaxation of Tetramethylurea in Benzene and Carbon Tetrachloride Solutions 79

.149 2

0

2

dWdX

dXdY

NdkTM

⎥⎥⎦

⎤

⎢⎢⎣

⎡⎟⎠⎞

⎜⎝⎛+=

πμ (9)

The energy parameters for the dielectric relaxation process of TMU in benzene and CCl4 and the corresponding energy parameters for viscous flow of benzene and CCl4 given in Table 2 have been calculated by using the following equations given by Eyring et al. [12], for the rate process:

( )RTFkTh /exp ετ Δ= (10)

εεε STHF Δ−Δ=Δ (11)

( )RTFVhN /exp ηη Δ= (12)

ηηη STHF Δ−Δ=Δ (13) where ΔFε, ΔHε and ΔSε are the free energy, enthalpy and entropy of activation for the dielectric relaxation process and ΔFη, ΔHη and ΔS η are the corresponding energy parameters for the viscous flow process. V is the molar volume of the solvent. The plots of log (τT) versus 103/T and log (η) versus 103/T according to equations (10) and (12) are found to be linear, which indicates that both dielectric relaxation process and viscous flow process can be treated as a rate processes. The ΔHε and ΔHη values were computed from the slope of linear plot of log (τT) versus 103/T and log (η) versus 103/T, respectively, using the relation: slope = ΔH / 2.303 R.

Variation of ε′ and ε″ with weight fraction of TMU in benzene and CCl4 is found to be linear. This shows that there is no change in the nature of the rotating molecular entities in the benzene solution. This ensures the applicability of Debye theory and hence that of Gopala Krishna’s method for calculating the dielectric relaxation time (τ) and dipole moment (μ) of TMU in non-polar solvents. It is observed that dielectric relaxation time of TMU in benzene and CCl4 decreases with increase in temperature. This behaviour may be explained on the basis of Debye theory of dielectric relaxation [14]. Dipole moment (μ) of TMU in benzene and CCl4 solutions is found to increase slightly with increase in temperature.

Table 1. Values of dielectric relaxation time (τ) and dipole moment (μ) for TMU in benzene and CCl4 solutions at different temperatures. _______________________________________

Solution Weight Temp. τ μ Fraction (0C) (10-12sec) (Debye)

_______________________________________ TMU in 0.0025 25 5.42 3.28 benzene 0.0043 30 5.08 3.32 0.0062 35 4.70 3.37 0.0081 40 4.31 3.42 TMU in 0.0010 25 9.25 3.54 CCl4 0.0018 30 9.11 3.57 0.0030 35 9.03 3.60 0.0040 40 8.94 3.62 _______________________________________ This shows that TMU exist in the monomer form in both the solutions. The values of dipole moment of TMU in CCl4 are slightly more as compare to benzene. This is due to the differences in the solvent effect in both solvents. The small variation in dipole moment value with rise in temperature may be attributed to the possible solute-solvent molecular association [15]. Solute-solvent molecular association for TMU in benzene solution arises because of the interaction of fractional positive charge at the site of nitrogen atom in TMU molecule with the π – delocalized electron cloud of the benzene ring (Fig.1). In case of carbon tetrachloride although there is no permanent dipole moment but short-range dipole moment arising from bond-moments always exists. It is perhaps the interaction of permanent dipole moment of TMU with this short-range dipole moment of CCl4, which is responsible to account for the solute-solvent association of TMU in CCl4 (Fig.2).

N

N

C

C

C

C

C

H

H

H

H

3

3O

3

3

Fig. 1. Solute-Solvent molecular association between TMU and benzene.


C

CC C

C

H

HH

H

NN

O

Cl

ClClCl

3 3

33δ δ -

δ +δ +

C

Fig. 2. Solute-Solvent molecular association between TMU and carbon tetrachloride. The energy parameters (ΔHε, ΔFε, ΔSε) for dielectric relaxation process and the energy parameters (ΔHη, ΔFη, ΔSη) for viscous flow process have been compared as shown in Table 2. For TMU in benzene and carbon tetrachloride, free energy of activation (ΔFε) for dielectric relaxation process is found to be less than the corresponding free energy of activation for viscous flow. Table 2. Free energies of activation i.e. ΔFε, ΔFη (kcal mole-1), enthalpies of activation i.e. ΔHε, ΔHη (kcal mole-1) and entropies of activation i.e. ΔSε, ΔSη (cal mole-1 deg-1 K-1) for TMU in benzene and CCl4 solutions. _________________________________________ Solution Temp. ΔFε ΔHε ΔSε ΔFη ΔHη Sη(0C) ________________________________________ TMU in 25 2.08 2.24 0.53 2.91 2.623 0.96 benzene 30 2.09 2.24 0.49 2.92 2.623 0.98 35 2.08 2.24 0.51 2.93 2.623 0.99 40 2.07 2.24 0.54 2.94 2.623 1.01 TMU in 25 2.40 0.19 7.41 3.18 2.273 3.04 CCl4 30 2.44 0.19 7.42 3.21 2.273 3.09 35 2.48 0.19 7.43 3.22 2.273 3.07 40 2.53 0.19 7.47 3.24 2.273 3.08 __________________________________

This is as expected, because process of viscous flow involves both rotational and translation motion of the molecules whereas dielectric process involves only rotational motion of the molecules.

The entropy of the system is a measure of its orderly nature. If the environment of the system is cooperative for a particular process the activated system becomes more stable than the normal system and change in entropy is negative. In the same sense, if the environment of the system is obstructive to the process, the activated system becomes more unstable than the normal system

and the environment of the system tends to bring the activated system back to the normal state. In this case the change in entropy for the process becomes positive. In the present case, change in entropy for dielectric relaxation process is negative for TMU in CCl4 whereas it is positive for TMU in benzene solution. This behaviour is due to different types of rotating entities formed because of different types of solute-solvent association in different solvents i.e. benzene and CCl4. In the case of TMU in CCl4, the entropies of activation for the viscous flow (ΔSη) are less negative as compared to those for dielectric process. This shows that the activated state is a state of greater order than the normal state in case of dielectric relaxation process as compared to the activated and normal states of viscous flow. In the case of TMU in benzene the entropy of activation (ΔSε) for dielectric process is found to be positive and for the viscous flow it is negative. This indicates that in dielectric process, the activated state is less ordered than that of viscous flow process. Enthalpy of activation (ΔHε) for dielectric relaxation process is found to be less than the corresponding enthalpy of activation (ΔHη) for the viscous flow process in both cases. Difference in enthalpies of activation for the dielectric relaxation process and viscous flow process indicates that dielectric relaxation process and viscous flow process involves the breaking of bonds with the neighbouring molecules in a different way and to a different extent. 4. Conclusions

The variation in dipole moment value for TMU in benzene and carbon tetrachloride (CCl4) solutions with rise in temperature may be attributed to the possible solute-solvent molecular associations. TMU exist in monomer form in both benzene and carbon tetrachloride (CCl4) solutions. The measured values of energy parameters for dielectric relaxation process and their comparison with energy parameters for viscous flow process shows that the dielectric relaxation process may be treated as a rate process just like the viscous flow process. Acknowledgements This work was supported in part by the University Grant Commission under grant F.16-33/2006 (SA-II).

Dielectric Relaxation of Tetramethylurea in Benzene and Carbon Tetrachloride Solutions 81

References [1] J. J. Logowski, The Chemistry of Non-

aqueous Solvents, Academic Press, New York, San Francisco, London 1976, p. 111.

[2] V. Sharma, N. Thakur, D. R. Sharma, N. S. Negi, V. S. Rangra, Indian J. Pure Appl. Phys. 45, 163 (2007).

[3] S. Kumar, D. R. Sharma, N. Thakur, N. S. Negi, V. S. Rangra, Z. Phys. Chem. 219, 1649 (2005).

[4] R. Kumar, V. S. Rangra, Z. Phys. Chem. 219, 169 (2005).

[5] V. S. Rangra, D. R. Sharma, Indian J. Phys. 78B (1), 111 (2004).

[6] S. L. Abd-El-Messieh, J. Mol. Liq. 95, 167 (2002).

[7] A. D. Vyas, V. A. Rana, Indian J. Pure Appl. Phys. 40, 69 (2002).

[8] A.Chaudhari, S. Ahire, S. C. Mehrotra, J. Mol. Liq. 94, 17 (2001).

[9] J. S. Dhull, D. R. Sharma, D. S. Gill, K. N. Lakshminarayana, Indian J. Phys. B56, 334 (1982).

[10] W. M. Heston (Jr.), A. D. Franklin, E. L. Hennely, C. P. Smyth, J. Am. Chem. Soc. 72, 3443 (1950).

[11] K. V. Gopala Krishna, Trans. Farad. Soc. 53, 767 (1957).

[12] H. Eyring, S. Glasstone, K. J. Laidler, Theory of Rate Process (McGraw-Hill, New York, 1941), p. 541.

[13] Hill Nora E, Vaughan W E, Price A H, & Davies M. Dielectric properties and molecular behaviour (Van Nostrand-Reinhold, London, 1969), p. 63.

[14] P. Debye. Polar Molecules (New York Chemical catalog, 1929).

[15] A. Sharma, D. R. Sharma, J. Phys. Soc. Japan. 61, 1049 (1992).

Dielectric Response of Nitrogen ion Implanted CR-39 Polymer

Nidhi*, Tanu Sharma*, Sanjeev Aggarwal*1, Annu Sharma* , S. Kumar*, S. K. Deshpande# and D. Kanjilal$ * Department of Physics, Kurukshetra University, Kurukshetra – 136 119, India

#UGC-DAE Consortium for Scientific Research, Mumbai Centre, Bhabha Atomic Research Centre, Mumbai 400 085, India

$Inter University Accelerator Centre, New Delhi – 110 016, India 1Email: [email protected]

Abstract

The effect of nitrogen ion implantation on dielectric properties of CR-39 polymer has been studied. Samples were implanted with 100 keV N+ ions to various doses ranging from1014 to 1016 cm-2. The frequency response of dielectric constant (ε׳), dielectric loss and conductivity has been studied both in the pristine and nitrogen ion implanted samples of CR-39 polymer in the frequency range 100 kHz to 100 MHz. Dielectric constant (ε׳) decreases with increase in ion fluence whereas dielectric loss and conductivity have been found to increase with ion fluence. The possible correlation between the modification of dielectric parameters and the changes induced due to nitrogen ion implantation in CR-39 polymer has been discussed.

1. Introduction

Ion beam treatment of polymers has been found to create chain scission, cross-linking, free radical formation, dangling pair production etc. which leads to structural and chemical modifications resulting observable changes in their physical properties [1-4]. Increase in hardness, strength, wear resistance, electrical conductivity, and improvements in the optical transmission properties of polymers have been reported following ion implantation under various conditions. The tailoring of dielectric properties of polymers is important for its technological applications point of view [4-7]. Many studies have reported the change in dielectric properties of different polymers due to interaction of ion beams [4-8]. In the present work we have examined the effect of 100 keV N+ implantation (at various doses) on the dielectric behavior of CR-39 polymer by using dielectric spectroscopy. Dielectric constant, dielectric loss tangent and a. c. conductivity measurements were carried out in the frequency range 105Hz to 108Hz at room temperature. 2. Experimental details

The samples of size 1 cm × 1 cm were cut from 1mm thick sheets of optically transparent CR-39 polymer supplied by M/S TASTRAK, Bristol, England. Some of these samples were implanted with 100 keV N+ ions having doses in

the range of 1014-1016 ions cm-2 under a vacuum of 10-6 Torr, utilizing Low energy beam facility (LEIBF) at Inter University Accelerator Centre (IUAC), New Delhi, India. The beam was electrostatiscally scanned over the entire surface at a beam current density of 160 nA cm-2. These samples were subjected to dielectric relaxation measurements over a frequency range of 105Hz to 108Hz using impedance Gain-Phase Analyzer (Model HP 4194A) available at UGC-DAE Consortium for Scientific Research, Mumbai Centre, BARC, Mumbai. All measurements were carried out at room temperature in rotary pump vacuum. 3. Results and discussion

The effect of nitrogen ion implantation on the dielectric constant (ε׳) has been shown in Fig.1 as a function of angular frequency ω for CR-39 virgin and nitrogen implanted samples with doses in the range of 1014-1016 ions cm-2. Fig. 1 clearly indicates that the value of Dielectric constant (ε׳) decreases with increase in ion fluence. . This may be due to the fact that as a result of ion implantation, the original bonds in the polymer are ruptured leading to chain scission, free radical formation etc. resulting in the formation of new bonds (double bonds, triple bonds, etc.).

Dielectric Response of Nitrogen ion Implanted CR-39 Polymer 83

106 1071.4

1.6

1.8

2.0

Diel

ectri

c co

nsta

nt


Virgin1×1014N+cm-2

1×1015N+cm-2

1×1016N+cm-2

Fig. 1. Variation in dielectric constant ε׳ with angular frequency ω for virgin and nitrogen implanted CR-39 samples.

The formation of double and triple bonds due to ion implantation in the polymer matrix becomes the source of the free carriers and as a consequence dielectric constant decreases with increase in ion fluence. On the other hand there is a slight decrease in the dielectric constant with increase in frequency. The slow migration of charge carriers may be assumed as the cause of decrease in the dielectric constant in higher frequency regions. With increase in the frequency, the charge carriers migrating through the dielectric get entrapped against a defect site and induce opposite charge in its vicinity. This difference in dielectric constant measured at low (105Hz) and high (108Hz) frequencies is called the strength of the relaxation in polymers.

The variation of dielectric loss (tan δ) as a function of angular frequencies at different doses is shown in Fig. 2. It is observed that the behaviour of tan δ with log frequency is similar in both pristine and implanted samples. The value of tan δ shows an increasing trend with increasing ion fluence. A dielectric loss peak has been observed for the virgin as well as nitrogen implanted CR-39 samples near 10 MHz. Since this peak appears to be associated with the vibrations of the main molecular chains, it is interesting to observe that even after implantation the main sources of this dielectric loss peak remain almost unaffected.

105 106 1070.010

0.015

0.020

0.025

Die

lect

ric lo

ss


Virgin1×1014N+cm-2

1×1015N+cm-2

Fig. 2. Variation in dielectric loss (tan δ) with angular frequency ω for virgin and nitrogen implanted CR-39 samples.

The variation in a. c. conductivity of the same samples was also studied. The conductivity of the virgin and irradiated samples has been found to increases with frequency and further, it increases with increase in ion fluence. This modification [5, 9] in the values of dielectric constant, dielectric loss and conductivity with ion fluences can be attributed to degradation process taking place inside the polymeric chains due to nitrogen ion implantation. When a polymeric sample is irradiated with ion beams, the process of scissioning and crosslinking of molecular chains takes place. These processes lead to the formation of π-electron clouds [5, 9], with the polarization aligned in the direction of the molecular chains. This chain in the direction of polarization lowers the dielectric constant in the direction perpendicular to the plane of paper.

The conductivity in the polymer is governed by the mechanism of hopping charge carriers. The number of sites available for hopping increases with implantation dose because large number of amorphous phase regions can be produced due to ion-polymer interactions. The number of electrons that can participate in the transition process increases with ion dose. As a result dielectric loss which is associated with electron transition also increases with increase in ion fluence. 4. Conclusion

It is concluded that the dielectric constant has been reduced as a result of 100 keV N+ implantation with increasing ion fluence in CR-39 polymer. Further dielectric loss and a.c. conductivity show an increasing trend with


increase in ion fluence. These measurements provide additional evidence as to how the electronic properties of polymers are modulated by ion implantation. It is believed that the study of electronic properties of such materials gives new evidences of the real degradation mechanism. References [1.] E. Yap, D. G. McCulloh and D. R.

McKenzie, M. V. Swain, L.S. Wielinski, and R. A. Clissold, Journal of Applied Physics, 83 (1998) 3404.

[2.] T. Sharma, S. Aggarwal, A. Sharma, S. Kumar, D. Kanjilal, S. K. Deshpande and P. S. Goyal, J. Appl. Phy., 102 (2007) 063527.

[3.] D. Fink, F. Hosci, H. Omichi, T. Sasuga, L. Amaral, Radiat. Eff. Deffects Solids, 132 (1994) 313.

[4.] T. Phukan, D. Kanjilal, T. D. Goswami, H. L.Das, Nucl. Inst. Meth. Phys. Res. B, 155 (1999) 116.

[5.] P. S. Alegaonkar and V. N. Bhoraskar, P. Balya and P. S. Goyal, Applied Physics Letters, 80, 640-642 (2002).

[6.] S. A. Nouh and A. S. Abdel-Naby, Radiat. Eff. Deffects Solids, 158 (2003) 553.

[7.] S. Eren San, J. Radiol. Prot., 25 (2005) 93. [8.] Tu De-Min, Xi Bao-Feng, Lin Rong-Sheng,

Wu Hong- Cai. Proc. of the 3rd International Conference on “Properties and Applications of Dielectric Materials”, Vol. 2, (1991)1076.

[9.] M. O. Aboelfotoh and C. Feger, Phys. Rev. B, 47 (1993) 13395.

Dielectric Relaxation of N, N-dimethylacetamide in Benzene Solution from Microwave Absorption Studies

Raman Kumar and V S Rangra

Electronics Research Laboratory, Department of Physics, Himachal Pradesh University, Shimla-05, India


Abstract

The dielectric constant(ε′) and dielectric loss(ε″) for dilute solution of N,N-dimethylacetamide (DMA) in benzene solution has been measured at 9.90GHz at different temperatures(250C, 300C, 350C,40 )C° by using standard standing wave microwave techniques. Following the Gopala Krishna’s single frequency concentration variational method, the dielectric relaxation time(τ ) and dipole moments(μ ) at different temperatures have been calculated. It is found that the dielectric relaxation process can be treated as the rate process, just like the viscous flow process. Based on the above studies, monomer structure of N,N-dimethylacetamide (DMA) in benzene solution has been inferred. The presence of solute-solvent molecular associations in benzene solution has been proposed. The energy parameters(ΔHε, ΔFε, ΔSε) for the dielectric relaxation process of N,N-dimethylacetamide (DMA) in benzene at different temperatures(250C, 300C, 350C,40 )C° have been calculated and compared with the corresponding energy parameters (ΔHη, ΔFη, ΔSη) for the viscous flow. 1. Introduction N,N-dimethylacetamide (DMA) is recognized as the non-aqueous dipolar, aprotic solvent having dielectric constant ε′ = 37.78[1] and dipole moment μ =3.81D[1]. The dimethyl compound has good solvent power for the polymers and co-polymers used in the spinning of artificial fibers, so most of the industrial applications have been along these lines[2]. This molecular aspect of the DMA motivated the authors to study the molecular behaviour of DMA in benzene solution. Dielectric relaxation data obtained from microwave absorption studies is expected to throw light on various types of molecular associations present in the solution, because of the capability of microwave to detect the weak molecular interactions. Dielectric relaxation studies of polar molecules in non-polar solvents using microwave absorption techniques have been frequently attempted by various researchers[3-5]. This paper is concerned with the dielectric relaxation studies of DMA in benzene at different temperatures(25, 30, 35 and 40°C) from microwave absorption studies at 9.90GHz. 2. Experimental Details N,N-dimethylacetmamide (A. R. Grade) from

Central Drug House (P) Ltd., Mumbai (India) was dried with 4A0 molecular sieves for 48h, respectively with occasional shakings and then distilled through long vertical fractionating column. The middle fractions were collected for use. Benzene (Central Drug House Pvt Ltd, New Delhi) was dried by refluxing over freshly cut sodium metal for 6-8 hours and then distilled through a long vertical fractionating column. The middle fraction of the distilled benzene was used. The X-band microwave bench was used to measure wavelength in the dielectric medium and voltage standing wave ratio(VSWR) using a short-circuiting plunger. The set up was tuned at microwave frequency 9.90GHz. The experimental techniques of Arrawatia et al used by Sharma & Sharma[6] for microwave measurements were used. A set of dilute solutions of DMA in the benzene solution was prepared and all the measurements were carried out at 250C, 300C, 350C and 400C by circulating water around the dielectric cell through a thermostat (LAUDA DR R WOBSER GMBH & CO. KG German made). Measuring the dipole moments of purified acetone, methanol and pyridine tested the precision and working of the equipment. The viscosity and densities of the solutions at various temperatures were measured by Ubbelohde viscometer and sealable type of pycnometer, respectively.


3. Result and Discussion Using standard standing wave microwave techniques and following the method of Heston et al.,[7] the dielectric constant(ε′ ) and the dielectric loss(ε″) for the dilute solutions of DMA in benzene have been calculated and given in table 1. Following equations have been used

22

⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=′

d

o

c

o

λλ

λλ

ε (1)

Table 1. Values of dielectric constant(ε′ ), dielectric loss(ε″ ), relaxation time(τ) and dipole moment(μ) for DMA in benzene solution at different temperatures. _________________________________ Temp. Weight ε′ ε″ τ μ (0C) fraction of (10-12sec) (D) solute in benzene _______________________________________ 25 .00278 2.310 0.0145 3.47 3.98 .00525 2.352 0.0235 .00648 2.380 0.0305 .00739 2.398 0.0341 .00956 2.443 0.0451 30 .00278 2.304 0.0134 3.39 3.91 .00525 2.345 0.0226 .00648 2.373 0.0291 .00739 2.388 0.0317 .00956 2.428 0.0415 35 .00278 2.293 0.0126 3.31 3.86 .00525 2.334 0.0217 .00648 2.362 0.0278 .00739 2.373 0.0300 .00956 2.413 0.0386 40 .00278 2.283 0.0119 3.24 3.78 .00525 2.327 0.0203 .00648 2.348 0.0256 .00739 2.362 0.0270 .00956 2.398 0.0359

⎟⎠⎞

⎜⎝⎛

⎟⎟⎠

⎞⎜⎜⎝

⎛=′′

dnd

d

g

d

o ρλλ

λλ

πε .2

2

(2)

where λo, λc, λg and λd are the wavelengths in free space, the cut-off wavelength, the waveguide wavelength and the wavelength in the waveguide filled with solution respectively. “ρ” is the inverse of voltage standing wave ratio(VSWR) and dndρ is the slope of the curve of ‘ρ’ versus n, where n is the integer (1,2,3,4,……) such that ( )2dnλ represents the length of the dielectric filled waveguide. The ε′ and ε″ values were reproducible within 5.0± % and 67.1± % respectively. Following the Gopala Krishna’s single frequency concentration variational method[8], the dielectric relaxation time (τ ) and the dipole moment(μ ) have been calculated.

( ) 22

22

22

εεεεε

′′++′−′′+′+′

=X (3)

( ) 2223

εεε

′′++′′′

=Y (4)

⎟⎠⎞

⎜⎝⎛=

dXdY

co

πλτ2

(5)

dWdX

dXdY

NdKTM

⎥⎥⎦

⎤

⎢⎢⎣

⎡⎟⎠⎞

⎜⎝⎛+=

2

0

2 149π

μ (6)

where d0 is the density of the solvent; M and W are the molecular weight and the weight fraction of DMA in the solution, respectively. K is the Boltzman’s constant and T is the temperature. N is the Avogadro’s number. The plot of ε′, ε″ versus weight fraction of DMA in benzene solution are found to be linear. The linear variations of ε′, ε″ ensures the applicability of Debye theory and hence, that of Gopala Krishna’s method for calculating relaxation time(τ ) and dipole moment(μ ) of polar substance DMA in non-polar solvents.

The energy parameters(ΔHε , ΔFε , ΔSε) for the dielectric relaxation process of DMA in benzene at 250C, 300C, 350C and 40OC and the corresponding energy parameters(ΔHη , ΔFη , ΔSη) for the viscous flow have been calculated by using Eyring et al.,[9] relations for the rate process. Following relations were used:


⎟⎠⎞

⎜⎝⎛ Δ=

RTF

kTh ετ exp (7)

εεε STF Δ−ΔΗ=Δ (8)

⎟⎟⎠

⎞⎜⎜⎝

⎛ Δ=

RTF

VhN ηη exp (9)

ηηη STF Δ−ΔΗ=Δ (10)

where V is the molar volume of the solvent and all other symbols have their usual significance. ΔHε, ΔFε and ΔSε are the enthalpy, free energy and entropy of activation respectively for dielectric relaxation process and ΔHη , ΔFη and ΔSη are corresponding parameters for the viscous flow. The plot of ( )Tτlog versus T310 and

( )ηlog versus T310 according to equations (7) and (9) were found to be linear, which shows that both relaxation and viscous processes can be considered as rate processes (Figure 1).

3.18 3.20 3.22 3.24 3.26 3.28 3.30 3.32 3.34 3.36

-8.994

-8.992

-8.990

-8.988

-8.986

-8.984

I/T x 103

log(τT

)

Fig.1. Plot between log(τ T) and 103/T.

The ΔHε and ΔHη values were computed from the slope of the linear plot of ( )Tτlog

versus T310 and ( )ηlog versus T310 respectively using the relation slope= RH 303.2Δ , where R is the gas constant. Both sets of energy parameters along with the dielectric relaxation time (τ ) of DMA in benzene solution at 250C, 300C, 350C and 400C have been summarized in Table 2.

Table 2. Relaxation time(τ), Free energies of activation(ΔFε, ΔFη in kJ mole-1), enthalpies of activation(ΔHε, ΔHη in kJmole-1) and entropies of activation(ΔSε, ΔSη in J mole-1 deg-1K-1) for DMA in benzene solution. _______________________________________ Temp. τ ΔFε ΔHε ΔSε ΔFη ΔHη ΔSη (0C) 25 3.47 7.60 1.02 -22.11 12.18 10.97 -4.06 30 3.39 7.71 1.02 -22.11 12.21 10.97 -4.09 35 3.31 7.82 1.02 -22.10 12.24 10.97 -4.11 40 3.24 7.94 1.02 -22.11 12.30 10.97 -4.25

The calculated value of dipole moment of DMA in benzene solution is found to be close to the literature value of the unassociated DMA molecule. This shows that, DMA exists in monomer form in benzene solution. It is interesting to note that, in the benzene solution, the dipole moment of DMA is found to decrease with the rise in temperature and approaches towards the literature value. This small variation in dipole moment value with rise in temperature may be attributed to the possible solute-solvent association. It is proposed that, solute-solvent association arises because of the interaction of fractional positive charge at the sight of N-atom of the DMA molecule and the π-delocalized electron cloud of the benzene ring of the benzene molecule is shown in Figure2.

C

C CH

H

H

O

N3

3

3C

δ+

Fig. 2. Solute-solvent association of DMA molecule in benzene.

Energy parameters for the dielectric relaxation process of DMA in benzene present an interesting behaviour. It is found that the free energy of activation(ΔFε) for the dielectric relaxation process is less than the free energy of activation(ΔFη) for the viscous flow process. This may be explained on the basis that the dielectric relaxation process involves the rotation


of molecular entities whereas in the viscous flow process, the rotational as well as the translational motion of the molecules is involved. Enthalpy of activation(ΔHε) for the dielectric relaxation process is less than the enthalpy of activation(ΔHη) for the viscous flow process. The enthalpy of activation depends upon the local environment of the molecules. Different values for the enthalpy of activation indicate that the dielectric relaxation process and viscous flow process involve the breaking of bonds with the neighbouring molecules in a different way and to a different extent. The entropy of a system is the measure of the orderly nature of the system. If the environment of the system is co-operative for the activated process, then the change in entropy(ΔSε) becomes –ve. Whereas the +ve value of the change in the entropy (ΔSε) for activated process indicates the non-cooperative environment of the system and the activated state is unstable. In the present case, it is observed that the change in entropy of the dielectric relaxation process is -ve, indicating that the environment of the system is co-operative like that of the activated viscous flow state.

References [1] A. K. Covington, T. Dickinson, Physical

Chemistry of Organic Solvent Systems, Plenum Press, London and New York, 1973.

[2] J. J. Lagowski, The Chemistry of Non- aqueous solvents, Academic Press, New- York and London, 1967.

[3] R. Kumar, V. S. Rangra, D. R. Sharma, N. Thakur, N. S. Negi, Z. Naturforsch. 62a (2007) 213.

[4] V. S. Rangra, D. R. Sharma, Indian J. Phys. 78B (2004) 111.

[5] T. Sato, R. Buchner, J. Chem. Phys. A 108 (2004) 5007.

[6] A. K. Sharma, D. R. Sharma, J. Phy. Soc. Japan. 53 (1984) 4471.

[7] W. M. Heston (Jr.), A. D. Franklin, E. L. Hennely, C. P. Smyth, J. Am. Chem. Soc. 72 (1950) 3443.

[8] K. V. Gopala Krishna, Trans. Faraday Soc. 53 (1957) 767.

[9] H. Eyring, S. Glasstone, K. J. Laidler, Theory of Rate Process, Mcgraw-Hill, New York, 1941.

Mixed Electronic-Ionic Conductivity in Copper Phosphate Glasses Doped with Sodium Oxide

P. S. Tarsikka and B. Singh*

Department of Mathematics, Statistics and Physics, Punjab Agricultural University, Ludhiana-141004. *Department of Physics, Punjabi University, Patiala-147002.

Email: pstarsikka007@yahoo. co. in

Abstract

Copper phosphate glasses behave as semiconductors, exhibiting electronic conduction by electron hopping from Cu+ to Cu++ ions. On the other hand, electrical conductivity of alkali containing copper phosphate glasses consists of both, electronic and ionic conduction. The aim of the study was to investigate electrical properties of the glasses of ternary system Na2O-Cu2O-P2O5 exhibiting, in general, mixed ionic-electronic conduction with predominance of ionic or electronic component depending on composition. The conductivity of four glass samples in 20%Na2O-(80-x)%Cu2O-x%P2O5 systems and two glass samples in 60%Na2O-40%P2O5 and 60% Cu2O-40%P2O5 systems was investigated as a function of temperature. In pure copper-phosphate glasses the conductivity is slightly lower than that for pure sodium phosphate glasses. The conductivity containing 20% fixed amount of sodium ions concentration decreases with the increase of copper ions concentration up to 30%Cu2O. It was found that conductivity for glasses containing ≤ 30% Cu2O is predominantly ionic, whereas glasses having Cu2O concentration above 30% exhibit predominant electronic conduction and is controlled by electron hopping between copper ions. In these glasses the sodium ions have low mobility, caused by ion-polaron interaction. For pure sodium phosphate glass or the glass containing lower concentration (up to 30%) of copper the plot of logσdc vs. 1000/T is linear, whereas for glasses having Cu2O concentration above 30 % the plots depart from linearity. These non-linear plots indicate a change in activation energy which suggests a contribution from electronic conduction. Ionic glasses exhibit only one conduction process and single activation energy in the whole temperature range. In transition metal oxide glasses different conduction processes contribute to electronic conduction at different temperature ranges leading to different values of activation energy. 1. Introduction

The electrical properties of semiconducting glasses have been of considerable interest in recent years. Transition metal oxide glasses having two different valence states of transition metal ions behave as electronic conductors. The loss of oxygen from the melt produces lower valence transition metal ions. The Conductivity in these glasses is described by small polaron hopping between such ions [1-2]. Several TMI glasses V2O5-P2O5 [3,4], V2O5-TeO2 [5], V2O5-B2O3[6], CuO-P2O5[7], Fe2O3-P2O5 [8] and WO3-P2O5 [9] have been studied and they exhibit electronic conduction. Electrical conductivity of alkali oxide glasses is known to be ionic [10,11]. When an alkali oxide is added, mobile alkali ions will contribute to the charge transport and mixed conductivity is observed. Generally, ionic conduction depends on the

alkali ion concentration and alkali ion mobility. Recently mixed electronic-ionic conductivity have been studied in vanadate oxide glasses containing alkaline ions and in iron-phosphate glasses containing sodium ions [12,13]. The aim of the present study was to investigate electrical properties of the molybdenum phosphate glasses containing sodium ions, exhibiting in general, mixed ionic and electronic conduction with predominance of ionic or electronic component depending on composition. 2. Experimental

Six samples of various compositions of sodium-cuprous phosphate glasses were prepared having the molar concentrations given in Table 1. Glass samples were prepared by heating the appropriate mixture of Cu2O, NaNO3 and P2O5 in a platinum crucible in air for 1 hour at


11000C. The melt was quenched on a brass plate preheated to about 1500C. X-ray diffraction studies showed that the samples were amorphous. Rectangular samples having thickness of about 1 mm were shaped by grinding and polishing with emery powder of 300 grades. Samples were given a coating of silver using conducting silver paint. Samples were then annealed at the temperature of 200 0C for 2 hours to stabilize the contacts, Linear V-I characteristics confirmed Ohmic contacts. A two electrode system was used for the dc conductivity measurements. DC conductivity was measured by Keithley 617 programmable electrometer. Temperature was recorded using a Newtronic S-96 temperature controller. Density of the samples was determined by the displacement method. 3. Results and Discussion

Figure 1 shows the plot of log d.c. x T against 1000/T for all the six samples. All the glass samples show a smooth variation of conductivity with temperature. It is clear from the plots that in the glasses containing 20mol%Na2O the electrical conductivity decreases with the increase of Cu2O concentration up to 30mol% and with further increase of Cu2O concentration the electrical conductivity increases. Also the electrical conductivity of pure sodium phosphate glass is more than pure copper phosphate glass. In alkali free transition metal oxide (TMO) glasses the conductivity increases if concentration of TMI increases. It is expected that the conductivity of alkali containing TMO glasses will consist of a mixture of ionic and electronic conduction. Ionic conduction depends on the alkali ions concentration and their mobility. Assuming that the motion of alkali ions and electrons is independent, the electrical conductivity should increase with an increase in alkali content. But such a behaviour is not observed in present study. Similar behaviour has been observed in sodium containing iron phosphate glasses [14]. It suggests that, unlike to typical mixed electronic-ionic conductive glasses, the sodium ions in copper phosphate glasses have lower mobility.

In Figure 1 the plots are linear for Cu2O concentration ≤ 30mol%, where as plots deviate from linearity for Cu2O > 30mol%. Nonlinear

plots indicate the temperature dependence of activation energy. Activation energy have been evaluated from the slopes of the plots in Figure 1. The nature of the temperature dependence of the activation energy and the fact that the activation energy at 450K for glasses having Cu2O concentration > 30mol% is less than that for the glasses having Cu2O concentration ≤ 30mol% indicate that the glass samples having Cu2O concentration ≤ 30mol%exhibit predominant ionic conduction, where as samples having Cu2O concentration above 30mol% exhibit predominant electronic conduction. This switchover from predominant ionic conduction to electronic conduction in Sodium- cuprous phosphate glasses occurs for Cu2O concentration > 30mol%.

Fig.1 The log of d.c. conductivity times temperature of all investigated samples versus reciprocal temperature

-10.5

-9.5

-8.5

-7.5

-6.5

-5.5

-4.5

2 2.5 3 3.5 4 4.5 5 5.5 6

1000/T [K-1]

log

d.c.

x T

[ohm

-1cm

-1K

]

sample1

sample2

sample3sample4

sample5

sample6

The transport of electrons in transition metal

oxide glasses is usually termed as small polaron hopping. A general formula for the electrical conductivity was proposed by Mott [15] in which the conductivity is given by

⎟⎠⎞

⎜⎝⎛ −−−=

KTWR

KTRe

cc eldc exp)2exp()1(

2

αν

σ

where α is the rate of the wave function decay, νel ≈1015 s-1 is the electronic frequency, C is the ratio of ion concentration in the low valence state to the total concentration of transition metal ions, N is the number of transition metal ion sites/cm3, R is the average site spacing and W is the

Mixed Electronic-Ionic Conductivity in Copper Phosphate Glasses Doped 91

activation energy arising from the electron-lattice interaction. The activation energy W is the sum of polaron hopping energy WH and disorder energy WD which might exist between the initial and final sites due to the variation in the local arrangements of ions. Austin and Mott [16] have shown that

W = WH + ½ WD for T > θD/2 W = WD for T < θ D/4 Values of activation energy W have been evaluated from the slopes of plots of log d.c.x T vs. 1000/T. Values of W are given in table 1. Values of electrical conductivity at 488 K and those of activation energy (W) are presented in Table 1.These values show that the conductivity in general tends to be lower in glasses with higher activation energy. The values of W decreases with the decrease of nearest neighbour cuprous ion separation. Table 1. The electrical properties of Na2O-Cu2O-P2O5 glasses. S* No

(mol%) 1*-2*-3*

R (oA)

d.c. conductivity (Ω-1 cm-1) at 488K

W (eV)

rp

(0A)

1 2 3 4 5 6

20-05-75 20-15-65 20-30-50 20-40-40 00-60-40 60-00-40

6.83 4.87 3.86 3.31 2.78 --

4.07x10-9 1.74x10-9 5.60x10-10 3.47x10-8 4.57x10-8 1.21x10-7

1.02 1.09 1.41 0.56 0.53 0.84

2.75 1.96 1.48 1.34 1.20 ---

S*- sample, 1-Na2O, 2-Cu2O, 3-P2O5 The large activation energy may suggest that conductivity is dominated by ionic transport. Values of R derived from the relation N = 1/R3 are also given in table1. Using this value of R, the polaron radius rp has been evaluated assuming a non–dispersive medium and using the simplified expression [10] rp = R/2(π /6)1/3

Values of rp are given in table 1.These values are less than 1.5 oA (for the samples 3-6) as expected for small polarons, suggesting that the polaron is strongly polarized and, therefore, the electronic conduction takes place by small polaron hopping.

4. Conclusions

The d.c. electrical conductivity of sodium copper phosphate glasses were investigated. The results show that the contribution of electronic conduction to d.c. conductivity increases with copper ion concentration. Whereas in glasses containing higher concentration of sodium ions the ionic conduction is predominance over electronic conduction. Polaron is strongly polarized and, therefore, the electronic conduction takes place by small polaron hopping. References [1] L. Murawski, C.H.Chung and J.D.

Mackenzie, J. Non-Cryst. Solids. 32(1979)91.

[2] M.Sayer, A.Mansingh, Non-crystalline Semiconductors, Vol. III, M.Pollak (Ed.) CRC Press Boca Raton, FL.USA,1987.

[3] M. Sayer and A. Mansingh, Phys. Rev..6B(1972) 4629.

[4] A. Mansingh, J.K. Vaid and R.P. Tandon, J. Phys. 8C (1975) 1023.

[5] A.Mansingh, V.K. Dhawan and M. Sayer, Philos. Mag 48B (1983) 215.

[6] B.K. Sharma, D.C. Dubey and A. Mansingh, J.Non-Cryst. Solids 65 (1984) 39.

[7] A. Dhawan, J.R. Jurado and J.M.F. Navarro, J. Non-Cryst. Solids 79 (1986) 353.

[8] K.W. Hansen and M.T. Splann, J. Electro.Soc. 113 (1966) 895.

[9] A. Mansingh, R.P. Tandon and J.K. Vaid, Phys.Rev.21B (1980) 4829.

[10] H. Nasu and N. Saga, J. Non-Cryst. Solids 53(1982) 123.

[11] M.D. Ingram, Phys. Chem. Glasses 28 (1987) 215

[12] R .J. Barczynski and L. Murawski, Material Sci.- Poland 24(1) (2006) 221.

[13] B. Kusz, K. Trzebiatowski and R.J. Barczynski, Solid State Ionics 159(2003) 293.

[14] L. Murawski, R. J .Barczynski and D. Samatowics, Solid State Ionics 157 (2003) 293

[15] N.F. Mott, J. Non-Cryst. Solids 1(1968) 1. [16] I.G. Austin and N.F. Mott, Adv. Phys.

18(1969)41.

Properties of Alumina Particulate Reinforced Aluminum Alloy Produced by Stir Casting

Ravindra Mamgain1 , H. S. Bains,2 A. Manna3 and S. N. Basu1

1Deptartment of Industrial & Production Engineering, Dehradun Institute of Technology Dehradun, Uttranchal.

2Deptt.of Mechanical Engineering, Sant Longowal Institute of Engineering and Technology Longowal, Sangrur, Punjab.

3Deptt.of Mechanical Engineering, Punjab Engineering College (Deemed University), Chandigarh.


Abstract

In this study, metal-matrix composites of aluminum (Al-6063) and alumina (Al2O3) particles with volume fraction of o.05, 0.10 and 0.15 were produced using stir casting technique and the microstructure and mechanical properties of the composite investigated. The stirring process was carried out at a speed of 220 rev per minute with a graphite impeller for 30 minute. Al2O3 particles in the composite exhibited a reasonably homogeneous distribution and were well wetted by aluminum. Hardness, tensile and impact strength were examined. The hardness of the composites increased with increasing particle volume fraction. The tensile strength of the composites decreased with increasing particle volume fraction. Similarly the impact strength of the composites decreased with increasing volume fraction.

1. Introduction

Metal matrix composite (MMC) are widely used in industries because of their excellent mechanical properties and wear resistance. Wear behavior of MMC materials is an interesting area of focus chiefly because relatively soft alloy such as aluminum can be made highly resistant to wear by introducing a predominantly hard but brittle particles, such as silicon carbide (SiC) or aluminum oxide (Al2O3) [1] . Aluminum-silicon alloys, as a matrix material, are characterized by light weight, good strength-to-weight ratio, ease of fabrication at reasonable cost, high strength at elevated temperature, good thermal conductivity, excellent castability, good weldability, excellent corrosion resistance and wear resistance properties [2, 7]. Most of the fabrication methods for the MMC have employed stirring to create a vortex. The negative pressure differential existing at the vortex helps to suck externally added particles into the liquid metal. How ever the vortex also sucks in air bubbles in the particle-melt slurry, resulting in large porosities in cast composites[3]. Broadly, there are two types of foundry methods for making composites with externally added particles, depending on the temperature at which the particles are introduced into the melt. In the liquid metallurgy process, the particles are added above the liquid temperature of the molten alloy, whereas in

compo casting the particles are introduced into the semisolid slurry temperature alloy [4, 8]. With regards to the material factors (such as volume fraction, type of reinforcement and size of reinforcement), the volume fraction of reinforcement (Vf) has the strongest effect on the mechanical properties of composites [5, 6]. Our focus in this study is to produce particulate reinforced aluminum-alumina matrix composites using stir casting method and also investigate the effect of the particle volume fraction on the mechanical properties of composites. 2. Experimental details

In the present investigation Al 6063 alloy is taken as the matrix material and Al203 of size 80µm as a reinforcement material. The chemical composition of Al (6063) alloy is shown in table 1. The studies were carried out using 5, 10, 15, vol% Al2O3/6063 aluminium based metal matrix composite.

Table 1. Chemical composition of Al (6063) alloy Element Si Mn Mg Cu Fe Ti Al

Al 6063 0.44 0.07 0.6 0.018 0.2 0.008 Rest

Properties of Alumina Particulate Reinforced Aluminum Alloy Produced by Stir Casting 93

3.1 Specimen preparation

In this experimentation work the Matrix alloy used is an Al-Mg-Si wrought alloy matrix (6063) reinforced with Al203 Particles of size 80µm. Commercial Al-6063 ( Al-98.52% Mg-0.649 Si-0.445 ) alloy reinforced with 5,10& 15Vol %. Al203 with a size of 80μm was used .The Matrix alloy was first melted in a graphite crucible in an electric furnace. The Matrix alloy (6063) was preheated at 300 °C for 1-2 hours before melting, and before mixing the Al203 particles was preheated at 300°C for 1 hour to make the surface of Al203 particle oxidized. The furnace temperature was first raised above the liquidus temperature to melt the alloy completely at 750°C and was then cooled down just below the liquidus temperature to keep the slurry in a semi solid state. At this stage the preheated Al203 particle were added & mixed manually. The mixing was done for a short time period of 1 to 1.5 minutes. The composite slurry was reheated to a fully liquid state and then automatic mechanical mixing was done for about 30 minutes at stirring rate of 220 rpm. In this experiment, the molten composite was transferred from the crucible into the mild steel mould.

Fig. 1. Stir casting set up

Fig.2. Mild steel mould

Fig. 3. Fabricated Al 6063/Al2O3 3.2 Testing

In present work the three composites, one by addition of 5%alumina, second by 10% alumina and other by 15% addition of alumina have been cast by stir casting technique and their mechanical properties like tensile strength, impact strength and hardness have been determined. (a) Tensile strength

The tensile strength of Al6063 base alloy, Al6063 +5% alumina, Al6063 +10%alumina and Al6063 +15% alumina by vol. was measured at room temperatures. The experiment result for different composition of alumina i.e. Al6063 base alloy, Al6063 +5%alumina, Al6063 +10%alumina and Al6063 +15%alumina by vol. percent are given in table 2 and table 3. Table 2. Variation of UTS with different alumina additions

Composition UTS Mpa Al 6063 base alloy 117.5 Al 6063 +5%alumina (Al2O3) 109.4 Al 6063 +10%alumina (Al2O3) 77.0 Al 6063 +15%alumina (Al2O3) 51.0

The UTS for Al 6063 base alloy, Al 6063

+5%alumina, Al 6063 +10%alumina and Al 6063 +15%alumina by vol. is 117.5 Mpa , 109.4Mpa, 77.0Mpa and 51.0 Mpa respectively, there is a decrease in UTS at room temperature for 5%,10% and 15% addition of alumina by vol. by 8.5 Mpa , 32.4 Mpa and 26 Mpa respectively. The tensile strength variation with extension (elongation) for different composition of alumina at room temperature are shown in tabular form.


Table 3. Variation of extension with different alumina addition at room temp.

Composition extension@ max %

extension@ break %

Al 6063 base alloy 11.13 11.67

Al 6063 +5%alumina

(Al2O3) 2.623 2.650

Al 6063 +10%alumina

(Al2O3) 3.955 4.777

Al 6063 +15%alumina

(Al2O3) 3.038 3.636

(b) Impact strength

Table 4 shows three consistent experiment result of impact strength for the different compositions of alumina. The impact strength for base alloyAl6063 is 1.480 Kgm/cm2. It has decreased marginally to -0.99181 Kgm/cm2 for 5% addition, impact strength further decreases to 0.6176 Kgm/cm2 for composite Al 6063 for 10 % and 0.18564 Kgm/cm2 for 15%. For 5 % addition there is no significant decrease in impact strength but with 10% and 15% addition the decrease is more as compared with 5% addition of alumina, this may be due to the brittle nature of the particles and the more segregation of the particles at some specific places. Table 4. Variation of impact strength with Alumina content

Composition Impact strength Al 6063 base alloy 1.480 Kgm/cm2 Al 6063 +5%alumina (Al2O3)

0.99181 Kgm/cm2

Al 6063 +10%alumina (Al2O3)

0.6176 Kgm/cm2

Al 6063 +15%alumina (Al2O3)

0.18564 Kgm/cm2

(c) Hardness

The hardness of Al 6063 base alloy is as low as 39 BHN and with 5% addition of alumina it increases up to 47BHN and further with 10% and 15% addition of alumina it reaches at 61BHN and 73BHN. The hardness of the composite increases with increase in vol. percent of alumina reinforced in the alloy. The hardness of the composite increases because hard nature of particles. With the 5% addition of particles the

hardness increased by 8BHN and with 10% and 15% addition it increases by 22 BHN and 34 BHN. (d) Microstructure

The result of microstructure investigation of 5, 10, 15 vol % Al (6063)/Al2O3 metal matrix composites are shown in figs. Significant microstructure variation can be seen in all the three specimens. More uniform distribution of particulate in the matrix can be seen in Al(6063) 5 vol % specimen while clustering of particulate is observed in each specimens. At high volume fraction particulate interact with each other increasing, in the process, setting velocity. This result in uneven distribution of particulate. The cluster formation takes place during mixing and stirring process. The rapid solidification rates in permanent mild steel mould inhibit cluster formation at solidification stage. 4. Results and discussion

The mechanical properties have been investigated for metal matrix composites of three compositions i.e. Al 6063 +5%alumina, Al 6063 +10%alumina and Al 6063 +15%alumina by vol. and compared with the base alloy Al 6063.

Fig. 4. Al (6063)5% Al2O3

Fig. 5. Al (6063) 10% Al2O3

Properties of Alumina Particulate Reinforced Aluminum Alloy Produced by Stir Casting 95

Fig. 6. Al (6063) 15% Al2O3

The ultimate tensile strength at room temperature for Al 6063 base alloy, Al 6063 +5%alumina, Al 6063 +10% alumina and Al 6063 +15% alumina by vol. are 117.5 Mpa, 109.4 Mpa, 77.0 Mpa and 51.0 Mpa respectively, there is a decrease in UTS by about 8.5Mpa, 32.4 Mpa and 26 Mpa for 5%, 10% and 15% addition of alumina by vol. in Al 6063 alloy respectively. The variations of ultimate tensile strength for different compositions of alumina at room temperature are shown in table 4. The ultimate tensile strength of the composites decreases with addition of alumina. The decrease in ultimate tensile strength may be due to the segregation of particles at some specific zones.

The second reason for decrease in tensile

strength may be due to the presence of the interfacial gaps between the matrix and the reinforcement, which is unable to transfer the load from the matrix to reinforcing phase. The hardness of the composites increases with the addition of alumina. Hardness of the Al 6063 base alloy is 39 BHN, with the addition of 5% alumina it increases to 47 BHN and with addition of 10% and 15% it increases to61 BHN and 73 BHN. The hardness of the Composite increases because hard nature of particles. With the 5% addition of particles the hardness increases by 8 BHN and with 10% and 15% it increases by 22 BHN and 34 BHN. This increase in hardness is attributed of the hard nature of particles as compared to base alloy.

The result show the average value of hardness there are variations in the hardness observed for same surface of composite, this may be due to the difference in the distribution of the alumina particles. The impact strength for the base alloy is 1.480 Kgm/cm2, for the addition of 5% alumina it decreases to 0.99181 Kgm/cm2 and with the addition of 10% and 15% it reduces to 0.6176 Kgm/cm2 and 0.18564 Kgm/cm2. For 5% addition there is no significant decrease in impact strength but with 10% and 15% the decrease is more as compared with 5% addition of alumina, this may be due to the brittle nature of the particles and more segregation of the particles at some specific places. References [1.] Tylore and Francis, Mechanics of

composite Material. [2.] T.W.Clyne and P.J.Withers. An

introduction to metal matrix composites. [3.] J.Hashim,L.Looney and M.S.J Hashmi,

“Particle distribution in metal matrix composites” J. of Material Processing Technology 92-93 (1999)1.

[4.] L.V.Ramanathan, P.C.R Nunes, “Effect of liquid metal processing parameters on microstructure and properties of alumina reinforced Al base MMC” Proc.of the 12th int Symp. On Material science Metal matrix Composites, Sept. 2-6, (1991) 611-616.

[5.] Foo KS, Banks WM, Craven AJ, Hendry A., “Interface characterization of an SiC Particulate/6061 aluminium alloy composites

[6.] K.M Shorowordi, T Laoui “Microstructure and interface characteristics of B4C, SiC and Al2O3 reinforced Al matrix composites: a comparative study” J. of Materials Processing Technology 142 (2003) 738-743

[7.] G.Guo, P.K.Rohatgi and S.Ray:’Aluminum composites for automotive applications’ Trans. Am. Foundarymen’s Soc. 19.

[8.] William F.Smith “Principal of material Sience and engineering”Vol.2, Issue 1, 1985, 478.

The Study of Polypyrrole Polymer Films for Sensing and Microwave Properties

D C Tiwari, Rishi Sharma and Vikas Sen

School of Studies in Electronics, Jiwaji Univeristy, Gwalior-474001 (MP), India. [email protected]

Abstract

Polymer samples were prepared by plasma and chemical polymerization technique. The stability of the polymer was enhanced by the doping of the polymer film by various dopants. The polypyrrole material was used for sensing property on various gases like Ammonia, Methanol, and Acetone. Remarkable change in the conductivity of polypyrrole was seen by the exposure of the gases at different concentration. Microwave properties of these polymers were also studied at 9.06GHz by using newly designed balance type microwave setup. 1. Introduction Recent years have witnessed phenomenal attraction in the field of conducting polymers and their composite leading to their application in molecular electronics, biosensors, actuators, super capacitors, electro magnetic interference (EMI) shielding, microwave absorbing materials and light emitting diodes [1-7]. Attractive features of conductive polymers are its variety of availability, they can be formed electrochemically, plasma polymerization, spin coating and by simple chemical methods [8-10]. Another attractive feature is its operation at room temperature. Conducing polymers are becoming very important because of sensor applications.

Polypyrrole (PPy) is considered as one of the most promising conducting polymers it exhibits relatively high conductivity, good environmental stability and versatility of synthesis. The polymers are generally poor conductors; Polypyrrole has been used extensively for chemical sensing and they achieve high electrical conductivity by doping it with organic acids [11-13]. The enhancement of electrical activity of the polymer depends on the chemical reactivity of the dopant with the polymer. Meyers [14] reported PPy having a conductivity of 45 S/Cm obtained by treating it with anhydrous ferric chloride in aether at 220C for 1 hour. Pomposo et. al. [15] have reported effect of changes in conductivity by using electron withdrawing substituents and proved that it improves both electrical conductivity and thermal stability of the polymer. Teh and Limei [16] have reported micro gap sensor, based on Polypyrrole. Andrew et. al. [17] has reported

capacitive humidity sensors. Zhu et. al. [18] reported gas sensing on ruthenium porphyrin and observed marked increase in electrical conductivity by exposing to NO2 gas.

Conductive polymers are the new materials used for microwave applications, depending on their conductivity and dielectric constant. They can be tailored for microwave applications. . Knowledge of the dielectric properties of films / substrate at microwave frequency is helpful in material characterization for device fabrication and for other applications. Substrate in microwave integrated circuits and packaging systems have dielectric constant in the range of 2-10 ohms and dielectric losses (tanδ) lower than 10-3. Nowadays work is going on in the area of microstrip sensors for investigation of materials [19]. There are some research work is going on gas sensing using the microwave conductivity of conducting polymer thin films, which leads to the possibility of using conducting polymer films as remote wireless gas sensors [20]. Microwave sensors are very important for remote sensing of earth resources in the tropical region. The capability of microwave to penetrate the cloud and rain allows for monitoring of the earth in all weather day and night [21]. There is various reflection and perturbation based techniques reported for evaluation of dielectric properties [22-24]. There are many complications for measuring dielectric constant of thin films in reflection mode techniques.

2. Experimental Setup

Sample Preparation

First of all Pyrrole (Himedia) was double

The Study of Polypyrrole Polymer Films for Sensing and Microwave Properties 97

distilled and then mounted in a crucible inside the reactor and the reactor was continuously evacuated with the help of Hind-Hivac rotary oil pump. On passing the high voltage discharge a clear golden yellow colour film of pyrrole was deposited on the electrodes. The same film was then doped with perchlorate by dipping it in by HClO4 solution. The colour of the film changes from golden yellow to dark brown. The contacts to the film are taken with the help of silver paint. The contacts are dried and flexible wires are connected to the sample. Chemical route In this case we have also prepared samples by oxidative polymerization of PPy with ferric chloride. A blue coloured precipitate was filtered and washed repeatedly with distilled water and dried under vacuum, the PPy powder is then prepared in pellet form. Sample for microwave Samples were prepared in rectangular shape as per the dimensions of the waveguide. 3. Conductivity Measurement

The I-V characteristic of the polymer is measured by connecting the polymer film in the circuit. The doped film shows a semiconducting nature. The current increases with voltage initially and then saturates after some time.

Fig. 1. I-V characteristic of PPy 4. Sensing Application The plasma prepared samples were grown on tin oxide coated glass slide and contacts were taken by silver paste. And the films were exposed to the ammonia gas. It was found that there is remarkable variation in the resistance of the film which indicates that these films can be used for gas sensing applications.

4

6

8

10

12

14

16

0 5 10 15 20

Time(min.)

R (M

-ohm

s)

Fig. 2. PPy exposed to Ammonia gas

Fig. 3. PPy pellet is exposed to ammonia gas

Microwave Measurement

We are using the technique developed by Tiwai et al. [25] for measurement of dielectric constant of material at microwave frequencies. We are using the mathematical the relation which can be simplified by taking following approximation is ½ c ( 22 βε −k ) <1 and using ε = ( )εε ′′−′ j and )( βββ ′′−′= j

( )

⎥⎥⎦

⎤

⎢⎢⎣

⎡×

+

++′′−′=′

cdaaaaaa

k21

24

23

4231222 ββε

⎥⎦

⎤⎢⎣

⎡×

+−

−′′′=′′cdaa

aaaak

2212

42

3

41322 ββε

where

( )21

221 Re β−⋅= kda , ( )2

122

2 Im β−⋅= kda ,

010 20 30 40 50 60

0 100 200 300 400

I (μA

)

-5 0 5 10 15 20 25 30 35 4020

22

24

26

28

30

32

34

36

38

40

10ml NH37.5ml NH3

5ml NH3 Imic.amp.

I(μΑ

)

Time(min)


( ) ( ) 122

13 222exp2exp

22aCosaa

aSina+−+

= ,

( ) ( )( ) ( ) 122

224 222exp2exp

2exp2expaCosaa

aaa+−+−−

= ,

Where gg l

Shiftλπ

λπβ 2*2' += (shift introduced after

sample insertion)

k=0

2λπ , )(1

2

1

vvLn

l==′′ αβ , l = length of the

sample, 1v = voltage without sample, 2v = attenuated voltage after sample insertion.

Procedure of microwave measurements

In the initial step measurements are taken from the detector output voltage without sample in then sample in the holder. Then the sample is inserted in the holder on either left or right side, readings are taken. If the sample is grown on any substrate then the substrate of same dimensions will be placed as a reference on any one of the sample holder, and measurement are taken. The SWR plot is drawn and attenuation in the voltage and the shift in the standing voltage pattern are noted; by using above formula dielectric constant is calculated.

The dielectric constant for many known materials like sodalime glass, teflon and polymer materials, PPy and PMMA are also measured at the frequency 9.06 GHz. We have taken commercially available PVC material of varied thickness having same length and the dielectric constant is measured it was found to be consistent up to up to 2 decimal places after repeated measurements (table 1).

Table 1. Dielectric measurement of PVC

Sno. Thickness ε' ε" tan δ

1. 1.83mm 3.47 0.26 0.0749

2. 1.9mm 3.37 0.26 0.0771

3. 1.98mm 3.27 0.25 0.0764

4. 2.07mm 3.17 0.27 0.0851

5. Conclusion The polypyrrole acts as a gas sensing material also its was found that the dielectric constant of the PVC material matched with the standard data.

References [1] E. Smela, J. Micromech. Microeng. 9

(1999) 1. [2] C.O. Yoon, H.K. Sung, J, H. Kim., E.

Barsoukov, J. Hyun Kim, H. Lee, Synth. Met. 99 (1999) 201.

[3] C. Adachi,M. E. Thompson, S.R. Forrest, IEEE J. Sel. Top. Quantum Electron 8 (2002) 372.

[4] J. H. Sinfelt, Science 195 (1977) 641. [5] T. J. Sokothein (Ed.), Handbook of

Conducting Polymer, Mercel Dekker, New York, 1986.

[6] M. Brie, R. Turcu, C. Neamtu, S. Pruneanu, Sens. Actu. B 37 (1996) 119.

[7] N. M. White, J. D. Turner, Mes. Sci. Technol. 8 (1997) 1.

[8] H. K. Jun, Y. S. Huh, B. S. Lee, S. T. Lee, J. O. Lim., D. D. Lee, J. S. Huh, Sens. Actu. B 96 (2003) 576.

[9] E. Segal, R. Tchoudakov, /m. Narkis, A. Siegmann, Y. Wei, Sens. Actu. B 104 (2005) 140.

[10] J. H. Cho., J. B. Yu, J. S. Kim, S. O. Sohn, D. D. Lee, J. S. Huh, Sens. Actu. B 108 (2005) 389.

[11] E. Nilella, F. Musio, M. B. Alba, Thin Solid Films 284-285 (1996) 908.

[12] J. E. G. de Souza, F. L. dos Santos, B. B. Neto, C. G. Dos Santos, M. V. B. dos Santos, C. P. de Melo, Sens. Actu. B 88 (2003) 246.

[13] C.P. de Melo, B. B. Neto, E. G. de Lima, L.F.B. Lira, J.E.G. de Souza, Sens. Actu. B 109 (2005) 348.

[14] R. E. Meyers, J. Electron. Mater. 15 (1986) 61.

[15] Calvo P. A., Rodriguez J., Grande H., Mecerreyes D., Pomposo J. A., Synth. Met., 126, (2002) 111.

[16] The K S and Lin. L, 7th International conference on miniaturized chemical and biological analysis system, Oct. 5-9, 2003, USA.

[17] Ralston Andrew R. K., Tobin J. A., Bajikar Sateesh S. and Debnice D., Sen. and Actu. B 22, (1994) 139.

[18] Zhu D. G., Cui D. F. and Petty M. C., Sens. and Actu. B, 12, 111-114, 1993.

[19] B.A Balyaev., N.A. Dorokin and A. A Leksikov., 49 (2006) 952.

[20] Alexey K., Lintao C. and Theresa M., APS 13-17 March meeting, 2006.

[21] Chung B. K., Prog. in Elect. Res., PIER 75, (2007), 239.

The Study of Polypyrrole Polymer Films for Sensing and Microwave Properties 99

[22] Harrington R. F., Time Harmonic Electronic fields, MacGraw-Hill, New York, 1961.

[23] Qian C. and Dou W. B., J. Elect. Wave and Appl. 19, (2005), 795.

[24] Roberts S. and Von-Hippel, J. Appl. Phys., 17, (1946), 610.

[25] D. C. Tiwari, P. B. Patil, R. Sharma, Inter. Conf. on Micro. and Optoelex., 17-20 Dec. 2007, Aurangabad, India.

Synthesization and Characterization of Co-Doped ZnO Diluted Magnetic Semiconductor

A. P. Singh1, R. Kumar1, P. Thakur2, K. H. Chae2, Basavaraj Angadi2 and W. K. Choi2

1Materials Science Division, Inter-University Accelerator Centre, New Delhi-110067 2Materials Science and Technology Research Division, Korea Institute of Science and Technology,

Seoul 136-791, Korea E-mail: [email protected]

Abstract

Structural, transport, magnetic, and electronic studies of Co doped ZnO thin films are presented here.

The measurements show that we have dissolved the Co clusters which were present in as implanted films by irradiating them with 200 MeV Ag+15 ions. The samples show semiconducting as well as ferromagnetism above room temperature. These results are further strengthened by the XANES and XMCD studies done on the samples. 1. Introduction

Spintronics is one of the emerging technologies [1], which promises to add multifunctionalities in the existing electronics. This is to be done using the spin degree of freedom of the electrons. Development of dilute magnetic semiconductors (DMS) is crucial for realization of Spintronics. Following the theoretical prediction of the possibility of room temperature ferromagnetism (RTFM) in ZnO doped with transition metal, there has been extensive research on transition metal doped ZnO [3,4,5]. There are numerous reports on doped ZnO which claim RTFM. But the main issue remains as to whether the observed ferromagnetism (FM) is carrier-mediated or is due to clusters/secondary phases present in the system.

Swift heavy ions (SHI) can be used to deposit a large amount of energy in the system through electron-phonon coupling. This heats up the material which is highly localized, both in space as well as time. Hence it can be used for dissolving any clusters that may be present in the system [6,7]. In this letter we report on the synthesis of Co-doped ZnO DMS prepared by ion-implantation and irradiation with SHI. Thin films of ZnO of thickness 400 nm were deposited on α-Al2O3 (0001) single crystal by plasma-assisted molecular beam epitaxy with a substrate temperature of 720° C. The base pressure was 2 x 10-9 torr. The well characterized films were implanted with 80 keV Co ions to doses of 1 x 1016 to 5 x 1016 ions/cm2 at 300° C to recover the implantation damages. The implantated samples were then irradiated with

200 MeV Ag+15 ions with a fluence of 1 x 1012 ions/cm2. 2. Results and discussion

The structural characterization of the films was done before and after irradiation using Bruker D8 X-ray Diffractometer. The electrical resistivity as a function of temperature was measured using the standard four probe resistivity measurement set-up. The isothermal magnetization hysteresis measurements were performed at 300 K using alternating gradient force magnetometer (AGFM) (micromag-2900, Princeton Measurements Co.) with a sensitivity of 10−8 emu. In these measurements, the magnetic contribution from the substrate was subtracted from the measured data. The Co L2,3 x-ray absorption spectroscopy (XAS) and x-ray magnetic circular Dichroism (XMCD) measurements were carried out at 2A MS beamline of Pohang Accelerator Laboratory (PAL). This beamline is an elliptical undulator with greater than 90% degree of circular polarization. All these spectra were collected in total electron yield mode with resolution of 0.15 eV and the base pressure of the experimental chamber was better than 3 x 10-10 torr. The XMCD spectra were taken for fixed helicity of light by reversing the magnetic field for each hν. The spectra were normalized to incident photon flux.

Fig. 1 shows the XRD pattern of the as-implanted and the irradiated films. From the figure it is evident that the implanted films show peak around 2θ = 44.35°. This broad peak corresponds to some nano-size cluster of Co

Synthesization and Characterization of Co-Doped ZnO Diluted Magnetic Semiconductor 101

(111) peak. It can be seen that with implantation dose this peak is getting sharp. This indicates that the size of the cluster is increasing with implantation dose. Also there is a small shoulder like structure at around 2θ = 73.5°. This has been identified as CoO (311). As can be seen from fig 1(b), after irradiation these Co clusters are dissolved, along with the secondary phase of CoO. Thus we can say that irradiation with swift heavy ions has dissolved the Co clusters and the other secondary phases present in the system. The reason for this to happen can be explained on the basis of thermal spike model. As the ion passes though the material it loses its energy via two processes: nuclear energy loss and electronic energy loss. At 200 MeV energy for Ag+15 ions the dominant process is the electronic energy loss and nuclear energy loss is negligible. In this process the swift ion deposits its energy via inelastic processes. It transfers its energy to electrons, causing excitations and ionizations. Then that energy is transferred to the lattice via electron-phonon coupling in the system. This heats up the lattice well above its melting temperature followed by rapid thermal quenching (1013–1014 K/s). So as the SHI passes through Co implanted ZnO, a solid solution of ZnO and Co is formed followed by the rapid thermal quenching. Co will substitute for Zn in ZnO depending on the solubility of Co in ZnO. Our XRD results confirm the formation of single phase Co substituted ZnO. (a) (b) Fig. 1. XRD of (a) undoped and 80 keV Co implanted at 300° C ZnO thin films, (b) after irradiation with 200 MeV Ag+15 ions with 1 x 1012 ions/cm2 fluence. The result of the transport measurements done on the samples is shown in fig. 2 for one of the samples. It clearly shows a semiconducting behavior. As can be seen from the figure that after irradiation, the resistivity of the films

decrease. The Co clusters present in the samples before irradiation act as scattering centers for the charge carriers. After irradiation these clusters are dissolved and Co now substitutes for Zn in ZnO. This makes the films more ordered and decreases the scattering processes and hence a decrease in resistivity. Here we have shown result for only one sample, all the samples are showing similar trends.

50 100 150 200 250 30017

18

19

20

21

22

23

24

25

26

Co implanted ZnO with 5x1016 @ 300 0C

Res

istiv

ity (m

Ω-c

m)

Temperature (K)

Unirradiated Irradiated by 1x1012 ions/cm2

of Ag+15 with 200 MeV

Fig. 2. Resistivity as a function of temperature for unirradiated and irradiated thin films implanted with Co fluence of 5 x 1016 ions/cm2 at 300° C. One of the most important properties we are looking for in these systems is ferromagnetism. Fig. 3 shows the isothermal magnetization hysteresis curves at 300° C for irradiated samples implanted with 1 x 1016 and 5 x 1016 ions/cm2 fluence of Co ions. The samples are clearly showing room temperature ferromagnetism with magnetization increasing with the Co content in the system.

-2800 -2100 -1400 -700 0 700 1400 2100 2800

-4

-2

0

2

4

Co - 5 x 1016 ions/cm2

Co - 1 x 1016 ions/cm2

T= 300K

Co implanted ZnO irradiated with 200MeV Ag ions fluence 1x1012ions/cm2

M(e

mu/

cc)

H(Oe) Fig. 3. Isothermal dc magnetization as a function of magnetic field for 200 MeV Ag+15 irradiated ZnO thin films implanted with Co ions with a fluence of 1 x 1016 and 5 x 1016 ions/cm2.

40 50 60 70 80

Unimplanted

2θ (degree)

Co - 1x1016ions/cm2

Al 2O

3 (103

)

CoO

(311

)

Inte

nsity

(A.U

)

ZnO

(004

)

Co

(111

)Al 2O

3 (006

)

ZnO

(002

)

Co - 2x1016ions/cm2

Co - 5x1016ions/cm2

40 50 60 70 80

Inte

nsity

(a.u

.)

2θ (degree)

Unimplanted

Co - 1x1016ions/cm2

Co - 2x1016ions/cm2

Al 2O

3 (103

)

Co - 5x1016ions/cm2

ZnO

(004

)

Al 2O

3 (006

)

ZnO

(002

)


To further confirm that Co has substituted for Zn in ZnO the samples were characterized for XAS and XMCD at Co L3,2-edge. This gives the local geometry and the magnetic contribution coming specifically from Co ions. Fig.4 shows the Co L3,2–edge XAS spectra of Co-doped ZnO thin film at room temperature, which is primarily due to Co 2p -3d hybridization and strongly influenced by the core-hole potentials.

770 775 780 785 790 795 800 805 81

5x1016 ions/cm2

2x1016 ions/cm2Inte

nsity

(Arb

. Uni

ts)

Photon Energy (eV)

Co L3,2 edgeCo: ZnO

RT

1x1016 ions/cm2

200 MeV Ag15+ ion irradiated Films

Fig. 4. Co L2,3–edge XAS spectra at room temperature. The XAS at Co L3,2–edge provides information about the 3d occupancy of the Co ions and the valence state of Co ions. At higher valence state of Co ions, the peak edge shifts towards higher energy. The spectral features between 775–784 eV show multiple absorption peaks and are assigned to Co 2p3/2 -3d (L3 - edge) transitions, and those in the region of 790–798 eV to Co 2p1/2 –3d (L2-edge) transitions. The overall multiplet spectral features are similar to Co2+ ions coordinated tetrahedrally to four Oxygen atoms. Moreover, the observed spectra are similar to atomic multiplet configuration interaction calculations confirming that Co occurs in Co2+ state in tetrahedral crystal field.

XMCD measurements were performed to element-specific local magnetic interactions by detecting XAS at Co L3,2 – edge by reversing the applied magnetic field for a fixed photon helicity. Fig. 5 shows the XAS and XMCD of Co L3,2 – edge at 300 K, 150 K and 80 K of

irradiated ZnO thin film implanted with fluence 5 x 1016 ions/cm2 of Co ions. From the XMCD data, the characteristic contributions, negative features at L3 – edge and positive features at L2 – edge, can be clearly identified. The relative intensity of the XMCD signal remains constant with temperature, indicating that the ferromagnetic ordering is due Co2+ ions and not due to any ferromagnetic Co clusters.

770 775 780 785 790 795 800 805

RT

Co L3,2edge

Co: 5x1016ions/cm2

XAS/XMCD

Inte

nsity

(Arb

. Uni

ts)

150 K

Photon Energy (eV)

ρ+

ρ−

4x(ρ+− ρ−)80 K

Fig. 5. Co L2,3–edge XAS and XMCD spectra at different temperatures with applied magnetic field of 0.32 T. 3. Conclusions In conclusion, we have prepared Co implanted ZnO thin films in which Co is substituting for Zn in ZnO lattice, through ion implantation and then irradiating them with swift heavy ions. The impurity phases and clusters those were present after implantation has been dissolved using swift heavy ions. Films are semiconducting and ferromagnetic at room temperature. The magnetization in the films are increasing with the Co content which further supports the argument that Co is substituting for Zn in ZnO and the magnetization is due to Co substituting for Zn in ZnO. XAS confirms the divalent state of Co ions and XMCD at Co L3,2–edge confirms that the ferromagnetism in Zn–Co–O system is due to Co2+ ions substituted at the Zn site.

Synthesization and Characterization of Co-Doped ZnO Diluted Magnetic Semiconductor 103

References [1] G. A. Prinz, Science 282 (1998) 1660. [2] T. Dietl, H. Ohno, F. Matsukura, J. Cibert,

T. Fukumura and M. Koinuma, Science 291 (2001) 854 .

[3] K. Ueda, H. Tabata, and T. Kawai, Appl. Phy Lett. 79 (2001) 988.

[4] S-J Han, J. W. Song, C-H Yang, S. H. Park, J-H Park, Y-H Jeong, and K. W. Rhie, Appl. Phys. Lett. 81 (2002) 4212.

[5] P. Sharma, A. Gupta, K. V. Rao, F. J. Owens, R. Sharma, R. Ahuja, J. M. Osorio Guellen, B. Johansson, and G. A. Gehring, Nat. Mater. 2 (2003) 673.

[6] Basavaraj Angadi, Y.S. Jung, W.K. Choi, Ravi Kumar, K. Jeong, S.W. Shin, J.H. Lee, J.H. Song, M. Wasi Khan and J.P. Srivastava, Appl. Phys. Lett. 88, 142502 (2006).

[7] Ravi Kumar, F. Singh, Basavaraj Angadi, J.W. Choi, W.K. Choi, K.jeong, J.H. Song, M. Wasi Khan, J.P. Srivastava and R.P. Tandon, J. Appl. Phys. 100, 113708 (2006).

[8] M. Kobayashi, Y. Ishida, J. I. Hwang, T. Mizokawa, A. Fujimoro, K. Mamiya, J. Okamoto, Y. Takeda, T. Okane, Y. Saitoh, Y. Muramatsu, A. Tanaka, H. Saeki, H. Tababta, and T. Kawai, Phys. Rev. B 72 (2005) 201201.

Growth of Highly Oriented AgInSe2 Films for Solar Cell Applications

Dinesh Pathak1*, R.K. Bedi1 and Davinder Kaur2 1Deptt. Of Physics Guru Nanak Dev University Amritsar

2Deptt. Of Physics Indian Institute of Technology, Roorkee India. Email : [email protected]

Abstract

AgInSe2 (AIS) films were grown on glass substrate kept at 473K by thermal evaporation technique. The

starting material were stoichiometrically mixed Ag, In and selenium powders. The film thickness was controlled during deposition by means of quartz crystal monitor and was kept near 1.5 μm. The structural properties were investigated by means of X ray diffractometer using Cu Kα (wavelength =1.5405 A0) radiations in 2θ range 200-800 by Bruker diffractometer .X ray diffraction pattern indicate that the film and as prepared powder of AgInSe2 have preferential texture in (112) direction .The prepared AgInSe2 films have single phase with predominant (112) orientation. Crystallite size has been calculated from FWHM value of the (112) peak by using Scherrer’s equation and is observerd 11 nm for the films prepared at 473 K. The surface morphology of the films studied from AFM . It depicts nano well type structures with depth of 250 nm. The variation of absorption coefficient (α) as a function of wavelength λ (nm) has been studied from optical absorption data 1. Introduction Thin film solar cell made from ternary chalcopyrite compounds, such as CuInSe2 and its analogous alloys, have begun to fulfill the promise of low cost , power generation from nonpolluting clean energy source. Because of broad range of optical band gap and carrier mobility and ability to form solid solution , they have recently led to their emergence as technologically significant material [1].They have direct band gap and high optical absorption coefficient [2]. These compounds ABX2 (A = Cu, Ag…, B = Ga, In…, X= Se, Te…) form a semiconducting group crystallizing with chalcopyrite structure. These compounds are being considered as interesting material because of the possibility for their potential application in solar cell and optoelectronic devices[3-5]. AgInSe2 belongs to I-III-VI2 family of semiconducting materials which crystallizes in tetragonal chalcopyrite lattice[6]. These are the ternary analogues of the AIIBVI compound with cubic Zinc Blende structure .The c/a ratio is not exactly equal to two because of different attractive forces between each kind of metal atom and the chalcogen atom in the chalcopyrite structure. Due to tetragonal distortion (2-c/a) this class of materials shows a great deal of application in the nonlinear optics (NLO) and photo-voltaic energy conversion[7-10]. They show their ability to form solid solution and to accommodate different dopants,

which led to their emergence as technological significant material [11]. However a little is known about the optical properties of AgInSe2 [12-13]

2. Experimental

The polycrystalline powder was synthesized by mixing stoichiometric amount of Ag, In and Se powder of 99.999 % purity ,followed by sealing in quartz tube in the vacuum. . At first furnace temperature was slowly raised (1K /min) to 1100 K with continuous vibrational shaking to ensure homogeneity of the sample. This temperature was maintained for 48 hours and then furnace was turned off and tube was left inside until it reached to room temperature .The ingots obtained were grinded into powder.

AgInSe2 films of thickness 1.5 μm were prepared by thermal evaporation technique using Hind-Hi vacuum coating unit 12A4H on cleaned glass substrate kept at 473 K. Vacuum system, which consist of an oil diffusion pump coupled with rotary pump ,was used for the preparation of films .Molybdenum boats were used for evaporation .The vacuum of the order of 10-5 m bar was maintained in the chamber throughout the deposition. Cleaned optically flat glass substrates were used for the preparation of films. The glass substrates (Blue Star PIC 1) were first cleaned with acetone and alcohol and then washed in an ultrasonic bath. The glass substrate was mounted on the substrate holder

Growth of Highly Oriented AgInSe2 Films for Solar Cell Applications 105

with heating arrangement and the temperature was measured with the help of K-type thermocouple obtained from Omega Engineering Inc.(USA). The film thickness was controlled during deposition by means of quartz crystal monitor. The X ray pattern of the film and as prepared powder was recorded by X Ray diffractometer using Cu Kα (wavelength =1.5405 A0) radiations. The absorption spectra of AgInSe2 films was recorded in the wavelength range 200- 1100 nm using UV-VIS spectrophotometer (UV-1601 PC (shimadzu, Japan).

3. Result and Discussion

XRD pattern of as prepared powder (Fig1.a) exhibit the prominent (112) peak of chalcopyrite phase in addition to other (220), (204), (312) and (116) peaks of very small intensity. However the AgInSe2 films prepared at 473 K indicate high orientation in the (112) plane parallel to the glass substrate (Fig1b). The crystallite size is calculated from the FWHM value of (112) peak by using Scherrer’s formula and found to be 11nm. While H. Matsuo et. al. [14] obtained the grain size in the range 45-85 nm for AgInSe2 films deposited at different temperatures on glass substrate.

The surface morphology of the films studied from AFM has been shown in Figure 2. It depicts nano well type structures with depth of 250 nm. Fig 3 (a) show the variation of absorption coefficient (α) as a function of wavelength λ (nm). It exhibit regular absorption edge and magnitude of the absorption coefficient above the fundamental edge is around to be 106 cm-1 This value is comparatively very large than that of other chalcopyrites like CuAlTe2 [15].

The linear nature of plot near the absorption edge confirms that AgInSe2 is semiconductor with direct band gap [16] which has been calculated using equation of Bardeen [17].

(αhν) = A (hν – Eg )1/2 (1)

Fig. 1. XRD pattern of (a) as synthesized powder (b) films prepared at 473 K.

Fig. 2. AFM of AgInSe2 films prepared at 473K.


Fig. 3. (a) the variation of absorption Coefficient (α) as a function of wavelength λ (nm) (b) variation in (αhν)2 with photon energy hν (eV)

The band gap energy estimated to be 1.19 ev [Fig 2(b)]. Various Authors reported different value of band gap of AgInSe2 films [18] which may be attributed due to the difference in Ag/In/Se Composition ratio.

4. Conclusion

High quality single phase Silver Indium Selinide films were successfully prepared by thermal evaporation technique Silver indium selinide films appear to have preferential textured in (112) direction. The crystallites of size 11nm has been obtained. AFM studies depict nano well type structures with depth of 250 nm. From optical measurements, the band

gap energy is estimated to be 1.19 eV. The sample exhibit the absorption coefficient 106 cm-

1 which is larger as compared to other chalcopyrites showing AgInSe2 ideal material for solar cell applications.

Refrences

[1] A. H. Ammar, A. M. Farid and M. A. M. Seyam, Vacuum 66 (2002) 27-38

[2] J. L. Shay, J. H. Wernick, Ternary chalcopyrite semiconductor growth, Electronic properties and applications, Pargamon, Oxford 1975.

[3] Iseler G W, Kildal W, & Menyuk N , Inst. Phys Conf. Ser, 5 (1977) 73.

[4] Kazmerski L I, Phys.Conf.ser. 5 (1977) 217. [5] Romeo N, Jap. J. Appl. Phys., 3, 19

(1980) 5. [6] Jaffe J E, Zunger A. Phys Rev B, 28 (1893)

5822. [7] Shay J L & Wernick J H, Ternary

chalcopyrite Semiconductor: Growth Electronic Properties and application Pergamon, Oxford , 975

[8] Chemla D S, Kupcek D J, Robertson D S & Smith R C , pt. Commun. 3 (1971) 29.

[9] Boyd G D, Kaspoer H M & Mcfee J H, IEEE J. Quantum Electron. 7 (1971) 563.

[10] Wagner S, Shay J L, Migloirato P & Kasper H M, Applied Phys Letter , 25 (1974) 434.

[11] Yamamota et.al ,Jpn J Applied Phys 3 (1980)19.

[12] Yoshino K, Mitani N, Sugiyama M, Chichibu S, Koamki H & Ikari T,Physics B 302/303 (2001) 349.

[13] Albornoz J G, Serna R & Leon M, J.Appl.Phys, 97 (2005) 103515.

[14] H. Matsuo, K. Yoshino and T. Ikari, Phys. stat. sol (c)3, No 8, (2006) 2644-2647

[15] T. Bekkay, M. Boustani, K. El Assali, A. Khiara ,J. C. Bernede and J. Pouze, Internationl Journal of Electronics Vol 92, No 8, Aug 2005, 445-449

[16] J. I. Pankove, Optical Processes in semiconductors, New York ,Dover,1971

[17] J Berdeen, F J Blatt, L H Hall, R. Breekwuridge, B. Russel, T. Hahn, ( Eds.) Photoconductivity Conference, Wiley, New York 1956

[18] A. El-Korashy, M. A. Abdel-Rahim and H. El- Zahed, Thin Solid Film 338, (1999)

Electrical Properties of Amorphous GaxSe1-x Thin Films

Falah Ibrahim Mustafa, N. Goyal and S. K. Tripathi Department of Physics, Centre of Advanced study, Panjab University, Chandigarh -160014, INDIA.

E-mail: [email protected]; [email protected]

Abstract

The Temperature dependence of the DC conductivity (σ) of amorphous GaxSe1-x (x=40, 50, 60) thin films, prepared by thermal evaporation technique have been studied. The incorporation of Ga atoms in Se matrix leads to an increase in the electrical conductivity with increase in Ga content and decrease in the thermal activation energy in the temperature range (100-400 K), we found two temperature regions (two different activation energies) divided at approximately high ranges and low ranges temperatures in all samples. 1. Introduction

In recent years, a great deal of interest has been focused on semiconducting ІІІ-VI layered compounds. GaSe is a typical layer structure compound. A layer of GaSe consists of two Ga and two Se sublayers in the sequence of Se-Ga-Ga-Se, where the Se-Ga and Ga-Ga bonds are covalent in the layers and the Se-Se bond between adjacent four atomic layers is due to van der Walls forces [1]. Investigations on the structural, electrical and optical properties of GaSe compound revealed that it is attractive for heterojunction device applications [2] and photoelectronic devices in the visible range [3]. It has been used as a radiation detector that operates at room temperature [4].

Fundamentally, for amorphous materials the temperature dependence of conductivity is determined by three mechanisms [5], extended state conductivity, conduction in band tail and conduction in localized sites. In case of an extended state conductivity, it is assumed that beyond mobility edge the mean free path for conduction is short and nearly equal to the average separation between atoms. The conduction in this case is an activated process and the conductivity is given as σd = σ min exp [-(EC – EF ) / kT ] ..... (1)

In the second mechanism i.e., conduction in the band tails, the conduction occurs due to the tunneling of the carriers to the unoccupied of the nearest neighbour. Since the tunneling process involves the emission or absorption of phonons, it requires a tunneling energy ΔW1. The conductivity due to this mechanism is given as σd = σ1 exp [-E1 / kT ] ..... (2)

Where the activation energy (E1 ): E1=EF - E + Δ W1. The conduction in localized states at Fermi-energy is analogous to impurity conduction in heavily doped semiconductors. The carriers can move between the states via phonon assisted hoping (tunneling) process. The conductivity due to this mechanism can be written as σd = σ2 exp [-(Δ W1) / kT ] ….. (3)

Here σ2 is less than σ1. As the temperature is lowered, the number and the energy of phonons decreases, hence the hopping between the nearest neighbour will become less probable and instead the carrier will hop to larger distances i.e., to the sites within the range kT. For such conduction, Mott [5] proposed the Variable Range Hopping (VRH) in which the conductivity is proportional to T -1/4. At low temperature, the activation energy of the conductivity decreases gradually with decreasing temperature, the conduction may be mainly due to hopping. Below about 220 K, σd varies exponentially with T -1/4 , indicating that charge transport is due to variable range hopping in states close to the Fermi level Ef. Following the model proposed by Mott [6] the conductivity is given by σd = σo exp – ( To / T )1/4 ..... (4)

The pre-exponential σo depends mainly on frequencies, while To has been shown to be approximately To =Co α3 / kN (Ef ) ..… (5) Where N (Ef ) is the density of hopping sites at the Fermi level, and Co is a numerical constant which depends on the detailed assumptions.


2. Experimental

Glassy alloys of GaxSe1-x (x=40, 50, 60) were prepared by a melt-quenching technique. Materials of 99.999% purity were sealed in quartz ampoules (length ~ 12 cm, internal diameter ~ 0.8 cm) with a vacuum of about 10-5 Torr. The sealed ampoules were kept inside a furnace where the temperature was raised slowly up to 900 oC. The ampoules were stayed for 24 h at the maximum temperature to make the melt homogeneous. Quenching was done in ice water. Thin films of glassy alloys were prepared by vacuum evaporation technique using a standard coating unit. well degassed corning glass plates, having predeposited indium electrodes, were used as a substrate for depositing amorphous films in the planer geometry (length ~ 1.2 cm) and electrode gap (~0.8mm).

These films were deposited at room temperature and at a base pressure of about 10-5 Torr using molybdenum boat. The thickness of these films was ~ 500 nm. The films were kept in deposition chamber in the dark for 24 h before mounting them in the sample holder. This was done to allow sufficient annealing at room temperature so that a metastable thermodynamic equilibrium may be attained in the samples as suggested by Abkowitz [7] chalcogenide glasses. The deposition parameters were kept almost same for all the samples so that a comparison of results could be made for various glassy samples. The amorphous nature of the resulting glassy alloys was verified by X-ray diffraction.

For DC conductivity measurements, the samples were mounted in a specially designed metallic sample holder where a vacuum of about 10-2 Torr could be maintained throughout the measurements. A DC voltage was applied across the sample and the resulting current was measured by a digital electrometer (Keithley, model 6517A). The temperature was measured by mounting a calibrated Pt 100 sensor near the sample with digital readout. 3. Results and discussion

Fig. 1. shows the X-ray diffraction pattern of Ga40Se60 thin films deposited at room temperature in the glass slides . It is clear from the figure that there is no prominent peaks which verify the amorphous nature of those thin films. The similar curves have been obtained for other samples (i.e. x=50, 60) also.

Table 1. Activation Energy at high and low temperature.

Composition

Activation energy at

High temperature

Activation energy at

Low temperature

a-Ga2Se3 0.79 eV (250-400 K)

0.041 eV (100-250 K)

a-GaSe 0.74 eV (250-400 K)

0.034 eV (100-250 K)

a-Ga3Se2 0.066 eV (170-400 K)

0.0138 eV (100-170 K)

Fig. 1. X-ray diffractogram of a-GaSe thin films deposited at room temperature.

The Temperature dependence of the DC conductivity (σ) of amorphous GaxSe1-x (x = 40, 50 , 60 ) thin films, prepared by thermal evaporation technique, have been studied and plotted in figure 2-4. In all samples, the ln σ dc∝ 1000/T curves found to be straight lines indicating a thermally activated process for DC conduction in these samples. The incorporation of Ga atoms in Se matrix leads to an increase in the electrical conductivity with increase in Ga content and decrease in the thermal activation energy. In the temperature range (100-400 K), we found two temperature regions (two different activation energies) divided at approximately high ranges and low ranges temperatures in all samples. In Figs. 2 & 3, the thermal activation energy of conduction of a-Ga2Se3 and a-GaSe thin films are 0.79eV and 0.74eV which lies in the temperature range 250-400 K, where the conduction mechanism is due to excitation of the holes to the extended valence band with an activation energy. The conductivity in this region is associated with the transport of carriers through the extended states of the relative transport band of Eq. (1). The second region of thermal activation energy of conduction are 0.041eV and 0.034eV in the temperature range 100-250 K which is due to hopping conduction near the Fermi level EF that

10 20 30 40 50 60 2θ

Count 100 0

Electrical Properties of Amorphous GaxSe1-x Thin Films 109

obeys for low activation energy, and σd varies exponentially with T -1/4 following the model proposed by Mott [6], In Eq 4. the phonons do not have enough energy for transferring to a nearest neighbour atom and the charge carrier hops from a neutral atom to another neutral atom situated at the same energy level, which can be many interatomic distance away [8]. Other authors [9, 10], have also studied and they have found activation energy ~ 0.85 eV and ~ 0.42 eV for a-GaSe system.

2 3 4 5 6 7 8 9 10 11

-28

-26

-24

-22

-20

-18

-16

-14

-12

0.041 eV

0.796 eV

1000/T (K-1)

Lnσ

(ohm

.cm

)-1

Ga40Se60

Fig. 2. The In (σ)-1/T plots of a-Ga2Se3 thin film

2 3 4 5 6 7 8 9 10 11-26

-24

-22

-20

-18

-16

-14

-12

1000/T (K-1)

Lnσ

(ohm

.cm

)-1

0.0345 eV

0.743 eVGa

50Se

50

Fig. 3. The In (σ)-1/T plots of a-GaSe thin films

Figure 4 shows the two activation energies for a-Ga3Se2. The first activation energy ( 0.066 eV) lie in the range (170-400 K) and second activation energy ( 0.0138 eV) in the range (100-170 K).

The σ at high temperature obeys the law: ln σ ∝ 1/T and in the low temperature range obeys the law ln σ ∝ T-1/4 , indicating variable range hopping in localized states near the Fermi level in the later case[9].

1 2 3 4 5 6 7 8 9 10 11 12-7

-6

-5

-4

-3

-2

Ga60Se40

1000/T (K-1)

Lnσ

(ohm

.cm

)-1

0.0138 eV

0.0664 eV

Fig. 4. The In (σ)-1/T plots of a-GaSe thin films. 4. Conclusions

The dc conductivity measurements were made on the deposited GaSe films in the room temperature. The conduction in the low temperature range (< 250 K) exhibits relatively less thermal activation and was found to be due to variable range hopping, while in the high temperature range ( 250-400 K), the conductivity increases with increasing temperature according to the semiconductor behavior. Acknowledgements

This work is financially supported by CSIR, New Delhi. References [1] P.A. Lee, Physics and Chemistry of Materials

with Layered Structures: Optical and Electrical Properties, vol. 4, Reidel, Boston, 1976, p. 76.

[2] M. Di Giulio, G. Micocci, P. Sililian, A.Tepore, J. Appl. Phys. 62 (1987) 4231.

[3] B. M. Basol, Thin Solid Film 361-362 (2000) 514.

[4] C. Manfredotti, R. Murri, Quirini, L. Vasanelli, Nucl. Instrum. Methods 131 (1975) 457.

[5] N. F. Mott, Adv. Phys. 16, 49 (1967). [6] N. F. Mott, J. Non- Cryst. Solids 8-10,1(1972). [7] V.A. Twaddell, W.C. Lacourse, J.D.

Mackenzie, J. Non- Cryst. Solids 8–10 (1972) 831.

[8] P. C. Sikka, K. V. Ferdinand, C. Jagadish, and P. C. Mathur, J. Mater. Sci. 20, 246 (1985).

[9] Masanori Ohyama, Yasuhiko Fujita, Surface and Coating Technology 169-170 (2003) 620.

[10] M. Di Giulio, Micocci, P. Sicilian, J. Appl. Phys. 62 (10), 15 November 1987.

Limiting Behaviour of A.C. Conductivity in Nanoferroelectrics

S.K.S. Parashar, Kajal Parashar, R.N.P. Choudhary1 and B.S. Murty2

Department of Applied Sciences & Humanities, National Institute of Technology, Hamirpur–177005(HP) 1Department of Physics and Meteorology, Indian Institute of Technology, Kharagpur – 721 302, India

2Department of Metallurgical and Materials Engineering, Indian Institute of Technology, Chennai-600 036. E-mail:[email protected]

Abstract

The advent of nanoscience and technology has changed the scenario in recent years where the crystallite size of a material plays a significant role in defining structure/microstructure of the system that subsequently governs its physical properties. The control of physical properties and structure/microstructure by restricting crystallite dimension to the nanoscale is one of the novel approaches that have evolved as the new trend setter for achieving novelty in the performance characteristics of a material system and devices based on it. The nanocrystalline ferroelectrics Sm, Gd, Nd and Zn modified PZT has been synthesized by mechanical alloying method. X-ray diffraction studies show that all samples have Cubic structure at lower crystallite size 15nm. The ac conductivity of nanoferroelectric ceramics have been studied over a wide temperature range. In all the cases, limiting power-law behavior S < 1 in σT (ω) =σ0 +AωS is observed at relatively lower crystallite size. It is concluded that this limiting behavior is a universal phenomenon. Available theoretical explanations (quantum tunneling approach) have been examined and found to be unsatisfactory. A new theoretical explanation has been proposed and it is well correlated with the experimental data.

1. Introduction

Functional ceramics based on lead zirconate titanate solid solutions of PZT are the materials-of-choice in a wide range of applications such as piezoelectric transducers and actuators, telecommunication, medical imaging, ultrasonic devices, random access memory and sensors devices [1-3]. Due to very few reports on ac conductivity in Pb-based ferroelectric synthesized by high energy ball milling, so far, the nature of the polaronic conductivity mechanisms remained unclear. Earlier attempts have been made (to explain transport phenomena in BaTiO3 from the polaronic point of view [4]. The ac conductivity of many materials shows dispersion behaviour through a dependence of the electrical conductivity σ on angular frequency ω of the form [5 ] σT (ω) =σ0 +AωS (1) where σ0 is the dc (low-frequency) conductivity and the exponent S lies in the range of 0 to 1. Such behaviour has been so widely observed for highly disordered materials, e. g. , ionically conducting glasses, conducting polymers, and amorphous semiconductors, that it has come to be known as the “ universal dynamic process” [5]. Other manifestations of this relaxation- type behavior appear in the real part of the dielectric constant, ε′, which varies as :

1−∞ ∝′−′ Sωεε (2)

Equation (2 ) follows from (1) as a consequence of the Kronig_ Kramers relations. In the time domain, Eq. (1) is equivalent to

Stt −−=Φ 1)/exp()( τ (3)

where )(tΦ is the normalized decay of the transient current which follows the sudden application of a steady electric field and τ is the relaxation time. The expression in Eq. (3) is known as a “stretched exponential” or as the Kohlrausch-Williams-Watts (KWW) expression for )(tΦ [6]. There are single–particle hopping models, involving random barriers and /or trapping sites, so that the carriers move in percolating networks [7]. Alternatively, there are the theories that regard the hopping event as involving many-particle interactions, so that the motion of a carrier is greatly influenced by the relaxation of its neighborhood, and the exponent S is a measure of the degree of interaction [8]. In both cases, there is the question of how S varies with temperature (or frequency, over a wide frequency range) and whether S can reach the limiting value of unity. In some theories S = 1 is regarded as an unattainable limit [5]. In fact, for this case, there is no decay of transient current, as given by Eq. (3). Other theories suggest that S increases toward the limit of unity as the temperature decreases [9]. The experimental situation is somewhat unclear. While for amorphous semiconductors at low temperatures and ionically conducting crystals and glasses, the case of S = 1 has been widely reported [7], the appearance of this limiting case has not been established for nanoferroelectric materials. In this paper, we examine a number of nanoferroelectrics PZT based (doped with Sm, Gd, Nd and Zn) which are referred to as PSZT , PGZT, PNZT and PZZT) with different crystallite sizes and show that the case of S ≈ 1 is observed for some

Limiting Behaviour of A. C. Conductivity in Nanoferroelectrics 111

relatively simple defect systems, and also that it may appear as a limiting behaviour at relatively lower crystallite size systems in which, typically, S≠1. 2. Experimental

The nanoferroelectrics sample were prepared by high energy ball milling technique (mechanical alloying (MA) and the detailed of which are reported earlier [3, 10-13]. Figure 1 shows the XRD pattern of Zn modified PZT (PZZT). The XRD pattern of 0h (without ball milling) shows distinct peaks belonging to the starting oxides of PbO, ZrO2, TiO2, and ZnO, depending on the composition. It has been observed that the crystallite size decreases with increase in milling time in all the samples. The average crystallite size of all the 50h milled samples was found to be in the range of 15-20nm (calculated from x-ray peak (110) broadening using Viogt peak profile analysis and TEM). It was also observed that the 50h samples shows the cubic structure with the lattice parameter 0.40254nm.

Fig.1. XRD patterns of mechanically alloyed Pb0.92Zn0.08(Zr0.53Ti0.47)O3 after different milling hour. Figure 2 shows the bright field TEM micrographs with selected area electron diffraction (SAED) pattern (inset) of the 50h milled powder of PZZT, which clearly establishes the nanocrystalline phase of these materials. MA synthesized powders are uniformly distributed and are approximately spherical.

Fig. 2. TEM micrographs and SAED pattern (inset) of 50 h milled powder of Zn modified PZT (PZZT) 3. Results and discussion

It is observed in Fig 3 that on increasing the temperature the dispersive component shifts towards the higher frequency region, first apparently as a moderate frequency independent component at the low frequency region, and at the higher temperature (close to 5250C) the curves become merge.

Fig. 3. Variation of ac conductivity of PZZT with frequency for (a) 16nm, (b) 29nm, (c) 78nm, and (d) 88nm crystallite size.

The existence of conductivity dispersion in the low frequency region above 4000C shows the presence of space charge. The plateau region of the conductivity at high temperature and high frequency could be due to the disappearance of the space charge effects, because the space charge effect vanishes at higher temperature and frequency All samples with small nanocrystallite size have higher ac conductivity than the larger ones. This could be partly attributed to the tetragonal phase at higher crystallite sizes, which appears to have a lower conductivity in comparison to cubic phase present in smaller crystallites. The saturation of σac(ω) at high


frequency may take place when ω is closed to the maximum jump frequency [14].

The frequency at which the change in slope is observed corresponds to polaron hopping of charged species. The polaron hopping frequency is observed to shift to lower frequencies with increasing temperature indicating increasing conductivity with temperature. With increasing temperature the charged species that are accumulated at grain boundaries have sufficient energy to jump over the barrier, thereby increasing the conductivity. Thus the grain boundary resistance decreases beyond this frequency and temperature. The ac conductivity of ceramics can arise from both bound as well as free-charge carriers. If the free carriers (in nonlocalized states) carry the current, the frequency dependence of conductivity

)ωσ ( can be written as [15,16]:

( )221( τωσωσ + = ) dc where τ is the relaxation time. This equation suggests that the conductivity due to free carriers must decrease on increasing the frequency. In the present study, the ac conductivity increases with increase in frequency for all the samples, and therefore, the observed ac conductivity is related to the hopping of bound carriers trapped in the nanocrystalline samples. Such bound charge carriers on the different lattices can be considered as polarons.

Two major physical models have been developed to account for sublinear frequency-dependence of conductivity: (i) quantum-mechanical tunneling (QMT) [17,18] and (ii) correlated barrier hopping (CBH) conduction model [7,19]. The polarization tends to follow the electron as it moves through the lattice; the combination of the electron and its induced polarization field can be considered as a quasi-particle, which is generally called polaron. S is calculated from the slope of log σ(ω) vs log (ω) plot (Fig. 4a). It is found that S increases on decreasing temperature (Fig. 4a). It has been observed that S is related to doping and stoichiometry of nanoceramics. This parameter shows maximum near Tc, after which, it decreases as the temperature rises. This behavior has been observed in a wide variety of materials [20]. The conductivity increases monotonically with rise in frequency. It has been observed that S, frequency dependent parameter, is related to crystallite size of materials. It has been noted that the varied nature of S with all modified PZT ceramic indicates either a broad distribution of relaxation times [21] or multiple site polaronic hopping [22].

The value of S less than unity is associated with charge carriers or extrinsic dipoles arising from defects or impurities. The value of S has a decreasing trend with rising temperature,

indicating the interaction between the different charged species. The nature of variation of ac conductivity with frequency as shown in Figure1A, is represented as ‘‘universal’’ power law, σ(ω) =AωS, (0<S<1). This is based on rigorous many-body dielectric interactions [5].

(a)

(b)

(c) Fig. 4. (a) Temperature dependence of S parameter for PSZT (19nm), PGZT (18nm), PNZT (17nm ) and PZZT (16nm) (b) Minimum hopping distance (Rmin) as a function of crystallite size (c) Polaron binding energy as a function of crystallite size of nanocrystalline PSZT, PGZT, PNZT, and PZZT at 10 KHz at ferroelectric- paraelectric phase transition. An attempt to use the QMT model has given the value of S (closer to 1) and suggests an unreasonably high value of phonon frequency

Limiting Behaviour of A. C. Conductivity in Nanoferroelectrics 113

(1039 Hz) at a measuring frequency of 105Hz. It is concluded that the extension of this theory is inadequate to cover nanocrystalline ferroelectric ceramics. In hopping conduction, localized charge carriers of each atomic site jump to another site by thermal energy from time to time. In this case, each jump of a charge carrier is considered to occur completely at random, without any correlation between one jump and the next. An electric field, however, affects the probability of jumps such that carriers move on average along the field direction [17].

Assuming that the defect centers are distributed randomly in space, the activation energy and hopping distances are estimated from the frequency exponent S. Figure 4b shows the crystallite size dependence of minimum value of hopping distance (Rmin) [23] of PSZT, PGZT, PNZT, and PZZT at 10kHz. The poloron binding energy (Wm) showed maxima with a minimum of corresponding Rmin for all the systems in Figure 4c. These Figure 4 (b,c) show once again the correlation between the activation energy, and the cutoff hopping distance (i.e., the higher the polaronic coupling with the lattice, the lower the hopping distance). Thus a gradual decrease of Wm with decrease in crystallite size could be visualized in terms of strong polaronic coupling between the charged carriers and the charge defects, i.e., they suffer an appreciable charge –carrier- lattice (phonon) interaction. Similar is the case of the reduction of the hopping distance with crystallite size, weak polaronic energies might be responsible for an increase in hopping distance. It is also observed that as the crystallite size decreases the symmetry changes from C1 to C4 i.e., tetragonal to cubic and the polarization is less in cubic system. Hence the higher the dielectric constant in tetragonal phase higher the activation energy. Activation energies are found to be maximum in the lanthanide dopant and less in Zn doped PZT. Such behaviour could be attributed to either the formation of defect complexes or some sort of charge-carrier compensation phenomenon at the lower crystallite size in all the samples. 4. Conclusions

The limiting behaviour appears (S ≈ 1) appears to be a universal phenomenon, yet none of the models present in the literature offers a satisfactory explanation for this distinctive behaviour. The ac electrical characterizations of modified PZTs have shown that a polaronic related charge-hopping mechanism is responsible for charge–carrier transportation in these PZT based nanoferroelectric ceramics.

References 1. J. F. scott and Paz de Aranjo, Science 246,

(1989) 1400. 2. Bog-Gi Kim, Seong M. Cho, Tea-Yong Kim,

and Hyun M. Jang, Phys. Rev. Lett. 52, (1984) 69.

3. S. K. S. Parashar, P. Padhi, R. N. P. Choudhary, Awalendra K. Thakur and B.S. Murty, Materials and Manufacturing Process, 22, (2007) 337.

4. P. Gerthen, R. Groth, and K. H. Hardlt, Phys. Status Solidi 11, (1965)303.

5. A. K. Jonscher, Dielectric Relaxation in Solids (Chelsea Dielectric Press, London, 1983).

6. G. Williams and D. C. Watts, Trans. Faraday Soc. 66, (1970) 80.

7. S. R. Elliott, Adv. Phys. 36, (1987) 135. 8. R. G. palmer, D. L. Stein, E. Abrahams, and

P. W. Anderson, Phys. Rev. Lett. 53, (1984) 958.

9. W. Dieterich, Solid State Ionics, 40/41, (1990) 509.

10. S. K. S. Parashar, R. N. P. Choudhary and B. S. Murty, J. Appl. Phys. 94, (2003) 6091.

11. S. K. S. Parashar, R. N. P. Choudhary and B. S. Murty, Mater. Sci. Eng. B. 110, (2004) 58.

12. S. K. S. Parashar, R. N. P. Choudhary and B. S. Murty, Ferroelectrics, 325, 65 (2005).

13. S. K. S. Parashar, R. N. P. Choudhary and B. S. Murty, J. Appl. Phys. 98, (2005) 1043051.

14. J. C. Dyre, J. Appl. Phys. 64, (1988) 2456. 15. A. I. Lakatos and M. Abkowitz, Phys. Rev. B

3, (1971) 1791. 16. S.S.N. Bharadwaja, P. Victor, P.

Venkatesarulu, and S. B. Krupanidhi, Phys. Rev B, 65, (2002) 174106.

17. M. Pollak and T. H. Geballe, Phys. Rev. 122, (1961) 1742.

18. M. M. El-Nahass, H. M. Zeyada, M. M. El-Samanoudy, and M. M. El-Menyawy, J. Phys.: Condens. Matter.18, (2006) 5163.

19. G. E. Pike, Phys. Rev. B 6, (1972) 1572. 20. A. Pelaize Barranco, F. Calderon Pinar, O.

Perez Martinez, E. Torres Garcia, and A. Huanosta-Tera, Acta. Mater. 49, (2001) 3289.

21. A. Mansingh, J. M. Reyes, and M. Sayer, J. Non-Cryst. Solids 7, (1972)12.

22. M. Pollak and T. H. Geballe, Phys. Rev. 122, (1961) 1742.

23. M. H. A. Pramanik, P. N. Butchier, and I. D. Cox, Philos. Mag. B 47, (1983) 437.

Effect of Annealing on the Optical Constants of (Ge20Se80)90Ag10 Thin Films

Akshay Kumar, F.I. Mustafa, Shikha Gupta and S.K. Tripathi

Department of Physics, Centre of Advanced Study in Physics, Panjab University, Chandigarh-160014. E-mail: [email protected], [email protected]

Abstract

Thin film of chemical composition (Ge20Se80)90Ag10 is prepared by thermal evaporation technique. The optical properties of film before and after annealing are determined by a method, based only on the transmission spectra at normal incidence, measured over the 400-1100 nm spectral range. This method takes into account non-uniform thickness of thermally evaporated thin film. The dispersion of refractive index n(λ) is discussed in terms of the single-oscillator Wemple and DiDomenico model. The optical absorption edge is described using the non-direct transition model proposed by Tauc and the optical band gap (Eg

opt) is determined from the absorption coefficient (α) by Tauc’s extrapolation procedure. It has been found that the value of refractive index (n) and oscillator strength (Ed) increases while oscillator energy (Eo) and optical band gap (Eg

opt) decrease after annealing. The decrease of (Egopt) has been

explained on basis of structural changes arising due to annealing. 1. Introduction

The amorphous chalcogenide thin films exhibit interesting optical properties such as high refractive index n, good transmission in the IR range [1]. The study of optical constants of materials is interesting for many reasons. Firstly, the use of these materials in optical fibers and reflected coating requires accurate knowledge of their optical constants over wide range of wavelength. Secondly, the optical properties of all materials are related to their atomic structure, electronic and band structure. These glasses are very interesting materials for infrared optics. The knowledge of accurate values of the wavelength dependent refractive index of thin film is very important, both from fundamental and technological viewpoints. It yields fundamental information on the optical energy gap, defect levels, phonon and plasma frequencies, etc. Moreover, the refractive index is necessary for the design and modeling of optical components and coating such as interface filters.

Ge-Se is known to be very good covalently bonded glass former. Various physical properties of these glasses show discontinuity when the average coordination number is <r>=2.4. Coordination number of Ge is 4 and Se is 2, so at x=20, the value of average coordination number is <r>=2.4 in a-Ge20Se80 system. In the Se-rich zone, the structure consists of Se chains linked by Ge atoms tetrahedrally coordinated by Se atoms i.e., the structure consists of chains of corner shared GeSe4/2 [2, 3]. As Ge concentration increases, the corner shared tetrahedrons give

place to edge shared ones [4]. It has been seen that the addition of Ag has dual role as an additive in Ge-Se chalcogenide glasses. In Se-rich compositions (x<1/3), Ag acts as a network modifier and phase separates into Ag2Se-rich glass, leaving the GetSe1-t backbone Se deficient (t > x). However, in Ge-rich compositions (x≥2/5), Ag becomes a network former [5]. As a network modifier, a bimodal glass transition temperature (Tg) is reported and attributed to a phase separation of Ag concentrated structures (resembling α-Ag2Se) from the host matrix [6].

We have focused our investigations to see effect of annealing on the optical properties of such a technically important material. In this paper, a relative simple method for determination of optical constants has been used. The optical parameters like refractive index n, absorption coefficient α, oscillator strength Ed, oscillator energy E0, and optical band gap Eg have been determined. It has been observed that these parameters changes after annealing. 2. Experimental Details

Gassy alloy of (Ge20Se80)90Ag10 is prepared by melt quenching technique. Materials (5N pure) are weighted according to their atomic percentages and sealed in quartz ampoules in a vacuum 2×10-5 mbar. The sealed ampoules are kept inside a furnace where the temperature is increased up to 1000oC at a heating rate of 3-4 oC/min. the ampoules are frequently rocked for 12 hours at the highest temperature to make the melt homogeneous. The quenching is done in ice cold water. Thin film of alloy are prepared by vacuum

Effect of Annealing on the Optical Constants of (Ge20Se80)90Ag10 Thin Films 115

evaporation technique on well-degassed Corning 7059 glass substrate at room temperature and base pressure of 2×10-5 mbar using a molybdenum boat. Amorphous nature of the film is confirmed by absence of any sharp peak in the X-ray diffractogram of the film.

The sample is annealed at 130oC for 2 hours under vacuum (10-3 mbar). The normal incidence transmission spectra of both as deposited and annealed films taken over spectral range 400-1100 nm using SOLAR TII, MS 2004. The spectrometer is set with a suitable slit width of 1 nm, in the spectral range. All optical measurements have been performed at room temperature 300K. 3. Results and discussion

The model behind Swanepoel’s method [7] assumes that the sample is a thin film of uniform thickness deposited on a transparent substrate having a refractive index‘s’. The system is surrounded by air, whose refractive index is no=1. The film has a complex refractive index n*= n-ik, where n is refractive index and k the extinction coefficient, which is related to the absorption coefficient (α) through the relation, k=αλ/4π. The optical constants are obtained by using only the transmission spectrum. The refractive index in the region where the absorption coefficient, α is ≈0 is calculated by equation [8]

22 SNNn −+= (1)

where

212

2

minmax

minmax ++

−=

sTT

TTsN (2)

Tmax and Tmin are the envelope values at the wavelengths in witch the upper and lower envelopes and the experimental transmission spectrum are tangent respectively, as shown in Fig.1. The accuracy in λ measurements is ±1 nm. Above Fig.1 shows the normal incidence optical transmission spectra of as deposited and annealed a-(Ge20Se80)90Ag10 thin films. Using equation (1), the values of n are calculated at wavelengths corresponding to the tangent points. If n1 and n2 are the refractive indices at two adjacent tangent points at λ1 and λ2, then according to basic equation for interference fringes λmnt =2 (3) where m is an order number. The thickness (t) is given by

( )nnt

1221

21

4 λλλλ

−=

(4)

4 0 0 5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 1 0 0 0 1 1 0 0 1 2 0 0-0 .1

0 .0

0 .1

0 .2

0 .3

0 .4

0 .5

0 .6

0 .7

0 .8 T m a xa s d e p o s ite da n n e a le d

Tran

smis

sion

T

W a v e le n g th (n m )

a s d e p o s ite d

a n n e a le d 2 h /1 3 0 oC

T m in

Fig. 1. Transmission spectrum of as deposited and annealed a-(Ge20Se80)90Ag10 thin film.

Using equation (3), new more precise values of refractive index and film thinness are determined by a procedure which is explained in detail in [7,8]. The calculated values of refractive index at different wavelengths for both, as deposited and annealed films have been plotted in Fig.2. The data on dispersion of refractive index, n(λ) have been calculated using the single-effective-oscillator model proposed by Wemple and DiDomenico [9,10]. They found that all the data can be described to an excellent approximation by the following relation:

22

0

02

)(1)(

ωω

ηη

−+=

EEE

n d (5)

where ħω is the photon energy, E0 is the oscillator energy and Ed is the so-called dispersion energy. Plotting (n2 – 1)-1 vs. (ħν)2 allows us to determine the oscillator parameters, by fitting a linear function to the smaller energy data. Figure 3 shows the plot of (n2-1)-1 vs. (hν)2, which is a straight line. Ed and E0 can be directly determined from the slope, (EdE0)-1 and the intercept, E0/Ed, on vertical axis.

400 500 600 700 800 900 1000 1100 12001.61.82.02.22.42.62.83.03.23.43.63.84.04.24.4

as deposited annealed

Ref

ract

ive

inde

x (n

)

W avelength (nm)

annealed 130oC/2h

as deposited

Fig. 2. Variation of n with wavelength (λ) in as deposited and annealed film.


3 4 5 6 7

0.020.040.060.080.100.120.140.160.180.200.220.240.260.280.300.320.34

(n

2 -1)-1

(hv)2(eV)2


annealed 130oC/2h

as deposited

Fig. 3. Plot between 1/ (n2-1) and (hν)2. After annealing, the values of n(0) and Ed increases from (1.914±0.004) – (2.268±0.004) and (8.44±0.01) – (12.36±0.01) respectively. The value of E0 decreases from (3.23±0.01) – (2.99±0.01) eV.

Finally, the optical gap (Egopt) is calculated

from the intersection of the plot (αħω)1/2 vs. ħω with the abscissa axis as shown in Fig. 4. The value of Eg

opt decreases from (1.80±0.01) – (1.72±0.01) after annealing the a-(Ge20Se80)90Ag10 thin film at 130oC for 2 hours under vacuum (10-3 mbar). The observed increase in refractive index n(0) and oscillator strength (Ed) can be explained on the basis of fact that upon annealing Se-Se bridges linked with edge sharing GeSe4/2 units, are converted into Se-Se linear chains. Upon annealing, this material may phase separates into an Ag2Se rich glass and decreasing the disorder in the material as suggested by Mitkova et al [5]. The decrease of (Eg

opt) from (1.80±0.01) – (1.72±0.01) after annealing is explained by fact that the binding energy of the Ag-Se bond (202.5 kJ mol-1) is smaller than that of the Ge-Se bonds (484 kJ mol-1). Therefore, there is a smaller energy splitting between the states of the valence and conduction band takes place.

1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6

50

100

150

200

250

300

350

400

450

500

550

600

as deposited

annealed 130oC/2h


(αhv

)1/2 (c

m-1eV

)1/2

hv(eV)

Fig. 4. Plot between (αħν)1/2 and ħν.

4. Conclusions

The effect of annealing on the optical properties of a-(Ge20Se80)90Ag10 thin film have been studied. It is found that values of n, Ed are increased after annealing. The value of (Eg

opt) and E0 decreases after annealing. The changes occur are explained on the basis of structural changes. Acknowledgements

This work is financially supported by CSIR (Major research project), N. Delhi. One of the authors A. Kumar is thankful to CSIR New Delhi, for providing financial assistance. References

[1] P. Nernec, M. Frumar, B. Frumarova, M. Jelinek, J. Lancok, J. Jedelsky, Opt. Mat. 15 (2005) 191.

[2] X. Feng, W.J. Bresser, P. Boolchand, Phys. Rev. Lett. 78 (1997) 4422.

[3] E. Sleeckx, L. Tichy, P. Nagels, R. Callaerts, J.Non-Cryst. Solids 198 (1996) 723.

[4] A. Kumar, S. Goel, S.K. Tripathi, Phys. Rev. B 38 (1988) 13432.

[5] M. Mitkova, Yu. Wang, P. Boolchand, Phys. Rev. Lett. 83 (1999) 3848.

[6] Y. Wang, M. Mitkova, D.G. Georgiev, S. Mamedov, P. Boolchand, J. Phys.: Condens. Mat. 15 (2003) S1573.

[7] R. Swanepoel, J. Phys. E: Sci. Instrum. 17 (1984) 896.

[8] R. Swanepoel, J. Phys. E 16 (1983) 1214. [9] S. H. Wemple, M. DiDomenico, Phys. Rev.

B 3 (1971) 1338. [10] S. H. Wemple, Phys. Rev. B. 7 (1973) 3767.

Effect of UV- Irradiation on Chemically Deposited Nanocrystalline CdSe Films

Charita Mehta, Jasim M. Abbas, G.S.S. Saini and S.K. Tripathi*

Department of Physics, Centre of Advanced Study in Physics, Panjab University, Chandigarh-160 014 India. E-mail: [email protected]; [email protected]

Abstract

Nanocrystalline films and nanocrystals of cadmium selenide are chemically deposited from an alkaline

bath using sodium selenosulphate as Se2- ion source. The formation of CdSe has been confirmed with the help of infrared spectroscopy by observing bands corresponding to the multiphonon absorption. The samples characterized by optical spectroscopy and X-ray diffraction have the nanocrystal radii 2-4 nm. Optical gaps (Eg) higher (up to 0.5 eV) than in single crystal samples are observed and explained in terms of quantum size effect. Increase in Eg depends strongly on deposition temperature with a greatest increase obtained at lowest temperature. Structural studies of as-deposited layers showed them to be composed of nanocrystalline cubic CdSe depending on deposition temperature. We demonstrate the effect of UV- irradiation on the optical properties of nanocrystalline films. 1. Introduction Nanocrystalline materials have opened new chapter in the field of electronic application since they exhibit scientifically interesting phenomena and desirable engineering characterstics. Properties of the materials in the nanometer scale could be changed by changing the crystallite size and/or shape of the film. Semiconductor nanoparticles have attracted great interest in both theoretical and applied research areas [1-2]. CdSe an important member of luminescent II-VI family having luminescence in the visible range of optical spectra, has shown potential to be used in nanocrystalline form in biological field [3], TFTs [4], diodes and lasers [5], solar cells [6], and other nanoscale devices [7]. By appropriate choice of CdSe nanocrystal (NC) size, the absorption edge can thus be made to fall anywhere in the visible region. In recent years, major attention have been given to the investigation of electrical and optical properties of CdSe thin films in order to improve the performance of the devices and also for finding new applications [8]. CdSe nanocrystalline films have been prepared by various techniques including chemical bath deposition (CBD) [9]. CBD aside of its industrial use, is also a cheap technique for preparation of semiconductor crystallites with dimensions of a few nanometers. Under certain preparation conditions, the chemically deposited films are nanocrystalline and exhibit quantum size effect (confinement of carriers within individual NCs). The quantum confinement

effects strongly the optical properties of the film in the same way as in the case of individual nanocrystals. The optical properties of NCs can be changed by varying the parameters of chemical bath deposition and/or by subsequent heat or chemical treatment of the films [9, 10-11]. The ability to control the optical properties of NCs has made this field of material science very interesting for fundamental understanding and technological applications. CdSe thin films investigated in this paper consist of closely spaced NCs, properties of which show many similarities with those of isolated NCs. 2. Experimental

Substrate cleaning plays an important role in the deposition of thin films. Commercially available glass slides were boiled in chromic acid for 2 h, washed with detergent, rinsed in acetone and finally cleaned with double distilled water before use. Cadmium selenide thin films were deposited on glass substrates by the method of CBD. In this method, the solution chemistry is chosen such that a spontaneous reaction from liquid phase is possible. When the ionic product exceeds the solubility product, precipitation occurs and ions combine on the substrate and in the solution to form nuclei which result in thin film formation on the substrate and the precipitation in the solution. The pH value of the reaction system is of prime importance for the chemical deposition of CdSe thin films. The precursor of selenide ions, sodium selenosuphate, was used in the form of solution, which has been obtained by adding selenium powder to a hot solution of sodium

mailto:[email protected]


sulphite, magnetically stirring this mixture for several hours at 80oC and filtering the excess of selenium. This solution is relatively unstable and therefore it must be freshly prepared prior to thin film deposition process. The optimal chemical composition of reaction system for preparation of highly reflecting CdSe nanocrystalline films were obtained by mixing the following solutions; 10 ml of 0.3M cadmium acetate solution taken in 100 ml capacity glass beaker and to it, 10 ml of 0.13M sodium citrate, 10 ml of sodium selenosulphate and 20 ml of double distilled water was slowly added with constant stirring. The final ph of the chemical bath was 8.0± 0.1. The solution was stirred for sometime and then transferred into another beaker containing cleaned glass substrates. The glass substrates were held vertically in the bath by placing them against the walls of beakers containing the deposition mixture. The bath solution was also formed at different temperatures at constant pH. After about 6 h the deposited films were thoroughly washed with double distilled water and dried in air. The CdSe thin films were uniform, well adherent to the substrates and red-orange in color. Crystallographic study was carried out using a Phillips PW-1710 X-ray diffractometer using CuKα radiation in the 2θ range from 10 to 70. To study the optical properties of n-CdSe thin films, the transmission spectra were recorded using Monochromator-spectrograph [SOLAR TII, MS 2004] in the transmission range 400-1000 nm for all samples. FTIR spectra have been recorded using Perkin Elmer PE-RX 1 FTIR spectrophotometer. The spectral resolution of the IR spectrophotometer was 1 cm-1 throughout the experiment. 3. Results and discussion Fig. 1 shows the diffraction spectrum of n-CdSe thin film. In this spectra, there is a highest intensity reflection peak at 2θ = 25.3 [111], with two small another intensity peaks at 2θ = 41.8 [220] and 50 [311]. The comparison of observed ‘d’ values with standard ‘d’ values [12, 13] confirms that the deposited film is having sphalerite cubic (zinc blende type) nanocrystalline structure without an interfacial layer. Information of the strain and the particle size are obtained from the full width at half maximum (FWHMs) of the diffraction peaks. The FWHMs (β) can be expressed as a linear combination of the contributions from the strain

(ε) and particle size (L) through the following relation [14].

λ

θελ

θβ sin1cos+=

L (1)

Fig. 2 represents the plots of (βcosθ)/λ versus (sin θ)/λ for n-CdSe thin film which is a straight line. The slope of the plot gives the amount of residual strain, which turns out to be -8.11×10-2 for n-CdSe thin film. The reciprocal of intercept on the (βcosθ)/λ axis gives the average particle size as ~ 3.3 nm. The negative value of residual strain for the as-deposited film indicates the compressive strain. If the film is deposited free from impurities, the compressive strain is generated at the thin film substrate interface, when the very small crystallites are bonded to substrates due to surface tension effect.

0 10 20 30 40 50 6010

20

30

40

50

60

70

80

90

100

Inte

nsity

2θ

[111

]

[220

]

[311

]

n-CdSe

Fig. 1. The XRD pattern of CdSe thin films deposited at 300 K at pH 8.

0.20 0.25 0.30 0.35 0.40 0.450.00

0.01

0.02

0.03

0.04

0.05

β co

s θ

/ λ

sin θ / λ

Fig. 2. Plot of sinθ/λ vs. βcosθ/λ for CdSe thin film.

Effect of UV- Irradiation on Chemically Deposited Nanocrystalline CdSe Films 119

4000 3500 3000 2500 2000 1500 1000 5000

20

40

60

80

100

619

921

1113

1387

1561

Tran

smitt

ance

(%)

Wavenumber (cm-1)

3368

519

n-CdSe

Fig. 3. shows the FTIR spectrum of CdSe nanocrystals. The presence of the band at 1387 cm-1 and 1561cm-1 confirms the presence of capping agent trisodiumcitrate used for above study. The former band can be assigned to symmetric stretching of COO-, while the later band can be assigned to the asymmetric COO - [15]. Another band at 3368 cm-1 can be assigned to OH stretching of trisodium citrate The presence of above mentioned bands shoulders around 1420 cm-1 shows that trisodium citrate is bounded to the CdSe nanocrystals and it is arresting the growth of bulk crystals of CdSe.

Fig. 3. FTIR spectrum of CdSe nanocrystals. Optical properties are studied by recording the transmission spectra of the films. Fig. 4 shows the transmission data of n-CdSe thin films deposited at different temperatures of bath solutions i.e., 300 K, 323 K, 353 K. From the transmission data, nearly at the fundamental absorption edge, the values of absorption coefficient (α), are calculated in the region of strong absorption using the relation

⎟⎠⎞

⎜⎝⎛=

Td1ln1α (2)

The fundamental absorption, which corresponds to the transition from valence band to conduction band, can be used to determine the band gap of the material. The relation between α and the incident photon energy (hν) can be written as [16]

( )

νν

αh

EhA ng−

= (3)

where A is a constant, Eg is the band gap of the material and the exponent n depends on the type of transition. The n may have values 1/2, 2, 3/2 and 3 corresponding to the allowed direct, allowed indirect, forbidden direct and forbidden indirect transitions, respectively.

300 400 500 600 700 800 900 1000 1100 1200

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

T% λ(nm)

pH 8 300 K 333 K 353 K

Fig. 4. Plot of transmission vs. wavelength for n-CdSe thin films deposited at different temperatures.

The value of Eg is calculated by extrapolating

the straight line portion of (αhν)1/n vs hν graph to hν axis taking n = 0.5. Fig. 5 shows the plots of (αhν)2 vs hν for as-deposited and annealed thin films.

1.0 1.5 2.0 2.5 3.0

0.00E+000

1.00E+010

2.00E+010

3.00E+010

4.00E+010

5.00E+010

6.00E+010

(αhν

)2

hν(eV)

pH 8 300K 333K 353K

n-CdSe

Fig. 5. Plot of (αhν)2 vs. hν for n-CdSe thin films. The correct values of the optical gap calculated from the plots are (2.25 ± 0.01) eV,


(2.00 ± 0.01) eV and (1.90 ± 0.01) eV for the films deposited at different temperatures. The value of Eg is found to decrease with the increase in the deposition temperature. These values of optical gap are inserted in Table 1. Clearly, the observed values of Eg are higher than the value of bulk optical gap of CdSe [(1.74 ± 0.01) eV] due to quantum confinement in the CdSe nanocrystallites. In order to explore further the properties of CdSe NC films, we studied the effect of UV irradaiation at different time intervals on the optical properties of the same. Fig. 6 shows the transmission data of n-CdSe thin films irradiated by UV light at different time intervals of 1 h, 4 h,and 5h.

300 400 500 600 700 800 900 100011001200

0.2

0.3

0.4

0.5

0.6

0.7

0.8

T%

λ(nm)

no UV irradiationUV irradiation for 1 hrUV irradiation for 4 hr UV irradiation for 5 hr

Fig. 6. Plot of transmission vs wavelength for CdSe NC films irradiated with UV light at different time intervals. Fig. 7 shows the plots of (αhυ)2 vs energy for n-CdSe films. Band gap decreases from (2.17 ± 0.01) eV, (2.1 ± 0.01) eV, (2.07 ± 0.01) eV and (2.0 ± 0.01) eV when UV light is irradiated for time intervals of 1 h, 4 h and 5 h respectively. Thus on irradiating UV light at increasing time intervals particle size of CdSe NCs increases. 4. Conclusions We have observed that the n-CdSe films deposited by chemical bath technique at different temperatures grow with nanocrystalline

1.0 1.5 2.0 2.5 3.0 3.5

0.00E+000

5.00E+009

1.00E+010

1.50E+010

2.00E+010

2.50E+010

3.00E+010

3.50E+010

4.00E+010

(αhν

)2

hν(eV)

no UV irradiation UV irradiation for 1 hrUV irradiation for 4 hrs UV irradiation for 5 hrs

n- CdSe

Fig. 7. Variation of energy with (αhυ)2 for n-CdSe films. phase. The optical study shows that the CdSe thin films are size quantized i.e., nanocrystals behave as quantum dots with blue shifted band gap energy of 0.5 eV in comparison to the bulk value. Furhter on irradiating UV light at different time intervals over n-CdSe films band gap decreases. Acknowledgements This work is financially supported by CSIR (major research project), New Delhi. References [1] V.I. Klimov, A.A. Mikhailovsky, S. Xu, A.

Malko, J.A. Hollingsworth, C.A. Leatherdale, H.J. Eisler, M.J. Bawendi, Science 290 (2000) 314.

[2] C.H. Lu, B. Bhattacharjee, C.H. Hsu, S.Y. Chen, R.C. Ruaan, W.H. Chang, J. Electroceram 17 (2006) 21.

[3] W. Chan, S. Nie, Science 281 (1998) 2016. [4] X.F. Duan, C.M. Niu, V. Sahi, J. Chen, J.W.

Parce, S. Empedocles, J.L. Goldman, Nature 425 (2003) 274.

[5] J.H. Park, J.Y. Kim, B.D. Chin, Y.C. Kim, J.K. Kim, O.O. Park, Nanotechnology 15 (2004) 1217.

[6] Y. Kim, S.H. Kim, H.H. Lee, K. Lee, W. Ma, X. Gong, A. J.Heeger, Adv. Matter 18 (2006) 572.

[7] W. Cai, D.W. Shin, K. Chen, O. Gheysens, Q. Cao, S.X. Wang, S.S. Ghambhir, X.X. Chen Nano lett. 6 (2006) 669.

Effect of UV- Irradiation on Chemically Deposited Nanocrystalline CdSe Films 121

[8] M.J. Lee, S.C. Lee, solid state electronics 43, (1999) 833.

[9] G. Hodes, Chemical Solution Deposition of Semiconductor films, Marcel Dekker, New York, Basel, 2003.

[10] S.K. Sarkar, N. Chandesekharan, S. Gorer, G. Hodes, Appl. Phys. Lett. 81 (2002) 5045.

[11] L. Xu, K. Chen, H.M. El-Khair, M. Li, X. Huang, Appl. Surf. Sci. 172 (2001) 84.

[12] JCPDS Data File No. 8-459.

[13] JCPDS Data File No. 19-191. [14] S.B. Qadri, E.F. Skelton, D. Hsu, A.D

Dinsmore, J. Yang, H.F. Gray, B.R. Ratna, Phys. Rev.B 60 (1999) 9191.

[15] X. Zou, E. Ying, S. Dong, Nanotechnology 17 (2006) 4758.

[16] J.I. Pankove, Optical Processes in Semiconductors, Englewood Cliffs. NJ: Prentice-Hall (1971).

.

Optical and Structural Investigations of TiO2 Films Deposited on Transparent Substrates by Sol-Gel Technique

Sudhir Kumar Sharma†, M. Vishwas*, K. Narasimha Rao†, S. Mohan† and K. V. A. Gowda*

†Department of Instrumentation, Indian Institute of Science, Bangalore, (Karnataka State)-560012 *Department of Physics, M. V. J. College of Engineering, Channasandra, Bangalore-560067

E Mail: [email protected]. [email protected]

Abstract

An inexpensive and effective method for the preparation of nano-crystalline TiO2 (anatase) thin films at room temperature on different transparent substrates is presented. This method is based on the use of peroxo-titanium complex i.e. Titanium isopropoxide as a single initiating organic precursor. Post-annealing treatment is necessary to convert the deposited amorphous film into TiO2 crystalline (anatase) phase. The films obtained are almost uniform, compact and free of pinholes. These films have been characterized for optical studies like transmittance-reflectance measurements, X-ray diffraction studies (XRD) and atomic force microscopic (AFM) studies. The optical constants such as refractive index and extinction coefficient have been estimated by using envelope technique. Also, The energy gap values have been estimated using Tauc’s formula for the glass and quartz substrate and found to be 2.35 eV and 2.39 eV approximately. 1. Introduction

Titanium oxide is one of the most extensively studied transition metal oxide. TiO2 possesses excellent properties such as chemical resistance, good mechanical strength, transparency, as well as insulating properties. Titanium dioxide (TiO2) films are is widely used in many optical, biomedical, and microelectronic applications because of low cost material with a high refractive index and high dielectric constant [1]. These TiO2 films carried out for the different applications as in opto-electronic devices, sensors, dye sensitized photo voltaic cells, electro chromic displays, planner wave guides and photo-catalysts etc. [2,3]. The preparation of TiO2 thin films has received great attention during the past several decades because of its remarkable optical, photo-catalytical and electronic properties. Titanium dioxide films can be synthesized by various thin film deposition techniques, such as thermal evaporation [4,5] chemical vapor deposition [6], Metal organic chemical vapor deposition [7], pulsed laser deposition [8] and sol gel process [9]. Thin films of TiO2 are frequently used for coatings due to their high refractive index, and high stability, and its properties are easily affected by the technological conditions of the deposition process such as the substrate temperature and oxygen partial pressure as well as by the post deposition heat treatment [10]. Tain et al [11] have demonstrated that the quantum confinement effect in their TiO2 thin films prepared by electron beam evaporation. They

observed with increasing the grain size, the band gap shifts to from 3.4 eV to 3.2 eV. Optical properties of TiO2 thin films synthesized by sol–gel process [12], hydrazine processes [13] have been reported by several investigators.

Apart from these investigations, oxide films are synthesized by sol–gel method because of several advantages, such as low processing temperature, homogeneity, possibility of coating on large area substrates, and the most important cost effective. Unlike physical vapor deposition (PVD) or chemical vapor deposition (CVD) coating technologies, sol–gel technology does not require any high vacuum equipment. Sol–gel / oxide films have many applications as functional materials in optical, microelectronics, and opto-electronics and for the purpose of corrosion, scratch, abrasion resistance [14]. In the present study, TiO2 thin films were prepared by spin coating of untreated sol on to uncoated glass and quartz substrates. The influence of annealing temperature was investigated on the structural and optical properties of TiO2 thin films.


The deposition of the TiO2 films was accomplished by sol gel technique at room temperature. The TiO2 films have been prepared by dissolving the titanium alkoxide precursor i.e. titanium isopropoxide



Optical and Structural Investigations of TiO2 Films Deposited on Transparent Substrates 123

[Ti(OC3H7)4] in an alcoholic bath. To avoid the early precipitation of the oxides 5 ml of concentrated HCl is used per 100 ml of ethyl alcohol. After the addition of the concentrated HCl the solution was stirred vigorously up to one hour. The clear solution obtained was kept in airtight beaker for one hour for the gel formation. Then the gel has been spin coated on glass and quartz substrates. The films were preheated in air at 60O C for 4 hours. After preheating, the films were annealed at different temperatures like 250O C, 300O C and 350O C for 8 hours respectively.

The optical transmittance was measured by single beam spectrophotometer (Ocean optics, USA) after annealing at different temperatures. The optical constants like refractive index and extinction coefficients were estimated by envelope technique [15]. XRD spectrum was recorded by using Shimatzu SPM 6000 in grazing angle mode. AFM images were recorded using the Nanoscope IIIa scanning probe microscope in a contact mode.

3. Results and Discussion 3. 1. Optical Studies

The optical transmission spectrum for the spin coated TiO2 films have been recorded in the UV and visible-IR regions. The transmittance spectrum obtained for the films deposited on glass and quartz substrate after annealing at 350oC is shown in Fig.1.

It is evident from Fig. 1 that the percentage transmission is approximately 90% for both the films deposited on transparent (glass and quartz) substrates in the visible region. High transmittance inferred that the sol-gel deposited films could be used for optical coating applications. The absorption edge of the films without annealing is at approximately 350 nm. Annealing at different temperatures lead the

400 600 800 1000 1200 14000

20

40

60

80

100

Quartz

Glass

Perc

enta

ge T

rans

mitt

acne

(%T)

Wavelength (nm) Fig. 1. Measured spectral transmittance of the TiO2 film on glass and quartz substrates annealed at 350oC

200 400 600 800 1000 1200

2.06

2.08

2.10

2.12

2.14

2.16

Glass

Quartz

Ref

ract

ive

Inde

x

Wavelength (nm) Fig. 2. The plot of refractive index of the TiO2 film on with wavelength estimated by envelope technique on transparent substrates increase in the percentage transmission in the region 340-400 nm range, but the absorption edge does not shift with annealing temperatures. It can be inferred from this that TiO2 films exhibit similar absorption behavior at wavelengths below 350 nm, either as deposited or annealed. The thickness of the films have been estimated by envelope technique [15] and found to be the range 100 nm to 150 nm.

The estimated variation of optical constants like refractive index and extinction coefficient with wavelength for the TiO2 films deposited on glass and quartz substrates after annealing at 350oC are shown in Fig. 2 and Fig. 3 respectively. The refractive index plot follows almost exponential decrease in the refractive index with increase in wavelength in the beginning and the variation is less at higher wavelengths. The estimated value of refractive index for glass and quartz substrate was found to be 2.11 and 2.08 at 510 nm respectively. Also, the decrease extinction coefficient with increase in wavelength has been

300 400 500 600 700 800 900 1000 1100-0.005

0.000

0.005

0.010

0.015

0.020

0.025

0.030

0.035

Quartz

GlassExtin

ctio

n co

eff.

Wavelength (nm)

Fig. 3. The plot of extinction coefficient of the TiO2 film with wavelength for glass and quartz substrates.


2.00 2.25 2.50 2.75 3.00 3.25 3.50 3.75 4.00-500

0

500

1000

1500

2000

Quartz Glass

(ahn

)2 x1012

(m-2 e

V2 )

Energy (eV) Fig. 4. The calculated plot of (αhν)2 vs (hν) for TiO2 films for glass and quartz substrates estimated by envelope technique [15]. The films deposited on glass and quartz substrates show vary low value of extinction coefficient in the visible region. The band gap estimation of TiO2 thin films deposited on quartz and glass using the Tauc’s Equation [16] is given as below has performed glass substrates

where a, absorption coefficient, Eph, incident photon energy, Eg, band gap of the material, k, is the proportional constant and a = 1 for direct transmission. A plot of (αhν)2 vs (hν) for TiO2 films deposited on glass and quartz substrates is shown in Fig. 4. The band gap (Eg) estimations for the TiO2 thin films have been extrapolated from linear part of the curve as shown in the Fig. 4. The optical band gap for the films annealed at 350o C found to be 3.35 eV and 3.39 eV for quartz and glass substrates respectively. The estimated band gap is compared with the bulk TiO2 (Eg = 3.2 eV) resulted an blue shift, which could be assigned due to decrease in the grain size of the deposited and treated films compared to the grain size of the bulk TiO2.

3. 2. XRD Studies X-ray diffraction technique has been employed to identify the structure of the films. The XRD pattern recorded for the TiO2 films deposited on glass substrate as deposited condition and annealed at 350o C are shown in Fig.5. It is clear from the Fig. 5 that before annealing, no diffraction peaks of TiO2 have observed indicating that the film is amorphous. Broad hump in the XRD pattern is characteristic of amorphicity and it is seen in all the

samples, in the low 2θ region extending from 15o to 40o, irrespective of the annealing temperature/ duration.

20 30 40 50 60 70-20

0

20

40

60

80

100

120

140

160

180

Anealed at 350o C

As deposited Film

(111

)

(200

)

Inte

nsity

(a.u

.)

2 Theta

Fig. 5. X-ray diffraction pattern for TiO2 films for glass substrate annealed at different temperatures This broad hump defines an amorphous phase in these films and is assigned to the underlying glass substrate. Also big hump recorded from 2θ=20 to 40o indicates the presence of glass substrate [17]. However annealing at 350oC, two sharp diffraction peaks at 2θ=28.37o and 47.35o are observed corresponding to inter-planner spacing of 3.14 Å and 1.89 Å indicates (111) and (220) planes.

These peaks clearly indicate the presence of anatase (crystalline) phase of TiO2. The threshold for the appearance of crystallinity in the films seems to be in the temperature range of 250 to 300o C. The films treated at 350o C and above are thus characterized by the existence of a crystalline phase [18]. The crystalline phase is identified as the anatase phase (a = 0.37 nm, c = 0.95 nm) with the single observed diffraction peak oriented along the (111) crystallographic plane, which corresponds well with the JCPDS data file number 21-1272. Annealing TiO2 films in the range from 250 to 500o C, the intensity of the diffraction peaks has been increased, which inferred that TiO2 film becoming more crystalline. The crystallite sizes have shown an increase as a function of annealing temperature. The variations in the grain size with temperature/duration of annealing treatment are well known. 3. 3. AFM Measurements

Fig. 6 illustrates the two dimensional (2D) atomic force microscopic images (AFM) of

Optical and Structural Investigations of TiO2 Films Deposited on Transparent Substrates 125

TiO2 thin film deposited on glass substrates annealed at 350o C for six hours. It exhibits almost uniform spherical grain like structure different sizes ranging from 20 to 120 nm.

Fig. 6. The morphological image (AFM) recorded of TiO2 film on glass substrates annealed at 350o C

Dense and compact surface morphology is

evident from the Fig. 6. As the annealing temperature increased further, it leads to the coalescence of grains to form ripple morphology not shown here. The wrinkled ripple pattern is due to minimization of surface energy to form a stable self-assembled pattern. The root mean square (RMS) roughness of 9 nm was evaluated from 9 mm2 surface area. The roughness in the films with thickness decreases from about 9.0 to 3.6nm (in RMS), when the annealing temperature was increased from 100 to 500o C [19].

4. Conclusions

To summarize, we have successfully prepared nano-crystalline TiO2 thin films on glass/quartz substrates using Titanium isopropaoxide precursor by employing a simple and inexpensive sol-gel technique. The optical spectral transmittance studies indicate that these films could be used for good optical coating applications after annealing at 350o C. The energy band gap estimation for TiO2 films was estimated by using Tauc’s formula and band gap energy found to be almost equal to 3.4 eV, which is independent of both the nature of the substrate and annealing temperature. XRD shows that as-deposited films are amorphous and became polycrystalline with anatase crystal structure, oriented along (111) plane upon annealing at 350o C. The surfaces of spin-coated films appeared to be crack free on the explored section (in 5X5 mm2

area) and exhibited a better homogeneity in thickness.

Acknowledgements

The author Mr. S. K. Sharma is very grateful to Mr. Ravindra Sahoo, Dept. of Chemistry, IIT

Bombay, for the AFM investigations of sol gel deposited films.

References 1. K. Narasimha Rao, Opt. Engg. 41 (2002)

2357. 2. M. Flischer and H. Meixner, Sensors and

Actuators B 4 (1991) 437. 3. Y. Q. Li, S. Y. Fu, G. Yang and M. Lee, J.

of Non -crystalline Solids 352 (2006) 3339.

4. D. Bhattacharyya, N. K. Sahoo, S. Thakur and N. C. Das, Thin Solid Films 360 (2000) 96.

5. K. Narasimha Rao, M. A. Murthy and S. Mohan, Thin Solid Films 176(2) (1989) 181.

6. Wen-zhen Zhang, Tao Zhang, Wen Yin and Geng-yu Cao, Chin. J. Chem. Phys. 20(1) (2007) 95.

7. P. Babelon, A. S. Dequiedt, H. Mostéfa-Sba, S. Bourgeois, P. Sibillot and M. Sacilotti, Thin Solid Films 322 (1998) 63.

8. E. Gyorgy, G. Socol. E. Axente, I. N. Mihailescu, C. Ducu and S. Ciuca, Applied Surface Science 247 (2005) 429.

9. Jorge Medina-Valtierra, M.S.Cardenas, C.F-Reyes and S.Calixto, J. Mex. Chem. Soc. 50 (1) (2006) 8.

10. M. Radecka, K. Zakrzewska, H. Czternastek, T. Sapinski, Applied Surface Science 65 (1993) 227

11. G.L. Tain, H.E. Hong-Bo, J.D. Shao, Chinese Physics Letters 22 (2005) 1787

12. M. Koelsch, S. Caignon, J.F. Guillemoles, J.P. Jolivet, Thin Solid Films 430 (2002) 2002

13. K. Madhusudhana Reddy, S.V. Panorama, A. Ramachandra Reddy, Materials Physics and Chemistry 78 (2002) 239

14. D. Basak, G. Amin, B. Mallik, G.K. Paul, S.K. Sen, Journal of Crystal Growth 256 (2003) 73

15. R. Swanepoel, J. Phys. E: Sci.Instrum. 16 (1983) 1214

16. J. Tauc, Mat. Res. Bull. 5 (1970) 721 17. Amita Verma, S.A. Agnihotry,

Electrochimica Acta 52 (2007) 2701 18. H.P. Deshmukh, P.S. Shinde b, P.S. Patil,

Materials Science and Engineering B 130 (2006) 220

19. L. Sun, P. Hou , Thin Solid Films 455 –456 (2004) 525

Effect of Annealing Temperature on the Optical Properties of (Pb0.8Sr0.2)TiO3 Thin Films

K. C. Verma1, K. Singh1, Vinay Gupta2, and N. S. Negi1

1Department of Physics, Himachal Pradesh University Shimla-171005, India 2Department of Physics and Astrophysics, University of Delhi, India


Abstract Pb0.8Sr0.2TiO3 (PST20) thin films have been synthesized by Metallo-organic decomposition (MOD) technique. The films were deposited and annealed on fused quartz substrates at different temperatures 550oC-750oC. The films were characterized by using XRD and UV-VIS-NIR spectrometer to determine the structural and optical behaviors under different annealing conditions. All the PST20 films were crystallized in tetragonal structure. The optical properties showed that the film annealed at 650oC was comparatively more transparent. 1. Introduction

(Pb1-xSrx)TiO3 thin films are of having high dielectric constant and have been studied for dielectric devices such as a capacitor material for dynamic random access memory and tunable microwave devices [1-3]. In addition, the large electro- optical refractive index of this material exhibits interesting applications in surface acoustic wave (SAW) and integrated acoustic-optic devices [3-5]. In telecommunication devices, electrical interconnect are limited to their speed, packaging, and power dissipation [6]. In order to overcome these limitations, optical interconnects are now being in use either in long distance networks such as fiber-based optical networks or in onboard and intrachip connections. Thin film of materials offer a greater flexibility than bulk materials for the fabrication of integrated optical devices. For example, by using thin films it is possible to achieve a high refractive index contrast between the film and the cladding layer. This makes possible fabrication of highly confined wave-guides and therefore of dense optical wiring. Polycrystalline PST20 thin films exhibit optical dispersion over the refractive index in the ultraviolet, visible and near infrared (UV-VIS-NIR) regions similar to BST thin films [7].

In the present work, we report on systematic study of the influence of annealing temperatures on the structural and optical properties of spin-coated PST20 thin films processed by Metallo-organic decomposition (MOD) technique. The optical dispersion in films has been analyzed over different optical regions. 2. Experimental

Pb0.8Sr0.2TiO3 formulation were prepared using Lead 2-ethylhexanoate (C7H15COO)2Pb, Strontium 2-

ethylhexanoate (C7H15COO)2Sr and tetra-n-butyl orthotitanate as precursors in xylene. The coating solution was prepared by mixing the precursors in the molar ratio of Pb:Sr:Ti :: 0.8:0.2:1 in xylene. The solution was refluxed at 120oC with constant stirring for 12h to homogenize and polymerize. The solution was coated on fused quartz substrate by spin-coating technique with 4300rpm for 60s. The spun-films were subjected at 350oC for 5min baking to remove the solvent and organic residuals. In the present work the spin-on coating and baking steps were sequentially repeated for four times. A post-deposition isothermal annealing of spin-on coating at 550oC, 650oC and 750oC for 3h in O2 ambient leads to formation of crystallized PST20 films. The crystalline structures of the PST20 thin films were analyzed by X-ray diffraction (XRD) using a Philips (model PW3710). The optical properties were carried out using a Parkin Elmer UV-0637 spectrometer. 3.Results and Discussion

The crystalline structures of the PST20 films on the fused quartz substrates with varying annealing temperatures from 550oC to 750oC are illustrated in fig.1. In case of the film annealed at 550oC, the XRD spectrum indicates a low crystallization having tetragonal perovskite phase without any secondary phase formation. The film annealed at 650oC exhibits better crystallization with well-defined perovskite phase. At higher annealing temperature of 750oC, film shows improvement in (001) and (101) orientations.

The optical properties of PST20 films have been studied at different annealing temperatures by measuring transmittance versus wavelength spectra in ultraviolet, visible and near infrared (UV-VIS-NIR) regions and are presented in fig.2.


Effect of Annealing Temperature on the Optical Properties of (Pb0.8Sr0.2)TiO3 Thin Films 127

2 0 3 0 4 0 5 0 6 0

111

2 θ

5 5 0 oC

211

112

201/

210

102

200

002

110

101

100

001

Inte

nsity

(arb

.uni

t)

6 5 0 o C

7 5 0 oC

Fig. 1. XRD Pattern of PST20 thin films The transmission spectra indicate abrupt

decrease in transmission near the fundamental absorption region. The films annealed at lower temperatures exhibited less transmission in the visible region. The percentage of transmission improves with the increase in annealing temperature. The film annealed at 650oC reveals highest transmission (~77%) than films annealed at other two temperatures i.e. 550 and 750oC. The transmission percentage decreases steadily on further increase in annealing temperatures to 750oC. The absorption line also shifts to the U-V region. The lower transmission observed at 550oC is due to lower crystallization observed in the XRD results and the grains are highly dens. The minimum transmission for PST20 film annealed at 750oC is due to its large grain size.

400 500 600 700 800 900 1000 1100 12000

10

20

30

40

50

60

70

80

90

100

Tran

smitt

ance

(%)

Wavelength (nm)

550oC 650oC 750oC

Fig. 2. Transmittance versus Wavelength of PST20

The wavelength dependence refractive index was evaluated from the interference zone observed in the transmittance spectra using the relation [8,9].

n = [N + (N2 – no

2n12)1/2]1/2 ----- (1)

Where, N = 2 no n1 /Tm – (no2 + n1

2) /2 ----- (2) for transparent region

and N = 2 no n1 ( TM – Tm) / TM Tm + (no

2 + n12) / 2 --(3)

for weak and medium absorption region. TM and Tm are the values of maximum and minimum transmission at a particular wavelength. n1 and no are the refractive index of substrate and air respectively. The refractive index versus wavelength behaviors for PST20 film annealed at different temperatures is shown in fig. 3. The refractive index (n) decreases rapidly as the wavelength is increased. The value of refractive index is low at higher wavelength for

400 500 600 700 800 900 1000 1100

2.8

3.0

3.2

3.4

3.6

2.4

2.8

3.2

3.6

2.2

2.3

2.4

2.5

2.6

2.7

2.8

Wavelength (nm)

550oC

Refra

ctiv

e in

dex

(n)

650oC

750oC

Fig. 3. Refractive index versus wavelength of PST20 PST20 film annealed at 650oC, indicating an improvement in the transparency. The value of n for PST film annealed at 650oC measured at λ ~700nm is 2.3 which is lower than the films annealed at 550oC (n~2.8) and 750oC (n~2.6). At λ ~ 500nm, the highest value of refractive index was obtained for the film annealed at 650oC. This is due to the fact that the film grows in a closed-packed columnar grain mode with smooth surfaces of low rms roughness at 650oC.


4. Conclusion

The structural and optical properties of PST20 films have been studied at different annealing temperatures. The transmittance and refractive index indicate that the film deposited at 650oC is highly transparent.

References 1. J. H. Hao, Z. Luo and J. Gao, J.Appl.Phys.100,

(2006) 114107. 2. S. W. Liu, J. Weaver, Z. Yuan, W. Donner, C. L.

Chen, A. Bhalla el al Appl. Phys.Lett. 87, (2005) 142905.

3. Y. Somiya, A. S. Bhalla and L. E. Cross, J. Appl.

Phys. 98, (2005) 104103. 4. V. Gupta and A. Mansingh, J. Appl. Phys. 80,

(1996) 1063. 5. A. Mansingh, R. Nayak, V. Gupta and K.

Sreenivas, Ferroelectrics 224, (1999)275. 6. D. P. Seraphin and D. E. Barr, Proc. SPIE 1390,

(1990) 39. 7. D. Y. Kim, S. E. Moon, E. K. Kim, S. J. Lee, J.J.

Choi, and H. E. Kim, Appl. Phys. Lett. 82, (2003) 1455.

8. R. Swanepoel, J. Phys. E 16, (1983) 1214. 9. R. Swanepoel, J. Phys. E 17, (1984) 896.

Non-Linear Optical Property of CdS Nanomaterials Synthesized by Chemical Route

C. S. Tiwary, P. Kumbhakar, A. K. Mitra and R. N. Roy

Department of Physics, National Institute of Technology, Durgapur-713209 E-mail: [email protected]

Abstract

CdS nanostructure semiconductors materials have become a subject of intensive research for their extraordinary properties compared to their bulk counterparts .Blue shift of the optical absorption spectrum, size dependent luminescence, enhanced oscillator strength, nonlinear optical effect , the realization of thermally stable and frequency selective lasers and photo detectors are some examples of the interesting properties exhibited by these nanoparticles. CdS are potential candidates for optoelectronics devices and as a potential optical limiter. In few systematic studies of nonlinear absorption and scattering properties of 4.5 nm sized CdS nanocrystals dispersed in dimethylformamide with varying concentrations are reported. In these open aperture Z-scan measurements with 532 nm nanosecond and 790 nm femto second pulse excitations is used .Their Optical limiting studies of these nanocrystals show that nonlinear scattering (NLS) is comparable to the two-photon absorption TPA).In the current work ,we synthesized CdS with PVP as capping agent and study the optical property.

1. Introduction

Semiconductor nanocrystals (NCs) have received enormous attention because of their size-dependent electronic and optical properties, making them potential candidates for various technological applications such as light-emitting diodes, lasers, and biological labels. In CdS, quantum size effect is observed for crystallite dimensions below 50Å which is approximately the Bohr excitation diameter in CdS. This ability to tune the band gap of semiconductor to suit any specific application by tailoring the size of the particles has many exciting technological implications. However in order to optimize the properties, the nanocrystallite sizes should have a narrow distribution in size and shape. Different chemical routes have been followed in the literature to control the particle shape and more importantly their size distribution. These include the uses of surface clipping legends [7] reverse micellar method [10] polymers [12] glasses and crystalline hosts such as zeolites [14]. Here we will concentrate on optical properties of CdS nanoparticles with PVP as capping agent. In this paper, we present detailed study of the effect of concentration on the nonlinear optical properties of 4.5 nm sized PVP capped CdS nanocrystals and estimate the parameters such as the two-photon absorption coefficient(β), the nonlinear scattering coefficient ( αs) of CdS nanocrystals at different

concentrations (at 532 nm), thermally induced nonlinear refractive index change (Δnth), and second order hyperpolarizability (γ). From the variation of αs with concentration, we estimated the thermo-optic coefficient (dn/dT) of 4.5 nm size CdS nanocrystals. 2. Experimental Details

In our work clusters of CdS in the quantum confinement regime are prepared in a rational technique whereby the cluster size and its distribution are controlled by chemical means. Here CdS nanocrystallites with PVP capping were synthesized as follows: Stock solutions were prepared as: 0.1M cadmium salt solution [Cd 2+] was made by dissolving cadmium acetate in methanol and 0.1 Na2S solution [S2-] was made by dissolving 80gm of sodium sulphide in. 50ml of methanol. 0.2M solution was made from 2.2ml of PVP in 100ml of methanol. From these stock solutions, 50ml of sodium sulphide solution and 50 ml of PVP solution was mixed and stirred, to which 100ml of cadmium acetate solution was added while stirring the solution. This results in a cloudy yellow solution. Then the solution was stirred for some more time, filtered and suction dried. Dry yellow powder was obtained by this process. By changing the relative ratio of sulphide to PVP, clusters of different sizes can be obtained. For smaller size clusters the S: Ph = 1:2, (sample S1) and for larger size crystals it was 2:1, (sample S2) (where S stands for sulphide and Ph stands for


PVP). The crystal structure and size of the particles was determined by X-ray diffraction(name of machine). The diffraction peaks of the nanoparticles were found to be shifted slightly to larger angles than expected for the corresponding bulk crystals suggesting a possible lattice contraction In order to study the nonlinear optical properties of CdS nanocrystallites Z Scan characterization has been made. 3. Results and discussion

Figure 1 shows the room temperature absorption spectra. The absorption edges of the nanocrystalline samples are observed at lower wavelengths signifying a blue shift. Figure 2 shows the size dependence of the band gap, clearly showing a blue shift with decreasing crystalline size. It is clear from the figure that a

Fig. 1. Room Temperature absorption Spectra of CdS samples

Fig. 2. Size dependence of the band gap change in band gap could be achieved in the band gap from 2.405eV to 2.97eVas we narrow

down the crystalline size Figure 3 shows the photoluminescence (PL) studies of CdS nanocrystals at room temperature. It is clear from this figure that PL spectra shows three peaks at 670nm, 915nm, and 1180nm.These spectra were recorded using 488nm Ar+ laser. Figure 6 shows the Cd 3d core level spectrum of CdS nanocrystallite sample, suggesting a single environment for all C sites in the cluster and bonding around Cd sites similar to the bulk.

Fig. 3. Photoluminescence(PL) studies of CdS nano-crystals at room temperature.

It is clearly observed that at lower wavelengths nanocrystalline samples show a blue-shift. The sample (S1) is expected to be of smaller size, as it was prepared with a relatively higher proportion of the capping agent. This is supported by the shift of the absorption band edge to still lower wavelength compared to that of sample (S2). This clearly demonstrates a progressive increase of the band gap with decreasing cluster size in CdS nanocrystallites. In more qualitative terms, the sample of larger clusters (S2) has an absorption edge at 461nm (2.69eV) and another one with smaller clusters (S1) has an edge at 396nm (3.13eV). Thus the shift in the band gaps are 0.3eV and 0.73eV for clusters (S2) and (S1) respectively. These shifts corresponds to 40A0 and 20A diameter particles for these two nanocrystallite samples, when analyzed on the basis of experimental results as well as tight binding calculations reported by Wang and Herron[22]. However the excitonic feature is not pronounced in any of these samples. This may be due to a large size distribution of particles We know that the PL spectrum is due to the radiative recombination of the lowest lying excited states of the nanocrystallites, the semi quantitative estimation of the dependence of peak position on the average grain radius can be done. In general, the carrier

Non-Linear Optical Property of CdS Nanomaterials Synthesized by Chemical Route 131

confinement effect in the semiconductor nanocrystallites stabilizes both excitation and corresponding luminescence due to excitonic recombinations even at room temperature. With the gradual decrease of the sizes of the nanocrystallites, the luminescence has been found to be dominated by the band impurity and surface state transitions over the excitonic luminescence. However, the relative intensities depend on the excitation intensity, the transitions associated with the band to band free electron and hole would dominate in the PL spectra even for samples with smaller (5.7nm) grain size.

Fig. 4. XRD spectra of CdS nanocrystals

The size measurements and surface morphology of the CdS nanocrystals are carried out by powder x-ray diffraction spectrometer (INEL x-ray diffraction spectrometer with Co target) and transmission electron microscope (TEM) studies. From the x-ray diffraction (XRD)spectra of CdS, as shown in Fig. 5, it is very evident that the nanocrystals formed are crystalline in nature. From the spectra the (h,k, l) planes and average grain size are determined. The crystal size (D) is calculated using the Scherer formula, D=(0.9λ) / (B cos θ), where λ is the wavelength of the x radiation, B is the full width at half maximum (FWHM) of the XRD peak, and θ is the half of the Bragg angle at maximum peak height. Though one could get a rough estimate of the average grain size using the XRD data, a better estimate of the crystal size is obtained by TEM studies. The TEM pictures are taken on copper grid with JEOL (100CX) unit and the high-resolution TEM (HRTEM) pictures are taken on carbon coated copper grids with JEOL (JEM-2010) system.

The TEM pictures of the CdS nanocrystals are

as shown in Fig. 5. HRTEM images recorded for the CdS nanocrystals, we observe that the CdS nanocrystals have hexagonal crystal structures. The crystalline planes inside the PVP capped CdS nanocrystals. The observed interplanar distance d is 0.338 nm. The inset shows the diffraction pattern of CdS nanocrystals in which each circle corresponds to a single plane of the crystal. The (h,k, l) values of the planes calculated, taking into account the hexagonal crystal structure of the CdS, are (1,0,0), (1,1,0), (1,1,1), and (1,1,2) and agree with the XRD data. Fig. 6. Open aperture Z-scan curves of the CdS nanocrystals collected with two detectors; detector 1 gives the transmitted light and scattering, and detector 2 gives the transmitted light without scattering with 6 ns laser pulses. Open aperture Z-scan curve of the CdS nanoparticles with 100 fs laser pulses, and the solid line is the curve obtained by theoretical fitting.[2]

The open aperture Z-scan data recorded by

placing the detectors at two positions as explained in the experimental section is shown in Fig.6. The CdS nanocrystals show strong reverse saturable absorption (RSA) behavior at all intensities. Detector 1 gives the “whole transmitted light” due to the nonlinear two-photon absorption alone. Detector 2 kept at the far field gives the transmitted beam minus the scattered beam due to both two-photon absorption and the nonlinear scattering losses with 6 ns pulses. At peak intensities >100 MW cm−2 we see nonlinear scattering along with


strong TPA. This can be seen as an enhanced depletion in the transmitted beam collected with detector 2, which is reflected in the Z-scan curve shown in Fig.7. The transmittance goes to as low as 0.1 when the loss due to the nonlinear scattering is taken into account. However, Z-scan data taken with 790 nm, 100 fs pulses collected with detector 2 show negligible scattering. As the CdS nanoparticles show strong TPA, the scattering observed with the nanosecond pulses is attributed to the local heating of CdS nanoparticles. The nonlinear scattering behavior of the CdS nanocrystals as a function of the input intensity (Z position) in which the curves shown are of those obtained for three different forward scattering angles with beam propagation direction using detector 3. It is observed that the nonlinear scattering is conical in nature with the intensity decreasing as we go away form the center.

To account for the losses as seen in detector 2, we have introduced the nonlinear scattering losses αs as derived by Joudrier. From the absorption spectra of CdS nanocrystals we clearly see that the linear absorption coefficient is very small at the excitation wavelength of 532 nm. Thus in our analysis we have neglected the contribution of the linear absorption cross section. Since CdS is a direct band gap semiconductor, a two-level model corresponding to valence and conduction bands is used to explain the observed results. By using this model, we first estimated the two-photon absorption coefficient β using αs as zero by fitting the Z-scan curve, which we got from detector 1. We then used this value to estimate the nonlinear scattering coefficient αs from the curve obtained from detector 2. The treatment of this theoretical modeling is as reported. The differential Eqs. (1)and (2) are taken as the two rate equations for the valence band and the conduction band populations and the intensity transmitted through the sample is given by Eq.(3)

Here N0 and N1 are populations of valence band and conduction band, respectively, αs is the effective scattering coefficient, β is the two-photon absorption coefficient, gs is a parameter independent of intensities but depends on the size, shape, and concentration of crystals and wavelength of light, Δn˜ is the difference in the effective refractive indices of both linear and nonlinear components, Δnl is the difference in the linear refractive indices of CdS and DMF, Δnnl is the difference in nonlinear refractive indices of CdS and DMF, which is a function of intensity, and τl is the lifetime of the excited state and is taken as 114 ps. The differential equations are solved numerically using the Runge-Kutta fourth order method. The equations are first decoupled and then integrated over time, length, and along the radial direction. Assuming the input beam to be Gaussian, the limits of integration for r, t, and z are varied from 0 to ∞ , from -∞ to ∞ , and from 0 to L(length of the sample), respectively. Typical number of slices used for r, t, and z are 60, 30, and 5, respectively. gsΔnl, gsΔnnl, β, and αs are then estimated through least square fitting of the experimental data. From the calculated value for gsΔnl, we have determined the scattering constant gs. The two-photon absorption cross section σ TPA is then calculated

where N is the number of molecules per unit volume, D is the molar concentration of CdS, NA is the Avogadro constant, h is Planck’s constant, and γ is the frequency of laser beam used. 4. Conclusions

The CdS nanocrystalline were synthesized using thiophenol as capping agent. By changing the relative ratio of sulphide to PVP, clusters of different sizes were obtained. Optical properties of CdS nanocrystallites exhibit extraordinary behavior over that of bulk.

Nonlinear optical absorption and optical limiting behavior of thioglycerol capped CdS were studied systematically. Enhanced RSA was observed in the Z-scan data showing an added component of losses due to nonlinear scattering along with the TPA. The losses due to nonlinear scattering were significant at high intensities. The obtained data were theoretically fitted with the help of a two-level model from which the two photon

Non-Linear Optical Property of CdS Nanomaterials Synthesized by Chemical Route 133

absorption coefficient, two-photon absorption cross section σTPA, and nonlinear scattering coefficient αs were calculated. The variation of αs was discussed in detail by taking into account the refractive index variation due to n2eff, which resulted in the estimation of the thermo-optic coefficient (dn/dT) for the 4.5 nm sized CdS nanocrystals. References

[1] S.M. Sze, Physics of Semiconductor Devices, Wiley, New York, 1981.

[2] E.H. Rhoderick, R.H. Williams, Metal–Semiconductor Contacts, Clarendon, Oxford, 1988.

[3] H.C. Card, E.H. Rhoderick, J. Phys. D 4 (1971) 1589.

[4] A. Turut, N. Yalcin, M. Saglam, Solid State Electron. 35 (1992) 835.

[5] A.M. Cowley, S.M. Sze, J. Appl. Phys. 36 (1965) 3212.

[6] H.H. Tseng, C.Y. Wu, Solid State Electron. 30 (1987) 383.

[7] M.K. Hudait, S.B. Kruppanidhi, Solid State Electron. 44 (2000) 1089.

[8] S. Ashok, J.M. Borrego, R.J. Guttmann, Solid State Electron. 22 (1979) 621.

[9] S. Altindal, N. Tugluoglu, A. Tataroglu, Solid State Electron. 47 (2003) 1847.

[10] A. Singh, K.C. Reinhardt, W.A. Anderson, J. Appl. Phys. 68 (1990) 3475.

[11] P. Chattopadhyay, A.N. Daw, Solid State Electron. 29 (1986) 555.

[12] K. Hatori, Y. Torn, Solid State Electron. 34 (1991) 531.

[13] I. Dokme, S. Altindal, Semicond. Sci. Technol. 21 (2006) 1053.

[14] B. Akkal, Z. Benamara, L. Bideux, B. Gruzza, Microelectron. J. 30 (1999) 673.

[15] P. Cova, A. Singh, Solid State Electron. 33 (1990) 11.

[16] A.N. Saxena, Surf. Sci. 13 (1969) 151. [17] S. Chand, J. Kumar, Appl. Phys. A 63 (1996)

171. [18] J.H. Werner, Appl. Phys. A 47 (1988) 291. [19] S. Hardikar, M.K. Hudait, P. Modak, N.

Padha, Appl. Phys. A 68 (1999) 49. [20] Shikha Tiwari ,Sanjay Tiwari ; Electrical and

optical properties of CdS nanocrystalline Semiconductors, Cryst. Res. Technol. 41, No. 1, 78 – 82 (2006)

[21] N. Venkatram, R. Sai Santosh Kumar, and D. Narayana Rao, Nonlinear absorption and scattering properties of cadmium sulphide nanocrystals with its application as a potential optical limiter; JOURNAL OF APPLIED PHYSICS 100, 074309 _2006.

[22] Norman Henon,Y ing Wang, and Hellmut Eckert, Synthesis and Characterization of Surface-Capped, Size-Quantized CdS Clusters. Chemical Control of Cluster Size; J. Am. Chem. SOC. 1990, I1 2, 1322-1 326.

[23] Y. Wang and N. Herron, J. Phys. Chem. 95, 525(1991).

Optical Band Gap and Refractive Index of Vacuum Evaporated a-Ge10Se90-xTex (x = 0, 20) Thin Films

Pankaj Sharma*, Ishu Sharma and S. C. Katyal

Department of Physics, Jaypee University of Information Technology, Waknaghat, Solan, H.P. 173215 (India)


Abstract

Optical measurements viz. absorption coefficient, optical band gap, refractive index and extinction coefficient of the thermally deposited Ge10Se90-xTex (x = 0, 20) thin films have been reported in the present work. The straight forward technique proposed by Swanepoel has been used to measure the refractive index in the spectral range 400-1500 nm. The optical band gap has been measured using Tauc’s extrapolation method. The optical band gap has been found to decrease with the addition of Te content. This decrease has been explained on the basis of chemical disordering and the decrease in the energy of the system with the addition of Te. Optical band gap has also been calculated using Shimakawa’s relation.

1. Introduction

The chalcogenide glasses are one of the most widely known families of amorphous materials and have been studied extensively over the past few decades because of their interesting fundamental properties and wide commercial applications in integrated optics, optical imaging, optical data storage and infrared optics [1-4]. People are taking more interest in electrical, optical and the structural properties of chalcogenide glasses with the addition of impurities. Se based systems such as Ge-Se system has attracted much attention in recent years and this family of chalcogenide glasses can provide an ideal system for investigation. The chemical composition and energy band structure changes as a result of introducing Ge atoms into Se matrix. The variation of the Ge-Se structure is reflected in different properties such as the glass forming regions, glass transition temperature, photoluminescence, IR and Raman spectra and the optical properties [5-7]. In our present work, we have studied the effect of addition of 20 at. % of Te on the optical properties of Ge10Se90 glassy alloy. The optical band gap (Eg

opt), absorption coefficient (α), refractive index (n) and extinction coefficient (k) has been calculated by analyzing transmission spectrum of thin films. The experimentally calculated optical band gap is also compared with theoretical results obtained by Shimakawa relation.

2. Experimental Procedure

Glassy alloys of Ge10Se90 and Ge10Se70Te20 are prepared by melt quenching technique. Materials (5N purity) are weighed according to their atomic percentages and sealed in quartz ampoules in a vacuum ~ 2x10-4 Pa. The sealed ampoules are kept inside a furnace where the temperature is increased up to 1000 0C at a heating rate of 2-3 0C/min. The ampoules are frequently rocked for 24 hours at the highest temperature to make the melt homogeneous to avoid phase separation. The quenching was done in ice cold water. Thin films were deposited on glass substrates which were first cleaned with soap solution, then ultrasonically cleaned by trichloroethylene, acetone followed by methyl alcohol. Finally the substrate is washed by DI water and dried in oven at 110 0C. Thin films of the alloys are prepared by vacuum evaporation technique at room temperature and base pressure of ~ 10-4 Pa using a molybdenum boat. The normal incidence transmission spectra of thin films of the samples have been measured by a double beam UV-Vis-NIR spectrophotometer [Hitachi-330], in the transmission range 400-1500 nm. The spectrometer was set with a suitable slit width of 1 nm in the measured spectral range. All optical measurements have been performed at room temperature.

Optical Band Gap and Refractive Index of Vacuum Evaporated 135


X-ray study of the bulk samples as well as their thin films confirms the amorphous nature as no prominent peak has been observed in both cases. Figure 1 shows the optical transmission spectrum of Ge10Se90 and Ge10Se70Te20 thin films. The plot shows fringes at various wavelengths. The fringes are shifted towards the higher wavelength region. This may be due to the influence of the scattering of light by defects induced by the incorporation of Te in Ge-Se system.

Fig. 1. Transmission spectra of Ge10Se90-xTex (x = 0, 20) thin films.

The variation of absorption coefficient (α)

with energy for the deposited films is shown in Figure 2. The absorption coefficient is measured in high and intermediate absorption region not in the weak absorption region. The values of absorption coefficient (α) are calculated by using the following relation

( ) ( )Tt 1ln1=α (1) where t is the thickness of the film and T is the transmittance [11,12]. This is observed that the absorption coefficient decreases with increase of wavelength and also with the addition of Te content. This may be due to the increase of the defect states created by the incorporation of Te in Ge-Se base alloy and the scattering of light by these defect states. Figure 3 shows the variation of (αhυ)1/2 with hν which is used to calculate

optgE . The optical gap is determined by the

intercept of the extrapolations to zero with the

photon energy axis 0)( 2/1 →ναh (Tauc extrapolation [8]). The optical band gap is in good agreement with the band gap calculated theoretically by Shimakawa relation (Table 1)

)()1()())(( BEYAYEYABE ggg −+=

where )(AEg and )(BEg are the optical gap

components of A and B respectively, Y is the volume fraction of element A. Fig.2. Plot of α versus νh for Ge10Se90-xTex (x = 0, 20) thin films.

Fig. 3. Plot of 2/1)( ναh versus νh for

Ge10Se90-xTex (x = 0, 20) thin films.

400 600 800 1000 1200 14000

20

40

60

80

100

T %

λ (nm)

Ge10Se90 Ge10Se70Te20

1.4 1.6 1.8 2.0 2.2103

104

105

α (c

m-1)

hν (eV)


1.2 1.4 1.6 1.8 2.0 2.20

50

100

150

200

250

300

350


(αhν

)0.5 (

eV c

m-1)0.

5

hν (eV)


This is found that in both experimentally as well as theoretically, the optical band gap decreases. This may be due to the tendency of Te atoms to form chemical disordering and to create localized states in the band gap [9]. According to Mott and Davis [10] the width of localized states near the mobility edge depends on the degree of disorder and defects present in amorphous structure. In particular the unsaturated bonds are responsible for the formation of defects in amorphous solids. Such defects produce localized states in band gap. The presence of high concentration of localized states in thin films is responsible for low optical band gap. Table 1. Values of theoretical band gap (Egth), optical band gap (Egopt ), refractive index (n), absorption coefficient (α) for a-Ge10Se90-xTex (x = 0, 20) thin films.

x Egopt

(eV)

Egth

(eV)

n at 800 nm

α (cm-1) at 1200 nm

0

1.865

1.878

2.69

0.012 x 104

20

1.478

1.463

3.12

0.12 x 104

The addition of Te concentration increases

the concentration of these localized states leading to lowering the optical band gap. The decrease of optical band gap may also be explained by Kastner’s model [13], according to it the dominant contribution for states near the valence band edge in materials having chalcogen atoms as major constituents, comes from chalcogen atoms, especially from their lone-pair p-orbital. The lone-pair electrons in these atoms adjacent to electropositive atoms will have higher energies than those close to electronegative atoms. Therefore, the addition of electropositive elements to the alloy may raise the energy of some lone-pair states sufficiently to broaden further the band inside the forbidden gap.

Figure 4 shows the variation of refractive index (n) with wavelength. The refractive index is obtained by using the transmission spectrum of Ge10Se90 and Ge10Se70Te20 thin films. The envelops of transmission spectrum for the values of maximum and minimum transmission has been drawn from the Figure 1 and are used to calculate the refractive index using the method suggested by Swanepoel [14]. The refractive

index (n) has been obtained using the following expressions

2/12/122 ])([ sMMn −+= (2)

where 2

)1(2 2 +−=

sT

sMm

for transparent

region (3)

and 2

)1(22 +

+−

=s

TTTT

sMmM

mM for weak and

medium absorption region. (4)

Fig. 4. Variation of refractive index with wavelength for Ge10Se90-xTex (x =0, 20) thin films. TM and Tm are the values of maximum and minimum transmission at a particular wavelength; s is the refractive index of the substrate. The value of refractive index increases and this may be due to the change in stoichiometry [15], crystallite size [16] and internal strain [17] of the glassy alloy with incorporation of Te content in the base alloy. The values of n at 800 nm are given in table 1.

Figure 5 indicates the variation of extinction coefficient (k) with wavelength. It is found that the value of k decreases with the increase of wavelength. The extinction coefficient is calculated from [18].

παλ 4=k (5)

where α is absorption coefficient and λ is corresponding wavelength. The value of k goes to decrease with increase of wavelength. Hence the fraction of light lost due to scattering and absorption per unit distance in a participating medium decreases with increase of wavelength. This may be due to the decrease in the value of absorption coefficient (α) with increase of wavelength.

6 0 0 9 0 0 1 2 0 0 1 5 0 02 .0

2 .5

3 .0

3 .5

4 .0

4 .5

Ref

ract

ive

inde

x

W a v e le n g th (n m )

G e 1 0S e 9 0 G e 1 0S e 7 0T e 2 0

Optical Band Gap and Refractive Index of Vacuum Evaporated 137

Fig. 5. Variation of extinction coefficient with wavelength for Ge10Se90-xTex (x = 0, 20) thin films. 4. Conclusion

Optical Thin films were characterized Different parameters related to optical properties are calculated for Ge10Se90 and Ge10Se70Te20 thin films. Optical band gap found to decrease with addition of 20 at. % Te content. Refractive index found to decrease with wavelength and with addition of Te it increases. Extinction coefficient also decreases with the increase of wavelength i.e. the fraction of light lost decreases with the increase of wavelength. The addition of Te content increases the defect states leading to change in stoichiometry, crystallite size and internal strain which may be responsible for the increase in refractive index. 5. References [1] K. Schwartz, The Physics of Optical Recording,

Springer-Verlag, Berlin (1993).

[2] A. Bradley, Optical Storage for Computers Technology and Applications, Ellis Horwood Limited, New York, (1989).

[3] J. Bardangna, S. A, Keneman, Holographic recording Media, edited by H. M. Smith, Springer-Verlag, Berlin, (1977).

[4] J. S. Sanghara and I. D. Agarwal, J.Non-Cryst. Solids,

256 & 257 (1999) 6. [5] Y. Wada, Y. Wang, O. Matsuda, K. Inoue and K.

Murase, J. Non-Cryst. Solids, 198-200 (1996) 732.

[6] A. Feltz, H. Aust and A. Blayer, J. Non-Cryst. Solids, 55 (1983) 179.

[7] S. Asokan, G. Parthasarathy and E.S.R. Gopal, Phil. Mag. B, 57(1) (1988) 49.

[8] J. Tauc, A. Menth, J. Non-Cryst. Solids, 8, (1972) 569.

[9] P.Boolchand, Physical Properties of Amorphous Materials, edited by D. Adler, B. B. Schwartz and M. C. Steele, Plenum Press, New York. (1985).

[10] N. F. Mott, E. A. Davis, Electronic Processes in Non-Crystalline Materials, Clarendon Press, Oxford. (1979).

[11] Priyamvada Bhardwaj, P. K. Shishodia, R. M. Mehra, J. Optoelectron. and Adv. Mater., 3, (2001), 319.

[12] F. Skuban, S. R. Lukic, D. M. Petrovic, I. Savic, Yu. S. Tver’yanovich, J. Optoelectron. and Adv. Mater., 7, (2005) 1793.

[13] M. Kastner, D. Adler, H. Fritzsche, Phys. Rev. Lett. 37 (1976) 1504.

[14] R. Swanepoel, J. Phys. E, 16 (1983) 1214. [15] K. Yamaguchi, N. Nakayama, H. Matsumoto, S.

I. Kegami, Jpn. J. Appl. Phys., 16 (1977) 1203. [16] F. Tepehan, N. Ozer, Solar Energy Mater. Solar

Cells, 30, (1993) 353. [17] A. Ashour, N. El-Kdry, S. A. Mahmoud, Thin

Solid Films, 269, (1995) 117. [18] E. Marquez, J. Ramirez, P. Villares, R. Jimenez,

P. J. S. Ewen and A. E. Owen, J. Phys. D: Appl. Phys., 25 (1992) 535.

5 0 0 6 0 0 7 0 0 8 0 0 9 0 0 1 0 0 00 .0 0

0 .0 5

0 .1 0

0 .1 5

0 .2 0

0 .2 5

0 .3 0

Extin

ctio

n Co

effic

ient

(k)

W a v e le n g th ( n m )

a - G e 1 0 S e 9 0 a - G e 1 0 S e 7 0 T e 2 0

Light Induced Changes in Se-Te Thin Film

Vineet Sharmaa and Anup Thakurb aJaypee University of Information Technology, Waknaghat, Solan – 173 215, India.

bUniversity College of Engineering, Punjabi University, Patiala-147 002, India. E-mail: [email protected]; [email protected]

Abstract

The chalcogenide glasses present interesting advantages as compared to the crystalline semiconductors.

The Se-Te chalcogenide alloy is one of the most widely studied of these amorphous glasses. The Se-Te alloy has more diverse scientific and practical applications in switching, memory devices, phase change recording devices etc as compared to a-Se, which crystallizes easily. On illuminating the Se-Te thin film with light, it shows changes in its various properties as structural, optical and electrical. The paper reports such changes observed on illumination the Se-Te thin film deposited by vacuum evaporation technique. As a result of illumination, the increased absorption coefficient causes a decrease in the optical band. The refractive index also changes along with the change in the dielectric constant values on illumination of the thin film. The change in properties of the thin film may be due to increase in the dangling bonds which increases the tailing in the band gap of the alloy

1. Introduction

The glasses made up of chalcogen elements of the VIth group of periodic table offer inetersting alternatives for the infrared detectors and fabrication of optical elements. The chalcogenide bonds allow to transmit far out into the infrared region due to their low characteristic vibrational frequencies [1]. The chalcogenide semiconducting materials have been observed to undergo various transformations on subjecting them to external electromagnetic radiations [2] in addition to their transformation from amorphous to crystalline phases. Generally, it is believed that the photoinstability is a privilege of the amorphous state and is due mainly to a rearrangement in the local order of the glassy network [3]. These glasses show a variety of photo-stimulated phenomena when exposed to light or other radiations [4, 5]. On irradiating these glasses with high energy particles or light, bond breaking and bond rearrangement can takes place, which results in the change of local structure of the glassy materials. These include subtle effects such as shifts in the absorption edge (photo-bleaching and photo-darkening), and more substantial atomic and molecular reconfiguration such as photo-induced refractive index changes and photo-doping effects [6]. Generally, these phenomena are associated with the changes in the optical constants [7] and absorption edge shift [8], allowing the use of these materials in the fabrication of a large number of optical devices. This clearly underlines the importance of these glassy materials by accurate determination of their optical parameters. Hence these glasses offer a nice option for these technical applications by an accurate determination of various optical parameters.

Therefore, the authors have decided to investigate the change in optical and structural properties before and after the UV irradiation in a-Se85Te15 thin film. An attempt has been made to analyze the change in optical parameters and structural characteristics of a-Se85Te15 thin film on UV irradiation. Swanepoel’s method [9, 10] of using the transmission spectrum has been used for determining the optical constants. 2. Experimental Procedure

Bulk alloy of a-Se85Te15 is prepared by melt- quenching technique as described elsewhere [11]. The constituent elements taken are 5 N pure. Thin film of the said alloy is prepared by thermal evaporation technique in vacuum of 2×10-5 mbar on cleaned glass substrate. The absence of prominent peak confirmed in X-ray diffraction analysis confirmed the amorphous nature of the thin film. The thin film is irradiated with UV light having intensity ~ 8000 Lux in vacuum. The transmission spectra of the a-Se85Te15 thin film before and after the UV electromagnetic irradiation have been measured by a double beam UV/VIS/NIR computer controlled spectrophotometer [Hitachi-330], in the transmission range 400-2000 nm. The spectrophotometer was set with a suitable slit width of 1 nm, in the spectral range. All optical measurements have been performed at room temperature. 3. Theory

To analyse the change in optical properties of the UV light irradiated sample with respect to the unirradiated thin film sample, Swanepoel’s method has been used. The Swanepoel’s method [9, 10]

Light Induced Changes in Se-Te Thin Film 139

assumes a non-uniform thickness of the thin film sample deposited on a transparent substrate having a refractive index ‘s’. The system is surrounded by air, whose refractive index is n0 = 1. The film has a complex refractive index n* = n – i k, where n is the refractive index and k the extinction coefficient, which is related to the absorption coefficient (α) through the relation, k = α λ / 4 π. The optical constants are obtained by using only the transmission spectrum. According to this method, which is based on the approach of Manifacier et al. [12], the refractive index in the region where α ≈ 0 is calculated by the following equation:

22 SNNn −+= (1) where

212

2

minmax

minmax ++

−=

sTTTTsN (2)

Tmax and Tmin are the envelope values at the wavelengths in which the upper and lower envelops and the experimental transmission spectrum are tangent respectively. The accuracy to which λ can be measured is ±1 nm. The maximum absolute accuracy of Tmax and Tmin is ±0.001. The values of n are calculated using Eq. (1) at wavelengths corresponding to the tangent points.

If n1 and n2 are the refractive indices at two adjacent tangent points at λ1 and λ2, then according to the basic equation for interference fringes

λmnt =2 (3) where m is an order number. The thickness is given by

( )1221

21

4 nnt

λλλλ−

= (4)

It should be noted that owing to optical absorption, this particular equation is not valid at the interference maxima and minima, but is valid at the tangent points referred to [9]. Using equation (3), new more precise values of the refractive index and the film thickness were determined by a procedure which was explained in detail in [9,10].

The absorption coefficient (α) [10] can be calculated from the relation

).exp( tx α−= (5) where x is absorbance, given by

)()1()()1(

23

42322

snnsnnEE

x MM

−−

−−−−= (6)

and

))(1(8 222

max

2snn

TsnEM −−+= (7)

In case of UV irradiated thin film there is no maxima and minima in the transmission spectrum.

Therefore, from the transmission data, nearly at the fundamental absorption edge, the values of absorption coefficient (α) are calculated in the region of strong absorption using the relation

⎟⎠⎞

⎜⎝⎛=

Td1ln1α (8)

The absorption coefficient of amorphous semiconductors in the strong-absorption region (α ≥ 104 cm-1), assuming parabolic valence and conduction band edges, is given by [13]

( )( )ωω

αη

η2opt

gEB −= (9)

where ħω, Egopt and B, represent photon energy,

optical gap and an energy independent constant, respectively. Finally, the optical gap is calculated from the intersection of the plot (αħω)1/2 vs. ħω with the abscissa axis. 4. Result and Discussion

The thickness of thin film has been calculated by using equation (4). The thickness has been found to be ~ 900 nm. Fig. 1. Transmission spectrum of virgin a-Se85Te15 thin film and UV irradiated in inset. The transmission spectra of both the unirradiated and the UV irradiated thin film sample have been analysed to look for the changes in the band structure and optical parameters. The optical properties show a transition on subjecting the a-Se85Te15 thin film to UV irradiation Figure 1 is a plot between percentage transmission and wavelength for unirradiated and UV irradiated thin

600 800 1000 1200 1400 1600 1800 20000

20

40

60

80

100

600 800 1000 1200 1400 1600 1800 2000 22000

5

10

15

20

25

30

35

40

45

Tran

smis

sion

(%)

W avelength (nm )

()

Wavelength (nm)

a-Se85Te15


film. The transmission in the irradiated thin film is very low as compared to pure film. This gives an indication of the crystalline state developed as a result of UV light exposure. The values of absorption coefficient (α) value shows an increase with increase in the frequency of the incident radiation in both cases of pure and UV irradiated thin film. The absorption coefficient increases after the UV irradiation as compared to the unirradiated thin film. The optical energy gap (Eg

opt) is calculated by extrapolating the curve between (αhν)1/2 vs. hν at the abscissa axis as shown in figure 2. The optical energy gap shows a decrease in its value from 1.40 to 1.30 eV. This decrease in the band gap may be due to an increase in the concentration of dangling bonds or surface voids on irradiating the a-Se85Te15 thin film sample with UV light. In other words, this implies an increase in the density of the surface voids of the thin film. The crystallised film developed due to the UV irradiation changes to micro and nano crystallites with progress in UV radiation exposure on the thin film. Due to the transition from amorphous to crystalline state, the tailing of the conduction and valence bands is increased. This causes a decrease in the optical band gap. There may be an increase in disorder at the expense of delocalized states near the band edges

Fig. 2. Plot of (αhν)1/2 vs. hν for pure a-Se85Te15 thin Film and UV irradiated in inset. .

5. Conclusions

The optical properties of a-Se85Te15 thin film on UV irradiation show a characteristic change. The optical energy gap Eg

opt decreases and the absorption coefficient (α) increases on UV irradiation. The film undergoes crystallization. This indicates an ordering of the bonds in thin film sample on UV irradiation. This may be due to the enhanced valence band tailing on UV irradiation. References 1. P. J. S. Ewen, C. W. Shinger, A. Zakery, A.

Zekak and A. E. Owen, SPIE Infrared Optoelectron. Mater Dev. 1512 (1991) 101.

2. DeNeufville J. P. In: Seraphin B. O., editor. Optical properties of solids-new developments. Amsterdam: North Holland, 1975. p. 437.

3. K. Tanaka, In: F. Yonezawa, editor. Fundamental physics of amorphous semiconductors. Berlin: Springer, 1981, p.104.

4. A. Arsh, M. Klebanov, V. Lyubin, L. Shapiro, A. Feigel, M. Veinger and B. Sfez, Optical Mat. 26 (2004) 301.

5. V. Lyubin, M. Klebanov, A. Feigel and B. Sfez, J. Non-Cryst. Solids 459 (2004) 183.

6. M. Mitkova, M. N. Kozicki, H. C. Kim and T. L. Alford, J. Non-Cryst. Solids 338-340 (2004)552.

7. E. Marquez, T. Wagner, J.M. Gonzalez-Leal, A.M. Bernal-Oliva, R. Prieto-Alcon, R. Jimenez-Garay and P.J.S. Ewen, J. Non-Cryst. Solids 274 (2000) 62.

8. V.M. Lyubin, M. Klebanov, B. Safe and B. Ashkinadze, Mat. Lett. 58 (2004) 1706.

9. R. Swanepoel, J. Phys. E: Sci. Instrum. 17 (1984) 896.

10. R. Swanepoel, J. Phys. E 16 (1983) 1214. 11. A. Thakur. P.S. Chandel, V. Sharma, N.

Goyal, G.S.S. Saini and S.K. Tripathi, J. Optoelect. & Adv. Mat. 5(2003)1203.

12. J.C. Manifacier, J. Gasiot, J.P. Fillard, J. Phys. E: Sci. Instrum. 9 (1976) 1002.

13. J. Tauc, J. Non-Cryst. Solids 8 (1972) 569. 14. Al. L. Efros and A. L. Efros, Sov. Phys.

Semicon. 16 (1982) 772.

0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

0

50

100

150

200

250

300

1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.01200

1400

1600

1800

2000

2200

2400

2600

2800

(α

hν)1/

2 (eVc

m-1)1/

2

hν(eV)

hν(eV)

(αhν

)1/2 (e

Vcm

-1)1/

2

Photo Bleaching in a-Ga50Se50 Thin Films.

Shikha Gupta, F.I. Mustafa, Akshay Kumar, N. Goyal, G.S.S. Saini and S.K. Tripathi* Department of Physics, Centre of Advanced Study in Physics, Panjab University

Chandigarh -160 014, INDIA. E-mail: [email protected], [email protected]

Abstract

In this paper, we report on photo-induced changes in optical properties of a-Ga50Se50 thin films prepared by thermal evaporation technique under vacuum. Thin film samples, on a microscope glass slide as substrate, were exposed to laser light of wavelength λ = 432 nm. Optical properties (absorption coefficient, α; optical gap, Eg

opt) of as- deposited thin films and their photo induced changes were studied at three different times of exposure (tE). The mechanism of optical absorption follows the rule of indirect allowed transition model proposed by Tauc and the optical band gap (Eg

opt) is calculated by Tauc’s extrapolation. The optical gap is calculated from the intersection of plot (αhν)0.5 vs. hν. It is seen that after illumination photo bleaching phenomenon is observed i.e. there is a shift of optical absorption edge to larger photon energy. The absorption coefficient, α, of GaSe thin film decreases on exposing the film to radiation. These results have been explained on the basis of some structural changes occurring after the laser irradiations. 1. Introduction

Interest in optical properties of chalcogenide glasses has recently increased as they are very promising materials for use in optoelectronic applications [1]. These glasses are very interesting materials for infrared optics. They have a large range of transparency (~ 0.6 to 30 μ m) and good mechanical and chemical properties, such as hardness, adhesion, low internal stress and water resistance. The low characteristic vibrational frequencies of chalcogenide bonds allow them to transmit far out into the infrared [2]. These glasses show a variety of photo-stimulated phenomena when exposed to light or other radiations [3–7]. When these glasses are irradiated with high energy particles or light bond breaking and bond rearrangement can take place, which result in the change in local structure of the glassy materials. These include subtle effects such as shifts in the absorption edge (photo-bleaching and photo-darkening), and more substantial atomic and molecular reconfiguration such as photo-induced refractive index changes and photo doping effects [8, 9]. In general, these phenomena are associated with changes in the optical constants [10, 11] and absorption edge shift [12]. These light-induced changes are favored in chalcogenide glasses due to their structural flexibility (low coordination of chalcogens) and also due to their high-lying lone-pair p states in their valence bands. In the last decades, many potential applications based on light-induced effects have been explored, for examples, in the

fields of sub micron photo resists, optical memories, diffraction elements, optical light-guide and optoelectronic elements and devices. Semiconductor optoelectronic devices can be operating under conditions of the action of different kinds of high energy radiations. Duration, intensity and energy of such radiations influences differently on parameters of these devices. Therefore, an establishment of the regularities in the changes of the device characteristics affected by radiation can be used for adjustment of their operating conditions

GaSe is a member of III-VI layered semiconductor family that has been a subject of several investigations in recent years. The layer structure is characterized by the covalent bond which is restricted in two dimensions in the plane of the layer and with Van der Waals forces in the third dimensions. These films have typical characteristics of semiconductor layers such as, (i) the low density of dangling bonds on the surface because of the almost complete chemical bonds within the layer [13], (ii) intercalation [14] and (iii) the mechanical weakness due to the weak Van der Waals force between the layers. The study of GaSe attracts broad attention in view of its number of positive properties for nonlinear applications in technology, the foremost among these being its extreme transparency. It has highly anisotropic transport, mechanical and optical properties [15], and high nonlinear optical coefficients in the infrared range, making it a candidate for second harmonic generation (SHG) materials [16]. GaSe is a candidate material to


form the heterojunction with a very low density of interface states. It has been used as photoelectric analyzers of polarized light [17].Therefore in the present paper the authors have decided to study the effect of irradiation on the optical properties of GaSe semiconducting material .Optical measurements have been taken at room temperature and the different parameters like absorption coefficient (α) and band gap (Eg) have been calculated. Section 2 describes the experimental details. The results have been presented and discussed in section 3. The last section deals with the conclusions of this work. 2. Experimental

Glassy alloys of Ga50Se50 are prepared by melt quenching technique. Materials (5 N pure) are weighed according to their atomic percentages and sealed in quartz ampoules in a vacuum ~ 2 × 10-5 mbar. The sealed ampoules are kept inside a furnace where the temperature is increased up to 1000 °C at a heating rate of 2–3 °C /min. The ampoules are frequently rocked for 24 h at the highest temperature to make the melt homogeneous. The quenching is done in liquid N2. Thin films of the alloys are prepared by vacuum evaporation technique on well-degassed Corning 7059 glass substrates at room temperature and base pressure of ~ 2 × 10-5 mbar using a molybdenum boat. The deposition rate is very slow so that the composition of the thin film is very close to the composition of the starting bulk material [18]. Amorphous nature of the samples has been checked by X-ray diffraction technique. The normal incidence transmission spectra of the substrate with and without a-Ga50Se50 thin films have been measured by a double beam UV/VIS/NIR computer controlled spectrophotometer [Hitachi-330], in the transmission range 400–1100 nm. All the optical constants measurements reported in this paper are performed at room temperature (300 K). 3. Results and discussion

Analysis of optical absorption spectra is one of the most productive tools for understanding and developing the band structure and energy gap of amorphous non-metallic materials. Optical properties are studied by recording the transmission spectra of the films. According to Tauc [19], it is possible to separate three distinct regions in the absorption edge spectrum of amorphous semiconductors. The first is the weak absorption tail, which originates from defects and

impurities, the second is the exponential edge region, which is strongly related to the structural randomness of the system and the third is the high absorption region that determines the optical energy gap. In the exponent edge, where the absorption coefficient, α, lies in the absorption region of 1 < α <104 cm-1, the absorption coefficient is governed by the relation:

The fundamental absorption, which corresponds to the transition from valence band to conduction band, can be used to determine the band gap of the material. The relation between α and the incident photon energy (hν) can be written as [20] where hν, Eg

opt and A, represent photon energy,

optical gap & an energy independent constant, respectively.

The spectral dependence of the transmittance, T, of amorphous GaSe thin films was measured in the wavelength range of 400-1100 nm. The data is illustrated in Fig. 1. The spectrum shows interference pattern with a sharp fall of transmittance at the band edge.

400 500 600 700 800 900 1000 11000.0

0.2

0.4

0.6

0.8

1.0

T%

Wavelength (nm)

0sec 500sec 1500sec 3000sec

GaSe

Fig. 1. Plot of T% vs. wavelength at different tE ( tE =0s, 500s, 1500s and 3000s). Figure. 1. shows the plot of T% vs. wavelength at different tE ( tE =0s, 500s, 1500s and 3000s). It is clear from figure 1 that transmission increases as exposure time increases The mechanism of optical absorption follows the rule of indirect allowed transition model proposed by Tauc and the optical band gap (Eg

opt) is calculated by Tauc’s

⎟⎠⎞

⎜⎝⎛=

Td1ln1α

hvhvA 2opt

g )E( −=α

Photo Bleaching in a-Ga50Se50 Thin Films. 143

extrapolation [21]. The optical gap is calculated from the intersection of plot (αhν)0.5 vs. hν with the abscissa axis. The correct values of the optical gap calculated from the figure are ( 1.34 ± 0.01) eV, ( 1.36 ± 0.01) eV, ( 1.43 ± 0.01) eV, ( 1.47 ± 0.01) eV for tE =0s, 500s, 1500s and 3000s respectively

1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.60

50

100

150

200

250

300

350

400

450

500

(α h

ν)1/

2 (cm

-1 e

V) 1

/2

hν (eV)

0 sec 500 sec 1500 sec 3000sec

GaSe

Fig. 2. Variation of (αhν)0.5 vs. hν at different exposure times tE ( tE =0s, 500s, 1500s and 3000s).

Fig.3. shows the plot of ΔE vs. tE. ΔE is the difference in the values of optical gap at different exposure times tE (tE = 500s, 1500s and 3000s) w.r.t the optical gap value of virgin film .

0 500 1000 1500 2000 2500 30001.32

1.34

1.36

1.38

1.40

1.42

1.44

1.46

1.48

ΔE

(eV

)

tE (sec)

GaSe

Fig. 3. Plot of ΔE vs. tE ( tE =0s, 500s, 1500s and 3000s).

From the figure it is clear that the value of optical gap is found to increase after illumination, which indicates that photo bleaching phenomenon is taking place in this material.

The value of absorption coefficient (α) is calculated at different energies and plotted in Fig.4

1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.80

20000

40000

60000

80000

100000

120000

α(c

m-1)

hν (eV)

0 sec 500 sec1500 sec 3000 sec

GaSe

Fig. 4. Plot of α vs. energy at different tE ( tE =0s, 500s, 1500s and 3000s).

The value of absorption coefficient, α, of GaSe thin film decreases on exposing the film to laser radiations. 4. Conclusions

Ga50Se50 thin films were prepared by thermal evaporation technique under vacuum on glass substrates kept at room temperature. The irradiation of a-Ga50Se50 thin film with laser light of wavelength λ= 432 nm has pronounced effect on the optical properties. There is increase of Eg

opt with irradiation in these thin films which shows photobleaching phenomenon. The absorption coefficient (α) decreases on irradiation. The results have been explained on the basis of some structural changes occurring after the laser irradiations. Acknowledgements This work is financially supported by CSIR (Major Research Project) New Delhi. References [1] A.M. Andriesh, M.S. Iovu, S.D. Shutov, J.

Optoelect. Adv. Mat. 4 (2002) 631. [2] P.J.S. Ewen, C.W. Shinger, A. Zakery, A.

Zekak, A.E. Owen, SPIE Infrared Optoelectron. Mater. Dev. 1512 (1991) 101.

[3] A. Ganjoo, N. Yoshida, K. Shimakawa, Recent Res. Dev. Appl. Phys. 2 (1999) 129.

[4] M.S. Iovu, S.D. Shutov, M. Popescu, J. Non-Cryst. Solids. 299 (2002) 924.


[5] S.K. Tripathi, A. Thakur, G. Singh, J. Sharma, V. Sharma, K.P. Singh, G.S.S. Saini, N. Goyal, J. Opt. Adv. Mater. 7 (2005) 2095.

[6] A. Arsh, M. Klebanov, V. Lyubin, L. Shapiro, A. Feigel, M. Veinger, B. Sfez, Opt. Mater. 26 (2004) 301.

[7] V. Lyubin, M. Klebanov, A. Feigel, B. Sfez, J. Non-Cryst. Solids. 459 (2004) 183.

[8] N. Goyal, A. Zolanvari, S.K. Tripathi, J. Mater. Sci.: Mater. El. 12 (2001) 523.

[9] M. Mitkova, M.N. Kozicki, H.C. Kim, T.L. Alford, J. Non-Cryst. Solids. 338 (2004) 552.

[10] L. Tichy, H. Ticha, P. Nagels, R. Callaerts, R. Mertens, M. Vlcek, Mater. Lett. 39 (1999) 122.

[11] J.M. Gonzalez-Leal, A. Ledesma, A.M. Bernal-Oliva, R. Prieto- Alcon, E. Marquez, J.A. Angel, J. Carabe, Mater. Lett. 39 (1999) 232.

[12] V.M. Lyubin, M. Klebanov, B. Sfez, B. Ashkinadze, Mater. Lett. 58 (2004) 1706.

[13] S. Gopal, C. Viswanathan, B. Karunagaran, Sa. K. Narayandas, D. Mangalraj and Junsin Yi, Cryst. Res. Technol. 40 (2005) 557.

[14] Falah I Mustafa, Akshay Kumar, N.Goyal, S.K.Tripathi. J. Opt. Adv. Mater. 9 (2007) 3210.

[15] F. Levy, Physics and Chemistry of Materials with Layered Structures: Structural Chemistry of Layer – type Phases (Reidel, Boston, 1976), Vol. 5, p.146.

[16] N.C. Fernelius, Prog. Cryst. Growth Charact. Mater. 28 (1994) 275.

[17] A.F.Qasrawi, Cryst. Res.Technol. 40 (2005) 610.

[18] P.J.S. Ewen, A. Zakery, A.P. Firth, A.E. Owen, Phil. Mag. B 57 (1988) 1.

[19] J. Tauc, A. Menth, J. Non-Cryst. Solids 8 (1972) 569.

[20] J.I. Pankove, Optical Processes in Semiconductors, Englewood Cliffs. NJ: Prentice-Hall (1971).

[21] R. Swanepoel, J. Phys. E 16 (1983) 1214.

Preparation of Nanoparticles by Laser Ablation in Solution: Influence of the Laser Wave Length and the Surrounding Liquid

Environment on Particle Size

R. Sarkar, P. Kumbhakar and A.K Mitra Department of Physics, National Institute of Technology,

Mahatma Gandhi Avenue, Durgapur-9, West Bengal. E-mail: [email protected]

Abstract

In this work we concentrate on preparation of metal colloids by laser ablation. Metal colloids are one

of the important nano-size materials. Through laser ablation in solutions is a new promising technique to obtain metal colloids. Laser ablation is the process of removing materials from a solid (or occasionally liquid) surface by irradiating it with a laser beam. At low laser flux, the material is heated by the absorbed laser energy and evaporates or sublimes. At high laser flux, the material is typically converted to plasma. The surrounding liquid environment affect the size and optical properties of nano particles prepared through laser ablation technique using a pulsed Nd: YAG laser. By laser ablation silver nano particles can be prepared. For this silver targets may be kept in acetone, deionised water, ethanol and others. Metal colloids in different solutions are found to have different size distributions. We shall investigate the influence of the ablation condition on the ablation efficiency and the size of the particles. The effect of absorption of the incident laser lights by the colloidal particles and its influence on the particle size can also be studied. The influence of the incident laser wavelength on the ablation efficiency can also be investigated. The wave length of the irradiating laser light will sufficiently affect the ablation efficiency because it is associated with the absorption of the metal targets at its surface and also with energy of the incident photons. In this context the control of particle size is a very important factor in colloid synthesis, because the character of nano-size metal particles is much sensitive to their size. It can be demonstrated that particle size of silver colloids decreases with increasing the intensity of irradiating laser light.

1. Introduction

Laser ablation is the process of removing material from a solid surface by irradiating it with a laser beam [1-9]. A laser is composed of an active laser medium, or gain medium, and a resonant optical cavity. The gain medium transfers external energy into the laser beam At low laser flux, the material is heated by the absorbed laser energy and evaporates or sublimates. At high laser flux, the material is typically converted to a plasma. Usually, laser ablation refers to removing material with a pulsed laser, but it is possible to ablate material with a continuous wave laser beam if the laser intensity is high enough. It is a material of controlled purity, size, concentration, and shape, which amplifies the beam by the process of stimulated emission as shown below: stimulated emission is the process by which, when perturbed by a photon, matter may lose energy resulting in the creation of another photon. The perturbing photon is not destroyed in the process and the second photon is created with the same

phase, frequency, polarization, and direction of travel as the original. Stimulated emission is really a quantum mechanical phenomenon. The process can be thought of as optical amplification and it forms the basis of both the laser and maser The gain medium is energized, or pumped, by an external energy source. Examples of pump sources include electricity and light, for example from a flash lamp or from another laser. The pump energy is absorbed by the laser medium, placing some of its particles into high-energy (excited) quantum states. Particles can interact with light both by absorbing photons or by emitting photons.

Emission can be spontaneous or stimulated. In the latter case, the photon is emitted in the same direction as the light that is passing by. When the number of particles in one excited state exceeds the number of particles in some lower-energy state, population inversion is achieved and the amount of stimulated emission due to light that passes through is larger than the amount of absorption. Hence, the light is amplified. Strictly speaking, these are the


essential ingredients of a laser. However, usually the term laser is used for devices where the light that is amplified is produced as spontaneous emission from the same gain medium as where the amplification takes place. Devices where light from an external source is amplified are normally called optical amplifiers. Fig. 1. Typical diagram for stimulated emission

An optical amplifier is a device that amplifies an optical signal directly, without the need to first convert it to an electrical signal. An optical amplifier may be thought of as a laser without an optical cavity, or one in which feedback from the cavity is suppressed. Stimulated emission in the amplifier's gain medium causes amplification of incoming light. Optical amplifiers are important in optical communication and laser physics. The depth over which the laser energy is absorbed, and thus the amount of material removed by a single laser pulse, depends on the material's optical properties and the laser wavelength. Laser pulses can vary over a very wide range of duration milliseconds to femtoseconds and fluxes, and can be precisely controlled. This makes laser ablation very valuable for both research and

industrial applications. The simplest application of laser ablation is to remove material from a solid surface in a controlled fashion. Laser machining and particularly laser drilling are examples; pulsed lasers can drill extremely small, deep holes through very hard materials. Very short laser pulses remove material so quickly that the surrounding material absorbs very little heat, so laser drilling can be done on delicate or heat-sensitive materials, including tooth enamel (laser dentistry).Also, laser energy can be selectively absorbed by coatings, particularly on metal, so CO2 or Nd:YAG pulsed lasers can be used to clean surfaces, remove paint or coating, or prepare surfaces for painting without damaging the underlying surface. High power lasers clean a large spot with a single pulse. Lower power lasers use many small pulses which may be scanned across an area. Another class of applications uses laser ablation to process the material removed into new forms either not possible or difficult to produce by other means. A recent example is the production of carbon nanotubes. Laser ablation has biological applications and can be used to destroy nerves and other tissues [1-3]. For example, a species of pond snails, Helisoma trivolvis can have their sensory neurons laser ablated off when the snail is still an embryo to prevent use of those nerves. 2. Experimental

Fig: 2 depict a configuration of experimental apparatus. The fundamental (1064nm) and the frequency multiplied outputs (355nm and 532nm) of an Nd: YAG laser were used as an irradiating source. The laser was operated at the power 170W. The central part of the laser beam was selected with an aperture to control the spot size of nonfocused laser light. Laser was conducted onto the target through the opening of the beaker. A lens (f=10mm) was placed above

Fig. 2. Typical set up for laser ablation

LASER LENS BEAKER SOLUTION TERGET

Preparation of Nanoparticles by Laser Ablation in Solution 147

the beaker when laser light was focused onto the target. To change the focusing condition of laser light, the relative position of the lens to the target was varied. Sliver target (99.99%) was washed with distilled water and placed in the beaker containing 10 ml distilled water. A yellowish colloidal solution of silver nanoparticles was obtained.

Fojtik and Henglein [7] were first to produce nanomaterials (Au, Ni, C) by Ruby laser ablation of thin 35-50nm films (Au and Ni) and suspended powders (C) in isopropanol, cyclohexane and water. The size of the produced nano particles varied from few nanometers to few ten nanometers (for Au and Ni). In 1993, Neddersen, Chumanov, and Cotton [8] used laser ablation technique to prepare solutions of different colloidal nanoparticles (Ag, Au, Pt, Pd, Cu). They evaporated solid targets by a Nd:YAG laser (1064) in water, methanol, and acetone. This technique consists of the employment of plasmon-related properties of metal nanoparticles to enhance Raman spectroscopy signals from different absorbed molecules. The method requires an ultrapure metal surface since the residual ions or other impurities can strongly influence the final signal. Laser ablation can be used to rapidly produce a variety of ultrapure materials.

The fabrication and stabilization of colloidal nanoparticles from a variety of materials was studied by Yang el al. [9]. They manage to synthesize diamond nanoparticles during laser ablation of carbon in water and acetone. Similar ablation in 25% ammoniac lead to the formation of C3N4 nanoparticles. Yeh et. al[10] managed to produce pure Cu nanoparticles during the ablation of CuO powders in isopropanol. Recently Mafune et. al. achieved significant progress in the development of chemical methods to control the nanofabrication during laser ablation in the liquids. The size of the nano particles can be drastically reduce by the use of aqueous solutions of surfactants, which cover the particles just after their ablation and thus prevent them from a further agglomeration. Sodium dodecyl sulfate (SDS). Two phenomena can be clearly distinguished, one of them the formation of plasma jet at the first moments after the laser action. This jet was attributed to radiation-related ablation of materials. The second one is the formation of the cavitations bubble, which grows first and then collapses. The collapse of bubble releases a significant amount of mechanical energy and can become an additional source of the material ablation.

The size of the particle is very important factor in colloid synthesis. The dependence of particle size on the number of irradiating laser pulses has been studied by M. Prochazka el.al. [1]. They demonstrated that the particle size of the silver colloids decrease with the increase of the irradiating laser pulses at 1064nm. A similar result has been obtained by Mafune et. al. [2] for gold colloids. They prepared nanocolloids with Nd:YAG fundamental i.e. 1064nm laser radiation and its second harmonic i.e. 532nm laser radiation. These results demonstrated the possibility of control of particle size of collides by changing the wavelength of the irradiating laser pulses. Particles become smaller from 29nm to 12nm with the decrease in laser wavelength under the incidence of laser light at a high influence of 36J/cm2 [3]. Semerok et. al [4] reported the wavelength dependence of ablation efficiency of various metals (Al, Cu, Mo, Fe, Pb, and Ni) in the atmospheric circumstances using laser light at 1064, 532, and 266 nm. The ablation efficiencies were higher at shorter wavelengths for all metal species.

A most remarkable feature by T.Tsuji et. al [3] of the laser ablation of silver and copper in water is the relation between the ablation efficiency and the wavelength of the focused laser light. The ablation efficiency was higher at longer wavelengths than that at shorter wavelengths. T. Tsuji et. al [3] reported that the two possible mechanisms for the self absorption process. One is inter-pulse self-absorption in which colloidal particles formed by former pulses absorb following pulses. The other is intra-pulse self-absorption, in which colloidal particles formed by former part of a laser pulse absorb later part of the same pulse. The latter may occur in the case of ablation using a nanosecond laser pulse. The process of inter-pulse self-absorption is essentially same as that of additional ablation of colloids, meaning that its effect depends on number of incident laser pulses. On the other hand, effect of intra-pulse self absorption is independent on the number of incident laser pulses, because it affects on colloidal particles only in the inertial process of particle formation.

The relation between particle size and the laser wave length is also studied by Jeon and Yeh [5]. They found that the silver particles prepared by 532nm light in water and isopropanol were larger than those prepared by 1064nm light. N. Koshizaki et. al [6] studied that the laser ablation of platinum target in pure water at 355nm as a function of laser energy. They


describe that three distinct reactions regimes between the ablated target species and water at different laser focusing conditions. At low laser influence (<10J/cm2), material removal is caused by laser heating of the platinum surface and the primary products are small clusters with a large percentage of platinum atoms in a nonzero oxidation state. At intermediate state, platinum nanoparticles are the primary products. At the high influence large faceted particles drying of gel formed by reactive plasma etching of the targets.

Metal nanoparticles synthesized by laser ablation were found to exhibit unique properties and characteristics, which make them very important for photonics and biological sensing, imaging application. References [1] M. Prochazka, J. Stepanek, B.Vlekova, P-Y,

Turpin, Anal.Chem. 69 (1997) 5103 [2] F. Mafune, J. Kohno, Y. Takeda, T.

Kondow, H.Sawabe , J. Phys. Chem. B 105 (2001) 5114

[3] T. Tsuji, K. Iryo, N. Watanabe, M. Tsuji, J. Photochemistry and photobiology A: Chemistry 145 (2001) 201

[4] A. Semerok, C. Chaleard, V. Detalle, J.-L Lacour, P Mauchien, P. Meynadier, C. Nouvellon, B. Salle, P. Palianov, M. Perdirix, G. Petite, Appl. Surf. Sci 138 (1999) 311

[5] J. S. Jeon, C. H. Yeh, J. Chin. Chem. Soc. 45 (1998) 721

[6] W. T. Nichols, t. Sasaki. N. Koshizaki, J. App. Phys 100 (2006) 114913

[7] A. Fojtik, and A. Henglein, G. Ber. Bunsen Phys. Chem. 97 (1993) 252

[8] Neddersen, J. Chumanov, G. and Cotton T, M., Appl. Spectrosc., 47 (1993) 1959

[9] Yang, G.W., Wang, J.B and Liu, Q,X,. J. Phy. Condens. Matter, 10 (1998) 7923

[10] M. S. Yeh, Y. S. Yang, Y. P. Lee et al. J. Phys. Chem. B 103 (1999) 6851

The Synthesis of Copper Oxide Nanoparticles by Nitrate-Citrate Gel Method

Iqbal Singh*, R.K.Bedi† and Rajeev Kumar†

*PG Department of Physics, Khalsa College, Amritsar-143005 (Punjab), INDIA India †Material Science Laboratory, Department of Physics, Guru Nanak Dev University,

Amritsar–143005 (Punjab), INDIA India. E-mail:[email protected]

Abstract

Cu2O ( Cuprous oxide ) and CuO ( Cupric oxide ) are two important compounds of copper. Cu2O has

a cubic crystal structure and presents a p-type semi conducting behavior with a band gap of 2.0 eV. CuO having monoclinic crystal structure, p-type semiconductor with band gap of 1.21 eV. It has attracted great interest in recent years for variety of applications, such as catalyst, batteries, varistors, photochemically active, photoconductive compounds and field emission etc. In the present work nitrate-citrate gel technique was used to synthesize the ultrafine particles of copper oxide. The structural and morphological characterization of the copper oxide nanoparticles were done by using X-Ray diffraction and Scanning electron microscopy technique respectively. X-ray studies showed that decomposed gel consists of Cu2O and CuO phases. Calcination of the decomposed gel at 500˚C produce single CuO phase, with average grain size of 32 nm. 1. Introduction

Cu2O and CuO are p-type semiconductors having applications in all branches of science such as in chemistry and electronics etc. [1-5]. Considering the potential applications of copper-based materials, many kinds of copper oxide morphologies have been reported, such as nanorods [4], nanowires [6], nanosphere [7], shuttle like structure [8], nanoribbons [9] and nanoparticles [10] and so forth. In all these morphologies, different techniques have been used and these are listed as thermal evaporation technique [11]; sol-gel spin coating [12]; chemical method [13]; hydrothermal route [14]; electrodeposition in an acetate bath [15]; vacuum annealing technique [16]; polymer precursor technique [17]; spray pyrolysis method [18-19]; vacuum evaporation [20] and so forth. However, there have been no reports on the preparation of copper oxide nanoparticle by sol-gel auto combustion method.

The sol-gel combustion synthesis has evolved as a standard technique for oxide nanoparticle fabrication. Combustion synthesis has been one of the methods most commonly used to obtain powders with compositional uniformity. Combustion synthesis strongly depends upon the molar ratio of metal nitrate to fuel [26-27]. This approach has a bright prespective for the large scale and controllable production of copper oxide nanoparticles.


In the present work, copper oxide nanoparticles have been prepared using copper nitrate (Cu(NO3)2.3H2O), hydrated citric acid (C(OH)(COOH).(CH3COOH)2.H2O) and deionized water as solvent, were used as starting material. Stoichiometric composition of the redox mixture is calculated on the basis of total oxidizing and reduction valancies of the oxidizer to the fuel keeping (O/F) ratio unity [26]. Appropriate amount of Cu(NO3)2.3H2O is dissolved in deionized water and citric acid was then added into the aqueous solution to form complex with Cu2+ ions. The thermal dehydration of the solution was done in an oven at 80˚C, resulted in a highly viscous bluish gel, hereafter named as precursor. As soon as this gel was formed, it was placed on preheated hot plate ( ~200˚C ). The gel at this temperature swelled and got ignited with an evolution of large volume of gaseous products. The combustion process resulted in a voluminous black powder, hereafter named as decomposed gel. The decomposed gel was calcined at 500˚C using muffle furnace in air atmosphere to remove residual organics. The decomposed gel and calcined powder were characterized by using a XRD diffractometer (XRD, Philips), scanning electron micrographs were taken using Jeol JSM-6100 (Japan) instrument after coating the samples with gold.


3. Result and Discussion 3.1 Analysis of the crystal structure of the

nanoparticles :

Fig. 1. X-ray diffraction pattern of decomposed gel of CuO

Figure 1(a,b) illustrates the XRD profiles of the copper oxide nanoparticles in decomposed gel and calcined powder at 500˚C respectively. Figure 1(a) shows that the decomposed gel, two phases Cu2O and CuO co-exist. All the peaks can be indexed and compared with those reported in literature. XRD pattern of the samples shows strong, sharp peaks, indicating that product after combustion has good crystallinity [23-25]. In figure 1(a) three peaks corresponding to the reflection from (110), (111) and (200) atomic planes of Cu2O, which are in addition to peaks of CuO corresponding to reflection from the atomic planes (110), (002), (111), (-202), (020), (202), (-311) and (220) [12-14,18, 28-29]. In the XRD pattern no peak from other impurities have been detected.

All the peaks can be ascribed as monoclinic crystal structure for CuO phase with lattice constant a=0.468 nm [12,14]. Figure 1(b) illustrate the XRD profile of the calcined

powder. This shows that calcinations at 500˚C give CuO phase only. It seems that the Cu2O phase is the intermediate phase and acts as catalytic site for the formation of CuO during calcinations. During calcination CuO is formed through oxidation reaction of Cu2O. The chemical reaction to synthesize the CuO nanoparticles can be described by following equation during calcination Cu2O + 1/2O2 2CuO

The calcination at 500˚C causes an increase in average crystallite size from 24.83-32nm. This increase in average crystallite size might be due to change in crystallographic phase from Cu2O to CuO. This change is might be due to the the quantity of the material to be oxidized that is why the phase conversion temperature was formed to be different in all studies. The formation Cu2O is strongly dependent on the temperature and its oxidation to CuO phase is independent of the oxygen partial pressure [32]. If we compare our results of the calcined powder with literature we found that the Cu2O phase conversion to CuO is almost at the same temperature (~500°C) [33,34]. The information on strain and crystallite size for copper oxide nanoparticles was obtained from the full-widths-at-half-maximum (FWHM) of the diffraction peaks. The FWHM (β) can be expressed as a linear combination of strain (ε) and crystallites size (L) through the following relation [35]

Fig. 2 The β cos θ/ λ vs sin θ/ λ plot for decomposed gel and calcined at 500˚C.

1(a)

1(b

The Synthesis of Copper Oxide Nanoparticles by Nitrate-Citrate Gel Method 151

Figure 2 represents a βcos θ/ λ vs sin θ/ λ plot and the slope of straight line gives the amount of strain and intercept on the βcos θ/ λ axis gives the crystallites size and the values are 32 nm and 0.9x10-3 respectively for calcined sample. 3.2 SEM study of the nanoparticles

The surface morphology of the CuO powder shown in figure 3(a,b) delineates the distribution and shape of the crystallites. The decomposed gel appear to very porous possibly due to gas evolution.

Fig 3 SEM microstructure of decomposed gel of CuO

The SEM analysis of the copper oxide particles reveals a nearly faceted geometry whereas the other chemical methods show the formation of CuO platelets[13], hollow microsphere [14] and nanorods [28-29]. It appears that the particles are dispersed with negligible agglomeration and show high surface area.

4. Conclusion The copper oxide powder can be synthesized by a simple sol-gel combustion method using citric acid as the complexing agent and as a fuel at low temperature. The X-ray diffraction patteren shows the formation of cubic Cu2O phase and monoclinic phase of CuO in the decompose gel, with crystallite size of 24.83 nm. The crystallite size grows to nm during calcination and Cu2O phase has been oxidized to CuO phase. The SEM analysis confirmed the formation of sub micron particle nature of copper oxide powder. The citric fuel method appears to be well suited for the preparation of copper oxide nanoparticles. References 1. H. Wang, J. Z. Xu, J. J. Zhu, H.Y. Chen, J.

Cryst. Growth 244 (2002) 88. 2. T. Ishihara, M. Higuchi, T. Takagi, M. Ito,

H. Nishiguchi, T. Takita, J. Mater. Chem. 8 (1998) 2037.

3. A.D. Berry, K. D. Gaskill, T. T. Holm, E. J. Cukauskas, R. Kaplan, R. L. Henry, Appl. Phys. Lett. 52 (1988) 1743.

4. W.Z. Wang, Y. J. Zhan, G. H. Wang, Chem. Commun. (2001) 727.

5. C. H. Lu, L. M. Qi, J. H. Yang, D. Y. Zhang, N. Z. Wu, J. M. Ma, J. Phys. Chem. B 108 (2004) 17825.

8. D. Chen, G. Z. Shen, K. B. Tang, Y. T. Qian, J. Cryst. Growth 254 (2003) 225.

9. B. Liu, H. C. Zeng,, J. Am. Chem. Soc. 126 (2004) 8124.

10. R.V. Kumar, R. Elgamiel, Y. Diamant, A. Gedanken, Langmuir 17 (2001) 1406.

11. L.S. Haung, S.G. Yang, T. Li, B. X. Gu, Y. W. Du, Y. N. Lu, S. Z. Shi, J. Cryst. Growth 260 (2004) 130.

12. H. Guan, C. Shao, B. Chen, J. Gong, X. Yang, Inorg. Chem. Comm. 6 (2003) 1409.

13. D. Li, Y.H. Leung, A. B. Djurisic, Z. T. Liu, M. H. Xie, J. Gao, W. K. Chan, J. Crystal growth 282 (2005) 105.

14. Y. Zhang, S. Wang, Y. Qian, Z. Zhang, Solid State Sci. 8 (2006) 462.

15. R. P. Wijesundera, M. Hidaka, K. Koga, M. Sakai, W Siripda, Thin Solid Films 500 (2006) 241.

16. S. Y. Lee, N. Mettlach, N. Nguyen, Y. M. Sun, J. M. White, Appl. Surface Sci., 206 (2003) 102.

17. A. I. Fernandez, A. Calleja, J. M. Chimenos, M. A. Fernandez, X. G. Capdevila, M.


Serarra, H. Xuriguera, F. Espiell, J. Sol-Gel Sci. & Tech. 36 (2005) 11.

18. J. Morales, L. Sanchez, F. Martin, J. R. Ramos-Barrado, M. Sanchenz, Electrochemica Acta 49 (2004)4589.

19. G. Papadimitropoulos, N. Vourdas, V. Em. Vamvaskas, D. DavaZoglov, Thin Solid Films, (2006).

20. T Kosugi, S Kaneku, J. Am. Ceram. Soc. 81 (1998) 3117.

26. K. C. Patil, S.T.Aruna, S. Ekambaram, Solid State & Mater. Sci. 2 (1997) 158.

27. Y. Zhang, S. Wang, Y. Qian, Z. Zhang, J. Crystal Growth 268 (2004) 590.

29. W. Wangab, Y. Zhana and G. Wang, Chem. Commun., (2001), 727.

30. Y. Wang, J. Zhu, X. Yang, L. Lu, X. Wang, Mater. Res. Bull. (2005).

31. B.J. Chung, J. Y. Park, S. M. Sim., J. Ceram. Processing Res. 4(3) (2003) 145.

31. K.C. Patil, S.T. Aruna, T. Mimani, Current opinion in Solid State & M. Sci. 6 (2002) 507.

32. T. Mimani, K.C. Patil, Mater. Phys. Mech. 4(2001)134.

33. S. R. Dhage, S. P. Gaikwad, V. Samuel, V. Ravi, Bull. Mater. Sci. 27 (2004) 221.

34. H. Li, X. Liu, L. Huang, Ceram Inter.(2006). 35. D Bahadur, S Rajkumar, A Kumar, J. Chem.

Sci. 118(2006)15. 36. Q. Liu, Y. Liang, H. Liu, J.Hang, Z. Xu,

Mater. Chem. And Phys. (2005). 37. O. A. Ileperuma, C. Vithana, K. Premaratne,

S. N. Akuranthilaka, S. M. Mcgregor, I. M. Dharmadasa, J. Mater. Sci. (1998) 367.

38. R. A. Singh, R. K. Gupta, S. K. Singh, Bull. Mater. Sci. 28(2005) 423.

39. R. S. Devan, Y. D. Kolekar, B. K. Chougule, J. Phys.: Cond. Mater 18 (2006) 9809.

40. R. Maity, K. K. Chattoopadhyay, Solar Energy Mater. & Solar Cells 90 (2006) 597.

Corrosion Resistance Properties of Surface Tolerant Coatings

Shyamjeet Yadav and D.D.N. Singh Applied Chemistry & Corrosion Division

National Metallurgical Laboratory Jamshedpur-831007 Email: [email protected]

Abstract The corrosion resistance performance of an organic coating formulated with vinyl based polymers, phosphate – vinyl ester, alkyd resin, micaceous pigments, catalyst and dryers has been studied under different test conditions. The coating is observed to exhibit high tolerance towards surface blemishes such as presence of oxides, mill scales etc. Two types of coated mild steel panels (sand blasted and mill scale covered) were evaluated in 3.5% sodium chloride for more than 4000 hours of exposure. It is observed that the coating maintains the polarization resistance in the range of giga ohm which is imparted by highly moisture resistant dielectric coatings. The water absorption tendency of the coating during the period of the test is low. The coated specimens exposed in salt spray chamber (as per ASTM B117) for 1000 hours did not show any trace of appearance of rusting. Adhesion tests (ASTM D 3359), prohesion tests ( ASTMG85-94 A5), cyclic humidity test (ASTMG60-86), and dilute acetic acid tests (ASTMD1308) were also performed and the coating passed the tests. Under all the conditions of the assessment it is observed that the coating applied on steel covered with black oxide scale performs superior than applied on blasted surface. Electrochemical impedance spectroscopy, D.C. cyclic polarization, Raman spectroscopy and Scanning electron microscopy studies have been performed to understand the mechanism involved in the protection afforded by the coating. 1. Introduction

Adherence of coatings to metals surfaces is pre-requisite for their good corrosion resistance performance in an aggressive environment. In order to achieve it, the coatings on application should wet the metals surface effectively. It is possible only if surface tension of the coating is very low in comparison to the substrate surface. Clean blasted metal surfaces possess very high surface tension whereas most of the organic coatings bear lower surface tension than the cleaned metal surface. This makes the coatings to adhere very effectively on metals surfaces. It is universally accepted that conventional coatings adhere well and perform satisfactorily when applied on sand/shot blasted surface free of oxides. Roughness of surface caused due to blasting provides good anchorage for the coating. The presence of any trace of oxide on metals makes their surface poor in accepting the coating. The presence of these oxides below the coat may act as active cathode for corrosion reactions in eventuality of development of defects. All the standards incorporating code of practices for application of coatings unexceptionally recommend an oxide free surface prior to the application of the coatings. In most of the

applications of heavy-duty paints, it is normal practice to apply a primer coat on sand /shot blasted surface followed by intermediate and top coats. The primer coats which are traditionally formulated comprising of red lead, zinc chromate, phosphates etc., act as inhibitive and bonding layer with the substrate and act as base for the other components of the coatings, forming a protective layer. The blasting of surface and application of primer coats not only make the process very expensive but also are concerned for pollution. Lead and chromium compounds are well known carcinogen and their uses are being discouraged all over the world. The development of a coating formulation free of lead and chromium compounds having anchoring tendencies with oxide covered surfaces therefore, is the need of time. The literature searches have shown that very serious efforts are being made all over the world to develop self-priming one step coatings.

The history for such developments goes back to 1931 when H.O. Albrecht filed a patent on single step formulation to control rust [1]. Thereafter, little efforts were put forward to improve this type of coatings. A two pack system was suggested for coating the metal substrate [2]. The first systematic step in this direction was reported by C.H.Hare in 1978 where In this


method, the wash primer having phosphoric acid as one of the components was slowly mixed with main component of coating having resins, The problem associated with this approach was that the components of the coatings got exhausted within a few hours of mixing of wash primer and they became unusable. Spadafora et.al. reported the application of primer less coating on aluminium substrate [3]. C.T. Lin et.al in 1992 [4] and Meldrum et.al in 1993 [5] published their results incorporating the formulations and performance of a single step phosphate / paint system. In 1993, Lin et.al reported the interfacial chemistry of a single – step phosphate/paint system and reported that enhanced coating adhesion was caused due to the formation of phosphorus –oxygen-carbon linkage bond [6]. It was suggested by Li and Lin that organic phosphatizing reagents were more dispersible in in-situ phosphatizing coatings and phosphate bonding was via acid (PO4

--)–base (Fe) type interaction rather than an induced dipole interaction between P=O and Fe complex type [7]. A series of papers had been published by these authors incorporating the results in this area [8-11]. Some patents on such coatings have also been filed recently covering the novelty of these types of coatings [12-13].

All the above developments have claimed that such coatings adhere well and provide equivalent or superior protection to the metal substrates in comparison to the traditional multi step coating process incorporating surface pretreatment steps. Most of the studies for the action of such coatings described in the literature pertain to their actions on oxide free surface. Since single step in situ phosphatizing coatings generally incorporate esters of phosphoric acid and organic alcohols and they are reported to interact more effectively with oxide than on iron surface [14], it was considered important to study the efficacy of such coatings on mill scale covered surface. In this communication, we report the results of performance of such a coating. The present communication attempts to answer the following quests: • Is there any deterioration or improvement in

performance of in-situ phosphatizing type of coatings when applied directly on mill scale covered steel surface? , and

• What is the mechanism of their bonding with oxide free and mill scale covered surfaces.

2. Experimental

The coated mild steel (composition: C=0.22, Mn=0.25, Si=0.65) panels of dimension 15x10x0.2 cm were supplied by the manufacturer of the

coating. All the test panels were prepared from the same lot of coil having identical composition. Two types of panels were prepared. One set was sand blasted before the application of the coating. The other set of panels were cut from the coil having intact black colour mill scale of thickness about 55-60 μm. Although the complete composition of the coating was not revealed by the manufacturer owing to the commercial considerations but a broad composition was available. Accordingly, one coat of primer comprising of copolymer of vinyl and alkyd with pendant reactive functional groups, randomly distributed along the chain length was applied on steel plates having thickness of about 150 μm. This copolymer was reacted with phosphoric acid to form ester. The top coat was acrylic based resin of thickness about 50 μm. The resins of both the top and primer coats were dissolved in solvent comprising of a mixture of xylene, toluene, naphtha and butanol. The primer and the top coats were applied by pneumatic spray in 3 coats with flash of interval of 15 minutes between coats, and air dried for 14 days before testing. Prior to the application of the coating, the panels were wiped with xylene to remove oil / grease. The total thickness of the coating as measured by thickness gauge meter was in the range of 200 ± 5 μm. The salt spray test was performed by maintaining the test chamber as described in ASTM B-117. To test the adhesion of the coating, the coated panels were hatched and evaluated as per ASTM D 3359. The coated panels were subjected to Electrochemical Impedance Spectroscopy (EIS) and cyclic polarization studies in 3.5% NaCl solution. The EIS studies were performed by imposing 10 mv of sinusoidal voltage ( with reference to open circuit potential) at the working electrode and varying the frequency from 100 KHz to 0.001 Hz. Cyclic polarization studies were performed at scan rate of 0.5mv / sec. The electrochemical studies were performed in a flex cell exposing the working electrode area of 8 cm2. Two graphite rods at two sides of the working electrode were fitted, which acted as auxiliary electrodes. The reference electrode was a saturated calomel electrode (SCE). A luggin capillary was used to provide electrolytic contact between the calomel electrode and electrochemical cell. All the electrochemical tests were performed at the temperature 30 ± 2oC by using a Gamry Potentiostat (supplied by M/S Gamry Instruments of USA).

Another set of experiments were performed to characterize the interface of the coating / steel by using analytical techniques. Raman spectroscopy studies surface were performed by using Almega

Corrosion Resistance Properties of Surface Tolerant Coatings 155

dispersive Raman Spectroscope by exciting the laser beam of He-Ne of 532 nm wavelength on the samples. The power of the laser was kept lowest possible (6 mw) to avoid the transformation of surface characteristics due to heating effect of laser. The locations of the specimens to be studied were focused through a Olympus microscope at the magnification of 50. The sample holder had motorized platform with Jockey to have a fine focusing at a suitable desired part of the sample. The grating was 672 lines / mm, 25 μm pinhole. Prior to analysis of samples, the instrument was calibrated by using pure Silicon at the peak of 522.28 cm-1. SEM and EDX analysis were performed by using Jeol make scanning electron microscope (JSM 840A). 3. Results and Discussion

The variation in module of polarization resistance (Rp) with passage of time for the coatings applied on blasted and mill scale covered (MSC) steel surface exposed in 3.5% sodium chloride, is shown in figure 1.

00.20.40.60.8

11.21.41.61.82

0 900 1800 2700 3600 4500

Time ( hours )

B

M SC

Fig. 1. Time vs Rp graph of coated samples exposed in 3.5% NaCl solution It is seen from this figure that during the initial period of exposure, the deterioration in Rp with exposure time is almost negligible in both the cases. At the time of termination of the experiments (4350 hours), the coating applied on MSC surface shows higher resistance than the blasted one, indicating that the coating adheres and forms more protective film on steel surface having oxide scales. This observation was also corroborated by the adhesion test performed as per ASTM D3359. The unexposed coating applied on

both types of surfaces passed the tests and no peeling off of the coating with adhesive tape took place. However, the test performed on exposed surface (exposed for 1000 hours in salt spray chamber) exhibited very distinct and reproducible results. The coating on blasted surface got completely peeled off with the tape, where as that on non -blasted surface, it remained intact. These observations very clearly demonstrate that the coating interacts very effectively with the surface having mill scale. The exposed surface after peeling off of the coating was subjected to Raman Spectroscopy and the results are shown in figures 2&3.

It is observed that the blasted surface had developed different type of oxides of iron below the coating and the oxidation was localized. In case of MSC surface, however, the observed oxide was only α-Fe2O3. It is well known that the mill scale is formed by combination of three oxides. The one very close to the steel surface is wustite (FeO) followed by magnetite (Fe3O4) and top one is hematite (Fe2O3). The observance of peaks of hematite in Raman spectra indicates that the layer of hematite did not transform to other form of oxide during the period of exposure.

These findings further confirm that the coating had interacted strongly forming impervious layer with mill scale covered steel than the surface of blasted steel. It is also seen from the Raman spectra of figure 2 that in addition to the peak of hematite, a prominent peak at 406 cm-1 is observed. This peak is ascribed to presence of phosphate ions. This indicates that the phosphate ions present in the coating have interacted well with the oxide. It is to be noted here that the ter-polymer blend of the coating is reacted with phosphoric acid and they are expected to react and form polyvinyl ester of phosphoric acid. This ester though phosphate groups perhaps interacts more strongly with oxide of iron than on the oxide free iron surface and form a layer of organic phosphate. This layer acts as a primer on steel and felicitates the coating to adhere over it. The Raman Spectroscopy study of the exposed surface after peeling off of the coating was also performed and the results are shown in figure 3. It is seen from this figure that the whole surface is covered with different layers of oxides of iron. Here very strong peaks of γ- and α- FeOOH are noted.


Table 1. Raman peaks and Identified compound

s.no.

Raman peak(cm-

1)

Identified compound

1 1317, 291.90

α-Fe2O3

2 659, 221 γ-FeOOH

Fig 2.

3 406.82 O2(PO4)3-

1 216 γ-FeOOH 2 392 α-FeOOH

Fig. 3

3 495 β-FeOOH

These forms of oxides of iron are quite stable and protect the steel's surface very effectively. Very weak peak of β- FeOOH is also recorded. This oxide on steel is formed in chloride bearing environments. These results suggest that the coating provides an effective barrier and controls the diffusion of water/ moisture/ chloride through it. The capacitance of the coating (Cc) with the passage of exposure time is shown in figure 4. The Cc value for both the coatings remains almost constant and identical up to the exposure period of about 2700 hours. After this period of exposure, the capacitances of both the coatings are observed to increase considerably. However, the MSC surface maintains lower Cc than the coatings applied on blasted surface. These results suggest that the top layers of the applied coating on blasted and MSC surfaces behave identically until the test electrolyte is reached to the substrate surface. Once the moisture/ Chloride reaches at the coating/steel interface, the innermost layer of the coating plays very decisive role on its performance. Since Phosphate ester of PVA forms stable and impervious primer layer with mill scale covered surface, it inhibits the reaction and affects the Cc values. The performance of any organic coating is strongly dependent on its stability to withstand the diffusion of moisture through it. Since the capacitance values of any coating are largely dependent on the quantity of water absorbed by it, the changes in these values were utilized to calculate water absorbed by coating. The capacitance of a coated surface exposed in aqueous solution is related by the equation: Ct/Co = 80x…………………….(1)

Fig. 2. Raman spectra of MSC exposed surface after peeling off of the coating

Fig. 3. Raman spectra of blasted surface applied coating after peeling off of the coating

Where Co and Ct are the capacitance value of the system at zero period of exposure (before exposure of the coating in the electrolyte) and after time t, respectively. 80 is dielectric constant of water and x is the volume fraction of water absorbed in the coating. The percentage volume fraction of absorbed water with passage of time calculated by using the above equation is shown in fig. 5. It is evident from the figure that the coating applied on both types of surfaces behaves identically up to the exposure period of about 2700 hours.

221.34241.78

291.90406.82

659.00

1317.29

2040.37 2680.782860.80

3154.363577.343737.833930.43

50

100

150

200

250

300

350

400

450

500

Ram

an in

tens

ity

1000 2000 3000 4000

Raman shift (cm-1)

124.15

216.02277.99

392.79

495.49

588.231292.25 2761.762846.50

2860.942882.35

2895.272981.273081.263183.50

3423.293443.213577.153621.693683.15

50

100

150

200

250

300

350

400

1000 2000 3000 4000

Raman shift (cm-1)

Ram

an in

tens

ity

Corrosion Resistance Properties of Surface Tolerant Coatings 157

-10

2

14

26

38

50

62

0 1000 2000 3000 4000 5000

Time (hours)

Cc

nF/c

m2 MSC

B

the exposure period of about 1500 hours. Beyond this period of exposure, the potential of MSC coated surface is again increased to its original potential of about -100mv (SCE). In case of the coating applied on blasted surface, the potential remains almost constant at about -500mv through out the test period. This potential corresponds to the reaction of steel with aqueous neutral chloride solution indicates that the electrolyte has reached on steel’s surface and corrosion reactions have set in. The potential of the MSC coated surface on the other hand, again enables up to exposure period about 2700 hours.

Fig. 4. Coating capacitance vs exposure time plot of coated surface exposed in 3.5%Nacl solution

00.20.40.60.8

11.21.41.61.8

2

0 900 1800 2700 3600 4500Time (hours)

X va

lue B

MSC

Fig. 5. Mole fraction of absorbed water vs exposure time plot for coated surfaces exposed in 3.5%NaCl solution

After this period, of exposure, an increase in

Cc of the coating applied on blasted surface is deserved. This increase cannot be attributed solely to the increase in water absorption. It is caused due to increase in conducting particles such as oxides and moistures at the steel’s surface below the coating. In contrast to this the Cc for MSC coated surface remains almost constant after 2700 hours of exposure. It is attributed to the presence of very effective and inhibitive layer of coating-phosphate-oxide which resists the development of conducting particles such as oxides etc. and maintains a constant value of Cc. These finding again fortify our earlier placed view that the coating/oxide interface forms an impervious layer of coating-phosphate-oxide The potentials of both the surfaces fall steeply after

Beyond this period of exposure the corrosion potential again drops down and attains the value of about -450mv (SEC) which is again closer to the potential of steel corroding in chloride solution [≈ -500mv (SEC)]. These results indicate that the diffusion of the test electrolyte commences from very beginning of exposure of the coating and it reaches near to the steel/ coating interface after the exposure of about 1500 hours. Owing to the presence of protective and adherent layer of polymer phosphate ester on MSC surface which acts as passivating primer, it reacts with water and develops a protective layer of iron oxide and phosphate, bringing the potential again in its original protective zone. When a substantial amount of chloride and water reaches near to this interface after the exposure of 2700 hours, the protective layer starts deteriorating and potential slowly drifts in active direction (figure 6) which inhibits the deterioration of substrate surface exposed to aqueous solution.

The corrosion potentials developed by the two-coated surfaces exposed in sodium chloride solution with passage of time are shown in fig. 6.

0

100

200

300

400

500

600

0 900 1800 2700 3600 4500

Time (hours)

Cor

rosi

on p

oten

tial (

mV)

, SC

E

MSC

B

Fig.6. Corrosion potential vs exposure time plot for coated samples exposed in 3.5%NaCl solution


Acknowledgements The authors express their sincere thanks to

the Director, National Metallurgical laboratory, Jamshedpur for granting permission to publish this work. We also acknowledge with thanks M/S Rotomac Electricals Pvt. Ltd. of Delhi for providing us the coated panels for the studies. References 1. H. O. Albrecht, Rust resisting coating

composition, U.S. Patent No. 1, 995, 954; 1931, June 10

2. C. H. Hare, corrosion and protection of metallic surfaces by painting, federation series on coating technology, Philadelphia, P A 1978, pp 5-50

3. S.J. Spadafora, C.R. Hegedus, D.J. Hirst, A. Eng, Primerlers finishing system for aluminium substrates, Mod- Paints coating, 1990, Sept pp 36-48

4. C. T. Lin, P. Lin, M.W. Hsiao, D.A. Meldrum, F.L. Martin, Chemistry of a simple step phosphate/ paint System, Ind. Eng. Chem. Res. 1992, 31, PP 424-30

5. D. A. Meldrum, C.T. Lin, AC impedance analysis and factorial designs of an In- Situ Phosphatizing coatings, S. Coat. Technol. 1993, 65, p 47.

6. C.T. Lin. P.Lin, F.Q. Puello, Interfacial Chemistry of a single- step Phosphate / paint System, Ind. Eng. Chem. Res. 1993, 32 pp 818-25

7. T. Yu , L. Li and C.T. Lin, Chemical affinity of In- situ phosphatizing reagents on cold rolled steel, J. Phys. Chem, 1995, 99, pp 7613 -20

8. T. Yu , C.T. lin, The performance of In- situ phosphatizing reagents in solvent borue paints, Ind. Eng. Chem. Res. 1997, 36, 368

9. T.Yu, C.T. Lin. In-situ Phosphatizing Coatings I, An air dried Lacquer System, J. Coat. Techn. 1999,77,61

10. Ibid, In-situ phosphatizing Coatings II, A high solids polyster baking enamels. J.Coat. Technol. 1999,77,61

11. M.C. Whitten, Y.Y. Chnang, C.T. Lin, Effect of catalyst and pigment on polyster- Melamine In- Situ phosphatizing coating on a cold rolled steel System, Ind Eng. Chem.Res. 2002,41 pp 5232-39

12. Dhrubo Bhattacharya, Self priming chromate free corrosion resistant coating composition and method, U.S. Patent 720, 8537, April24, 2007

13. C.T.Lin, Additive package for phosphatizing paint and method, U.S. Patent 5,322, 870 June 21, 1994

14. C.S. Saba, N.H. Forster, Tribology Letters, 2002, 12, pb 135-46

Apparent Molar Volumes and Viscosities of Amino Acids in Aqueous Sodium Nitrate Solutions at 298.15 K.

Tarlok S. Banipal, *Vaneet Dhir, * and Parampaul K. Banipal!

Department of Applied Chemistry* and Chemistry! , Guru Nanak Dev University, Amritsar – 143005, India Email: [email protected]

Abstract

Apparent molar volumes, V2,Φ and viscosities ,η , of Glycine , DL-α -alanine , DL-α -amino butyric acid , L-

valine and L-leucine in water and in (0.25.0.5,0.75,1.0,1.5,2.0) mol.kg-1 aqueous sodium nitrate solutions have been determined at 298.15K from densities and flow time measurements , respectively. The standard partial molar volumes, V2

0,Φ at infinite dilution, obtained from V2,Φ, have been used to calculate the corresponding volume of transfer, ΔV2,Φ from water to aqueous sodium nitrate solutions. B coefficients have been calculated.

1. Introduction

Proteins play a very important role in all the chemical and biological processes. In order to understand the proteins, simple low molecular weight compounds are often used as models. The most important behaviors of protein molecule is hydration which is very important in the structure and function of protein in aqueous solutions, in order to get an idea about the role of hydration in protein folding/unfolding, it is necessary to study both the native and denatured states of a protein. The use of low molecular weight compounds as models have two major importance’s. The first importance is the relative ease of the microscopic interpretation of experimental data. The second importance is by simple change of low molecular weight protein structure one can easily calculate the contribution of a chosen atomic group. The side chain groups of the amino acids residues provide a very important range of properties, from hydrophilic to hydrophobic groups. The side chain groups are involved in a wide range of interactions. For example :

(1) The phenomenon of electrostriction that is caused by the polar end groups. (2) The structure enforcing influences of the hydrophobic alkyl groups. (3) The interactions in between the hydrophilic and hydrophobic groups, hydrogen bonds that provides the peptide bonds in the polypeptide backbone. All these wide ranges of interactions are being are being affected with the change in the concentration of aqueous electrolyte that is being studied.

Attempts are being made for the systematic study of the volumetric and viscosity behavior of some amino acids in aqueous solutions of sodium nitrate at 298.15K. Salt induced electrostatic forces

are known to play a important role in modifying the protein structure by affecting the properties like solubility, denaturation, and activity of engymes1, 2.

Remarkable experimental work has been reported on thermodynamics of amino acids in aqueous alkali metal salts3-15. However, few studies on the thermodynamic properties of the amino acids, especially viscosity properties, have been carried out in aqueous alkaline earth metal salts solutions16, in spite of their biological importance17.Consequently, in the present paper, the apparent molar volumes, V2,Φ, and viscosities, η , of Glycine , DL-α -alanine , DL-α -amino butyric acid , L-valine and L-leucine in water and in aqueous sodiumnitratesolutions (0.25,0.5,.75,1.0,1.5,2.0)mol.kg-1 have been determined by measuring the densities using a vibrating -tube digital densimeter and an ubbelohde- type viscometer , respectively, at 298.15K. From these data, the partial molar volumes V2,Φ, and viscosity B coefficient have been calculated. The hydration number, nH, side chain contributions of amino acids, and concentration effect of sodium nitrate have been discussed in terms of various interactions.

2. Experimental

The amino acids selected for the present study Glycine , DL-α -alanine , DL-α -amino butyric acid , L-valine and L-leucine were obtained from sigma chemicals co. these along with sodium nitrate (AR, Thomas Baker, India), were used without further purification and dried over anhydrous CaCl2 in a vacuum desiccator before use. Deionized, doubly distilled degassed water with a specific conductance of less than 10-6 Ωcm-1 was used for all of the measurements. All solutions were prepared by mass


using a mettler balance with an accuracy of ± 0.01 mg. The solution densities were measured using a vibrating tube digital density meter (model DMA 60/602, Anton paar), accuracy of 3 x10-6 g.cm-3. Densities function was checked by measuring the densities of aqueous sodium chloride solutions, which agreed well with the literature values23. The temperature of water around the densimeter cell was controlled to within 0.01K.

Viscosities were measured using an ubbelohde-type viscometer, which was calibrated using the flow time of water from (288.15 to 318.15) K.The flow time of a constant volume of water through the capillary was measured with an electronic stop watch with a resolution of 0.01 s. The viscosity of a solution, η, is calculated by using following expression:

η=d(at–b/t) (1)

where d is the density of the solution, t is the flow time, and a and b are the viscometer constants. The measured viscosity values are accurate up to ± 0.001 m.pas.


The densities, ρ, and apparent molar volumes, V2,Φ, of amino acids in water and ion aqueous sodium nitrate solutions of various malalities (ms, molality of sodium nitrate solutions, mol.kg-1) at 298.15 K. Apparent molar volumes of amino acids have been calculated as follows:

V2,Φ = M/ρ – [ ( ρ – ρ0 )1000]/m ρ ρ0 (2)

Where M is the molar mass of amino acid, m (mol.kg-1) is the molality of amino acid, and ρ is the densities of solution and solvent, respectively. At infinite dilution, the apparent molar volumes, and partial molar volumes, are identical (V2,Φ =V2,

0Φ ). In

the case of negligible concentration dependence of V2,Φ , V2,Φ was determined by taking the average of all the data points. However, where finite concentration dependence was observed, V2,Φ was determined by least – squares fitting of the data using the following equation.

V2,Φ = V2,0Φ + Sv m (3)

where Sv is the experimental slope.

From the V20,Φ data, m the standard partial molar

volumes of transfer, ΔtV0, at infinite dilution from water to aqueous sodium nitrate solutions have been evaluated as follows ΔtV0 = V2,

0Φ (in aqueous

sodium nitrate) - V2, 0Φ(in water) (4)

The ΔtV0values for the amino acids are summarized in table 3 and illustrated in figure 1. The ΔtV0values for the studied amino acids are positive and increase almost linearly with the increase in concentration of aqueous NaNO3 solutions. The more positive ΔtV0values in the case of Glycine indicate the dominance of the effect of charged end groups (NH3+ and COO-) , whereas less positive ΔtV0values in the case of DL-α -amino butyric acid , L-valine and L-leucine indicate the effect of the hydrophobic parts.

McMillan and Mayer proposed a theory of solutions that permits the formal separation of the effects due to interactions between the pair of solute molecules and those due to interactions between two or more solute molecules. According to this treatment, at infinite dilution d ΔtV can be expressed as : ΔtV∞ = 2VXY mS + 3 VXY mS

2 (5)

Where X stands for amino acids and Y stands for sodium nitrate. Vxy and Vxyy are pair and triplet interaction coefficients, respectively. Vxy and Vxyy are positive and negative respectively respectively. The number of water molecules bound to the amino acids was calculated using the method reported by Millero etal.

nH=V2∞(elect)/Ve

∞-Vb∞ (6)

Where Ve is the molar volume of electrostricted water and Vb is the molar volume of bulk water. The value (25) of (Ve – Vb) is ~ - 3.3 cm3.mol-1 at 298.15K.

Fig. 1. Transfer viscosity coefficient, ∆trB /dm3mol-1, of some amino acids vs different molalities, mS, of sodium nitrate.

0

0.01

0.02

0.03

0 1 2 3mol.kg-1

tr B/d

m3 m

ol-1

Apparent Molar Volumes and Viscosities of Amino Acids in Aqueous Sodium Nitrate Solutions 161

Fig. 2. Standard partial molar volume of transfer, Δt V∞ , of some amino acids vs different molalities , mS, of sodium nitrate solutions at 298.15 K

Fig. 3. Contributions of :(◊) NH3+, COO- ;(ٱ) -CH2,

()-CHCH3; (x)- CHCH2CH3;( җ) -CHCH2CH(CH3)2.

Fig. 4. Partial molal specific volumes .

References [1] P. H. Von Hippel, T. Schleich, Acc. Chem.

Res.2 (1996) 257-265. [2] W. P. Jencks, Catalysis in Chemistry and

Enzymology, Mc Graw-Hill, New York, 1969, p.351.

[3] N. C. Dey, B. K. Saika, Can. J. Chem. 58 (1980) 1512-1515.

[4] R. K. Wadi, R. K. Goyal, J. Solution Chem.21 (1992) 163-170.

[5] A. Soto, A. Arce, M. Khoshkbarchi, J. H. Vera, Biophys. Chem.73 (1998) 77-83.

[6] R. Bhat, J. C. Ahluwalia, J. Phys. Chem. 89 (1985) 1099-1105.

[7] R. Bhat, J. C. Ahluwalia, Int. J.peptide protein Res.30 (1987) 145-152.

[8] R. K wadi, R. K. goyal, J.chem.Eng.Data. 37 (1992) 377-386.

[9] B. P. Kelley, T. H. Lilley, J. chem. Soc. Faraday Trans. I 74 (1978) 2771-2778.

[10] B. P. Kelley, T. H.Lilley, J.chem. Soc. Faraday Trans. I 74 (1978) 2779- 2785.

[11] I. N. Basumallick, R. Mohanty, Ind. J. chem.A 25 (1986) 1089-1091.

[12] T. Owaga, K. Mizutani,M. yasuda, Bull. Chem.Soc.Jpn.57 (1984) 2064-2068

[13] T. H. Lilley, E. Mosses, I. R. Tasker, J. Chem. Soc. Faraday Trans .I 76 (1980) 906-914.

[14] T. H. lilley, I. R. Tasker, J. Chem. Soc. Faraday Trans.I 78(1982) 1-6.

[15] M. Natarajan, R. K. wadi, H. C. Gaur, J. chem. Eng. Data.35 (1990) 87-93.

[16] R. Bhat, J. C. Ahluwalia, Int. J. peptide protein Res.30 (1987) 145-152.

[17] B. I. Kankia, Biophys. chem.84 (2000) 227-237.

[18] M. K. Khoshhkbarchi, J. H. Vera, Ind. Eng. chem. Res. 35 (1990) 87.

[19] M. Natarajan, R. K. wadi, H. C. Gaur, J. chem. Eng. Data 35 (1990) 87.

[20] T. S. Banipal, G. Singh, B. S. Lark, J. Solution Chem. 30, 2001,657-670.

[21] T. S. Banipal, G. Sehgal, Thermochem. Acta 262 , 1995 , 175-183.

[22] T. S. Banipal, D. Kaur, P. Lal, G. Singh. J. chem. Eng. Data 47, 2002,1391-1395.

[23] A. L. Surdo, E. M. Alzola, F. J. Millero. J. chem. Thermodyn. 14 ,1982 , 649-662.

[24] R. Bhat, J. C. Ahluwalia, J. Phys. chem. 89, 1985, 1099-1105.

[25] F. J. Millero, A. Surdo, C. Shin. J. Phys. Chem. 82 ,1978 , 784-792.

[26] M. Kikuchi, M. Sakurai, K. Nitta, J. Chem. Eng. Data 40 ,1995, 935-942.

[27] D. P. Kharakoz, D. P. Biophys. Chem. 34, 1989, 115-125.

[28] Z. Yan, J. Wang, Lu. W, J. thermochim. Acta 199, 334, 17-27 (29)

[29] J. Wang, Z. Yan, K. Zhuo, D. Liu, Z. Phys. Chem. 214 , 2000 , 333-345.

0

0.5

1

1.5

2

2.5

3

0 2 4 6n c

tV0 /c

m3 .m

ol-1

0

0.5

1

1.5

2

2.5

3

3.5

0 0.5 1 1.5 2 2.5

m s /mol.kg -1

ΔVt /c

m3 .m

ol-1

-3

-2

-1

0

1

2

3

4

m s /m ol. k g -1

ΔtV

0 ( N

H3+ , C

OO

- ,R)/c

m3 .m

ol-1

Effect of Diluents on the Thermal Behavior of Vinyl Ester Resins

Bharti Gaur Department of Applied Sciences and Humanities,

National Institute of Technology, Hamirpur, H.P. India E-mail: [email protected]

Abstract

Vinyl ester resins prepared from o-cresol epoxy novolac resin and methacrylic acid in the presence of

triphenyl phosphine as catalyst and hydroquinone as polymerization inhibitor. The vinyl ester resin having acid value of ~ 7 mg of KOH per gram of solid thus obtained was characterized by Fourier transform infra red spectroscopy. Four samples each were prepared at 30°C using styrene and methyl acrylate/ethyl acrylate/butyl acrylate in the weight ratios 40: 0, 30: 10, 20:20, 0: 40, respectively as reactive diluents. The curing behavior of these samples in the presence of benzoyl peroxide (2phr) as initiator was studied using DSC. The resin samples were cured at 90 ± 2°C for 2h. The decomposition behavior of all these samples was studied by thermogravimetric analysis (TGA) at 10°C min-1. It was observed that the addition of styrene to vinyl ester resin samples containing methyl, ethyl and butyl acrylates tend to decrease the reactivity of these samples during curing. It was also observed that the activation energy, calculated using Ozawa method for these samples increased on increasing the percentage of styrene in the formulations. The TG and DTG thermograms showed single-step degradation in the individual cases of styrene, methyl acrylate, ethyl acrylate, and butyl acrylate, whereas a two-step degradation process was observed when styrene was mixed with methyl acrylate, ethyl acrylate or butyl acrylate in any proportion. For all the samples, the order of the reaction was one for the first step. The value of activation energy (E) and the pre-exponential factor (Z) was calculated using Coats and Redfern equation. Vinyl ester resin sample containing ethyl acrylate gave a more thermally stable product than other acrylates as well as styrene, probably due to its more reactivity towards vinyl ester resin during cure. It was found that the cured samples showed decrease in the stability with the increase in concentration of styrene in the formulation. 1. Introduction

Vinyl ester (VE) resins are widely used thermosetting resins because of their low cost, excellent chemical and corrosion resistance, outstanding heat performance and favorable mechanical properties. These are the addition products of epoxy resins and α-β unsaturated carboxylic acids [1-2]. These resins incorporate excellent mechanical, chemical, and solvent resistance of epoxy resins and the processibility of unsaturated polyester resin. The aromatic rings provide good mechanical properties and heat resistance and the ether group attributes to good chemical resistance. Their advantage of being processable over a wide range of temperature is of vital importance for many end uses, such as in solvent storage tanks, sewer pipes, building and construction, coating, automobile structural parts, swimming pools, marine composites [3-7].

Vinyl ester resins in the neat or undiluted form vary from semisolid to solid. The viscosity of neat resin is considerably high (105cPs) and therefore utilizes reactive or nonreactive diluents to provide workable viscosity levels and enhanced reactivity [8-13]. These diluents also

control the crosslink density and affect strength, percent elongation, hardness, chemical resistance, scratch resistance and surface finish. Several investigators [11-15] have studied the thermal behavior of VE resins usually with styrene or substituted styrene as the reactive diluent. However, the effect of other diluents and their mixtures have not been emphasized much. In the present communication, the curing and decomposition behavior of VE resins in the presence of mixtures of styrene and methyl acrylate/ethyl acrylate/butyl acrylate as reactive diluent is being reported. 2. Experimental

O-Cresol (S.D. Fine Chemicals) and formaldehyde (37-41% solution, S.D.Fine Chemicals) were used for the preparation of novolacs. Epichlorohydrin L. R. grade (CDH), sodium hydroxide pellets (Merck), were used for the preparation of epoxy novolacs. Methacrylic acid (Merck), triphenylphosphine (Fluka AG), were used for the preparation of VE resins and methyl acrylate, ethyl acrylate, butyl acrylate (Merck) and styrene (Ranbaxy), were used as reactive diluents in the present study.

Effect of Diluents on the Thermal Behavior of Vinyl Ester Resins 163

Synthesis of o- cresol formaldehyde novolac O-Cresol and formaldehyde in the molar

ratio of 1:0.7 were used for the preparation of novolac resin. The pH was adjusted to 1.5 with conc. H2SO4/p-toluene sulphonic acid under a pH scan 3+ double junction pH meter. The reaction was carried out at 80º C with constant stirring and formaldehyde solution was added in a drop wise manner over a period of 3 hours. 40 ml of 10% sodium bicarbonate was added to neutralize the H2SO4 and arrest the reaction. The reaction mixture was washed with warm water to neutralize the pH by removing excess sodium bicarbonate and salt. The resin was finally dried at 80º C under reduced pressure (35 ± 5mm Hg). Synthesis of Epoxy novolac from o-cresol formaldehyde novolac resin

Epoxy novolac resin was prepared by reacting the above prepared novolac resin with epichlorohydrin (5 moles for every phenolic group of the novolac resin). Sodium hydroxide in the mole ratio with epichlorohydrin of 0.2:1.0 was used as a catalyst. Novolac resin along with epichlorohydrin were charged into the flask and heated under stirring to temperature of 112º C ± 1º C. While maintaining this temperature for epoxidation to proceed, sodium hydroxide (40% w/w) were added gradually to the reactants in the flask over a period of three and a half hours. After which the contents were dissolved in toluene and the solution filtered using Whatman filter paper no. 42 to remove the salts. Toluene was then removed by heating under reduced pressure. Synthesis of vinyl ester resin from o-cresol epoxy novolac resin

Vinyl ester resin was prepared using 1:0.9 mole ratios of o-cresol epoxy novolac resin prepared (epoxide equivalent weight: 258, determined by pyridinium chloride method [16]) and methacrylic acid in the presence of triphenyl phosphine (1 phr by weight of the epoxy resin) and hydroquinone (200 ppm) at 85º ± 1ºC. The esterification reaction was carried out for a three hours to obtain a product with an acid value of 20mg KOH/gm solids determined by the method given by of Ogg, Porter, and Willitis [17]. The VER samples were stored in a refrigerator at 10ºC. Structural Characterization

The structural characterization of o-cresol formaldehyde novolac, o-cresol epoxy novolac

resin, and vinyl ester resins based on o-cresol epoxy novolac resins was done using FTIR spectroscopy. The FTIR spectra of the samples were recorded by dissolving in chloroform/methanol and subsequent evaporation of solvent on KBr disc. Thermo Nicolet IR200 FT-IR Spectrometer was used for this purpose. Curing and decomposition behavior of vinyl ester resins

The samples for studying the curing and decomposition behavior were prepared using varied weight fraction of different reactive diluents by mixing the vinyl ester resin with diluent(s), and benzoyl peroxide (2phr) at 30°C until the mixtures became homogeneous in nature. Details of these samples have been given in Table-1.

DSC scans under dynamic conditions were obtained with program rates of 2, 5, 10, 15 and 20ºC min-1 from 40°C to the temperature at which the exothermic reactions were completed. From the DSC scans, the activation energy (E) with ±3 % accuracy was calculated by the Ozawa method [18]. Further refinements of the E values were carried out using a series of iterations until two successive values of E were almost identical. The frequency factor (Z) was calculated using Kissinger equation.

2

/1 )(min

RTEeZ

RTEβ=− ………………….(1)

where, E is the activation energy (cal/mol), T is the peak temperature (K), R is the gas constant, β is the program rate (deg min-1). Decomposition Behaviour

From dynamic DSC scans obtained at 5°C min-1, a curing temperature of 90°C was selected to obtain an appreciable rate of curing in all the samples. This temperature was used to cure the resin samples in an air-circulating oven (± 2°C) for subsequent studies on thermal stability and the kinetics of decomposition. Dynamic thermograms for the cured samples were obtained at the rate of 10°C min-1 from 50 to 700°C under a nitrogen atmosphere. The relative thermal stability of the resin was quantitatively estimated by comparing the temperature for a particular degree of weight loss. The values of activation energy (E) and pre-exponential factor (Z) were determined by the Coats and Redfern equation [19] as given below:


Table 1. Samples and their Corresponding Weight Fractions

Percent weight fraction Diluent VS40 VM40 VM20 VM10 VE40 VE20 VE10 VB40 VB20 VB10

Styrene 40 0 20 30 0 20 30 0 20 30 Methyl Acrylate 0 40 20 10 - - - - - -

Ethyl Acrylate 0 - - - 40 20 10 - - - Butyl Acrylate 0 - - - - - - 40 20 10

RTE

ERT

EZR

Tg

3.221log)(log 10210 −

⎭⎬⎫

⎩⎨⎧

⎥⎦⎤

⎢⎣⎡ −=⎥⎦

⎤⎢⎣⎡

βα ……...(2)

⎭⎬⎫

⎩⎨⎧

⎥⎦

⎤⎢⎣

⎡−−−

−=−

ng

n

1)1(1log)(

1

10αα for n ≠ 1…......(3)

( )αα −−= 1log)( 10g for n = 1 Equation 2 can be reduced to eq.4:

XBAY1000

+= ……………...……..……….. (4)

The parameters in eq. 4 are defined as follows for different values of n between 0 and 2.0:

⎥⎦⎤

⎢⎣⎡= 2

)(logT

gY α ………………………….… (5)

TX 1000= ………………...………………… (6)

⎭⎬⎫

⎩⎨⎧

⎥⎦⎤

⎢⎣⎡ −=

ERT

EZRA 21log10 β

……………….... (7)

REB3.2

−= …………...……………………… (8)

f

fT

WWWW

−

−=

0

α………………..…...……….. (9)

where α is the fractional weight loss, WT is the weight at temperature T, Wf is the final weight, W0 is the initial weight, and R is the gas constant. Best-Fit technique was used to calculate the value of order of reaction (n), which was further confirmed by regression analysis. A thermal analyzer (Thermal Analyzer 2000; TA Instruments), equipped with a Differential Scanning Calorimeter (DSC; 2910) was used to study the curing as well as decomposition behavior of these prepared resin samples. 3. Results and discussion Characterization of vinyl ester resin from o-cresol epoxy novolac resins

In the IR spectrum of vinyl ester resin the characteristic broad absorption band due to the secondary hydroxyl group was observed at 3423 cm-1 and the disappearance of the peak at

3001cm-1 due to the C-H stretching of the epoxy ring and the appearance of the peaks at 1715 and 1170 cm-1due to carbonyl groups in the methacrylate were observed. The absorption peaks at 1634 cm-1 and 946 cm-1 were due to the stretching and wagging vibrations, respectively, of the C=C in the methacrylate. Curing Behavior

The curing behaviour of vinyl ester resin varies with the structure and mixture of diluents added in different proportions. A typical DSC scan for the curing of vinyl ester resin containing styrene (40% w/w) at a program rate of 10°Cmin-1 is given in Figure 1. The onset exothermic temperature (Ti), peak temperature (Tmax), activation energy (E), and the frequency factor (Z) for various formulations of vinyl ester resin samples containing reactive diluents (40% w/w): styrene, methyl acrylate, ethyl acrylate, and butyl acrylate are given in the Table-2.

Fig. 1. Dynamic DSC for curing at 10°C/min of vinyl ester resin samples (a) VS 40

It is apparent from the table that the onset and peak temperatures of VE40 are lower than those of VM40, VB40 and VS40. DSC Scans of these samples obtained at 2°, 5°, 15°, and 20°C min-1 also show a similar trend. This result indicates that ethyl acrylate is comparatively more reactive than methyl acrylate, butyl acrylate as well as

Effect of Diluents on the Thermal Behavior of Vinyl Ester Resins 165

styrene during curing in the presence of free radical initiator benzoyl peroxide

Table 2. Curing behavior of vinyl ester resin sample containing styrene and methyl acrylate/ethyl acrylate/butyl acrylate at 10°C

The activation energy and frequency factor

for these samples given in the table also substantiate this fact. This behaviour is probably due to large difference in the polarities of acrylates. Ethyl acrylate being less polar reacts more readily with vinyl ester resin during curing. It is also apparent from these table that addition of styrene to vinyl ester resin samples containing methyl, ethyl and butyl acrylates tend to decrease the reactivity of these samples during curing. It can also be observed that the activation energy in the samples on increasing the percentage of styrene in the formulations also increased the activation energy. Thermogravimetric studies The thermogravimetric scans and dynamic thermal gravimetric (DTG) scans for the samples VS40, VM20, VM10 at a rate of 10°C min-1 from ambient temperatures to 700°C are shown in Figures 2-4.

Fig. 2. Dynamic TG scan of VS40 containing 40%w/w styrene at 10°C min-1

The thermograms of the samples VE40,

VB40 also showed single step degradation whereas VE20, VE10, VB20 and VB10 showed two-step decomposition pattern. Thermograms of all samples indicated weight loss of 1-4% between 50 to 250°C, which may be due to residual diluents, and byproducts of the curing reactions The relative thermal stabilities of all the samples were studied by comparing the temperatures of 1-10% weight loss.

Fig. 3. Dynamic TG scan of VM20 containing 20 %w/w styrene at 10°C min-1

On comparing the weight loss at a particular temperature it is observed that the thermal stability of VE40 containing ethyl acrylate was highest. The cured resin samples showed decrease in the stability with the increase in concentration of styrene in the formulation, which is in agreement to the results obtained from DSC studies. It was observed from DSC studies that

Sample design.

Ti (K)

Tmax (K)

Tf (K)

E (Kcal/mol)

Z min-1

VS40 373.05 392.85 408.50 20.163 -

VM40 356.90 377.50 400.00 20.180 5.268× 1011 VM20 359.45 378.00 405.45 21.970 2.926× 1012 VM10 361.50 379.75 407.00 22.627 8.910× 1012 VE40 340.00 372.00 412.00 16.50 2.963 × 109

VE20 342.20 374.50 418.00 17.79 1.705× 1010 VE10 346.50 375.50 422.00 19.63 1.852 × 1011 VB40 349.00 372.40 389.50 20.280 5.655 × 1011

VB20 352.65 371.50 392.45 21.197 1.482 × 1012 VB10 354.05 376.00 396.80 20.984 1.050 × 1013


the addition of styrene in the formulations affects the crosslink density of the cured product.

Fig. 4. Dynamic TG scan of VM10 containing 30%w/w styrene at 10°C min-1

The kinetic parameters of the decomposition

reaction of the samples were calculated using the method of Coats and Redfern. The values of activation energy and frequency factor for the decomposition reaction of the samples have been summarized in the Table-3. From the table it becomes apparent that vinyl ester resin sample containing ethyl acrylate gave a more thermally stable product than other acrylates as well as styrene, probably due to its more reactivity towards vinyl ester resin during cure. Table 3. Activation energy and frequency factor of decomposition of cured vinyl ester resin samples

S. No.

Sample designatio

n

Activation Energy (E) (kcal/mol)

Frequency factor (Z)

(min-1) 1 VS40 27.5579 1.0050×107 2 VM40 24.1800 9.9900×105 3 VM20 22.1088 - 4 VM10 22.2400 8.0780×105 5 VE40 32.3702 6.3196×108 6 VE20 29.9300 2.2263×108 7 VE10 25.1560 6.7419×106 8 VB40 25.0786 5.4183×106 9 VB20 23.9883 1.7365×106

10 VB10 23.9349 - 4. Conclusions

It was observed that the addition of styrene to vinyl ester resin samples containing methyl, ethyl and butyl acrylates tend to decrease the reactivity

of these samples during curing. The decomposition behavior of VE resin containing styrene, methyl acrylate/ethyl acrylate/ butyl acrylate and their mixtures showed single and two steps, respectively. Vinyl ester resin sample containing ethyl acrylate gave a more thermally stable product than other acrylates or styrene, due to its more reactivity towards vinyl ester resin during cure. References

[1.] Launikitis, M.B. Vinyl Ester Resins, In Handbook of Composites, Luben, G. Ed; Van Nostrand, Reinhold Co.: New York, 1982; 38-49.

[2.] Young, R.E. In Unsaturated Polyester Technology, Bruins, P.F. Ed.; Gordon and Breach: New York, 1976; 315-342.

[3.] Brown, J.R.; Mathys, Z. Composites Part A: Appl. Sci. Manu. 1997, 28, 675-681.

[4.] Lane, et.al. US Patent 6,187,442, 2001. [5.] Mouritz, A.P.; Mathys, Z. 1999, 47,

643-653. [6.] Love, J.L. US Patent 5,961,825, 2001. [7.] Zhang, S.; Ye, L.; Mai, Y-W. Appl.

Comp. Mater. 2000, 7, 125-138 [8.] Gaur, B; Rai, J.S.P. Eur.Polym. J.

1993, 29, 1149-1153. [9.] Varma, I.K; Rao, B.S.; Choudhry, M.S.;

Choudhry, V.; Varma, D.S. Die. Angew. Makromol. Chem 1985, 130, 191-199.

[10.] Choudhary, M. S., Varma, I. K. Die Angew. Makromol. Chem. 1993, 209: 33.

[11.] Gaur, B., Rai, J. S. P. Polymer, 1992, 33: 4210.

[12.] Malik, M., Choudhary, V., Varma, I. K. J. Appl. Polym. Sci. 2001, 82, 416-423.

[13.] Bhatnagar, R.; Varma, I. K. J. Therm. Anal., 1989, 35, 1241-1249.

[14.] Padma, A.; Rao, R.M.V.G.K.; Nagendrappa, G. J. Appl. Polym. Sci. 1997, 65, 1751.

[15.] Han, C.D; Lem, K.-W. Polym. Eng. Sci. 1984, 24, 175

[16.] Knoll, D.W.; Nelson, D.H.; Kehres, P.W. Am. Chem. Soc. Symp. 134th Meeting, 1958.

[17.] Ogg, C.L.; Porter, W.L.; Willitis, C.O. Ind. Eng. Chem. Anal. Edn. 1945, 17, 394.

[18.] Ozawa, T. J. Therm. Anal. 1976, 9, 217,369.

[19.] Coats, A. W.; Redfern, J. P. Nature, 1964, 68, 201.

Evaluation of Crystallinity of Graft Copolymers of Flax with Binary Vinyl Monomers

Susheel Kalia1*, B.S. Kaith2 and A.S. Singha3

1Deptt. of Chemistry, Singhania University, Pacheri Badi, Jhunjhunu – 333 515 (Rajasthan) India. 2Deptt. of Chemistry, National Institute of Technology (Deemed University), Jalandhar (Pb.) India

3Deptt. of Applied Sciences, National Institute of Technology (Deemed University), Hamirpur - 177 005. E-mail: [email protected], [email protected]

Abstract

In this paper, we graft copolymerized flax with binary vinyl monomer mixtures. Maximum grafting (86.02 %) has been found with MMA+EA binary mixture. Percentage crystallinity and crystallinity index of graft copolymers were measured with X-ray diffraction (XRD) technique. Crystallinity of flax decreases upon grafting with binary vinyl monomer mixtures. Flax showed the highest value of percentage crystallinity (76.96 %) and crystallinity index (0.7005) in comparison to Flax-g-copolymers.

1. Introduction

Since the possibility of production of newer monomers at low cost is very bleak, so the modification of a wide variety of existing synthetic and natural polymer through graft copolymerization technique for incorporating highly specific properties is of utmost importance. Improvement in dyeing, printing, chemical resistance, water repellency, fiber strength, abrasion resistance, crease resistance, impact strength of thermoplastics, improvement in properties of natural rubber and preparation of ion–exchange membranes are a few example of advantage of graft copolymerization. Different workers have studied the grafting of binary vinyl monomer mixtures on the back-bone polymer. It has been reported that the monomer reactivity ratios for the grafting process are completely different from the values observed for conventional solution polymerization, for example, binary mixtures of monomers: acrylonitrile/styrene and acrylamide/styrene [1-6].

Mechanical properties of polymers such as tensile strength, impact-strength and extensibility have a direct correlation with the percentage grafting (Pg). On grafting crystal lattice of the polymer is disrupted but the strength of the material may act to reinforce the structure [7, 8]. However, if crystallinity is not disturbed on grafting, then continuous increase in strength can be obtained with increase in Pg [9]. Most of the cellulosic fibers possess both crystalline and amorphous regions. The X-ray pattern of crystalline polymers show both sharp features associated with regions of three-dimensional order and more diffused features characteristics of molecularly disordered substances like liquid. The

occurrence of both types of features in the fibers indicates that ordered and disordered regions co-exist in crystalline polymers [10]. Lower crystallinity means higher amorphous regions, which are more accessible to chemicals and water. Crystallinity is also related to strength and generally, higher the crystallinity, higher is the strength of the fibers if the polymer structures are same [10, 11]. The present paper deals with grafting of flax with binary vinyl monomer mixtures. Percentage crystallinity and crystallinity index of synthesized flax-g-copolymers were measured using XRD technique.

2. Materials and Methods

Flax fiber (Linum usitatissimum) was obtained from the Department of Agronomy, CSK HP Agriculture University, Palampur (India). Monomers were washed with 5% sodium hydroxide followed by water and were dried over anhydrous sodium sulphate. The dried monomers were distilled and the middle fraction was used. Libror AEG-220 (Shimadzu make) electronic balance was used for weighing purpose. XRD studies of the samples were recorded with Bruker D8 advance diffractometer. 2. 1. Graft Copolymerization of binary vinyl monomer mixtures onto flax

Grafting of binary monomer mixtures onto flax was carried out as per method reported earlier [13].


2. 2. Percentage crystallinity and crystallinity index of graft copolymers

Percentage crystallinity (% Cr) and crystallinity index (C.I.) were calculated as per the methods reported earlier [24]. 3. Results and Discussion

C2, C3 and C6 hydroxyls and the C-H groups are the active cites for grafting in cellulosic fibers [12-15]. The grafting onto flax fibers in presence of FAS-H2O2 (Fenton’s reagent) takes place as per the mechanism proposed by Bhattacharya and Misra [16]. In case of graft copolymerization of MMA onto Flax fiber, optimum reaction conditions for getting maximum graft yield (41.74%) were : MMA (mol L-

1) = 1.96×10-3, FAS-H2O2 (molar ratio) = 1:6, temperature (oC ) = 55, time (minutes) = 120, pH = 7.0, respectively. 3.1 Effect of Concentrations of Binary Vinyl Monomer Mixtures on Percent Grafting

It is evident from Table 1 that in case of binary mixtures consisting of MMA (methyl methacrylate) with ethyl acrylate (EA), acrylonitrile (AN), vinyl acetate (VA) and acrylamide (AAm), higher percentage grafting has been found as compared to MMA alone (41.74%). The higher percentage grafting in case of these binary monomer mixtures can be explained by the fact that addition of electron acceptor monomers (EA, AN, and AAm) and electron donor monomer (VA) to MMA increase the reactivity of MMA towards grafting. The reactivity ratio values of these monomer mixtures [(MMA + AN: r1 = 1.09, r2 = 0.15); (MMA + VA: r1 = 2.75, r2 = 0.1) and (MMA + AAm: r1 = 2.53, r2 = 0.82)] indicates that r2 value of AN, VA and AAm is very small, which indicates that these monomers react with MMA in preference to their monomeric units thereby producing more of growing copolymeric chains and hence higher percentage of grafting. Whereas, higher reactivity ratio of MMA (r1) indicates that MMA radicals react with its own monomer thereby producing more homopolymer. This might be the reason that MMA produces lower percentage grafting when used alone in comparison to binary mixture [17-21]. The low graft yield with MMA/acrylic acid (AA) was due to the fact that AA is more associated with water thereby resulting in decreased free radical sites and hence resulted a low graft yield. In case of MMA/styrene (Sty) binary mixture, two monomers with electron accepting and electron donating ability enter into a charge transfer complex formation thereby reducing the activity of monomers towards

grafting. Moreover, both MMA and Sty are insoluble in water and hence accessibility of growing radicals to the active sites is poor. In addition to above factors, styrene also undergoes resonance stabilization, thereby resulting in low free radical sites on the monomeric units and hence a decreased graft yield has been observed [22, 23]. 3.2 Percentage crystallinity and crystallinity index of graft copolymers

It has been observed that percentage crystallinity and crystallinity index decreases with grafting (Table 2). The X-ray spectrum of Flax fiber is more convex than that of graft copolymers (Fig. 1). In case of Flax fiber, the incorporation of monomer chains to the back-bone of Flax had impaired the crystallinity of Flax fiber [24]. Therefore, on grafting percentage crystallinity decreases rapidly with reduction in its stiffness and hardness [25-27].

Table 1. Effect of conc. of different binary monomer mixtures on Pg

Monomers Concentrations (x 10-3 molL-1) Pg

MMA + EA 1.96 + 1.84 86.02

MMA + AN 1.96 + 3.79 68.66

MMA + AA 1.96 + 3.64 35.26

MMA + VA 1.96 + 2.70 71.44

MMA + AAm 1.96 + 1.40 68.04

MMA + Sty 1.96 + 1.30 37.48

Crystallinity index (C.I.) gives a quantitative measure of the orientation of the cellulose crystals in fibers. A lower crystallinity index in case of graft copolymers means poor order of cellulose crystals in the fiber. This is due to misorientation of the cellulose crystals to the fiber axis during grafting as indicated by the lower crystallinity index of graft copolymers. This clearly indicates that the cellulose crystals are better oriented in Flax fiber followed by Flax-g-copolymers.

Evaluation of Crystallinity of Graft Copolymers of Flax with Binary Vinyl Monomers 169

O p eratio ns : S mo o th 0 .1 2 9 | Sm oo th 0 .03 2 | B ack gro u nd 0.0 14 ,1.0 00 | Im po rtRF - F ile : R F.ra w - T ype : 2 Th /Th lo cke d - S ta rt: 5 .0 00 ° - En d : 5 0.0 0 0 ° - S tep : 0.0 20 ° - Ste p tim e : 1 .2 O p eratio ns : S mo o th 0 .1 7 7 | Sm oo th 0 .03 2 | B ack gro u nd 0.0 14 ,1.0 00 | Im po rtM MA -V A = IA - F i le: M MA -V A= IA .ra w - Type : 2Th /Th locke d - Sta rt: 5.0 00 ° - E nd : 50 .00 0 ° - S tep : 0.0 2O p eratio ns : S mo o th 0 .1 4 1 | Sm oo th 0 .14 1 | B ack gro u nd 0.0 14 ,1.0 00 | Im po rtM MA -S = IA - F ile : M M A -S = IA .r aw - Typ e: 2 Th/Th lo cke d - S tar t: 5 .00 0 ° - E n d: 5 0.0 00 ° - S te p : 0 .05 0 °

O pe ra tion s: Sm o oth 0 .14 1 | S mo oth 0.0 20 | S m oo th 0 .10 5 | B ackg ro un d 0 .0 14 ,1.0 00 | Im po rtM M A- EA = IA - F ile : MMA -EA = IA.r aw - Typ e: 2 Th/Th lo cked - S tar t: 5 .00 0 ° - E n d: 5 0.0 00 ° - Ste p: 0 .02O pe ra tion s: Sm o oth 0 .14 1 | S mo oth 0.1 41 | B a ckg rou n d 0 .01 4,1 .00 0 | Im p or tM M A- AN =IA - F il e: MM A -A N= IA.raw - Typ e: 2 Th/Th lo cked - S tar t: 5 .00 0 ° - E n d: 5 0.0 00 ° - Ste p: 0 .05O pe ra tion s: Sm o oth 0 .12 9 | S mo oth 0.0 32 | B a ckg rou n d 0 .01 4,1 .00 0 | Im p or tM M A- AA m =IA - F il e: MM A -A Am =IA.r aw - Typ e: 2 Th/Th loc ked - S tart: 5 .00 0 ° - E n d: 50 .0 00 ° - Ste p: 0O pe ra tion s: Sm o oth 0 .10 5 | S mo oth 0.0 32 | B a ckg rou n d 0 .01 4,1 .00 0 | Im p or tM M A- AA = IA - F ile : MMA -AA = IA.r aw - Typ e: 2 Th/Th lo cked - S tar t: 5 .00 0 ° - E n d: 5 0.0 00 ° - Ste p: 0 .02

Lin

(Cou

nts)

010 020 030 040 050 060 070 080 090 0

10 0011 0012 0013 0014 0015 0016 0017 0018 0019 0020 0021 0022 0023 0024 0025 0026 0027 0028 0029 0030 0031 0032 0033 0034 0035 00

2-The ta - S cale6 1 0 20 30

Fig. 1. X-ray diffraction studies of Flax-g- Copolymers vis-à-vis flax fiber. Table 2. Percentage crystallinity and crystallinity index of flax and graft copolymers.

Sample Pg %Cr C.I.

Flax fiber - 76.96 0.7005 Flax-g-

poly(MMA+EA) 86.02 60.31 0.3419

Flax-g-poly(MMA+AN) 68.66 62.64 0.4036

Flax-g-poly(MMA+AA) 35.26 64.10 0.4400

Flax-g-poly(MMA+VA) 71.44 62.51 0.4002

Flax-g-poly(MMA+AAm) 68.04 62.87 0.4094

Flax-g-poly(MMA+Sty) 37.48 65.81 0.4804

4. Conclusion

Graft copolymerization of vinyl monomers onto flax has decreases the crystallinity of flax fiber. Flax fiber showed the highest value of percentage crystallinity and crystallinity index in comparison to Flax-g-copolymers. References

1. Naguib, H. F.; Aly, R.O.; Sabaa, M.W.; Mokhtar, S.M.; Polym. Test. 2003, 22, 825.

2. Fineman, M.; Ross, S.D.; J. Polym. Sci. 1950, 5, 259.

3. Kelen, T.;Tudos, F.; J. Macromol. Sci. Chem. A 1975, 9, 1.

4. Behnken, D.W.; J. Polym. Sci. A 1964, 2, 645.

5. Tidwell, P.W.; Mortimer, G.; J. Polym. Sci. A 1965, 3, 369.

6. El-Naggar, A.M.; Zhody, M.H.; Sahar, S.M.; Allan, E.A.; Polym. Int. 2001, 50, 1082.

7. Patterson, G.S.; Hoffmann, A.S.; Merill, E.W.; J. Appl. Polym. Sci. 1960, 4, 159.

8. Sella, C.; Chapiro, A.; Matsumoto, A.; J. Polym. Sci. 1962, 57, 529.

9. Chapiro, A.; J. Polym. Sci. 1957, 23, 377.

10. Ishikawa, S.; J. Polym. Sci. Part B: Polym. Lett. 1965, 3, 959.

11. Tsukada, M.; J. Appl. Polym. Sci. 1988, 35, 965.

12. kaith, B.S.; Singha, A.S.; Kumar, S.; Kalia, S.; Int. J. Polym. Mater. In Press.

13. Kaith, B.S.; Kalia, S.; Polymer Composites 2007, In Press.

14. kaith, B.S.; Singha, A.S.; Kalia, S.; Autex Res. J. 2007, 7, 119.

15. Kaith, B.S.; Kalia, S.; Polymer Journal 2007, In Press.

16. Bhattacharya, A.; Misra, B.N.; Prog. Polym. Sci. 2004, 29:767.

17. Florja czyk, Z.; Krawiec, W.; Die Makromol. Chem. 1988, 189, 53.

18. Misra, B.N.; Sood, D.S.; Sharma, R.K.; Angew. Makromol. Chem. 1982, 102, 59.

19. Ng, L.T.; Swami, S.; Polym. Int. 2006, 55, 535.

20. Barsbay, M.; Can, H.K.; Rzaev, Z.M.O.; Guner, A.; Polym. Bull. 2005, 53, 305.

21. Brandrup, J.; Immergut, E.H.; Editors, Polymer Handbook, Second Edition, Wiley Interscience, NY, pp. 105-386 (1975).

22. Kaith, B.S.; Singha, A.S.; Kumar, S.; Int. J. Chem. Sci. 2006, 4, 195.

23. Kaplan, C.H.; Zakir, R.; Ali, G.; J. Mol. Liq. 2004, 111, 77.

24. Kaith, B.S.; Kalia, S.; Int. J. Polym. Analys. Charact. 2007, 12, 401.

25. Liu, Y.; Zhang, R.; Zhang, J.; Zhou, W.; Li, S.; Iran. Polym. J. 2006, 15, 935.

26. Billmeyer, F.W. Jr.; “Textbook of Polymer Sciences”, 3rd ed. Wiley, New York, 1984, p. 347.

27. Kaith, B.S.; Singha, A.S.; Chauhan, A.; Misra, B.N.; J. Polym. Mater. 2006, 23, 349.

Synthesis and Structural Studies of N-(2-oxo-3-oxa-4(phenyl)-Butanyl Benzene Sulphonamide and Related Compound as Potential Juvenile

Hormone Analogue

Pamita Awasthi, Shilpa Dogra and R.K Mahajan*

Department of Applied Sciences and Humanities, National Institute of Technology, Hamirpur – 177 005. *Department of Chemistry, Himachal Pradesh University, Shimla 171 005 (HP)

[email protected]

Abstract

Synthetic Juvenile H and its analogues is new and emerging area to counter the insect problem. The chemostirlization method to control the vast insect population is more effectual than the lethal action of classical insecticides. In this paper we report the synthesis of N-(2-oxo-3-oxa-4(phenyl)-butanyl) benzene sulphonamide and its related compounds as a potential pesticides by incorporating sulphonamide feature. In depth structural study on some of the selected compound has been done using 1D NMR and mass spectroscopic tools. 1. Introduction

Insects have tendency to multiply rapidly thereby controlling the population of insect itself a problem. The regular and continues use of classical insecticides has made a drastic impact on environment as well as on warm blooded animals. Since many years hormones are the subject of grate importance as an alternative and safer method to control pest/ insect populations. Growth and development of the insects are controlled by the physiological changes in their tissues and which are further regulated by three hormones- Brain Hormone, Moulting hormone and Juvenile hormone. Endocrine system of insect provides these hormones which control their reproductive, moulting, metamorphism and development [1-4].

Juvenile Hormone (JH) is the main hormone able to regulate all aspect of insect life before the synthetic juvenoides were known. It was an assumption that insects could hardly develop resistance against the JH. And as a matter of fact, JH activity is absolutely essential to any insect in order to develop into reproducible adults. Now numerous synthetic juvenile hormone analogues (JHA) with JH activity are known which posses different structural features and different properties from the natural JH. It has been suggested that insect might have developed resistance in contact with naturally occurring juvenoids during millions of years of insect evolution. Therefore tremendous interest has been generated to synthesize newer and newer compounds with different structural features and study the effect of such changes on insect

hormonal activity. This will prove to be very useful in evolving an alternative to the use of undesirable insecticides.

Large number of synthetic JH cyclic or acyclic are reported in literature along with descriptive bioassay. And it has been seen that small change in structural feature promotes the significant changes in its biological activity [4-5]. Although various structural variation in the compound of the type (Fig. 1)

R

R= Aromatic or heterocyclic ring

Fig. 1. Structure of JHA have been reported in the literature, not much work seems to have been reported in the direction of incorporating hetero atom or functional groups in the side change. In our laboratory, we have synthesized large number of juvenoids of the type (Fig.1) incorporating different hetero atoms like nitrogen, oxygen and sulphure and various structural features like amides, esters, oximes, hydroxamates and carbonates in the side chain [6-11] and some has shown excellent JH activities. In continuation of this work we have synthesized large number of JHA containing sulphonamide functional group [12-13] in the side chain and also incorporated the aromatic ring at the terminal (Fig. 2). We have also carried out the detailed structural studies using NMR and mass spectroscopic tools on one of the representative in order to confirm the structure.

Structural Studies of N-(2-oxo-3-oxa-4(phenyl)-Butanyl Benzene Sulphonamide 171

R

R'

R-Aromatic Ring; R’- Aromatic or Heterocyclic ring

Fig. 2. Structure of JHA

Experimental Benzene sulphonyl glycine (3) A mixture of glycine (7.5g; 0.1mol ), bezenesulphonyl chloride (1; 17.6g; 0.1mol) and NaOH solution (1N, 200 ml) was stirred together at 65-70 for two hours. A clear solution was obtained. The reaction mixture was cooled to 5 to 10° and treated with conc. HCl to make it slightly acidic (pH 6.5). Where by benzenesulphonyl glycine separated out as white crystals. It was recrystallized from water to give pure benzenesulphonyl glycine (3; 12.5g; 69%), m.p 167-169° Toluene- p-sulphonyl glycine (4) The reaction of glycine (7.5g; 0.1mol) with toluene p-sulphonyl chloride (2 : 19.09g; 0.1mol) as above furnished toluene -p-sulphonyl glycine. It was also recrystallized from water to give pure toluene-p-sulphonyl glycine (4; 3.6g; 71.5%), m.p 147-149° ,literature m.p 149-150°. Benzenesulphonyl glycine acid chlorides (5) Benzenesulphony glycine (3; 2.33g; 0.011mol) was dissolved in dry benzene and excess of freshly distilled thionyl chloride (5.0) was slowly added to it stirred at room temperature for 20 minutes. The reaction mixture was gently refluxed for three hours. The solvent and excess of thionyl chloride were then distilled off under reduced pressure to give acid chloride of benzenesulphonyl glycine (5). It was used for next reaction without further purification. Toluene- p-sulphonyl glycine acid chloride (6) : Toluene-p-sulphonyl glycine (4; 2.52g; 0.011 mol) was treated with freshly distilled excess of thionyl chloride (5.0g) as above to give acid chloride of toluene-p-sulphonyl glycine (6) . It was also used for next reation without further purification. KBH4 Reduction of acetophenone 1-( p-chloro phenyl) ethanol (7) A suspension of KBH4 (1.0g) in 3ml of 1N NaOH was added to a solution of p-chloro acetophenone (4.6g; 30mol) in 20ml of methanol. The reaction

mixture was refluxed for 4 hrs, cooled and acidified with 50% HCl. Water was then added until separation of phase began and the mixture was extracted with ether. The extract washed with 0.1 N NaOH solution followed by water and dried over anhydrous Na2SO4 . The solvent was removed and residue was distilled to give 1-( p-chloro phenyl) – ethanol (3.2g; 70%) b.p 110-115°c/ 12mm, literature b.p 140-142°c/ 31mm 1-( p-nitro phenyl) ethanol (8) : p- nitroacetophenone (4.95g; 30 mm) was taken in 20 ml methanol and to this was added a suspension of KBH4 (1.0g) and worked up as above to given 1-(p-nitro phenyl) ethanol (8; 2.8g; 56%) b.p 245-250/3mm. NaBH4 Reduction of acetophenone 1-(p-chloro phenyl) ethanol (7) A suspension of NaBH4 (1.0g) was added to a solution of p-chloro acetophenone (4.6g; 30mm) in methanol (20ml) . The reaction mixture was refluxed for 4 hrs. It was then worked up as above to give 1-(p-chloro phenyl) ethanol (7; 65%), b.p 110-115°c / 12mm identified with (7) prepared above by KBH4 reduction. 1-(p-nitrophenyl) ethanol (8; 245-250°c /3mm) was also obtained by NaBH4 reduction of p-nitroacetophenone, which was identified with (8) prepared above by KBH4 reduction. N-(2-oxo-3-oxa-4-methyl-4-(p-chlorophenyl)-butanyl) benzenesulphonamide (9) The acid chloride of benzesulphonyl glycine (5; 1.2g; 0.005ml) freshly prepared from benzene sulphonyl glycine was taken in dry benzene sulphonyl glycine was taken in dry benzene (10ml). It was constantly stirred while 1- (p-chorophenyl) ethanol (7; 0.799g; 0.005 mol) in 5ml of dry benzene was dropwise added to it. The reaction mixture was stirred for three hours at 35-40°c. It was further stirred at room temperature for 12 hours and kept stand over night. It was then poured over crushed ice and extracted with chloroform. The solvent was distilled off under reduced pressure leaving behind a solid residue which was recrystallized from chloroform to give N-(2-oxo-3-oxa-4-methyl-4-(p-chlorophenyl) –butanyl) benzenesulphonamide( 9; 0.900g; 49.72% ), m.p 53-55°c; Rf 0.84 (system ‘a’) . Anal. Found : C, 54.48; H, 4.45; N, 3.98% C16H16O4NSCl requires : C, 54.31; H, 4.52; N, 3.96% N-(2-oxo-3-oxa-4-methyl-4-(p-nitrophenyl)- butanyl) benzenesulphonamide (10) The acid chloride of benzenesulphonyl glycine (5; 1.2g, 0.005 mol) freshly prepared from benzenesulphonyl glycine was taken in dry benzene (10ml) . It was constantly stirred while 1- (p-nitrophenyl) ethanol (8; 0.855g; 0.005 mol) in 5ml of dry benzene was dropwise added to it. The reaction mixture was stirred for three hours at 35-40°C . It was further stirred at room temperature for 12 hours and kept standing over night. The reaction was worked up as above to give N-(2-oxo-3-oxa-4-methyl-4-(p-nitro phenyl)-butanyl) benzenesulphonamide (10; 0.800g; 42.78%), m.p 95-97°C; Rf 0.45 (system ‘a’) Anal. Found : C, 52.70; N, 4.31; N, 7.7% C16H16O6N2S requires : C, 52.74; H, 4.39; N, 7.69%


N-(2-oxo-3-oxa-4-(phenyl)-butanyl) benzenesulphonamide (11) To the acid chloride of benzenesulphonyl glycine (5; 1.2g; 0.005 mol) freshly prepared from benzene sulphonyl glycine was taken in dry benzene (10 ml) . It was constantly stirred while benzyl alcohol (d; 0.555g; 0.005 mol) in 5ml of dry benzene was drop wise added to it . The reaction mixture was further stirred at room temperature for 12 hrs and kept standing over night . It was then poured over crushed ice and extract with ether . The solvent was distilled off leaving behind a solid residue which was recrystallized from chloro form to give N-(2-oxo-3-oxa-4-(phenyl)-butanyl) benzenesulphonamide (11; 0.650g; 41.66%), m.p 41-43°C; Rf 0.50 (system ‘a’) Anal. Found : C, 59.15; H, 4.98; N, 4.35% C15H15O4NS requires : C, 59.01; H, 4.91; N, 4.59% N- (2-oxo-3-oxa -4-(phenyl)- butanyl) p- toluene sulphonamide (12) : The reaction of acid chloride of toluene – p-sulphonyl glycine (6; 1.2g; 0.004 mol) freshly prepared from toluene –p-sulphonyl glycine with benzyl alcohol (d, 0.523g; 0.004 mol) in dry benzene as above gave N-(2-oxo-3-oxa-4-(phenyl)-butanyl) p-toluene sulphonamide (12) m.p 62-65°C; Rf 0.65 (system ‘b’) Anal. Found : C, 60.09; H, 5.25; N, 4.31% C16H17O4NS requires : C, 60.18; H, 5.39; N, 4.38% N- (2-oxo-3-oxa-4-(p-nitro phenyl) – butanyl) benzenesulphonanide (13) : The reaction of acid chloride of benzene sulphonamide glycine (5; 1.2g; 0.005 mol) freshly prepared from benzenesulphonyl glycine with p-nitro benzyl alcohol (d; 0.786g; 0.005 mol) in dry benzene as above gave N-(2-oxo-3-oxa-4-(p-nitrophenyl)-butanyl) benzenesulphonamide (13), m.p 99-101°C; Rf 0.62 (system ‘a’) Anal . found : C, 51.49; H, 4.05; N, 8.05 % C15H14O6N2S requires : C, 51.42; H, 4.0; N, 8.0 % N- (2-oxo-3-oxa -4-(p-nitro phenyl) butanyl) p- toluene sulphonamide (14): The reaction of acid chloride of toluene-p-sulphonyl glycine (6; 1.2g; 0.004 mol) freshly prepared from toluene-p-sulphonyl glycine with p- nitro behzyl alcohol (d, 0.741g; 0.004 mol) in dry benzene as above gave N- (2-oxo-3-oxa-4-(p-nitro phenyl) butanyl) p- toluene sulphonamide (14), m.p 111-113°C; Rf 0.61 (system’b’) Anal. Found : C, 52.85; H, 4.30; N, 7.55 % C16H16O6N2S requires: C, 52.74; H, 4.39; N, 7.69% 2. Results and discussion Synthesis of N-(2-oxo-3-oxa-4-methyl-4-(p-substituted phenyl)-butyl benzene sulphonamides (9-10)

The synthesis of N- (2-oxo-3-oxa-4-methyl-

4-(p-chlorophenyl)- butyl) benzene

sulphonamides (9-10) were accomplished along

the following lines (Scheme-1)

R

SO2- Cl

2HNOH

O

a, b

S

O

ONH

OH

O

R(1- 2) (3.4)

i)

S

R

O

ONH

Cl

O(5-6)

R'

O

R'

OH

d

(7-8)

ii)

iii)

S

R

O

ONH

Cl

O

(5-6)

OH

R'

(7-8)

R

S

O

O

NH

O

O

R'

(9-10)

C

Scheme-1 R = H; R’ = Cl, NO2 Reagents : (a) NaOH; (b) HCl; (c)

SOCl2; (d) NaBH4 or KBH4

The reaction of benzene sulphonyl chloride (1) and p-toluene sulphonyl chloride (2) with glycine under alkaline condition gave benzene sulphonyl glycine (3) and p- toluene sulphonyl glycine (4) respectively. The treatment of these acids with thionyl chloride gave acid chlorides (5-6). 1- (p-substituted phenyl) ethanols, (7-8) were obtained by potassium borohydride reduction of the corresponding acetophenones. Reaction of acid chlorides of benzene sulphonyl glycine and p- toluene sulphonyl glycine (5-6) with 1- (p- substituted phenyl) ethanol (7-8) in dry benzene furnished the desired products. The PMR studies of above compounds (9-10) are discussed below PMR Studies The PMR spectra of the above compounds (9-10) were recorded in CCl4 on Jeol PMS 60 SI, 60 MHz spectrometer with TMS as internal standard and data are presented in Table (1).


R

S

O

ONH

H2C O

CH

CH3A

B

R'

'c''a'

'b'R = H ; R' = Cl

O

Fig. 3. The position of protons a, b, c in side hain and protons in ring A & B based on δ values in PMR spectra. In the PMR spectrum of N- (2-oxo-3-oxa-4-methyl-4-(p-chlorophenyl)–butyl) benzene sulphonamide (9), five protons of ring ‘A’ were observed around 7.43-8.1 δ as a complex. The four aromatic protons of ring ‘B’ appeared as a singlet at 7.1 δ (Fig. 3) The benzylic protons ‘a’ appeared as a multiplet at 5.66δ, while adjoining three methyl

protons ‘b’ were observed as a doublet at 1.44 δ. The singlet at 3.8δ can be attributed to methylene protons ‘c’ which are flanked by –O-CO-group on one side and -SO2NH group the other side. The compound (10) in which ring ’B’ contains a p-nitro group, 4-aromatic protons were split up into two doublets can seen at 8.2 δ and 7.4 δ . Mass spectral studies The structures of above compounds (9-10) were further supported by the mass spectrum of one representative (9) in above series which was recorded on Varian Mat CH-7 mass spectrophotometer. The most characteristic peaks along with their relative abundance observed for this compound are given in table 2.

Table 1. PMR data of N-(2-oxo-3-oxa-4-methyl-4-(p-chlorophenyl)- butyl) benzene sulphonamide and related compound- S.No Name of compound Compound

No. PMR data (δ value)

1 N-(2-oxo-3-oxa-4-methyl-4-(p-chlorophenyl)-butyl) bezene sulphonamide

9 8.1-7.43(complex, 5H, Ar-H, ring A ); 7.37( S, 4H, Ar-H, ring B ); 5.66(m, 1H, benzylic); 3.8 (S, 2H, -NH-CH2-CO-group); 1.44 (d, 3H,Ar-CH(CH3)-)

2 N-(2-oxo-3-oxa-4-methyl-4-(p-nitro phenyl)-butyl) benzene sulphonamide

10 8.2 (d, 2H, Ar-H, ortho to –NO2 group ); 8-7.3(comlex, 7H,(2H meta to –NO2 group ring ‘B’ ,5H,Ar-H, ring ‘A’); 5.9(m, 1H, benzylic); 3.9(S, 2H,-NH-CH2-CO-group); 1.53(d, 3H, Ar-CH(CH3)-group)

Table 2. Location of peaks and relative abundance of mass spectrum of compound (9)

S.n

o.

Ion Relative

abundance

(M/Z)

1

77 / 66.6

2

Cl

CH+

CH3

139 / 100.0

3

Cl

CH

CH3

O+

155 / 7.9

4

Cl

CH2O CH2

+

CH3 O

198 / 6.0

5 S

O

O

144 / 0.8

6 S NH

O

O

156 / 8.6

7 Cl

111 / 4.5

8 CH3- CH2- O- C - CH2

O

87 / 0.3


9 S

O

O

CH2

170 / 53.5

10

Cl

CH2O

C

OCH3

183 / 0.3

11 C- CH2 - NH - S

O O

0

198 / 6.0

12 CH3 - CH2 - O - C - CH2 - NH - S -

O

O

O

166 / 0.8

13 CH3 - CH2 - O - C - CH2 - NH

O

102 / 3.2

14 CH3 - CH2 - O -C - CH2 -NH2

O

103 / 29.8

15 O CH2-NH

O

73 / 0.8

16 O CH2-NH2

O

74 / 2.8

17

HO CH2 - NH2

O

75 / 6.8

Synthesis of N- (2-oxo-3-oxa-4-(phenyl)- butanyl) benzenesulphonamide and related compounds (11- 14) Synthesis of N – (2 –oxo- 3- oxa-4- (phenyl)-butanyl) benzene sulphonamide (4) and related compounds (12-14) were accomplished along the following lines (Scheme-2) The acid chloride of benzene sulphonyl glycine (5) and p- toluenesulphonyl glycine (6) prepared above were treated with benzyl alcohol at room temperature for 12 hours to give the desired products N- (2-oxo-3-oxa-4-(phenyl)- butanyl) benzene sulphonamide(11) and N- (2-oxo-3-oxa-4-(phenyl)-butanyl) p- toluene sulphonamide (12)

The similar reactions of these acid chlorides with p-nitrobenzyl alcohol gave the desired products N-(2-oxo-3-oxa-4-(p-nitro-phenyl)-butanyl benzene sulphonamide (13) and N-(2-oxo-3-oxa-4-(p-phenyl)-butanyl)p-toluene sulphonamide (14) respectively in good yields. PMR Studies The PMR spectra of above compounds (11-14) were recorded in trifluroaceteic acid on pms 60 SI, 60 MHz spectrophotometer with TMS as internal standard and data is presented in Table 3.

Table 3. PMR Data of N-(2-oxo-3-oxa-4-(p-phenyl)-butanyl) sulphonamide and related compounds

SN

Name of the compound No. PMR data (δ value)

1 N-(2-oxo-3-oxa-4-(phenyl)-butanyl) benzene sulphonamide

11 7.53-6.8 (complex, 10H, Ar-H from ring A & B); 5.23(S, 2H, O-CH2-Ar); 4.4 (S, 2H, -NH-CH2-CO-group)

2 N-(2-oxo-3-oxa-4-(phenyl)-butanyl) p- toluene sulphonamide

12 7.61(d, 2H .Ar-H,ortho to –SO2 –NH group, ring ‘A’); 7.36-6.94(complex, 7H, 2H, Ar-H, meta –SO2-NH group, 5H, Ar-H, ring ‘B’); 5.06 (S, 2H, -CO-O-CH2-Ar) 3.83 (S, 2H, -NH-CH2-CO- group); 2.41(S, 3H, Ar-CH3)

3 N-(2-oxo-3-oxa-4-(p-nitro phenyl)-butanyl) benzene sulphonamide

13 8.13(d, 2H, Ar-H, ortho to –NO2 group, ring ‘B’); 7.26-7.93(complex, 7H, 2H, meta to –NO2 group ring ‘B’, 5H, Ar-H, ring’A’); 5.17(S, 2H,-CO-O-CH2-Ar group); 3.9( S, 2H, -NH-CH2-CO-group)

4 N-(2-oxo-3-oxa-4-(p-nitrophenyl)butanyl) p- toluene sulphonamide

14 8.1(d, 2H, ortho to p-NO2 group, ring’B’); 7.13-7.9(complex, 6H, 2H, meta to –NO2 group, ring’B’, 4H, Ar-H, ring’A’); 5.23(S, 2H,-CO-O-CH2-Ar group ); 3.9 (S, 2H, -NH-CH2-CO-group); 2.44( S,3H, Ar-CH3)


S

R

O

ONH

CH2 OCH2

O

R'

A

B

(11)

R = H , CH3

R' = H , NO2

ab

Fig. 4. The position of protons a, b in side chain and protons in ring A & B based on δ values in PMR spectra. In the PMR spectra of N- (2-oxo-3-oxa-4-(phenyl) – butanyl) benzene sulphonamide, ten protons from ring ‘A’ as well as from ring ‘B’ appeared as complex around 7.53-6.8 δ (Fig. 4) The two distinct singlet one at 5.23 δ and other 4.4 δ can attributed to protons ‘b’ and ‘a’ respectively. p- nitro group in compounds (13-14) showed a strong deshielding effect on aromatic protons of ring ‘B’, by which two protons which are ortho to –NO2 group appeared at 8.13 δ while two protons which are meta to –NO2 group will merge into the complex. Signal of ring ‘A’ protons, around 7.26 – 7.93 δ . The compound (12-14) showed an additional signal at 2.35 δ.

R

SO

Ocl OH

O

a , b

R

SO

ONH

OH

OH2N

S

R

O

ONH

Cl

O

(1- 2)

(5-6)

d

R

SO

ONH

O

O

R"

R = H , CH3 ;

R" = H, NO2 :Reagents : (a) NaoH ; (b) HCl ; (c) SOCl2 ; (d) OH

(3-4)

R''

(11-14)

+

S

Scheme- 2

Acknowledgment

Authors are grateful to Prof. Lalit Awasthi, Head of Computer Science & Engineering Department National Institute of Technology for

providing us with Computational facilities at computer graphic laboratory of the Computer Center in framing out this manuscript.

References

1 Altstein M, Aharorison N & Menn J J, Arch. Insect Biochem Physiol, 22, 1993, 5.

2 Stall G B, Insect control with insect growth regulators based on insect hormones, In: Marini-bett’olo, G B [Ed] Natural Products and the Protection of Plants, Pontific Academia Scientiarum, roma, Italy, 217, 1977, 253.

3 Williams C M, Third Generation Pesticides, 217, 1967, 12.

4 Terry Paul H, Berkrrec A B, J Agr Food Chem, 21, 1973, 3.

5 Yu Wu, Parthasarthy. R, Hua Bai, Palli R. Subba, Mechanism of Development Elsevier 123, 2006, 530-547.

6 Mahajan R K, Gupta Neelam & Uppal Satinder K, Coll Czech Chem Commun, 50, 1985, 690.

7 Mahajan R K, Gupta Neelam & Uppal Satinder K, Coll Czech Chem Commun, 51, 1986, 2879.

8 Mahajan R K, Anand Shalu, Proc Nat Acad Sci India, 68(A) 1998, 1.

9 Mahajan R K, Pal Yogender & Anand Shalu, Indian J Chemistry, 35(B), 1996, 333.

10 Mahajan R K, Sharma G C, J Ind Chem Soc of Chemist, 15, 1998, 51.

11 Mahajan R K, Gupta Neelam, Uppal Satinder K & Bhardwaj Rakesh, Ind J Expertl Biology, 25, 1987, 86.

12 Mahajan R K, Patial Ved Prakash & Sharma Pamita, Indian J Chemistry, 41B, 2002, 2635.

13 Awasthi Pamita, Mahajan R. K. and Barthwal Ritu Current trends in Solid State NMR Methodology and Practice and 13th

National Magnetic Resonance Society Meeting, National Chemical Laboratories (NCL) Pune, P-31, 5-8 Feb. 2007.

Gelatin Grafted Polypropylene: Kinetics of Biodegradation

Inderjeet Kaur and N. Deepika Khanna Department of Chemistry, Himachal Pradesh University, Shimla-171005, India


Abstract

Gelatin based graft copolymers of polypropylene (PP), has been synthesized by chemical method using benzoyl peroxide (BPO), as radical initiator. Biodegradation studies of pristine PP and gelatin grafted PP have been carried out by soil burial test containing simple soil and soil enriched with nitrogenous content by adding urea. The microbial degradation was substantiated by the direct attack of the microbes on the grafted samples. On comparison it was observed that the rate of degradation by the direct attack was fast in comparison to the degradation in soil burial studies. The biochemical tests performed on the organisms isolated from the soil, identified these organisms as Bacillus circulans, Kurthia gibsonii and Flavobacterium sp. which helped biodegradation of PP-g-Gelatin samples. The degradation of the grafted samples was further confirmed by carrying out the physical characterization of the original samples and the degraded samples involving scanning electron microscopy, X-ray diffraction and thermogravimetric analysis studies. A contrast difference in the topological morphology, the crystallinity and the thermal behavior between the original and the degraded samples was observed. The overall mechanism of degradation was evaluated by kinetic parameters derived from thermogravemetric data. All the materials decompose by a type R3 mechanism. 1. Introduction Development of biodegradable polymers is a newly emerging field. A vast number of biodegradable polymers have been synthesized recently and some microorganism and enzymes capable of degrading them have been identified. In developing countries, environmental pollution by synthetic polymers such as polyethylene (PE) and polypropylene (PP) has assumed dangerous proportions. As a result, attempts have been made to solve these problems by introducing biodegradability into these polymers. Biodegradation is a process whereby bacteria, fungi, yeasts and their enzymes consume a substance as a food source so that its original form disappears. Under appropriate conditions of moisture, temperature and oxygen availability, biodegradation is a rapid process. Biodegradation for limited periods is a reasonable target for the complete assimilation and disappearance of an article leaving no toxic or environmentally harmful residue. Gelatin, an animal protein is a water soluble, biodegradable polymer with extensive industrial, pharmaceutical and biomedical uses. It can be hydrolyzed by a variety of the proteolytic enzymes to yield its constituent amino acids peptide components [1]. This nonspecificity is a desirable factor in intentional biodegradation. A method was developed to prepare a simple,

flexible gelatin film-based artificial skin that could adhere to an open wound and protect it against fluid loss and infection. The films were tough and adhered to open wounds spontaneously [2]. Kuwajima et al [3] grafted methyl methacrylate onto gelatin by radical initiators and studied these in aqueous solution at temperatures between 60 and 800C. Kumar et al [4] prepared gelatin-g-poly(ethyl acrylate) in an aqueous medium, using K2S2O8 as an initiator. The copolymer was tested for their microbial susceptibility. The effect of γ-sterilization on the biodegradation of polyolefins was studied by Alariqi et al [5] under composting and fungal culture environments. The changes in functional groups surface morphology and chain scission were characterized by various physical methods. Thermal properties and enzymatic degradation of blends of poly(ε-caprolactone) with gelatin was investigated by Rosa et al [6]. Morancho et al [7] studied the biodegradability of mixtures of PP and gelatin by colorimetry and thermo-gravimetric analysis and observed an increase in the thermal stability of the Gelatin units, PP remains unaffected. Specimen in film shape as well as in powder shape were subjected to the biodegradation tests by Yang et al [8] to investigate dependence of test results on the shape of specimens. In the present manuscript we report on the biodegradation studies and mechanism of thermal

Gelatin Grafted Polypropylene: Kinetics of Biodegradation 177

degradation. Biodegradation was done by soil burial test, microbial studies of the soil containing samples in Nutrient Agar and Czapec Dox and periodic hydrolysis of the samples placed for degradation was also carried out. Various kinetic parameters[9], energy of activation Eact, entropy of activation S*, free energy of activation G* and specific reaction rate constant Kr for pure and degraded samples is evaluated by applying Coats and Redfern[10] integral method for non isothermal processes to different thermal degradation kinetic models. 2. Experimental 2.1. Materials Commercial polypropylene (PP)( Thukral Trading Co. Delhi, India) was recrystallized from p-xylene using methanol and irradiated from Co60 source housed in Gamma Chamber-900 at a constant dose rate of 3.40 kGy/h. Gelatin (Merck) and benzoyl peroxide (BPO) (Merck) were used as received. Isolation of bacteria and fungus from soil was made through Nutrient Agar and Czapec Dox enrichment culture. 2.2. Graft Copolymerization Gelatin was grafted onto preirradiated PP using BPO as the radical initiator by the method reported earlier [11]. Maximum percentage of grafting (120%) of Gelatin onto PP was obtained at optimum conditions of [BPO] =4.132 X 10-2 mol/L at 700C in 120 min. using 30 ml of water. Characterization of PP-g-Gelatin was carried out through FTIR, thermogravimetric analysis, scanning electron micrographs, swelling and water retention studies. 2.3. Biodegradation Studies 2.3.1. Soil Burial Method Soil of known moisture content (17.5%) and pH (8.78) was taken in different pots. A weighed amount (0.500 g) of each of the samples i.e. pristine PP, PP-g-Gelatin both composite (containing the true graft and unreacted PP and gelatin) and true graft (unreacted PP and gelatin removed) wrapped separately in loose knitted synthetic net were placed in each pot. Care was taken that the samples were completely buried in soil beds and kept at room temperature (250C-300C). Samples were removed after a specific number of days for assessment of changes for weight loss. Percent wt. loss determined as a function of number of days was calculated as [12]

W100

0_ W

W0X% Wt. loss =

Where Wo is the initial weight in the beginning, W is the weight after a specific number of days i.e. 10, 20, 30, 45, 60, 75, 90 and 120 days. The soil beds are also supplemented with organic fertilizer to encourage an active microbial flora (6g of urea per Kg of soil) to check the effect of nitrogenous compounds on the degradation of samples of PP. Finally, the samples can be used to ‘bait’ microorganisms involved in the degradation process. These microbes can be isolated and characterized by incorporated into the petri dish screen method. 2.3.2. Petri Dish Screen The presence of microorganisms such as bacteria, fungi was checked by microbial study of the soil containing the samples for biodegradation. Isolation of bacteria and fungus from soil was made through enrichment culture, Nutrient agar and Czapec Dox media by Serial dilution method [12]. The growth pattern was observed and the number of colony forming units was counted. 2.3.3. Identification and Direct Microbial Studies Cultures for identification were submitted in IMTECH (Institute of Microbial Technology), Chandigarh. Direct microbes study of PP and PP-g-Gelatin (both composite and true graft) with isolated culture was studied by using a 20ml of medium which is prepared by adding 1.25g peptone and 0.75g beef extract in 250ml of water. The reaction was carried out for 7 days in Orbitek shaker (RPM = 160) at 300C. After completion of reaction the sample was collected by centrifuging it in Remi Cooling Compufuge CPR 23 by setting at 7000 RPM for 10 min. and filtered it very carefully. Sample was washed with water and dried at 400C until it gives constant reading. Dried sample was weighed and % weight loss was calculated as follows:

W100

0 d_ W

W0X% Wt. Loss =

Where Wo is the initial weight of sample and Wd is the weight of dried sample after degradation. 2.3.4. Hydrolysis Studies Samples of PP and PP-g-Gelatin (both composite and true graft) (0.500g) were separately wrapped in synthetic net and buried in soil having good plantation. The samples were removed from the pots at a predetermined time and dried at 450C for 12h. Dried samples were weighed and hydrolyzed with 20 ml of 6N HCl


for 4h at 800C. After the hydrolysis, solution was filtered the residue i.e. PP was dried and weighed. Degradation of PP was calculated from the % of PP left from the grafted samples after hydrolysis as follows.

W100

0_ W

W0X% PP left =

H

Where WH is the wt. of grafted sample left after hydrolysis and Wo is the initial wt. of sample taken. 2.3.5. Characterization of PP and PP Grafted Samples Physical characterization of PP and PP grafted samples (pure and degraded) has been done by scanning Electron Micrographs (SEM), X-ray diffraction (XRD) and thermogravimetric analysis (TGA). The overall mechanism of degradation was studied by kinetic parameters. 3. Results and Discussion 3.1. Biodegradation Studies 3.1.1. Soil Burial Studies Biodegradation of gelatin grafted samples (both composite; GC and true graft; GT) of PP buried under the natural soil and urea enriched soil was monitored as a function of number of days and the results are presented in Fig. 1. It is observed from the figures that percent weight loss of both the samples increases with increase in the number of days indicating that the samples continuously degrade with increase in the length of time. PP-g-Gelatin (composite) showed 31% wt. loss whereas PP-g-Gelatin (true graft) showed 28% wt. loss in urea enriched soil which is higher than the wt. loss observed in natural soil (28% and 26% respectively) in 120 days. In comparison to the grafted samples, pristine PP showed 0% wt. loss in 120 days in both soil systems. As more % wt. loss is observed in urea enriched soil than in simple soil this indicates that the microorganisms present in urea enriched soil are more numerous than those found in natural soil because of the additional nourishment provided in a urea enriched soil environment. 3.1.2. Microanalysis The microanalysis of the soil samples containing PP-g-Gelatin samples both composite and true graft placed for degradation was carried out in Nutrient Agar and Czapec Dox. A thick growth of colonies was observed in the soil containing grafted sample than in the soil containing pristine PP or the soil containing no sample. On comparison of the maximum count of

colonies in Nutrient Agar and Czapec Dox on particular number of days and the maximum drop in the loss of % wt. loss during degradation (Fig. 2) it is observed that maximum growths of colonies between 20-30 and 60-90 days corresponds to maximum drop in the wt. loss during that period. This implies that the growth of colonies is due to the attack of microorganisms which feed upon the material their by growing their colonies and simultaneously degrading the material. The cultures (I-V) were isolated for direct attack on the samples. On the basis of biochemical tests performed on the organisms (I-III) these are identified as Bacillus circulans, Kurthia gibsonii and Flavobacterium sp. respectively.

0

5

10

15

20

25

30

35

0 30 60 90 120

------------Number of Days---------->

-----

-----

% W

t. Lo

ss--

-----

-->

GC (Simple soil)GT (Simple soil)GC (Urea enriched soil)GT (Urea enriched soil)

Fig. 1. Percent weight loss of PP grafted samples with number of days by Soil Burial Method.

0

1

2

3

4

5

6

7

0 30 60 90 120-----------Number of Days---------->

----

-----

-% W

t. Lo

ss--

----

--->

GC (Natural Soil)GT(Natural Soil)GC (Urea enriched Soil)GT (Urea enriched Soil)

Fig. 2. Percent weight loss of PP grafted samples within consecutive number of days by Soil Burial Method. The microbial degradation was further substantiated by the direct attack of the isolated microbes from microanalysis studies of soil on


the grafted samples. It was observed that the rate of degradation by direct attack was faster than to the degradation rate in soil burial studies. The results of direct microbial interaction by isolated culture I-V with pristine PP and PP-g-Gelatin samples for 7 days are presented in Table I. Table 1. Biodegradation study by isolated culture from Soil (In 7 days)

Percent weight loss Degradation with

Cells

Samples *Control

sample I II III IV V

PP 0 20 10 5 10 0 GC 30 80 70 80 60 80 GT 15 45 40 30 30 60

*Degradation without Cells

3.2. Hydrolysis Studies The hydrolysis studies of the samples kept for degradation was carried as a function of number of days. On perusal of hydrolysis studies it is observed that the amount of % residue left continuously decreases with increasing number of days both in the case of PP-g-Gelatin composite and PP-g-Gelatin true graft. The amount of the residue left in the respective samples is 45% and 51%. 3.3. Characterization of PP and PP Grafted Samples 3.3.1. SEMs Scanning electron micrographs of the pure and grafted samples were taken before and after the soil burial test exposures at 2000X magnification on LEO vp 435 instrument. Prior to the test exposure the pristine PP (Fig. 3) sample exhibited a relatively continuous morphology of the surface, which do not change even after a soil test exposure of 120 days i.e. it showed no sign of degradation.

Figs. 4 and 5 present images of the surface of PP-g-Gelatin before and after the degradation studies. Before the biodegradability test, the PP-g-Gelatin (Fig. 4) was characterized by contiguous surface of PP in which gelatin appears as incorporated onto the surface of PP upon grafting. The biodegraded surface after 120 days (Fig. 5) of soil test exposure presented larger holes and more numerous surface irregularities compared to that of before 120 days in the soil burial test.

The samples of PP and PP-g-Gelatin collected after 7 days of direct microbes study were also examined by their SEM and results are presented in Fig. 6. Results indicate that

microorganisms introduced into the medium degraded PP-g-Gelatin (both composite and true graft) more than by soil burial test as observed from the SEM.

Fig. 3. SEM of PP

Fig. 4. SEM of PP-g-Gelatin

Fig. 5. SEM of PP-g-Gelatin (after soil burial studies)

Fig. 6. SEM of PP-g-Gelatin(after microbial studies)

3

4

5

6


Table 2. TGA of PP and PP grafted samples

From the XRD spectra (Fig. 7) of PP-g-Gelatin samples following conclusions are drawn: After four months of biodegradation, the intensity of almost all the peaks is reduced except that at 12.70 and 27.40. This decrease of intensity is due to existence of distortion of long-range order in which structural units show statistical fluctuations in their positions. It is observed from the diffraction pattern of degraded PP-g-Gelatin that there occurs a decrease in the peak intensities and increase in the full width at half maximum and decrease in peak intensity is generally associated with the decrease in crystalinity of the polymer. The larger the crystals of a given component, the sharper are the peaks on the XRD pattern for each crystal plane. Thus, the breadth of the peak can be related to the crystal size. Reduction of peak intensities is followed by their decrease of FWHM (Full Width at Half Maximum). This decrease of FWHM results in the increase of their respective crystallite size. 3.3.2. XRD

Fig. 7. XRD of GC (before and after degradation)

PPH= Hydrolyzed product of PP-g-Gelatin; * after 120 days

This may be due to reduction of short-range order or small fluctuations in the separation distance between monomeric units. 3.3.3. TGA The thermogravimetric analysis of pristine PP, PP-g-Gelatin and their respective samples kept for degradation undisturbed for 120 days was carried out in nitrogen (200ml/min.) atmosphere at a rate of 100C/min on Perkin Elmer (Pyris Diamond, IIT Rorkee, India) and presented in Table II.

However, on comparison of the TG data of the grafted PP before and after degradation it is observed that the DT values at every 10% wt. loss of the degraded samples is higher than that of the pure grafted PP. Further these DT values of the degraded sample beyond 50% wt. loss are parallel to those of the DT values of pristine PP. These observations indicate that the degradation begins at the grafted gelatin chains and approaches towards the PP chains. 3.3.3. Kinetic Models of Degradation Different kinetic models were applied in the present study and interestingly both the samples follow the same mechanism R3 (spherical symmetry) irrespective of different stages of decomposition. The order of stability of the samples is explained on the basis of E* in KJ/mol as follows: PP (758.263) > PPa (373.540) > PPHb (150.461) > PP-g-Gelatin (98.748) > PP-g-Gelatina (65.506). Therefore, pristine PP shows much higher thermal stability as indicated by order of E* than PP-g-Gelatin. a After degradation of 120 days, b Hydrolyzed product of PP-g-Gelatin after

hydrolysis of 120 days with 6N HCl

4. Conclusion Polypropylene has been successfully intercross linked with gelatin through chemical method of graft copolymerization. The biodegradation of the graft copolymer was

DT (oC) at every 10% wt. loss Sample IDT

(0C) FDT (0C) 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

PP 425.6 460.5 414.1 428.8 436.0 440.8 444.5 447.8 451.1 454.7 459.1 470.6

PP-g-Gelatin 410.6 462.9 321.3 401.9 418.5 427.6 434.1 439.1 444.1 448.9 454.5 463.9

PP* 432.5 471.5 421.4 434.3 444.5 446.3 450.1 453.7 457.4 461.3 466.2 485.2

PPH* 415 462 379.8 409.4 423.0 431.9 438.3 444.5 449.2 454.6 461.2 653.0

PP-g-Gelatin* 414.2 464.1 355.1 408.7 423.7 433.1 439.9 445.8 451.2 457.4 466.9 -


monitored as a function of time during soil burial investigation. The introduction of gelatin moieties into the PP backbone has triggered degradation into the otherwise resistant PP after initial degradation beginning at the gelatin sites. The effect of degradation on the surface morphology and thermal properties has been investigated by SEM and TGA studies thus substantiating the degradation of the modified grafted polypropylene samples. Acknowledgement

Miss Naisergik Deepika Khanna, JRF is thankful to the Council of Scientific and Industrial Research, New Delhi, for providing financial assistance to carryout this work. Refrences [1] Eastoe JE, Leach AA. In: Ward AG, Courts

A, editors. The science and Technology of Gelatin. New York: Academic, 1977. p. 73.

[2] Shinde BG, Nithianandam VS, Kaleem K, Erhan S. Bio-Medical Mater Eng. 1992;2:123.

[3] Kuwajima T, Yoshida H, Hayashi K. J Appl Polym Sci. 1976;20:967.

[4] Kumar GS, Kaplagam V, Nandi US. J Appl Polym Sci. 1981;26:3633.

[5] Alariqi SAS, Kumar AP, Rao BSM, Singh RP. Polym Degrad Stab. 2006;91:1105.

[6] Rosa DS, Lopes DR, Calil MR. Polymer Testing. 2005;24:756.

[7] Morancho JM, Vlles X, Contat L, Ribes A. Polym Degrad Stab. 2006;91:44.

[8] Yang HS, Yoon JS, Kim MN. Polym Degrad Stab. 2005;87:131.

[9] Chauhan GS, Bhatt SS, Kaur I, Singha AS, Kaith BS. J Polym Mater. 1999; 16; 245.

[10] Coats AW, Redfern JP. Nature (London). 1964;201;68.

[11] Kaur I, Gautam N, Khanna ND. J. Appl. Polym. Sci. 2007;107:2238.

[12] Chandra R, Rustgi R. Prog. Poym. Sci., 1998; 23; 1273.

Crystal and Molecular Structure of 1-(4-chlorophenoxy) 3, 3-dimethyl-1-4(1, 2, 4-triazole-1-Y-1)-2-butanone

Rachana Tiwari, Anjana Chauhan and R.K. Tiwari

SOS in physics, Jiwaji University, Gwalior (M.P.)-474011 E-mail: [email protected], [email protected]

Abstract

The crystals of the title compound were grown from a solution of cyclohexane at 278˚K. The crystals are rectangular in shape with unit cell parameters a = 8.16(10)Å, b = 16.81(3)Å, c = 22.05(2)Å, β = 92.37˚ and Z = 8. The space group is P21/c. The structure was solved by SHELX-97 using diffractometer data and finally refined to an R value of 0.0569. The molecule is found to adopt a conformation such that the triazolyl ring is inclined at an angle of 72.9˚ to the aromatic ring. The resulting arrangement leads to the close approach of the ortho-H, H(2A) to the triazolyl atoms N(1A) and N(2A) such that both N—H distances lie within the sum of the Vander Walls radii of N and H. The benzene and triazolyl rings are partially planer.

1. Introduction

Triadimefon is the common name for a triazole group fungicidal compound originally synthesized in 1970 by Meiser at research centre of Bayer AG in Wuppertal Elberfield, West Germany which was initially tested under the code name Bay MEB 64671. It is now known under the trade names Amiral and Bayleton®. Chemically it is 1-(4-chlorophenoxy)3,3-dimethyl-1-H(1,2,4-triazole-1-Y-1)-2-butanone. Its empirical formula is C14H16ClN3O2 and molecular weight is 293.7 g/mol.

2. Experiment

White crystals were obtained by slow evaporation from a solution of cyclohexanone at 278˚K temp. The crystals are rectangular in shape. The unit cell parameters were determined by automatic computerized4-circled Enraf-Nonious CAD-4 Diffractometer. These data showed a = 8.16(10)Å, b = 16.81(3)Å, c = 22.05(2)Å, β = 92.37(10)˚ & Z = 8.

Thus space group was determinated to be P21/c . The crystal density was calculated by floatation method in the mixture of benzene and carbon-tetrachloride. Benzene was added to the solution until the crystal floated in the middle of mixture. Thus the crystal and solution are of same density and density of solution was measured with pyknometer. The measured density is 1.291μg/m3 and calculated density is 1.295μg/m3. The preliminary data about the crystal is given in Table 1.

Structure determination The crystal was solved using SHELXS-972

program for crystal structure solution. The positional co-ordinates, which were obtained from SHELXS-97 and isotropic temperature factors, were subjected to refinement by SHELXL3 refinement program. After so many cycles of refinement the R factors dropped to 0.123. Further refinement of the structure was carried out with individual an isotropic temperature factors of the exponential form. -2P1^2[h^2a*^2U11+-----------+2hKa*b×U12] reduced R factor to 0.101. Refinement of the structure was terminated after two more cycles when all the deviations in parameters became much smaller than the corresponding estimated standard derivations. The final R value was 0.0569 for all 20912 reflections collected. The final positional and thermal parameters of non-hydrogen atoms are listed in Table 2. 3. Results and discussions

The perspective view of the molecule and numbering scheme is shown in Fig. 1. The bond length and bond angles involving all atoms are listed in Table 3. and Table 4. And ORTEP4 drawing is shown in Fig. 2 respectively.

Fig. 1. Molecular no. scheme of triadimefon

Structure of 1-(4-chlorophenoxy) 3, 3-dimethyl-1-4(1, 2, 4-triazole-1-Y-1)-2-butanone 183

The average bond distance of C-H is 0.95Å. The bond lengths and angles in the benzene ring show regular features in both the molecules. The Cl(1A)-C(4A) and Cl(1B)-C(4B) distances are 1.733Å and 1.738Å comparable to other structure5-8. These distances are short and may be due to delocalization of electrons from the benzene rings. The whole molecules appeared to be twisted and folded and reason may be stacking constraint. The bond distances around C(7A) and C(7B) are usual shorter than single bond values.

Fig. 2. ORTEP drawing.

They may also appear to bear a partial double bond character. The O(1A)-C(7A) and O(2B)-C(7B) distances are 1.423(2)Å 1.411(2)Å respectively. These distances do not change significantly in similar structures, despite variable intermolecular interactions through them. The bond distances in the triazolyl rings are comparable to corresponding distances is hetrocyclic rings 1.339(Å)9. The average set of data by Spencer10 are 1.377Å and 119˚ respectively. The dimensions of the methyl groups are normal and comparable with those in o-methyl obtusaquinone11 and moscaline hydrobromide12.

The molecule is found to adopt a conformation such that the triazolyl ring is inclined at an angle of 72.9(9)˚ to the aromatic ring and at an angle of 61.5(9)˚ to C(11A), C(10A), O(1A), C(7A) grouping. The resulting arrangement leads to the close approach of the ortho-H, H(2A) to the triazolyl atoms N(1A) and N(2A) such that both N—H distances lie within the sum of the Vander Walls radii of N and H. There was an accompanying distortion of the exocyclic angles at C(1A) with the C(2A)-C(1A)-O(1A) bond angle of 124.65(17)˚ being

considerably larger than the value found for O(1A)-C(1A)-C(6A) 114.97(17)˚.

The triazolyl ring is planner with C(7A) lying only 0.063(7)Å from the mean plane. Although the C(8A)-N(1A) and C(9A)-N(3A) distances are somewhat larger than C(8A)-N(3A) and C(9A)-N(2A), in keeping with the uncharged canonical valance form.

All four C-N distances are shorter than a normal single bond (1.47Å). The N(1A)-N(2A) bond is also shorter than a normal single bond (1.45Å). The three atoms bonded to N(1) are almost coplanar with it. Taken together these data indicate extensive delocalization within the hetrocyclic ring. The most note worthy feature of the hetrocyclic ring is the asymmetry of the exocyclic angles at N(1A) [120.28˚, 130.80˚]. We have observed a similar pattern in related triazole systems and it appear to be a function of a triazolyl ring itself rather than the influence of any inter or intramolecular interactions.

Table 1. Crystal data and structure refinement for triadimefon

Identification code TRIADIMEFON

Empirical formula C14H16ClN3O2

Formula weight 293.75

Temperature 293(2)K

Wavelength 0.71073Å Crystal

system, space group

Monoclinic, P21/c

Unit cell dimentions

a = 8.16260(10)Å alpha = 90 deg.b = 16.8106(3)Å beta = 92.37(10)

deg. c = 22.0520(2)Å gamma = 90 deg.

Volume 3023.35(7)Å3

Z, Calculated density 8, 1.291 Mg/m3

Absorption coefficient 0.257 mm-1

F(000) 1232

Crystal size 0.48 × 0.38 × 0.22 mm

The C(11A), C(10A), C(7A), O(1A), C(1A) backbone is rather compressed resulting in the main from the orientation of the tert-butyl group,


the C(11A)-C(10A)-C(7A)-O(1A) torsion being only 99.17(2)˚.

From the least square plane equation by Blow’s13 method, the benzene and triazolyl rings are partially planner since the atomic displacements are much less than their e.s.d’s. The triazolyl ring is inclined to the aromatic ring at an angle of 72.9(9)˚. Torsion angles calculated are given in Table 5.

Table 2. Atomic coordinates (× 10 ^ 4) and equivalent isotropic displacement parameters (A ^ 2 × 10 ^ 3) for TRIADIMEFON. U (eq) is defined as one third of the trace of the orthogonalized Uij tensor.

Atom x y z U(eq)

Cl(1A) O(1A) O(2A) N(1A) N(2A) N(3A) C(1A) C(2A) C(3A) C(4A) C(5A) C(6A) C(7A) C(8A) C(9A)

C(10A) C(11A) C(12A) C(13A) C(14A) Cl(1B) O(1B) O(2B) N(1B) N(2B) N(3B) C(1B) C(2B) C(3B) C(4B) C(5B) C(6B) C(7B) C(8B) C(9B)

C(10B) C(11B) C(12B) C(13B) C(14B)

-80(1) 3916(2) 5953(2) 3876(2) 3675(3) 2306(3) 3027(2) 2635(3) 1691(3) 1145(3) 1556(3) 2495(3) 4853(2) 3040(3) 2750(4) 6245(3) 7866(3) 8705(3) 7585(4) 8935(4) 60(1)

-3826(2) -5981(2) -3838(2) -3753(3) -2356(3) -2944(3) -2586(3) -1673(3) -1111(3) -1463(3) -2388(3) -4779(2) -2864(4) -2984(3) -6189(3) -7764(3) -8625(4) -8879(4) -7432(4)

6330(1) 3437(1) 2041(1) 2622(1) 2790(1) 1670(1) 4131(1) 4468(1) 5156(1) 5488(1) 5156(1) 4480(1) 3134(1) 1958(1) 2195(2) 2685(1) 3113(1) 3096(2) 3982(2) 2690(2) 1302(1) 4234(1) 5624(1) 4991(1) 4827(1) 5949(1) 3531(1) 3153(1) 2461(1) 2166(1) 2540(1) 3226(1) 4493(1) 5427(2) 5653(1) 4946(1) 4493(1) 4421(2) 4958(2) 3657(2)

1095(1) 1028(1) 1004(1) 1898(1) 2494(1) 2225(1) 1087(1) 1634(1) 1636(1) 1097(1) 551(1) 544(1)

1535(1) 1753(1) 2656(1) 1217(1) 1179(1) 1812(1) 977(1) 725(2)

4111(1) 4083(1) 4026(1) 3183(1) 2583(1) 2819(1) 4047(1) 3513(1) 3536(1) 4085(1) 4619(1) 4602(1) 3572(1) 2398(1) 3310(1) 3870(1) 3953(1) 3314(1) 4347(2) 4187(2)

90(1) 51(1) 64(1) 46(1) 61(1) 80(1) 41(1) 50(1) 55(1) 53(1) 55(1) 51(1) 42(1) 69(1) 74(1) 46(1) 57(1) 80(1) 80(1)

103(1) 91(1) 52(1) 68(1) 44(1) 62(1) 81(1) 42(1) 48(1) 52(1) 51(1) 56(1) 51(1) 42(1) 76(1) 69(1) 48(1) 63(1) 94(1)

125(2) 110(1)

Table 3. Bond length [Å] with estimated standard deviation in parentheses for TRIADIMEFON.

Cl(1A)-C(4A) O(1A)-C(1A) O(1A)-C(7A) O(2A)-C(10A) N(1A)-C(8A) N(1A)-N(2A) N(1A)-C(7A) N(2A)-C(9A) N(3A)-C(8A) N(3A)-C(9A) C(1A)-C(2A) C(1A)-C(6A) C(2A)-C(3A) C(3A)-C(4A) C(4A)-C(5A) C(5A)-C(6A) C(7A)-C(10A) C(10A)-C(11A) C(11A)-C(12A) C(11A)-C(14A) C(11A)-C(13A) Cl(1B)-C(4B) O(1B)-C(1B) O(1B)-C(7B) O(2B)-C(10B) N(1B)-C(9B) N(1B)-N(2B) N(1B)-C(7B) N(2B)-C(8B) N(3B)-C(9B) N(3B)-C(8B) C(1B)-C(2B) C(1B)-C(6B) C(2B)-C(3B) C(3B)-C(4B) C(4B)-C(5B) C(5B)-C(6B) C(7B)-C(10B) C(10B)-C(11B) C(11B)-C(13B) C(11B)-C(14B) C(11B)-C(12B)

1.733(2) 1.383(2) 1.423(2) 1.200(2) 1.339(3) 1.361(2) 1.440(2) 1.312(3) 1.315(3) 1.336(3) 1.382(3) 1.386(3) 1.390(3) 1.371(3) 1.381(3) 1.372(3) 1.555(3) 1.512(3) 1.529(3) 1.530(3) 1.542(3) 1.738(2) 1.388(2) 1.411(2) 1.201(2) 1.337(3) 1.356(2) 1.443(2) 1.316(3) 1.314(3) 1.331(3) 1.381(3) 1.386(3) 1.381(3) 1.370(3) 1.376(3) 1.378(3) 1.548(3) 1.512(3) 1.504(4) 1.517(4) 1.553(4 )

Structure of 1-(4-chlorophenoxy) 3, 3-dimethyl-1-4(1, 2, 4-triazole-1-Y-1)-2-butanone 185

Table 4. Bond angles (degree) with estimated standard deviation in parentheses for TRIADIMEFON.

C(1A)-O(1A)-C(7A) C(8A)-N(1A)-N(2A) C(8A)-N(1A)-C(7A) N(2A)-N(1A)-C(7A) C(9A)-N(2A)-N(1A) C(8A)-N(3A)-C(9A) C(2A)-C(1A)-O(1A) C(2A)-C(1A)-C(6A) O(1A)-C(1A)-C(6A) C(1A)-C(2A)-C(3A) C(4A)-C(3A)-C(2A) C(3A)-C(4A)-C(5A) C(3A)-C(4A)-Cl(1A) C(5A)-C(4A)-Cl(1A) C(6A)-C(5A)-C(4A) C(5A)-C(6A)-C(1A) O(1A)-C(7A)-N(1A) O(1A)-C(7A)-C(10A) N(1A)-C(7A)-C(10A) N(3A)-C(8A)-N(1A) N(2A)-C(9A)-N(3A) O(2A)-C(10A)-C(11A) O(2A)-C(10A)-C(7A) C(11A)-C(10A)-C(7A) C(10A)-C(11A)-C(12A) C(10A)-C(11A)-C(14A) C(12A)-C(11A)-C(14A) C(10A)-C(11A)-C(13A) C(12A)-C(11A)-C(13A) C(14A)-C(11A)-C(13A) C(1B)-O(1B)-C(7B) C(9B)-N(1B)-N(2B) C(9B)-N(1B)-C(7B) N(2B)-N(1B)-C(7B) C(8B)-N(2B)-N(1B) C(9B)-N(3B)-C(8B) C(2B)-C(1B)-C(6B) C(2B)-C(1B)-O(1B) C(6B)-C(1B)-O(1B) C(3B)-C(2B)-C(1B) C(4B)-C(3B)-C(2B) C(3B)-C(4B)-C(5B) C(3B)-C(4B)-Cl(1B) C(5B)-C(4B)-Cl(1B) C(4B)-C(5B)-C(6B) C(5B)-C(6B)-C(1B) O(1B)-C(7B)-N(1B) O(1B)-C(7B)-C(10B) N(1B)-C(7B)-C(10B) N(2B)-C(8B)-N(3B) N(3B)-C(9B)-N(1B) O(2B)-C(10B)-C(11B) O(2B)-C(10B)-C(7B) C(11B)-C(10B)-C(7B) C(13B)-C(11B)-C(10B) C(13B)-C(11B)-C(14B) C(10B)-C(11B)-C(14B) C(13B)-C(11B)-C(12B) C(10B)-C(11B)-C(12B) C(14B)-C(11B)-C(12B)

119.69(15) 108.92(18) 130.80(19) 120.28(16) 101.36(19) 101.6(2)

124.65(17) 120.37(19) 114.97(17) 119.45(19) 119.8(2) 120.6(2)

120.15(19) 119.27(18) 120.1(2) 119.7(2)

111.06(16) 101.45(15) 113.11(16) 111.2(2) 116.9(2) 124.6(2)

118.52(19) 116.86(18) 107.47(18) 109.9(2) 110.0(2) 110.3(2) 109.7(2) 109.4(2)

119.10(15) 109.03(17) 130.16(18) 120.79(16) 101.42(18) 101.8(2)

120.65(19) 124.74(17) 114.60(17) 119.35(19) 119.9(2) 121.1(2)

119.72(18) 119.18(17) 119.5(2) 119.5(2)

111.33(16) 101.97(15) 112.92(16) 116.7(2) 111.1(2) 123.6(2)

119.35(19) 117.01(18) 110.3(2) 112.8(3) 111.5(2) 107.7(3)

106.50(19) 107.8(2)

Table 5. Torsion angles (deg) for TRIADIMEFON C(8A)-N(1A)-N(2A)-C(9A) C(7A)-N(1A)-N(2A)-C(9A) C(7A)-O(1A)-C(1A)-C(2A) C(7A)-O(1A)-C(1A)-C(6A) O(1A)-C(1A)-C(2A)-C(3A) C(6A)-C(1A)-C(2A)-C(3A) C(1A)-C(2A)-C(3A)-C(4A) C(2A)-C(3A)-C(4A)-C(5A) C(2A)-C(3A)-C(4A)-Cl(1A) C(3A)-C(4A)-C(5A)-C(6A) Cl(1A)-C(4A)-C(5A)-C(6A) C(4A)-C(5A)-C(6A)-C(1A) C(2A)-C(1A)-C(6A)-C(5A) O(1A)-C(1A)-C(6A)-C(5A) C(1A)-O(1A)-C(7A)-N(1A) C(1A)-O(1A)-C(7A)-C(10A) C(8A)-N(1A)-C(7A)-O(1A) N(2A)-N(1A)-C(7A)-O(1A) C(8A)-N(1A)-C(7A)-C(10A) N(2A)-N(1A)-C(7A)-C(10A) C(9A)-N(3A)-C(8A)-N(1A) N(2A)-N(1A)-C(8A)-N(3A) C(7A)-N(1A)-C(8A)-N(3A) N(1A)-N(2A)-C(9A)-N(3A) C(8A)-N(3A)-C(9A)-N(2A) O(1A)-C(7A)-C(10A)-O(2A) N(1A)-C(7A)-C(10A)-O(2A) O(1A)-C(7A)-C(10A)-C(11A) N(1A)-C(7A)-C(10A)-C(11A) O(2A)-C(10A)-C(11A)-C(12A) C(7A)-C(10A)-C(11A)-C(12A) O(2A)-C(10A)-C(11A)-C(14A) C(7A)-C(10A)-C(11A)-C(14A) O(2A)-C(10A)-C(11A)-C(13A) C(7A)-C(10A)-C(11A)-C(13A) C(9B)-N(1B)-N(2B)-C(8B) C(7B)-N(1B)-N(2B)-C(8B) C(7B)-O(1B)-C(1B)-C(2B) C(7B)-O(1B)-C(1B)-C(6B) C(6B)-C(1B)-C(2B)-C(3B) O(1B)-C(1B)-C(2B)-C(3B) C(1B)-C(2B)-C(3B)-C(4B) C(2B)-C(3B)-C(4B)-C(5B) C(2B)-C(3B)-C(4B)-C(1B)

1.2(3) -179.4(2) -16.5(3)

164.77(17) -178.10(19)

0.5(3) 0.7(3) -1.5(4)

178.23(17) 1.1(3)

-178.63(17) 0.1(3) -0.9(3)

177.82(18) 87.8(2)

-151.70(17) 57.8(3)

-121.47(19) -55.5(3) 125.2(2)

0.0(3) -0.8(3)

179.9(2) -1.3(3) 0.8(3)

-78.6(2) 40.4(3)

99.17(19) -141.81(19)

-108.0(3) 74.3(2) 11.6(3)

-166.0(2) 132.4(2) -45.2(2) -1.2(3)

177.4(2) 16.8(3)

-164.68(18) 0.4(3)

178.92(19) -1.1(3) 1.1(3)

-179.17(16)


C(3B)-C(4B)-C(5B)-C(6B) Cl(1B)-C(4B)-C(5B)-C(6B) C(4B)-C(5B)-C(6B)-C(1B) C(2B)-C(1B)-C(6B)-C(5B) O(1B)-C(1B)-C(6B)-C(5B) C(1B)-O(1B)-C(7B)-N(1B) C(1B)-O(1B)-C(7B)-C(10B) C(9B)-N(1B)-C(7B)-O(1B) N(2B)-N(1B)-C(7B)-O(1B) C(9B)-N(1B)-C(7B)-C(10B) N(2B)-N(1B)-C(7B)-C(10B) N(1B)-N(2B)-C(8B)-N(3B) C(9B)-N(3B)-C(8B)-N(2B) C(8B)-N(3B)-C(9B)-N(1B) N(2B)-N(1B)-C(9B)-N(3B) C(7B)-N(1B)-C(9B)-N(3B) O(1B)-C(7B)-C(10B)-O(2B) N(1B)-C(7B)-C(10B)-O(2B) O(1B)-C(7B)-C(10B)-C(11B) N(1B)-C(7B)-C(10B)-C(11B) O(2B)-C(10B)-C(11B)-C(13B) C(7B)-C(10B)-C(11B)-C(13B) O(2B)-C(10B)-C(11B)-C(14B) C(7B)-C(10B)-C(11B)-C(14B) O(2B)-C(10B)-C(11B)-C(12B) C(7B)-C(10B)-C(11B)-C(12B)

-0.4(3) 179.84(17)

-0.2(3) 0.2(3)

-178.40(19) -89.6(2)

149.76(17) -54.7(3)

127.01(19) 59.3(3)

-119.0(2) 0.9(3) -0.2(4) -0.6(3) 1.2(3)

-177.2(2) 82.9(2) -36.7(3) -96.4(2)

143.98(19) -9.8(4)

169.5(3) -135.9(3) 43.4(3)

106.7(3) -74.0(3)

References

[1] Kolbe, W. (1976), poflanzenschutz-Nachrichten Bayer 31, 163-180.Sheldrich, G.M. (1997), SHELXS-97, Program for crystal structure determination.

[2] Sheldrich, G.M. (1997), SHELXS-97, Program for the solution of crystal structure.

[3] Jolmson, C.K. (1965), ORTEP, Report ORNL-3794, Oak Ridge National laboratory, Temessee, U.S.A.

[4] Nowell, I.W. and Walker, P.E.(1982) Acta Cryst. B38, 1857-1859.

[5] Bucheuauer, H. (1976). Z.P. Ilanzeskar. Pflanzenschutz. B3, 368-369.

[6] Martin, T.J. & Morris, D.B.(1979) Pflanzenschutz Nachr. Am. Ed. 32, 31-79.

[7] Sanger, Jyotsna, Ph.D. Thesis, Jiwaji University, Gwalior India(2000).

[8] Haridas, M., Kulkarni, N.R., Tiwari, R.K. and Singh T.P.(1982), Curr. Sci. (India) 51(23), 1111.

[9] Spencer, M. (1959), Acta Cryst. 12, 50. [10] Palmer K.J., Wang, R.Y. and Jurd, L.

(1973), Acta Cryst. B29, 1509. [11] Ernst, S.R. and Cogle Jr. F.W. (1973), Acta

Cryst. B29, 1543. [12] Blow, D.M. (1960), Acta Cryst. 13, 168.

Influence of Brass as a Filler on Load-Speed Sensitivity of Polymer Based Friction Composites

Mukesh Kumar and Jayashree Bijwe

Industrial Tribology Machine Dynamics and Maintenance Engineering Centre, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110 016, INDIA


Abstract

Polymer based composite friction materials are widely used in brake systems of automobiles, locomotives etc. Metals are important fillers in friction materials since they control the thermal conductivity (TC) of composites apart from additional functions such as wear resistance, strength etc. Low TC of composites renders the tribo-surface vulnerable for degradation of organic ingredients affecting the braking capability adversely while too high TC on the other hand, results in adverse effect on brake-fluid. The metallic fillers also affect the tribo performance of composites. Hence the contents are to be optimized to achieve best combination of performance properties and thermal conductivity. The optimum combination hence is to be tailored with right contents of metallic fillers. Hence in this paper the effect of brass particles in increasing amount on friction and wear performance of polymer based NAO friction composites was studied. Four friction composites with varying amount of brass (0, 4, 8 and 12 wt.% and barite in 35, 31, 27, and 23 wt.% in complementary manner) were developed as brake-pads. These were characterized for physical, chemical and mechanical properties. These brake pads were evaluated for their friction sensitivity towards load and speed in simulated braking conditions under variable operating parameters such as speed and pressure on a reduced scale prototype (RSP). Composite with 8% (by wt.) brass fibers proved to exhibit best combination of performance parameters related to friction and wear in this testing mode. Scanning electron microscopy (SEM) were employed to understand wear mechanisms.

1. Introduction

Non-asbestos organic fiber reinforced-low metallic composites are increasingly being used in automotive brake disc pads, shoes, linings, blocks, clutch facings etc primarily because of awareness of health hazardness of asbestos. They are essentially multi ingredient systems in order to achieve the desired amalgam of performance properties [1, 2] and more than several hundred ingredients have been reported being used for friction composites in the literature. These are categorized as binder, fibers, friction modifiers and fillers based on the major function they perform. Binder is a resin mainly phenolic based, whose function is to hold all the ingredients very firmly so that they can contribute towards their major functions. While fibers such as mineral, ceramics, organic and metallic types provide mainly strength, wear resistance and raise the friction level, in general. Friction modifiers such as abrasives and solid lubricants are used to achieve the desired range of friction with minimal fluctuations. The class of fillers is again subdivided as functional fillers (to enhance the specific function such as resistance to fade, thermal conductivity etc) and space/inert fillers (mainly to cut the cost). A lot is reported on the influence of these ingredients in friction composites on tribo -performance in various testing situations such as fade and recovery, load, speed etc [3-7].

Metallic fillers contents are important in friction materials since they control the conductivity of composites apart from additional functions such as wear resistance; strength etc. Comparatively little is reported on metallic contents in non-asbestos friction materials in this context [8-10]. Thermo-physical properties such as specific heat, thermal conductivity (TC), diffusivity, thermal expansion etc. play vital roles in performance of brake pads, especially when braking is severe. Low TC of composite renders the tribo-surface vulnerable due to accumulation of frictional heat which leads to degradation of organic ingredients which in turn, affects the braking capability adversely. Too high TC on the other hand, results in adverse effect on brake-fluid leading to “spongy brakes” which is an unwanted phenomenon. The optimum TC of a composite hence is also very much essential feature. In spite of this fact, not much is reported in the literature on this aspect. Handa and Kato [8] studied the influence of variation of 3 ingredients (Cu powder, CNSL powder and BaSO4) in quaternary compositions. The studies were targeted to examine the influence of their variation on friction and wear properties on a reduced scale tribometer. The composites were of academic interest and testing conditions were also not realistic. No efforts were made to correlate the performance with


increase in thermal conductivity or changes in interface temperature as a result of inclusion of more Cu powder. Jang et. al [9] have observed the effect of metallic fibers such as low carbon steel, Al and Cu on performance of NAO friction composites using a small-scale friction tester. The studies were again focused on investigating the influence of addition of these fibers individually in composites under various operating conditions. Overall benefits or limitations of each fiber were discussed. The issues of optimal amount of fibers or thermal conductivity or effect on interface temperature, however, were not touched. In the present work four NAO composites were developed by varying brass fibers (0, 4, 8 and 12 %) and barite (inert filler) in a complementary manner keeping other ingredients unaltered. These composites were evaluated for their friction sensitivity towards load and speed in simulated braking conditions against a commercial disc under variable operating conditions such as speed and pressure on a reduced scale prototype (RSP). SEM studies were done to understand wear mechanisms. 2. Experimental and Results 2. 1. Fabrication of the composites

The fabrication of composites containing 12 ingredients was based on keeping parent composition of 10 ingredients (65 % by wt) constant and varying two ingredients viz brass fiber and barite in a complementary manner as shown in Table 1 based on systematic increase in brass fibers ( 0, 4, 8 and 12 wt% ) and proportionate decrease in barite contents. The parent composition contained straight phenolic resin (10 %), functional fillers such as alumina, graphite, vermiculite, cashew dust (35 %) and fibers such as glass, PAN, Lapinus, Aramid and steel wool. These composites were designated as B0, B1, B2, and B3 accordingly (Table 1).

Table 1. Details of the formulated composites based on the variation in amount of brass fibers and barite

The total metallic contents (steel and brass) in

B0, B1, B2 and B3 thus were; 8, 12, 16 and 20 %. The ingredients were mixed in a plough type shear mixer to ensure the macroscopic homogeneity. The mixing schedule was of ten minutes duration. The mixture was then placed into a four-cavity mould supported by the adhesive-coated back plates. Each cavity was filled with approximately 80g of the mixture and heat

cured in a compression-molding machine under a pressure of 8 MPa for 7-8 minutes at a curing temperature of 150oC. Three intermittent ‘breathings’ were also allowed during the initiation of curing to expel volatiles. The pads were then removed and were post-cured in an oven at 100° C for 8 hours. The surfaces of the pads were then polished with a grinding wheel to attain the desired thickness and surface finish.

2. 2. Characterization of the composites

For mechanical strength testing, test bars were fabricated as per ASTM standards. Composites were characterized for physical (density, water swelling, heat swelling and void contents) and chemical (acetone execration) properties. Details of measurement procedure of these properties are discussed elsewhere [11]. Void contents were calculated by measuring actual density and theoretical density. Hardness was measured by a Rockwell hardness tester on R scale. Thermo-physical properties were measured as per ASTM-E1461-01 standard on FL-3000 Flash line instrument supplied by Anter Corporation, USA. Samples of square size (10 x 10 mm) and thickness 2-2.5 mm were used for these measurements at room temperature.

2. 3. Test set-up and procedure for studying sensitivity of friction to load and speed

The friction and wear tests were done on a horizontal loading assisted pad-on-disc type reduce scale prototype (RSP) shown in Fig.1. The RSP essentially consists of a pearlitic grey cast iron rotor disc of the passenger car. The disc was connected disc was connected through an interchangeable flange to the flywheel of mass of 100 kg (moment of inertia 6 kg m2) on the shaft.

Fig. 1. Schematic of reduce scale proto-type(RSP)

The flywheel was connected to a 7.5 HP AC motor via a cross coupling. The disc rpm was controlled by controlling the AC motor input into the drive, which was pre-set through the control panel capable of imparting a speed up to 1400 rpm to the disc. A pair of samples of square size (24 x 24 mm2) was cut from the brake pads along with the back

Ingredients composition by wt. Composites designation B0 B1 B2 B3

Space filler- BaSO4 35 31 27 23 Metallic filler- Brass 0 4 8 12 Total amount of two 35 35 35 35

Influence of Brass as a Filler on Load-Speed Sensitivity of Polymer Based Friction Composites 189

plates and push fitted in the sample holders connected to a pressure actuator. The specimens were placed in diametrically opposite locations on the same side of the disc. The load on the pads could be manipulated by controlling the applied pressure on the pads ranging between 1 and 6 MPa in the steps of 0.1 MPa. The load cell attached to the frame carrying the specimen holder measured the frictional force. The operating variables conforming to a particular experimental design such as applied pressure, rubbing speed, number of braking cycles and braking duration could be pre-set in the programmer on the controller. The specimens were ground to a thickness of approximately 1.5 cm (including back plate). The uniform contact of the friction surface was assured through a few cycles of initial bedding-in operation under a nominal pressure of 1 MPa and at a linear speed of 5.03 m/s on the disc. This was done to ensure more than 80% of conformal contact. The samples were cleaned to remove the loose wear debris followed by the initial weighing of the sample after cooling. The specimens were then subjected to braking cycles on the RSP under a series of braking pressure and speed as described in Table 4. Braking duration (touch time) and number of braking in each cycle were constant as 1 sec and 25 respectively.

Table 2. Experimental design for tribo-evaluation

Initial temperature of the disc was 250 C. The temperature of specimen and disc was measured with laser gun (Quick temp 860-T3, Testo Co. Germany make) after each experiment to have a rough idea about trends in increase in temperature after completion of braking cycles and are shown in Table 5. The frictional force was recorded continuously on the PC and the built in software calculated the average value. In the final display 25 values of µ were recorded as a function of 25 cycles. The µ stabilized after first 10 cycles in general. Hence the mean µ of last 15 cycles was calculated and considered as representative value (µm) for that experiment. The variations in µm at various speeds and pressures is shown in Table 6. The sensitivity of µm to speed was analyzed by calculating the extent of decrease in µm with respect to transitions in speed from one level to the successive level. These were designated as ∆µm

v1–2 and ∆µm

v2–3, which were the

decrements in µm with the increase in speed from 10 m/s (v1) to 12.5m/s (v2) and 12.5m/s (v2) to 15.0m/s represented in Fig. 2a.

(1) ∆µm

v1–2

(2) ∆µm

v2–3

Fig. 2a. Speed sensitivity -Variation in ∆µmv of the

composites corresponding to the sliding speed transitions (1) from v1 to v2 (10.0 to 12.5 m/s) and (2) from v2 to v3 (12.5 to 15.0 m/s). Similarly sensitivity of µm to pressure was analyzed by calculating the extent of decrease in µm with respect to transitions in pressure from one level to the successive level. These were designated as ∆µm

p1–2

and ∆µmp2–3, which were the decrements in µm with

the increase in pressure from 2MPa (p1) to 3MPa (p2) and 3MPa (p2) to 4MPa (p3) m/s respectively at a constant speed and the data are represented in Fig. 2b.

(1) ∆µm

p1–2

(2) ∆µm

p2–3

Fig. 2b. Pressure sensitivity- Variation in ∆µmp of the

composites corresponding to the pressure transitions (1) from P1 to P2 (2 to 3 MPa) and (2) from P2 to P3 (3 to 4 MPa).

Operating variables Experimental design Pressure (MPa) 2, 3 and 4 Linear Speed 10.0/36, 12.5/45 and 15.0/5


The data on wear volume as a function of PV value (pressure and velocity) are are shown in Fig.3. Wear was calculated after each experiment of 25 cycles and plotted individually. Studies on worn surfaces by SEM are shown in Fig.4. 3. Discussion on Results Physical and mechanical properties As seen from the Table 2, it was observed that the density of composites showed increasing trend because of addition of brass fibers which is heavier than the filler Barite. Acetone exaction indicates amount of uncured resin in the composites which was negligible in all the composites except B0 where it was highest (1.1%).

Table 2. Physical and mechanical properties of the selected composites

Void contents increased slowly with increase in brass fibers because of larger size of brass fiber as compared to barite powdery particles. The heat swelling also showed the same trend as expected because of higher expansion of brass fibers. Mechanical properties, (tensile and flexural) decreased with increase in contents of brass basically because of increase in filler content of bigger size with simultaneous decrease in a filler content of smaller size. Similarly thermal conductivity, diffusivity and specific heat increased with increasing amount of brass fibers. Similar trends were observed in our earlier investigations on similar series of composites with increasing steel wool [11]. Inclusion of metal fibers with reduction of equivalent amount of powdery filler (barite in both the cases) reduced the strength, modulus and hardness while increased the density, void contents, heat swelling and thermo-physical properties. The water absorption and amount of uncured resin, however, did not show

Fig. 3- Histogram showing the variation in wear of the composites at various PV (pressure & velocity) values. regular trends in both the series of steel wool and brass fibers. 3. 1. Load- speed sensitivity studies on Reduced Scale Prototype (RSP) Tribo-performance As seen from Table 5, it was observed that pad temperature was higher than the disc temperature. Moreover, with increase in speed and pressure temperature of the surfaces of discs and pads increased as expected.

It was, however, interesting to note that as brass contents in composites increased, the temperature on surfaces of discs and pads decreased slowly, in general with few exceptions which are underlined in Table 5. In case of pad surface exceptions are very few. Thus the basic aim of the research work to minimize the temperatures at the surface of pad and disc with the help of metal contents was fulfilled. However, it was also observed that the interface temperature is not the only criteria to decide the overall performance of the composites. Had it been true, B3 should have shown the best performance in wear mode also. This confirms that the metal contents which are added to control the interface temperature interfered with other performance properties also making the formulation-design task more difficult. Hence selection of ingredients in formulation in optimized contents becomes more imperative. As seen in Table 6, it was observed that with increase in pressure μ of composites decreased In case of speed increase, no fixed trends were observed because speed is known to have more complex influence on μ rather than the pressure. Some general trends emerged from this table as follows. • Friction behavior at lowest speed (10m/s) –B2 and

B1 showed higher μ and B3 showed lowest..

Properties B0 B1 B2 B3 Density (g/cc) 2.27 2.29 2.33 2.38 Acetone extraction (%) 1.18 0.40 0.35 0.60 Water absorption (%) 0.73 0.32 0.56 0.70 Void content (%) 1.98 2.08 3.03 3.52 Heat swelling (%) 1.47 1.78 2.23 2.52 Tensile strength (MPa) 11.6 10.9 10.0 8.2 Tensile Modulus (GPa) 0.42 0.35 0.39 0.33 Flexural strength (MPa) 37.3 29.6 28.9 27.2 Flexural Modulus (GPa) 7.35 6.31 5.97 5.77 Rockwell Hardness (R-scale)

118 116 116 114

Diffusivity (cm2/sec) x 4

92 100 105 108 Sp. Heat (J/kg K) 730 737 833 855 Conductivity (W/m K) 1.53 1.69 2.04 2.20


• Friction behavior at moderate speed (12.5m/s)–B3 started showing higher μ especially at higher pressures, followed by B2. B0 showed the lowest μ.

• Friction behavior at highest speed (12.5m/s)- B3 and B2 showed highest μ in general . B0 showed the lowest μ.

These trends indicate that inclusion of metal contents and hence TC of pads increased the μ of composites in general. Increase in TC reduced the interface temperature of the tribo-couple and hence the fading tendency in μ. Composite B0 which had no brass fibers hence showed very low μ as operating conditions became more severe (high P-V conditions). Composites B2 and B3 thus proved more efficient in protecting μ from lowering down under severe operating conditions.

Fig 2 indicates the sensitivity of μ towards operating conditions. For ideal materials such curves should be as straight, parallel and very near to x-axis confirming least sensitivity towards operating parameters. As seen from Fig 2a1, for lower to moderate speed transition, performance in general followed the order; B1 =B0 > B3 > B2 For moderate to high speed transition (Fig 2a2), B3 >B2 > B0> B1 For sensitivity to pressure as seen from Fig 2b1, performance order in general was; B3 =B0 > B2 > B1 For moderate to high speed transition (Fig 2b2), performance order was; B3 >B1 > B2 > B0

Table 5. Increase in the temperature of surfaces of discs and composites tested on RSP under various conditions of loads and speeds

Table 6. Stabilized coefficient of friction (μ) for the selected composites.

Thus, overall composite B3 appears to be the best from least sensitivity of friction coefficient to load –speed variation. B2 was moderate performer in this respect. Figure 3 shows wear of composites at different PV values. In general it was observed that B0 and B3 showed higher wear than other two composites. B2 appeared to be most wear resistant in most of the cases. SEM studies

Surfaces of specimens worn under highest PV condition (60 MPa-m/s) were selected for SEM studies. The micrographs are arranged in the order of their increase wear performance (B2> B1 > B3 > B0).

As seen from micrograph (4B0a-X 500), the surface of B0 (which showed lowest wear performer and highest wear) was fully covered with secondary plateaus. The micrographs 4B3a and 4B3b (BS) (both X 500) indicated the dominance of secondary plateaus which support higher wear of B3. Micrograph 4B3c (BS) (X 500) at different location confirms the existence of various ingredients and few primary plateaus. Micrographs 4B1a and 4B1b (BS) (both X 500) confirms more amount of primary plateaus supporting its good wear resistance. Surface of B2, which showed lowest wear, however, was distinctly different from other surfaces as seen in micrographs 4B2a and 4B2c. SEM and BS images

Disc Temperature (0C) Parameters /samples Speed-10 m/s (rpm800) Speed-12.5 m/s (rpm1000) Speed-15 m/s (rpm 1200)

2MPa 3MPa 4MPa 2MPa 3MPa 4MPa 2MPa 3MPa 4MPa B0 56 64 67 63 83 83 75 107 108 B1 55 65 76 71 84 104 83 106 115 B2 53 62 63 65 89 95 82 110 110 B3 48 57 67 66 83 91 74 91 103 Pad Temperature (0C)

B0 108 122 114 113 152 191 138 179 239 B1 100 114 144 138 168 203 165 210 251 B2 65 112 123 114 180 190 158 206 243 B3 80 103 118 116 174 183 139 201 234

Speed-10 m/s (rpm800) Speed-12.5 m/s (rpm1000) Speed-15 m/s (rpm 1200) Parameters/ Samples 2MPa 3MPa 4MPa 2MPa 3MPa 4MPa 2MPa 3MPa 4M

B0 0.401 0.388 0.372 0.410 0.377 0.357 0.409 0.334 0.31B1 0.431 0.394 0.381 0.449 0.382 0.379 0.432 0.359 0.34B2 0.410 0.397 0.382 0.440 0.396 0.377 0.450 0.365 0.35B3 0.396 0.385 0.376 0.417 0.396 0.381 0.423 0.362 0.35


(4B2a and 4B2b –X 100) indicate wavy nature of the surface and micro-voids. The primary plateaus (left corner) are also seen. The secondary plateaus are small in size and uniformly distributed. 4. Conclusions

Based on studies conducted on composites with increasing brass contents under various operating conditions in P-V sensitivity mode, following conclusions were drawn • Increase in brass contents led to deterioration in

strength and modulus, hardness etc and increase in density, void contents, heat swelling, water

• absorption, specific heat, thermal diffusivity and conductivity (TC).

• Higher TC lead to lower interface temperature. μ decreased with load. However, no fixed trends were observed for speed.

• Composite with 8 % (B2) proved to have best combination of performance properties.

• It was thus finally concluded that enhancement in TC of a composite does not necessarily improve its fade behavior and overall tribo-performance. Tribo-performance, in general, improves up to a certain level of brass contents (in this case up to 8%) and then drops down. Various mechanisms were found to be responsible for these observations.

4B0a 4B3a

4B3b 4B3c

4B1a 4B1b


Fig. 4. SEM studies on the surfaces of composites worn under highest PV condition (60 MPa.m/s) in RSP testing References [1] J. Bijwe, “Composites as friction materials:

Recent developments in non-asbestos fiber reinforced friction materials-A review”, Polym. Compos., 18, 3 (1997), 378-396.

[2] D. Chan and G. W. Stachowiak, “Review of automotive brake friction materials”, Proc. Instn. Mech. Engrs. Part D: Jr. of Automobile Engineering, 218 (2004), 953-966.

[3] F. Dong, F. D. Blum and L. R. Dharani, "Lapinus fiber reinforced phenolic composites: flexural and frictional properties," Polym. & Polym. Compos., 4 (1996) 155-161.

[4] B. K. Sathapathy, “Performance Evaluation of Non-Asbestos Fiber Reinforced Organic Friction Materials “Ph .D. Thesis, Indian Institute of Technology Delhi, (2003).

[5] L. Gudmand-Hoyer, A. Bach, G.T. Nielsen and P. Morgen, “Tribological properties of automotive disc brakes with solid lubricants”, Wear, 232 (1999) 168-175.

[6] P. Gopal, L. R. Dharani and F. D. Blum, "Load, speed and temperature sensitivities of a carbon fiber reinforced phenolic friction material," Wear 181-183 (1995) 913-921.

[7] S.J. Kim, M.H. Cho, D.S. Lim, and H. Jang, “Synergistic effect of aramid pulp and potassium titanate whiskers in the automotive friction materials”, Wear, 251 (2001) 1484-1491.

[8] Y. Handa and T. Kato, “Effects of Cu powder BaSO4 and cashew dust on the wear and friction characteristics of automotive brake pads”, Tribol. Trans., 39, 2 (1996) 346-353.

[9] H. Jang, K. Koa, S. J. Kim, R.H. Basch, and J.W. Fash, “The effect of metal fibers on the friction performance of automotive brake friction materials”, Wear 256 (2004) 406-414.

[10] X. Qu, L. Zhang, H. Hing, and G. Liu, “The effect of steel fiber orientation on Frictional properties of asbestos-free friction materials”, Polymer Comp., 25, 1 (2004), 94-101.

[11] J. Bijwe, M. Kumar, “Optimization of Steel Wool Contents in Non-Asbestos Organic (NAO) Friction Composites for Best Combination of Thermal Conductivity and Tribo-performance”, Wear 263 (2007) 1243-1248

4B2a 4B2b

Reinforcement of Graft Copolymers of Binary Vinyl Mixtures of Methyl Meth Acrylate (MMA) onto Saccharum Cilliare Fiber

with Urea-Formaldehyde Matrix and Evaluation of their Mechanical Properties

A.S. Singha and Anjali Shama

Department of Applied Sciences, National Institute of Technology (Deemed University) Hamirpur– 177005 (H.P.), INDIA

E-mail: [email protected] Abstract

Present investigation has revealed that binary vinyl monomer mixtures of Methyl meth acrylate (MMA/AN, MMA/MA, MMA/AA, MMA/AAm) were graft copolymerized onto Saccharum Cilliare fiber in air in the presence of ferrous ammonium sulphate - potassium persulphate (FAS - KPS) as redox initiator. Various reaction parameters such as solvent, time, temperature, pH and concentration of monomer and initiator were optimized in order to get maximum graft yield. Raw fiber and its graft copolymers were then subjected for evaluation of different properties and urea- formaldehyde matrix based composites were prepared by using graft copolymers as reinforcing material. Then composites were tested for different mechanical properties such as wear resistance, tensile strength and compressive strength. It has been observed that the properties such as wear resistance, tensile strength and compressive strength were increased for the composites reinforced with graft copolymers of Saccharum. cilliare fiber.

1. Introduction

In recent years polymers are being considered as an ideal material for meeting the demands in specialized fields. Commonly available natural and synthetic polymers although having large advantages yet are not suitable for use under hostile conditions which may include action of acids , bases, solvents etc. Hence there is a need to modify these polymers so that these could be used in specialized fields. One of the best method to incorporate new functionalities into these polymers is the graft copolymerization. Modification of starch, fibrous proteins and cellulosics by using variety of initiating systems has been extensively studied by various workers [1-7]. These cellulosic materials are being preferred for synthesis of natural fiber reinforced composites. The composites are the material composed of two or more distinct components. Composites are divided into two basic forms: composite materials and composite structure. Composite materials are composed of reinforcing structures which enhance the properties and is surrounded by a continuous matrix. It has been observed that natural fibers such as flax on reinforcement improved the mechanical properties of polystyrene composites [8-9]. Fibers are the most important class of reinforcements, as they

satisfy the desired conditions and transfer strength to the matrix constituent, influencing and enhancing their properties as desired. These natural fillers are especially being required since the production of composites using natural substances as reinforcing fillers is not only inexpensive but also able to minimize the environmental pollution caused by the characteristic biodegradability [10], enabling these composites to play an important role in resolving future environmental problems .With emphasis of environmental awareness and consciousness, academic and industrial needs for developing environmentally friendly composite materials have recently been increasing with significant attraction, based on renewable resources like natural fibers as alternatives for glass fiber reinforcement in traditional glass fiber-reinforced polymer matrix composites [11-15]. In 1908, the first composite material was applied for the fabrication of big quantities of sheets, tubes and pipes in electrochemical usage (paper or cotton as reinforcement in sheets made of phenol or melamine-formaldehyde resin). In 1896, for example, aero-plane seats and fuel tanks were made of natural fibers with a small content of polymeric binders [16].The benefits offered by lingo cellulosic materials include making the final product light [17], decreasing the wear of the machinery used, low cost,

Reinforcement of Graft Copolymers of Binary Vinyl Mixtures of Methyl Meth Acrylate 195

biodegradability [10]. Joshi et al [18] proposed that composites reinforced with natural fiber are likely to be environmentally superior to those reinforced with glass fiber in many applications. It has been reported that polypropylene (PP) composites reinforced with wood fiber have properties similar to glass fiber reinforced PP composites [19]. Since trough grafting it is easy to impart new properties into the polymers so an attempt has been made to graft co-polymerize Saccharum cilliare fiber with methyl methacrylate MMA and graft co-polymers thus prepared are used as reinforcement in the preparation of urea-formaldehyde (UF) composites. 2. Experimental Methods Graft copolymerization

The 0.5 gm of natural fiber S. Cilliare was immersed in a known amount of distilled water for 24 hrs. A known amount of initiator (FAS-KPS ) was added to the reaction medium which was followed by the addition of known amount of binary monomer mixtures (MMA/AN, MMA/MA, MMA/AA and MMA/AAm) containing MMA as principal monomer (for which optimum reaction conditions were initially worked out) into the reaction flask containing fiber. The homopolymer formed was removed by extraction with suitable solvents. Acetone was used to remove homopolymer of MMA and MA, DMF was used to remove polyacrylonitrile (PAN) and homopolymers of AA and AAm were removed by extraction with hot water. The graft

copolymers were then dried in an hot air oven at 50oC to a constant weight. Percentage of grafting (Pg) and percentage efficiency (Pe) were then calculated by the method reported in the literature [20]. Optimization of different reaction parameters for grafting of MMA onto S. Cilliare fiber The optimization of various reaction parameters like amount of the solvent, reaction time, temperature, ratio of the initiator (FAS-KPS ratio) and concentration of the monomer was carried out for graft co-polymerization of MMA onto S. Cilliare fiber. The optimum conditions for maximum graft yield (58.0 %) were : solvent-125 ml; time-90 min.; temperature-350C; FAS:KPS-1:1 and monomer- (7.35×10-3 mol/L). Synthesis of urea – formaldehyde resin Urea and Formaldehyde solution in different molar ratio (1.0:.1, 1.0 :0.3, 1.0 : 1.5, 1.0 : .0.7, 1.0 : 0.9) were mixed with stirrer. Then 50% solution of NaOH was added till the pH of mixture adjusted 7.5. Initial temperature maintained between 50- 60 0C till dimethylol urea was formed (ppt are present on flask). Then the contents were cooled and acetic acid (50%) was added till pH 6.5. Continue heating at the same temperature till complete resinification. Then resin was cooled at room temperature and ammonium chloride was mixed. Then contents were poured into greased aluminum chambers of cross – section 5x5 mm and specific length.

Table 1. Optimization of U-F resin for mechanical properties

SR. No. Ratio of U-F Tensile strength (N)/

extension (mm) Compressive strength (N)/

compression (mm) Weight loss (gm) at 1,2, 3kg

1. 0.1:1.0 8/.221 10/.662 0.0291, 0.0384, 0.0493 2. 0.3:1.0 32/1.310 690/2.55 0.0052, 0.0079, 0.010 3. 0.5:1.0 65/2.0 810/3.3 0.0029,0.0042,0.0074 4. 0.7:1.0 45/.887 360/2.551 .0068, 0.0083, 0.0106

5. 0.9:1.0 30/1.3 220/1.1 0.0097, 0.0124, 0.0169 The resin was Characterized by FTIR, Scanning Electron Micrographs, Thermo gravimetric analysis (TGA), Differential thermal analysis (DTA) and Derivative thermogravimetry (DTG) techniques. Synthesis of Composites

The best ratio optimized was 1.0:0.5, which

show better tensile, compressive and wear resistance as compared to other ratio. Saccharum cilliare fiber was dried at 100 0C for 24 h to remove the moisture content Then fiber was

mixed with resin and the mixture were put into specially made chambers. These chambers were then cured at 100-1500C in compression moulding machine. The composites prepared were cut into different lengths (8 cm for tensile strength, 4 cm for compressive strength, 5 cm for wear resistance and 5 x 5 mm cross- section).


The cured samples were then subjected to various mechanical studies. 3. Results and discussions

As reported earlier [20] in polymeric materials containing cellulose, C2, C3 and C6 hydroxyl groups and C-H groups are the active sites for the incorporation of polymeric chain through graft co-polymerization in S. cilliare fiber. Potassium persulfate takes part in the redox reaction with Fe2+, which is generated in the reaction mixture as per the equation given below: Fe2+ + -O3S-O-O-SO3

- ---- Fe3+ + SO42- + SO4

-* In aqueous medium OH* are generated by

interaction of SO4-* with H2O which is

responsible for graft copolymerization. Graft copolymerization onto S. Cilliare is supposed to take place as per the mechanism given below : SO4

-* + H2O --- HSO4- + OH* (1)

SO4-*+ScCell-OH ---- HSO4

- + ScCell-O* (2) ScCell-OH + OH* ----- ScCell-O* + H2O (3) nM M + OH* -- OH-M* ---- OH-(M) *n+1 (4) OH-(M) * n+1 + Sc Cell-OH ----- Sc Cell-O* + OH-(M) n+1 -H (5) nM Sc Cell-O* + M ----- Sc Cell-O-M* ----- Sc Cell-O-(M) *n+1 (6) Sc Cell-O-(M) *n+1 + OH- (M) *n+1 ---- Sc Cell-O-(M)2n+2-OH (7) Sc Cell-O-(M) *n+1 + Fe 3+ --- --- Sc Cell-O-(M) n+1 + Fe 2+ (8) HO – (M*) n+1 + M* ----------- HO – (M) n+1 - M (9) Where Sc Cell-OH = Saccharum cilliare cellulose and M = Monomer Optimization of Urea– Formaldehyde resin Wear Resistance

It has been observed that wear rate of

samples of ratio 1.0:0.5 was less as compared to any other samples. Loss of material was due to abrasion and friction of samples with disc (Fig.1).

Compressive Strength

It is evident from the table that the sample of

ratio 1.0:0.5 undergoes more compression but the load bearing capacity was less. Where as samples of ratio1.0: 1.5 could bear a load of 810 N at a compression of 3.3 mm.

Tensile Strength It has been observed that U-F samples of

ratio 1.0:0.5 bear more load at a particular applied load as compared to samples of other ratios. This ratio (1.0:1.5) could bear a load of 65 N force with an extension of 2.0 mm. On the other hand, samples of other ratios bear low loads. As evident from the table samples of ratio 1.0:0.5 show maximum tensile and compressive strength. Moreover, at this ratio wear rate was also very less. Therefore, the 1.0:0.5 ratio was taken for further preparation of composites of urea-formaldehyde resin with the graft copolymers of MMA and its binary mixtures Scanning electron micrographs

Scanning Electron Micrographs of raw

fiber were on Leo 435 VP. On comparing the scanning electron micrographs of raw fiber, urea formaldehyde resin and its graft copolymer with MMA (fig. 1.), it has been found that upon graft co-polymerization, a considerable amount of monomer gets deposited onto the fiber backbone.

Fig. 1. (a) SEM of raw fiber

Fig. 1. (b) SEM of urea formaldehyde

Fig. 1. (c) SEM of graft copolymer


Thermal behavior of ungrafted U-F matrix, S. cilliare fiber and its graft copolymer with MMA

Thermo-gravimetric analysis (TGA) of raw fiber and grafted fibers was studied as a function of % weight loss with the increase in temperature (Table 2.). In case of raw fiber, in the beginning depolymerization, dehydration and glucosan formation took place between the temperature range of 260C to 2010C followed by the cleavage of C-H, C-C and C-O bonds. Initial decomposition (IDT) temperature is 2620C and final decomposition temperature is 4610C breakdown of the crystalline region took place. On the other hand, in case of grafted fiber it is single stage decomposition and the initial decomposition temperature is 2700C, the final decomposition of the grafted fiber took place at 3920C with the total loss of crystalline structure. Initial decomposition (IDT) temperature, final decomposition temperature (FDT) and decomposition temperature (DT) at every 10% weight loss are presented in the Table 3. The percent residue of grafted fiber left is 2.06% while for S. cilliare fiber is 5.52%

As shown in the Table 3. Differential Thermal Analysis (DTA) of raw fiber shows

exothermic peaks at 3280C (81µV), 4330C (203µV) and there has been a continuous exothermic combustion of the sample at the furnace temperature in the presence of atmospheric oxygen, which indicates the complete breakdown of C-C and C-O bonds of the crystalline region with total evolution of -6882 mJ/mg of energy (∆H = -ve). In case of graft copolymer, a continuous exothermic rise in temperature has been observed. The exothermic peak is observed at 3810C (271µV) shows the loss of amorphous and crystalline structure of grafted fiber in presence of atmospheric air, with total evolution of -4369 mJ/mg of energy (∆H=-ve).

Further DTG analysis also shows that grafted sample exhibit double decomposition behavior. The first peak is observed at 3510C with maximum decomposition of 3.102mg/min. Similarly second peak is observed at 3810C with maximum decomposition of 0.83mg/min, whereas in case of raw fiber first peak is observed at 3130C with maximum decomposition of 0.732 mg/min and second peak is observed at 4290C with maximum decomposition of 1.159 mg/min (Table 4.).

Table 2. Thermo-gravimetric analysis of Saccharum cilliare fiber, urea-formaldehyde resin and composite of Poly-g- MMA

Sample IDT (0C)

FDT (0C) DT (0C) at every 10 % weight loss

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

% Resid

ue Raw fiber 262 461.0 10 222 275 292 305 319 337 375 420 439 5.52 Poly-g- MMA 270 392.0 40 242 284 302 320 344 350 362 380 396 2.06

U-F resin 1780 530.0 56 175 210 250 268 277 299 340 530 12.41 Table 3. DTA of raw S. cilliare fiber and its graft copolymer

Sr. No. Sample Exothermic peak at Temperature (µV) 1. Raw fiber 328.0 (81) ; 433.0 (203) 2. Poly-g- MMA 381.0 (271)

3. U-F resin 271.0 (27.9) Table 4. DTG of raw S. cilliare fiber and its graft copolymer

Sr. No. Sample Exothermic peak at Temperature (mg/min)

1. Raw fiber 313.0, (.732), ;429.0 (1.159) 2. Poly-g- MMA 351 (3.102), 381 (0.83) 3. U-F resin 175 (0.298), 272 (1.172)


The IR spectra of raw Saccharum cilliare fiber showed a broad peak at 3370 cm-1 due to bonded OH groups and at 2922.4, 1438 and 1043 cm-

1respectively due to –CH2, C-C, and C-O stretching. While in case of grafted fiber the additional peak at 1735.6 cm-1 was seen due to the presence of carbonyl group of vinyl monomer (MMA). In case of U-F matrix peak was found at 1020.5 cm-1 due to ether linkage, 1637.9 cm-1 due to C=O group, 2927.2 cm-1 due to C-H stretching and at 3365.6 cm-1 due to –NH2 group of the matrix. Effect of concentrations of Binary Monomer mixtures on percent grafting

It has been observed that low percent grafting (49 %) has been observed with binary monomer mixture of MMA/MA (7.35 ×10-3 and 2.08 ×10-3 mol/L). The percentage of grafting has been found to be 65.0 % with MMA/AN (7.35 × 10-3 and 2.83 ×10-3 mol/L) this is expected as unlike MA, AN is a strong acceptor monomer and creates lot of free radical sites which results in higher graft yield.

When MMA/AAm was used as binary monomer mixture, the Pg has been found to be 59.0 % (7.35 ×10-3 and 1.32 ×10-3), Whereas with MMA/AA the maximum percent grafting has been found to be 57 % (7.35 ×10-3 and 4.21 ×10-3 mol/L. The decreased rate of grafting with acrylic acid (AA) occurred due to the fact that AA has got high affinity for the solvent (water) unlike AAm and it exists in associated form. Due to its association as dimer it produces lesser number of free radicals and result in lower graft yield. Effect of reinforcement on the Mechanical properties of the urea-formaldehyde composites using raw Saccharum Cilliare fiber and graft copolymers Wear Test

It is evident from the Fig that wear rate of urea-formaldehyde matrix decreases appreciably as it is reinforced with Saccharum cilliare fiber. Also it has been found that reinforcement with the graft copolymers of MMA/MA shows better wear resistance as compared to other graft copolymers of MMA, raw fiber and urea-formaldehyde matrix. Loss of material is due to abrasion which further enhances on load increase and frictional heat effects. Thus it is quite

evident that wear rate is reduced when urea-formaldehyde matrix is reinforced with the fiber Saccharum cilliare (Fig.2).

Effect of raw and graft copolymers ofMMA (air) onto S. cil liare fiber on wear rate reinforced with UF

0

0.0021

0.0042

0.0063

0.0084

0.0105

F/Uraw MMA

MMA/AN

MMA/AA

MMA/AAm

MMA/MA

Wei

ght l

oss

(gm

)

1kg2kg3kg4kg

Fig. 2.

Effect of graft copolymeres of MMA (air) on Tensile strength reinforced w ith UF

0

100

200

300

400

500

600

0 1 2 3 4

Extension (mm)Fo

rce

(N)

F/UrawMMAMMA/ANMMA/AAMMA/AAmMMA/MA

Fig. 3.

Effect of raw and graft copolymers of MMA (air) onto S.cilliare fiber on compressive strength reinforced with UF

0

200

400

600

800

1000

1200

0 1 2 3 4 5

Compression (mm)

Forc

e (N

)

F/UrawMMAMMA/ANMMA/AAmMMA/AAMMA/MA

Fig. 4.

Tensile Strength

It has been observed that the tensile strength of graft copolymers of MMA/MA breaks after extension of 1.0 mm on applying force of 478.3 N shows better results as compared to the other graft copolymers. Whereas urea-formaldehyde matrix breaks on applying 65 N force with the extension 2.6mm (Fig.3). Compressive Strength

It has been found that graft copolymer of MMA/MA shows better compressive strength (1120 N) as compared to composite with raw (324.8 N) and urea-formaldehyde matrix (265.7 N) (Fig.4).


4. Conclusions

It has been found that upon graft copolymerization of methyl methacrylate onto natural fiber Saccharum Cilliare, reinforced with urea-formaldehyde matrix the mechanical properties like tensile strength, compressive strength and wear resistance has been found to be improved. It has been observed that composites of graft copolymer of MMA/MA shows better results. Also, it has been observed during thermo gravimetric analysis that the difference in FDT and IDT for grafted fiber is found to be less than that of raw fiber. Therefore rate of decomposition of raw fiber is slower as compared to the modified fiber, indicating that raw fiber is thermally more stable than modified fiber. References [1] I.. M.. Trivedi . and P.C. Mehta Cell. Chem.

Technol 9 (1975) 583. [2] M. M. Haque, M. Habibuddowla and A. J.

Mahmood .J. Polym. Sci. Chem., 18 (1980) 1447.

[3] M. Misra, A. K. Mohanty and B.C. Singh J. Appl. Polym. Sci. 33 (1987) 2809.

[4] G. S. Chauhan, B. S. Kaith and L. K. Guleria Res. J. Chem. Environ. 4 (2000) 35.

[5] G. S. Chauhan , H. Lal , A. S. Singha and B. S. Kaith .: Ind. J. Fiber Text. Res., 26 (2001) 302.

[6] G. F. Fanta, R. C. Burr, C. R. Russell, C. E. Rist, J. Appl. Polym. Sci. 15 (1971) 1889.

[7] B. N. Misra, R. Dogra, I. K. Mehta and A. S. Singha, Die Angewandte Markromolekulare Chemie, 90 (1980) 83.

[8] B. S. Kaith, A. S. Singha, D. K. Dwedi, Sanjeev Kumar, D. Kumar and A. Dhemeniya, International Journal of Plastic Technology, 7 (2003) 119-125.

[9] D. K. Dwedi , A. S. Singha, ,Sanjeev Kumar and B. S. Kaith International Journal of Plastic Technology, 8 (2004) 299-304.

[10] HGB Premalal, H. Ismail, A. Baharin. Comparison of the mechanical properties of rice

[11] Mohanty AK, Misra M, Hinrichsen G. Biofibres, biodegradable polymers and biocomposites: An overview. Macromol Mater Eng; 276/277 (2000) 1–24.

[12] Mohanty AK, Misra M, Drzal LT. Surface modifications of natural fibres and performance of the resulting biocomposites: An overview. Comp Interfaces; 8(5) (2001) 313–43.

[13] Netravali AN, Chabba S. Composites get greener. Materialstoday 2003:22–9.

[14] Baillie C. Eco-composites. Comp Sci Technol; 63 (2003)1223–4.

[15] Wambua P, Ivens J, Verpoest I. Natural fibers: can they replace glass in fiber reinforced plastics Comp Sci Technol; 63 (2003) 1259–64.

[16] A. K. Bledzki, J. Izbicka and J. Gassan, Kunststoffe-Umwelt-Recycling, Stettin [Poland], September 27- 29 (1995).

[17] RE Jacobson, MS Engineer, DF Caulfield, RM Rowell, AR Sanadi. Recent developments in annual growth ligno-cellulosics as Reinforcing fillers in thermoplastics. In: Proceedings of 2nd Biomass Conference of the Americas: Energy, Environment, Agriculture and Industry, August 21–24, Portland OR. Golden, CO: National Renewable Energy Laboratory, (1995) 1171–1180.

[18] Joshi, S.V., Drzal, L.T., Mohanty, A.K., Arora, S., Composites: Part A, 35 (2004) 371. [19] Krishnan, M., Narayan, R., Mat., Res. Soc. Proceedings, 266 (1992).

[20] B. S. Kaith and A. S. Singha and Sanjeev K. Sharma, J. Polym. Mater., 20 (2003) 195.

Applications of Waste Bio-Mass as Reinforcing Material in P-R-F Based Composites and Study of their Mechanical and Thermal Behavior

A.S. Singha and Ashwarya Jyoti Khanna

Department of Applied Sciences, National Institute of Technology, Hamirpur-177 005 (HP), India. E-mail: [email protected]

Abstract

Present investigation has revealed that Phenol-Formaldehyde resin in 1.0 : 1.5 ratio exhibits optimum

mechanical behaviour whereas in case of phenol-resorcinol- formaldehyde resin, the best mechanical behaviour was shown by 1.0 : 1.0 : 1.5 ratio .FTIR, SEM and TGA/DTA of these samples were taken. It has been observed that tensile strength, compressive strength and wear resistance of the P-R-F resin increases many folds by reinforcing with pine needles. However, reinforcing of these resins with Pine-needles in three different forms i. e. particle size, short fibre and continuous fibre and evaluation of their mechanical properties showed that particle reinforcement is more effective than short and continuous fibre reinforcement. These results were further supported by the thermal studies and SEM of these composites.

1. Introduction

In 1908, the first composite materials were applied for the fabrication of big quantities of sheets, tubes and pipes in electro-technical usage (Bledzki et. al., 1995) since before all resources for the production of commodities and technical products were derived from wood and natural fibres. Composites, the wonder material, composed of at least two elements, working together to produce material properties that are different to the properties of those elements on their own. Most composites consist of a bulk material(matrix) and a reinforcement of some kind, added to increase the strength and stiffness of the matrix, is usually in the fibre form. These days, material scientist all over the world, have focused their attention on composites reinforced with either synthetic or natural fibers. Researchers in different laboratories have studied the reinforcement of composites with different natural fibres such as flax, hemp, pineapple etc. (Kohler and Wedler, 1995; Mieck et. al., 1994; Mukherjee and Satyanarayana, 1986 ). India endowed with the abundant availability of natural fibrs such as jute, coir, sisal, pineapple, ramie, bamboo, banana, pinus etc. has focused on the development of natural fibre composites to explore value added application avenues (Kaith et al., 2003, 2004). Today the renaissance of these materials as reinforcing fibres in technical applications is taking place mainly in automobile industry and for packaging purposes such as egg-cartons and fruit-cartons. Natural fibre reinforced composites could be used for the manufacturing of car parts,

door panels etc. (Wittig, 1994; Mieck and Reβ mann, 1995).

Pinus, with over 100 species, is the largest genus of conifers and the most wide-spread genus of trees in the Himalayan range. Many pine trees are fast growing species, tolerant of poor soils and relatively arid conditions. Important pine products include wood, turpentine and edible seeds. But, pine needles are the most common waste bio- product of pine trees found in the jungles of Himachal Pradesh. Keeping in view the easy availability and the abundance of this waste bio-mass, we have proposed to use these pine needles as reinforcing material for the preparation of P-R-F based composites. Moreover, pine needles have more tensile strength at break-point as compared to other natural fibre.

Since phenolic resins are hard, rigid and strong materials and they have excellent heat, moisture, chemical and abrasion resistance so in the present work efforts have been made to develop Phenol-Resorcinol-Formaldehyde matrix, by mixing resorcinol in P-F resin and reinforcing this matrix with raw pine needles of different sizes and to study their mechanical and thermal behaviour. 2. Results and discussions

2.1. Optimization of Phenol – Formaldehyde Resin Wear Resistance: It has been observed that wear rate of samples of ratio 1.0:1.5 was less as compared to any other samples. Wear rate was

Applications of Waste Bio-Mass as Reinforcing Material in P-R-F Based 201

0.0031g at 1.0 kg load, 0.0058g at 2.0 kg load and 0.0073g at 3.0 kg load. Loss of material was due to abrasion and friction of samples with disc (Fig.1). Compressive Strength

It is evident from the (Fig. 2) that the sample of ratio 1.0:2.5 undergoes more compression but the load bearing capacity was less. Whereas samples of ratio 1.0: 1.5 could bear a load of 1748 N at a compression of 2.5 mm. Tensile Strength

It has been observed that P-F samples of ratio 1.0 : 1.5 bear more load at a particular applied load as compared to samples of other ratios. This ratio (1.0:1.5) could bear a load of 380.8 NF with an extension of 2.5 mm (Fig. 3). On the other hand, samples of other ratios bear low loads. As is evident from (Figs.1-3) samples of ratio 1.0:1.5 show maximum tensile and compressive strength. Moreover, at this ratio wear rate was also very less. Therefore, this ratio was taken for further preparation of phenol resorcinol formaldehyde resin. 2.2 Optimization of Phenol- Resorcinol- formaldehyde Resin Wear Test

Wear rate of P-R-F resin samples of all ratios has been found to be high as compared to P-F resin. P-R-F samples of ratio: 1.0:2.0:1.5 and 1.0:2.5:1.5 showed a maximum wear rate. Whereas, the sample of ratio: 1.0:1.0:1.5 exhibited less wear rate at all loads (Fig. 4).


As is evident from (Fig. 5) that compressive strength of P- R- F resin with ratio: 1.0:1.0: 1.5 could bear maximum force of 449.5 N at a compression of 1.5 mm. However, the sample of ratio: 1.0:1.5:1.5 showed a compression of 2.0 mm but it failed to bear higher loads.

Tensile Strength

Tensile strength of P-R-F resin samples has been found to be low as compared to the samples of P-F resin. Sample of ratio 1.0:1.0:1.5 show maximum tensile strength at a force of 202.8 N with an extension of 2.5 mm. Whereas, other samples either failed to bear higher load or had

low extension capacity (Fig. 6). As evident from Figs. (4-6), mechanical properties of P- R- F resin samples with ratio: 1.0:1.0:1.5 were found good as compare to other samples. So, this ratio was taken further to prepare composites.

2.3 Effect of Reinforcement on the Mechanical Properties of P- R-F Based Composites Compressive Strength

Compressive strength of P-R-F matrix has been found to increase as reinforcement with pine needles. It has been found that on particle reinforcement compressive strength increase by 1.7 times, 1.32 times with short-fibre reinforcement and 1.2 times on continuous fibre reinforcement (Fig. 7). Tensile Strength

It has been observed that composites with particle reinforcement showed more tensile strength which was followed by short fibre and continuous fibre reinforced composites (Fig. 8).

Wear Test

As evident from (Fig. 9) that wear rate of P-R-F matrix decreases appreciably as reinforcement with pine needles. It was observed that particle reinforcement decreases the wear rate by 1.5 times whereas; short-fibre reinforcement decreases the wear rate by 1.2 times. Fig.1. Wear rate of PF resin samples.

Fig.2.Compressive strength results of PF resin.

Fig.3. Tensile strength results of PF resin.

Fig.4. Wear rate of PRF resin samples.

0

0.002

0.004

0.006

0.008

0.01

0.012

0 2 4

LOAD (kg)

WEI

GH

T LO

SS (g

)

PF-1:1PF-1:1.5PF-1:2.0PF-1:2.5PF-1:3.0

0

500

1000

1500

2000

0 1 2 3C OM PR ESSION ( mm)

FOR

CE

(N)

PF-1:1.0PF-1:1.5PF-1:2.0PF-1:2.5PF-1:3.0

0100200300400

0 1 2 3

EXTENSION (mm)

FOR

CE

(N)

PF-1:1.0PF-1:1.5PF-1:2.0PF-1:2.5PF-1:3.0

0

0.01

0.02

0.03

0.04

0 5

LOAD (kg)

FOR

CE

(N)

PFR-1:1.5:0.5PFR-1:1.5:1.0PFR-1:1.5:1.5PFR-1:1.5:2.0PFR-1:1.5:2.5


Fig.5. Compressive strength results of PRF resin.

Fig.6. Tensile strength results of PRF resin.

Fig.7. Compressive strength results of PRF composites.

Fig.8. Tensile strength results of PRF composites.

Fig.9. Wear rate of PRF composites. Infra-red spectrophotometric studies

IR spectrum of Phenol– Formaldehyde resin sample showed a broad peak at 3404cm -1 due to OH groups19,20 and at 2927 and 1375 cm -1 arising from C-H stretchings19. Peak at 1020 cm-

1 was due to ether linkage and peak at 1601and 1475 cm -1 was due to phenyl nucleus 21. However, in case of Phenol-Resorcinol-Formaldehyde additional peaks at 1225 and 1087 cm -1 was due to –O- streching.

2.4 Thermal Behavior of P-F Resin, P-R-F Resin, Pine Needles and P-R-F Composites

It is quite evident from Table 1 that P-F resin (1.0:1.5) exhibits higher initial and final decomposition temperatures (IDT & FDT) as compared to P-R-F resin (1.0:1.0:1.5). This shows that three dimensional bonding structures in case of phenol-formaldehyde resin is more stable as compared that of phenol-resorcinol-formaldehyde resin. It has further been observed

that particle reinforcement slightly increases the FDT of the resin which is strongly supported by the mechanical behaviour of these various reinforced composites. These studies are further supported by differential thermal analysis (DTA) as the P-F resin has been found to exhibit exothermic peak at 749.4 oC with release of 20.4μ V energy (Table 2). TABLE 1. Thermogravimetric Analysis of P-F, P-R-F, PN and PN-rnf-P-R-F Composites Sr. No.

Sample Code

IDT (0C)

% wt. loss

FDT (0C)

% wt. loss

Final Residue (%)

1. PN 236.1 0.86 509.7 3.17 32.83 2. P-F 473.5 0.75 972.1 2.33 48.50 3. P-R-F 224.7 0.25 589.7 3.47 38.00 4. P-rnf-P-R-

F 238.5 1.37 593.1 3.90 12.16

5. SF-rnf-P-R-F

224.6 0.40 589.9 4.32 21.33

6. CF-rnf-P-R-F

222.9 0.03 589.6 4.95 17.00

Table 2. Differential Thermal Analysis of P-F, P-R-F, PN and PN-rnf-P-R-F Composites Sr. No.

Sample Code

Exothermic peaks 0C(μ V)

1. PN 330.7 [9.9]; 473.3 [29.1]

2. P-F 152.8 [8.8]; 507.9 [3.6]; 646.2 [8.8]; 749.4 [20.4]

3. P-R-F 505.9 [20.8] 4. P-rnf-P-R-F 492.0 [22.6] 5. SF-rnf-PRF 482.8 [30.3] 6. CF-rnf-PRF 509.1 [20.0]

Where P-F = phenol formaldehyde resin; P-R-F = phenol-resorcinol-formaldehyde resin; PN = pine needle; rnf = reinforced SEM studies SEM micrographs Fig.10.(a-e) of PF, PRF, PRF-Prnf, PRF-SFrnf and PRF-CFrnf shows the difference between the loaded and unloaded matrices and the matrix-fiber adhesion.

0

100

200

300

400

500

0 1 2 3

COMPRESSION (mm)

FOR

CE

(N)


0

50

100

150

200

250

0 1 2 3

EXTENSION (mm)

FOR

CE

(N)


0100200300400500600700800900

0 2 4

Compression (mm)

Forc

e (N

)

P :R :F -1 :1: 1.5(resin sample)P art ic lere info rcementSho rt f ibrere info rcement C o ntinuo us f ibrere info rcement

0

50

100

150

200

250

300

350

400

0 2 4

EXTENSION (mm)

FOR

CE

(N)

Sample R es in

P art icleR einfo rcementSho rt f ibreR einfo rcementC o nt inuo us F ibreR einfo rcement

00 . 0 0 20 . 0 0 40 . 0 0 60 . 0 0 8

0 . 0 10 . 0 120 . 0 140 . 0 160 . 0 18

0 . 0 2

0 1 2 3 4

LOA D ( k g)

P:R:F- 1:1:1.5(sample resin)ParticlereinforcementShort fibrereinforcementContinuous fibrereinforcement

Applications of Waste Bio-Mass as Reinforcing Material in P-R-F Based 203

Fig.10. (a) SEM of PF sample Fig.10. (b) SEM of

PRF sample

Fig.10. (c) SEM of PRF-Prnf composite

Fig.10. (d) SEM of PRF-SFrnf composite

Fig.10. (e) SEM of PRF-CFrnf composite 3. Conclusion

In case of mechanical behaviour particle reinforcement of the P- R- F resin has been found to be more effective as compared to short fibre reinforcement. This could be due to more matrix- particle interfacial interaction as compared to short fibre/ continuous fibre– matrix interfacial interaction. The mechanical behaviour

has been strongly supported by the thermal analysis Acknowlegement

We express our sincere thanks to faculty and technical staff of Department of Mechanical Engineering NIT Hamirpur for their assistance and help in conducting mechanical studies. Refrences [1] A. K.Bledzki, J. Izbicka J. and Gassan

Kunststoffe – Umwelt – Recycling, Stettin [Poland ], (1995) 27-29 September.

[2] B. S. Kaith, A. S. Singha, D. K. Dwedi, Sanjeev Kumar, D. Kumar and A. Dhemeniya, International Journal of Plastic Technology, 7(2003) 119-125.

[3] B. S. Kaith, A. S. Singha, D. K. Dwedi, Sanjeev Kumar, D. Kumar, and A. Dhemeniya, International Journal of Plastic Technology, 8(2004) 299-304.

[4] R. Kohler and M. Wedler, Landinfo, 3 (1995) 33.

[5] K.P. Mieck, A. Ncchwatal and C. Knobelsdorf, Melliand Textiberichte, 11(1994) 892.

[6] K. P Mieck, and T. Reβmann, Kunststoffe, 85(1995) 366.

[7] P. S. Mukherjee and K. G. Satyanarayana, Journal of Material Science, 21(1986) 51.

[8] W. Wittig Kunststoffe im Automobilbau, VDI- Verlag, DUSSELDORF (1994).

[9] M. Kacurakova, A. C. Smith, M. J. Gidley, R. H. Wilson and P. S. Belton, Carbohydr. Res. 337 (2002)1145

[10] F. W. Billmeyer, Jr., “Textbook of Polymer Sciences”, 3rd ed. Wiley, New York, 1984. p. 347.

[11] S. Borysiak and J. Garbarczyk, “Fiber and Textile in Eastern Europe”, 11 (2003) 44.

Characterization and Salt-Resistant Study of Gum Arabic and Methacrylic Acid Based Hydrogel

B. S. Kaith* and Shabnam Ranjta

*Department of Chemistry, National Institute of Technology (Deemed University), G. T. Bypass road, Jalandhar (Punjab) 144 011, India

Applied Chemistry Research Laboratory, National Institute of Technology (Deemed University) Hamirpur - 177005 (India)


Abstract

In this paper an attempt has been made to check the salt-resistant behavior of Gum arabic and methacrylic acid based hydrogel, using different salt solutions. Characterization of Gum arabic and GA-cl-poly(MAA) was carried-out with different techniques like FTIR spectroscopy, SEM and TGA-DTA techniques. The hydrogel under study was subjected to salt-resistant swelling in different salt solutions as a function of concentration, temperature and pH. The synthesized polymer exhibited salt resistant swelling trend in the following order: NaCl > ZnCl2 > MgCl2 > CaCl2 > BaCl2 > FeCl3.

1. Introduction

Hydrogels are three dimensional, hydrophilic, polymeric network structures which are capable of imbibing large amounts of water or biological fluids [1-2]. These networks are composed of homopolymer or co-polymer which are insoluble in water due to the presence of cross linking. These hydrogels respond to a greater extent to external stimuli like temperature [3] and pH [4]. There are numerous applications of these hydrogels such as in medical, pharmaceutical [5], agriculture [6] and in water treatment sectors [7].

Gum arabic is a natural polysaccharide comprising mainly of rhamnose (12-14%), arabinose (24-29%), galactose (36-42%) and glucuronic acid (16-17%) [8]. The gum is built upon a chain of D-galactose units with side chains of L- rhamnose or L-arabinose and D-glucuronic acid as terminal units [9]. Grafting and network formation of Gum arabic with different monomers and cross-linkers under variable synthetic conditions is an important method to develop a range of polymers with improved properties and applications [10].

In the present paper, a cross-linked three dimensional network of Gum arabic with methacrylic acid has been subjected to salt-resistant studies in different salt solutions such as NaCl, MgCl2, CaCl2, FeCl3, ZnCl2 and BaCl2 as a function of concentration, temperature and pH.

2. Experimental 2.1. Materials and Methods

Gum arabic GA (MERCK, Pvt. Ltd.), methacrylic acid MAA (LOBA Chemie Pvt. Ltd.), potassium persulphate KPS (S. D. Fine) and hexamethylene tetramine HMTA (LOBA Chemie Pvt. Ltd.) were used as received.

Weighing was done on electronic balance (LIBROR AEG-220 Shimadzu), FTIR spectra of the gels were recorded in KBr pellets using Perkin Elmer spectrophotometer, SEM of the samples were taken on Jeol Steroscan 150 Microscope and TGA/DTA studies were carried-out on Leinsis thermal analyzer in air at a heating rate of 10°C/minute. 2.2. Salt-resistant swelling behavior GA-cl-poly(MAA) was studied for its salt resistant behaviour by taking different salt solutions such as NaCl, MgCl2, CaCl2, FeCl3, ZnCl2 and BaCl2 of varying concentrations (1%, 5%, 10%, 15% and 20%). GA-cl-poly(MAA) sample was immersed in each salt solution for 24h at 25oC. The swollen samples were wiped, weighed and the percent swelling (Ps) was calculated. Further optimization of temperature dependent swelling of crosslinked sample was carried-out by immersing the sample in optimized concentration of each salt solution for 24h at 15, 25, 35, 45 and 55 oC. Similarly, pH dependent swelling was carried-out at optimized temperature in 0.5N HCl, distilled water and 0.5N NaOH.

Study of Gum Arabic and Methacrylic Acid Based Hydrogel 205


The –OH groups present on the back bone polymer act as the active sites for the graft co-polymerization of poly(MAA) onto it. The mechanism for the same is discussed as follows: Initiation ¯ O3S-O-O-SO3¯ 2 SO4¯ * (1) SO4¯ * + H2O HSO4 - + *OH (2) GA-OH +*OH GA-O * + HSO4¯ (3) GA-OH +*OH GA-O* + H2O (4) M + *OH *M-OH (5) M + SO4¯ * * M-SO4¯ (6) Propagation GA-OH + *M-OH GA-O-M* + H2O (7) GA-O-M* +nM GA-O-(M)n-M* (8) GA-O* + nM GA-O-(M)n-1M* (9) *M-OH + nM HO-(M)n-M* (10) Termination GA-O-(M)n-M* + M*-(M)n-O-GA GA -O-(M)n-M2-(M)n-O-GA (11) (Graft co-polymer) GA-O-(M)n-1M* + M* -(M)n-1-O-GA GA-O-(M)n-1-M2-(M)n-1-O-GA (12) (Graft co-polymer) GA-O-(M)n-M* + *OH GA-O-(M)n+1 –OH (13) HO-(M)n-M* + *M-(M)n-OH HO-(M)n-M2-(M)n-OH (14)

M* = monomer free radical; GA-O* = Backbone free radical SO4

-* radicals are produced by the dissociation of persulphate which further reacts with water to give *OH. The generated SO4

-* attacks Gum arabic backbone and results in the formation of free radical-sites on it. *OH attacks the backbone polymer and the monomer (MAA) leading to the generation of free radical sites on both of them which on further reaction gives rise to a graft copolymer. However, the termination of the reaction takes place either by reaction between *OH and a free radical or the reaction between two activated chains.

The three - dimensional cross-linked network structure of GA-cl-poly(MAA) is shown in Fig.1.

NN

N

N

°°

°°°°°

°

° °O

O

Backbone

Backbone

n

°°

nOOH

O OH

Fig. 1.Crosslinked network of GA-cl-poly(MAA) 3.1. Characterization 3.1.1 FTIR spectroscopy

Broad peaks are obtained in the IR spectrum of Gum arabic at 3365.2 cm-1 (O-H stretching of carbohydrates), 2939.1 cm-1 (CH2 asymmetric stretching), 1379.3cm-1 (CH, CH2 and OH in-plane bending in carbohydrates), 1042.9 cm-1 (C-O stretching region as complex bands, resulting from C-O and C-O-C stretching vibrations), 704.8 cm-1, 641.7 cm-1 and 603 cm-1 (pyranose rings). On the other hand, IR spectrum of GA-cl-poly(MAA) showed peaks at 2364.1 cm-1 (overtones and combinations of OH in- plane bending and C-O stretching vibration), 1725.1cm-1 and 1631.2cm-1 (C=O stretching in acids), 1405.7 cm-1(coupled OH in-plane bending and C-O stretching) in addition to peaks obtained with that of Gum arabic. 3.1.2 SEM studies of the gels

Comparison of the scanning electron micrographs of Gum arabic and GA-cl-poly(MAA), reveals a clear-cut distinction between the un-grafted and grafted samples (Figs. 2 and 3).

Fig. 2. SEM of Gum arabic


Fig. 3. SEM of GA-cl-poly(MAA) 3. 1. 3 Thermal behavior of gels TGA/ DTA studies of both the backbone and functionalized polymer were performed as a function of percent weight loss vs temperature (Figs. 4 and 5). Gum arabic showed initial decomposition temperature (IDT) at 227 °C and final decomposition temperature (FDT) at 517°C while GA-cl-poly(MAA) showed initial decomposition temperature at 200°C and final decomposition temperature at 461°C. Two-stage decomposition in both the cases was observed. Gum arabic showed three exothermic peaks at 292 (30 μV), 470 (131 μV ) and 513 °C (212 μV). On the other hand GA-cl-poly(MAA) showed three exothermic peaks at 349oC (41 μV ), 459oC (213 μV) and 642oC(18μV).

Fig. 4. TGA/ DTA of Gum arabic

Fig.5. TGA/ DTA of GA-cl-poly(MAA) 4. Salt resistant swelling behavior 4. 1. Effect of concentration of salt onto Ps

The Ps of the hydrogel was studied at different salt concentrations (1, 5, 10, 15 and 20%). It was found that Ps decreases with increase in salt concentration. It can be due to reverse osmosis process. Maximum swelling has been found at 1% concentration (Fig.6).

Fig. 6. Effect of salt concentration onto percent swelling (Ps) of GA-cl-poly(MAA) in different salt solutions 4. 2. Effect of temperature onto Ps

Swelling temperature was varied from 15-55oC (Fig.7). It was observed that Ps increases with increase in temperature and maximum Ps was found at 45 oC. However, further increase in temperature resulted in decreased Ps. This may be due to desorption at higher temperature[11].

Study of Gum Arabic and Methacrylic Acid Based Hydrogel 207

Fig. 7. Effect of temperature onto percent swelling (Ps) of GA-cl-poly(MAA) in different salt solutions. 4. 3. Effect of pH onto Ps

The water absorbing capacity of the polymer was investigated in different media (0.5N HCl, distilled water and 0.5N NaOH). It was found that the hydrogel swells to very less extent in acidic and basic media but showed appreciable swelling in distilled water (Fig.8). This could be due to disintegration of the candidate polymer both in acidic and basic media [12].

Fig.8.Effect of pH onto percent swelling (Ps)

of GA-cl-poly(MAA) in different salt solutions.

4. 4. Effect of charge of cation onto Ps It was found that the swelling capacity of

hydrogels in saline solution appreciably decreased as compared with swelling in distilled water. It has been observed that the swelling capacity decreased with increasing charge of the cation. It could be explained on the basis of charge screening effect of additional cations causing non-perfect anion-anion electrostatic repulsion, leading to decreased osmotic pressure (ionic pressure) difference between the polymer network and the external solution [12]. Therefore, the swelling capacity

would be in the order : Na+> (Zn2+, Mg2+, Ca2+, Ba2+) > Fe3+ 4. 5. Effect of size of cation onto Ps

It was also found that among the same valent ions, lesser the size of cation , more the swelling capacity i. e. swelling capacity is in the order: Zn2+

> Mg2+ > Ca2+ > Ba2+. So keeping in view the above two aspects, the trend in the percent swelling is:

Na+> Zn2+ > Mg2+ > Ca2+ > Ba2+ > Fe3+ 5. Conclusion

The Gum arabic and methacrylic acid based hydrogel has been found to be pH as well as temperature sensitive. Appreciable swelling of the hydrogel was noticed in different salt solutions. Hence the polymer synthesized is quite important from technological point of view. . References

[1] N.A. Peppas, A.G. Mikos, Hydrogels in Medicine and Pharmacy Vol.1, CRC Press, Boca Raton FL, 1986.

[2] L. Brannon – Peppas, Absorbent Polymer Technology, Elsevier, Amsterdam,1990.

[3] R.S. Zhang, Polymer 46 (2005) 2443. [4] J.H. Kou, G.L. Amidon, Pharm. res. 5 (1988)

592. [5] N.A. Peppas, R.Langer, Science 263 (1994)

1715. [6] W.E. Rudzinski, A.M. Dave, U.H. Vaishnav,

S.G. Kumbar, A.R. Kulkarni, T.M. Aminabhavi, Des. Monomers Polym. 5 (2002) 39.

[7] F.E. Okieimen, C.E. Sogbaike, J.E. Ebhoaye , Sep. Purif. Technol. 44 (2005) 85

[8] A.R. Menzies, M.E. Osman, A.A. Malik, T.C. Baldwin, Food Addit. Contam. 13(8) (1996) 991.

[9] Y. Dror, Y. Cohen, R. Yerushalmi –Rozen, J. Polym. Sci. 44 (2006) 3265.

[10] M.J. Zohuriaan- Mehr, Z. Motazedi, K. Kabiri, A. Ershad-Langroudi, J. Macromol. Sci. Part A 42(12) (2005) 1655.

[11] B.S. Kaith, K. Kumar, e-Polymers no.002 (2007).

[12] B.S. Kaith, K. Kumar, Bull. Mater. Sci. 30 (2007) 387.

[13] A. Pourjavadi, G.R. Mahdavinia, Turk. J. Chem. 30 (2006) 595.

Evaluation of Mechanical Properties of Grewia Optiva Fiber Reinforced Polymer Composites

A.S. Singha and Vijay Kumar Thakur

Department of Applied Sciences and Humanities, National Institute of Technology (Deemed University) Hamirpur (H.P.) 177 005 INDIA


Abstract

The lack of resources and increasing environmental pollution; has led great interest in the research of materials that are friendly to our health and environment. Polymer composites prepared from natural materials is currently the most promising area in polymer science. In particular, many synthetic polymeric materials are being produced by combining with various reinforcing fillers to improve their mechanical properties and obtain the desired properties. Among these reinforcing fillers, active research is under way concerning the use of ligno cellulosic materials, which are among the most environmentally friendly biomasses, as a substitute for synthetic materials. The cost of producing composites comprising natural products such as ligno cellulosic materials as the reinforcing filler and thermosetting polymer as the matrix polymer is quite low. Furthermore, these materials can easily be obtained from waste products and have a minimal effect on the environment, due to their enormous properties such as low density and high specific properties. These are biodegradable and non-abrasive; thus, in recent years, the emphasis has increasingly been positioned on these composites, which may well play a major role in resolving some of the pressing environmental issues with which we are confronted in the future. Instead of the inorganic materials and synthetic fibers which were previously added to plastics as fillers, ligno cellulosic materials offer many environmental benefits when used as reinforcing fillers for polymers, including their making the final product lightweight, decreasing the erosion of the manufacturing machinery, low cost, biodegradability, and absence of production of residue or toxic by-products when burnt . Concerning these advantages a study on the growth of green polymer composites using Grewia Optiva fiber reinforced Urea-Formaldehyde resin has been made. Present work reveals that mechanical properties such as: tensile strength, compressive strength and wear resistance etc.of the urea resin increase by many folds when reinforced with natural fiber (Grewia Optiva). However, reinforcing of this resin (U-F) with fiber in two different forms i.e. particle size and short fiber with suitable loading and assessment of their mechanical properties showed that particle reinforcement is more effective than short fiber reinforcement. These results were further supported by the thermal studies and SEM studies of these composites. 1. Introduction

Increased environmental awareness and consciousness for developing environmentally friendly composite materials has demonstrated the way to significant attraction, based on renewable resources like natural fibers as alternatives for glass fiber reinforcement in traditional glass fiber-reinforced polymer matrix composites [1–5]. Advantages of natural fibers over usual reinforcing fibers such as glass and carbon fibers are low cost, low density, high toughness, acceptable specific strength, enhanced energy recovery, recyclability, biodegradability, eco-friendly nature et [6]. Therefore, natural fibers can serve as smart reinforcing materials for improving the strength and stiffness of biocomposites. The properties of

fibers vary with their sources and treatments [7–9]. It has been observed that natural fibers such as flax on reinforcement improved the mechanical properties of polystyrene composites [10-11]. In the present study we have used polymer resin (Urea-Formaldehyde) as the matrix and a ligno-cellulosic material (Grewia Optiva fiber) as the reinforcing material to prepare particle and short fiber reinforced composites. 2. Experimental

Urea (Qualigens Chemicals Ltd), formaldehyde solution (Qualigens Chemicals Ltd.), and sodium hydroxide (Qualigens Chemicals Ltd.), Grewia Optiva fiber were used as received. Weights of the samples were taken

Evaluation of Mechanical Properties of Grewia Optiva Fiber 209

on Shimadzu make electronic balance (LIBROR AEG- 220). ), testing of samples for tensile and compressive strengths was made on Computerized Universal Testing Machine (HOUNSFIELD H25KS).

The ligno-cellulosic material used as the reinforcing filler in the composite was Grewia Optiva fiber. Grewia Optiva fibers were dried at 100 0C for 24 h to adjust it to a moisture content of 1–2% and then stored over desiccant in sealed container These were used in two forms as shown: 1. Particle Reinforcement: Grewia Optiva fiber was grinded to a powder and filtered through a sieve of pore size 200 microns. 2. Short- Fiber Reinforcement: Grewia Optiva fiber was chopped into 0.3 cm size. This fiber was used as short fiber.

Urea-Formaldehyde Resin was prepared by taking different molar ratio in a reaction kettle using different customary conditions such as temperature (70-800C for U-F), pH7-8, and speed of the mechanical stirrer (9-10rpm.). Samples were then prepared in compression molding machine at required temperature after mixing of the resins and Grewia Optiva fiber both in short and particle size with 10% loading in specially made chambers .The cured samples were then subjected to various mechanical studies. The composites prepared were cut into different lengths (10 cm for tensile strength, 5 cm for compressive strength, 5 cm for wear resistance and 5 x 5 mm cross- section. The cured samples were then subjected to various mechanical and thermal studies. 3. Results and discussions

It has been observed that U-F samples of ratio 1.2.5 shows optimum mechanical properties such as tensile strength, compressive strength and wear resistance as compared to other ratios as shown in Figs (1-3). Therefore, this Ratio was taken for further preparation of Urea- formaldehyde resin.

0

20

40

60

80

100

120

140

0 2 4

EXTENSION(mm)

FORCE(

N)

UF-1.0:1.0UF-1.0:1.5UF-1.0:2.0UF-1.0:2.5UF-1.0:3.0

Fig. 1. Tensile strength of UF resin

0

200

400

600

800

1000

1200

0 2 4 6

COMPRESSION (mm )

FORCE

( N )

UF-1.0:1.0

UF-1.0:1.5

UF-1.0:2.0

UF-1.0:2.5

UF-1.0:3.0

Fig. 2. Compressive strength of UF resin

0

0.01

0.02

0.03

0.04

0.05

0.06

0 2 4

LOAD (kg )

WEI

GHT

LOSS

U:F -1.0:1.0

U:F -1.0:1.5

U:F -1.0:2.0

U:F -1.0:2.5

U:F -1.0:3.0

Fig. 3. Wear resistance of UF resin Experimental results obtained through tensile test, compressive test and wear resistance of composites are shown in Fig. (4-6).

050

100150200250300350400

0 2 4

EXTENSION (mm)

FORCE

(N ) UF-

1.0:2.5

P-Rnf

SF-Rnf

Fig. 4. Tensile strength of G.O.Rnf composites.

0

500

1000

1500

2000

2500

3000

3500

0 5COMPRESSIO

N (mm)

FOR

CE

(N )

UF-1.0:2.5

P-Rnf

S-Rnf

Fig. 5. Compressive strength of G.O. Rnf

Composites


0

0.002

0.004

0.006

0.008

0.01

0.012

0 2 4

LOAD (kg )

WEI

GH

T LO

SS UF

P-Rnf

S-Rnf

L-Rnf

Fig. 6. Wear resistance of G.O.Rnf composites.

The analysis of composites results reveals that wear rate of U-F matrix (Fig.6) decreases appreciably as reinforcement with Grewia Optiva fiber. It was observed that particle reinforcement decreases the wear rate many times. While compressive strength (Fig.5) and tensile strength. (Fig.4) of UF matrix has been found to increase as reinforcement with Grewia Optiva fiber. Present study revealed that particle reinforcement is more effective than short fiber reinforcement. This may be due to larger surface area of particle reinforcement as compared to short fiber reinforcement. Thermal studies

Thermal analysis of materials gives us good account of their thermal stability. It has been observed that particle reinforcement changes the IDT as well as FDT of the resin as shown in table1. Table 1. Thermal analysis of Resin, Fiber and P-Rnf composite

Morphological studies

Morphological analysis of different samples was carried out by studying SEM micrographs as shown in Fig (7-9). These SEM micrographs of the samples give us information about the morphology of the resin and its respective biocomposite. These micrographs clearly show the difference between loaded and unloaded matrix.

Fig. 7. SEM of P-Rnf UF Resin

Fig. 8. SEM of SF-Rnf UF Resin

Fig. 9. SEM of UF Resin

Sr. No

Sample Code

IDT (0C)

% wt. loss

FDT (0C)

% wt. loss

Final Residue

(%)

1. GO 200 7.25 501 87.44 12.66

2. U-F Resin 238 22.48 993 87. 51 12.49

3. P-rnf-UF 234 20.51 815 83.33 16.37

Evaluation of Mechanical Properties of Grewia Optiva Fiber 211

4. Conclusions

Mechanical properties of randomly oriented intimately mixed Grewia Optiva fiber reinforced UF composites were investigated with special reference to the size of the fiber. Various test methods were used for complete mechanical characterization of natural fiber reinforced composites. In case of mechanical behaviour particle reinforcement of the UF resin has been found to be more effective as compared to short reinforcement. The mechanical behaviour has been strongly supported by the SEM analysis. Finally, it can be concluded that by utilizing Grewia Optiva fiber, we can prepare user-friendly and cost-effective composite materials possessing appropriate mechanical properties. References [1] A.K. Mohanty, M. Misra, G. Hinrichsen,

Macromol Mater Eng 276/277(2000) 1–24. [2] A.K. Mohanty, M. Misra, L.T.Drzal. Comp

Interfaces; 8(5), (2001)313–43.

[3] A.N. Netravali, S Chabba.Materialstoday 2003, 22–9.

[4] C. Baillie. Comp Sci Technol; 63 (2003)1223–4.

[5] P. Wambua, J Ivens, I. Verpoest, Comp Sci Technol; 63 (2003)1259–64.

[6] B. N Misra, J.K Jassal, R. Dogra and D. Sood, J. Macromol. Sci.Chem., A-1980.

[7] L.Y. Mwaikambo, M.P. Ansell, Die Angewandte Macromol Chem, 272 (1999) 108–16.

[8] A.K. Mohanty, D. Hokens., M. Misra, L.T. Drzal. In: Proceedings (in CD) of 16th Ann Tech Conf., Am Soc Comp., septenser 9–12., Blackburg., VA; 2001.

[9] K. Oksman, M. Skrifvars, S.F. Selin, Comp Sci Technol; 63(2003)1317–24.

[10] B. S. Kaith., A. S. Singha., D. K. Dwedi., Sanjeev Kumar., D. Kumar and A. Dhemeniya., International Journal of Plastic Technology., 7(2003) 119-125.

[11] D. K. Dwedi., A. S. Singha,Sanjeev Kumar and B. S. Kaith International Journal of Plastic Technology., 8(2004) 299-304.

Synthesis and Evaluation of Green Composites Based on Hibiscus Sabdariffa Reinforced Thermosetting Resin

A.S. Singha and Vijay Kumar Thakur

Department of Applied Sciences and Humanities, National Institute of Technology (Deemed University), Hamirpur (H.P.) 177 005 INDIA


Abstract

Natural fibers have received much more attention than ever before from the research community all over the world during the past decade. Natural fibers are now considered as serious alternative to synthetic fibers for use in composite materials as reinforcing agents. These days various synthetic polymers are being prepared and combined with various reinforcing fillers in order to improve the mechanical properties and obtain the characteristics demanded in actual application. Studies are ongoing to find ways to use lignocelluloses bio fibers in place of synthetic fibers as reinforcing fillers. It has been observed that biofibers such as grewia optiva, hibiscus sabdariffa and flax etc. on reinforcement improved the mechanical properties of polymer composites. These bio fibers are especially being required since the production of composites using natural substances as reinforcing fillers is not only inexpensive but also able to minimize the environmental pollution caused by the characteristic biodegradability, enabling these composites to play an important role in resolving future environmental problems. These are biodegradable and non-abrasive. A study on the synthesis of new series of green composites using Hibiscus Sabdariffa fiber reinforced Urea-Resorcinol-Formaldehyde resin has been made. Urea-Resorcinol-Formaldehyde resins have been prepared by employing Phenolic resin (Resorcinol). Initially Urea resin prepared was subjected to evaluation of its optimum mechanical properties. The reinforcing of the resins with Hibiscus Sabdariffa fiber was accomplished in two different forms: particle size and short fiber by employing optimized resin. Present work reveals that mechanical properties such as: tensile strength, compressive strength and wear resistance etc. of the Urea resin increase by many folds when reinforced with hibiscus fiber. The assessment of the mechanical properties of the green composites prepared showed that particle reinforcement is more efficient than that with short fiber reinforcement. These results were further supported by the SEM studies of these composites. 1. Introduction

These days various synthetic polymers are being prepared and combined with various reinforcing fillers in order to improve the mechanical properties and obtain the characteristics demanded in actual application. Studies are continuing to find ways to use lignocelluloses fibers in place of synthetic fibers as reinforcing fillers. It has been observed that natural fibers such as flax on reinforcement improved the mechanical properties of polystyrene composites [1-2]. These natural fillers are especially being required since the production of composites using natural substances as reinforcing fillers is not only inexpensive but also able to minimize the environmental pollution caused by the characteristic biodegradability [3], enabling these composites to play an important role in resolving future environmental problems . The need for materials that are non-toxic to the human body

and have appropriate characteristics for specific purposes is ever increasing due to the lack of resources and increasing levels of environmental pollution. Thus, research is going on to develop composites using various recycled wastes [4,5], especially in developing composites using most environmentally friendly agro-wastes (lignocelluloses materials) as reinforcing fillers and thermosetting polymers as matrixes. The easiness of these composites lies in the fact that the ingredients are obtained easily from natural wastes and hence the composites can be made relatively easily. They can be used to resolve environmental problems and to produce products with various physical properties and effective functions. Ligno-cellulosic materials as reinforcing fillers in plastics, in place of the previously used inorganic substances and synthetic fibers, offer a major benefit in terms of environmental protection. The benefits offered by ligno cellulosic materials include making the final product light [6], decreasing the wear of the

Synthesis and Evaluation of Green Composites Based on Hibiscus Sabdariffa Reinforced 213

machinery used, low cost, biodegradability [3], and absence of residues or toxic by products. In the present study we have used a thermosetting polymer (URF Resins) as the matrix and a ligno-cellulosic material (Hibiscus Sabdariffa fiber) as the reinforcing filler to prepare particle and short fiber reinforced green composites. 2. Experimental

Material and methods

Urea (Qualigens Chemicals Ltd), formaldehyde solution (Qualigens Chemicals Ltd.), Resorcinol (CDH) and sodium hydroxide (Qualigens Chemicals Ltd.), Hibiscus Sabdariffa fiber (Local Sources) were used as received.

Weights of the samples were taken on Shimadzu compose electronic balance (LIBROR AEG- 220), wear testing was made on Wear & Friction Monitor (DUCOM- TR-20L), testing of samples for tensile and compressive strengths was made on Computerized Universal Testing Machine (HOUNSFIELD H25KS) and thermal studies were conceded on Thermal Analyzer (LINSEIS, L 81-11 Matrix polymer

Polymeric resin, Urea-Resorcinol-Formaldehyde (U-R-F) was used as Thermosetting Matrix Polymer Reinforcing filler

The ligno-cellulosic material used as the reinforcing filler in the composite was Hibiscus Sabdariffa fiber.

This fiber was used in two forms as shown below: 1. Particle Reinforcement

Hibiscus Sabdariffa fiber was grinded to a powder and filtered through a sieve of pore size 200 microns. 2. Short- Fiber Reinforcement

Hibiscus Sabdariffa fiber was chopped into 0.3 cm size. This fiber was used as short fiber. Methods

1. Synthesis of Resins: Polymeric resin URF was synthesized by the

standard method developed in our Applied Chemistry Research Laboratory by taking different molar ratio (1.0:1.0, 1.0:1.5, 1.0:2.0, 1.0:2.5 and 1.0:3.0)of Resorcinol to the optimized ratio of UF resin(1:2.5) by weight, in a reaction kettle using different customary

conditions such as temperature(55-600C) for reactions to occur, amount of NaOH as base catalyst(0.5g of wt.of URF), pH(7.5-8) and speed of the mechanical stirrer (9-10rpm.) Samples were then prepared in compression molding machine at 50-700C. The cured samples were then subjected to various mechanical studies.

2. Synthesis of Composites:

Hibiscus Sabdariffa fiber was dried at 100 0C for 24 h to adjust it to a moisture content of 1–2% and then stored over desiccant in sealed container. Required level of filler loading (wt %) was designed in the sample preparation. After mixing of the resins and Hibiscus Sabdariffa fiber (HSF) both in short and particle size with 10% loading the mixture were put into specially made chambers. These chambers were then cured at 50-700C in Compression molding machine. The composites prepared were cut into different lengths (10 cm for tensile strength, 5 cm for compressive strength, 5 cm for wear resistance and 5 x 5 mm cross- section. The cured samples were then subjected to various mechanical and thermal studies. 3. Results and Discussions

Optimization of Urea-Resorcinol– Formaldehyde Resin Wear Resistance

It has been observed that wear rate of samples of ratio 1.0:1:2.5 was less as compared to any other samples. Loss of material was due to abrasion and friction of samples with disc (Fig.1).

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0 2 4

LOAD (kg )

WEI

GHT

LOSS

URF-1:0.5:2.5URF-1:1.0:2.5URF-1:1.5:2.5URF-1:2.0:2.5URF-1:2.5:2.5URF-1:3.0:2.5

Fig. 1. Wear resistance of URF Resin


It is evident from the (Fig.2 that the sample of ratio 1.0:1:2.5 bears more load as compared to other ratios.


Tensile Strength

It has been observed that U-R-F samples of ratio 1.0:1:2.5 bear more load at a particular applied load as compared to samples of other ratios (Fig3). On the other hand, samples of other ratios bear low loads.

As is evident from Figs.1-3, samples of ratio 1.0:1:2.5 show maximum tensile and compressive strength. Moreover, at this ratio wear rate was also very less. Therefore, this ratio was taken for further preparation of Urea-Resorcinol -formaldehyde resin.

0

200

400

600

800

1000

1200

1400

0 2 4COMPRESSION (mm )

FOR

CE

(N )


Fig. 2. Compressive strength of URF resin

0

100

200

300

400

500

600

0 2 4 6EXTENSION (mm )

FOR

CE

(N )


z

Fig. 3. Tensile strength of URF resin Effect of Reinforcement on the Mechanical Properties of U- R-F Based Composites

0

0.002

0.004

0.006

0.008

0.01

0.012

0 2 4LOAD ( k g )

URF-1:1.0:2.5

P-Rnf

SF-Rnf

Fig. 4. Wear resistance of URF Composites

Wear Test: As evident from (Fig. 4) that wear rate of

URF matrix decreases appreciably as reinforcement with Hibiscus Sabdariffa fiber. It was observed that particle reinforcement decreases the wear rate by slightly more times than short- fiber reinforcement


Compressive strength of URF matrix (Fig.5) has been found to increase as reinforcement with Hibiscus Sabdariffa fiber. It has been found that composites with particle reinforcement showed more compressive strength which were followed by short fiber.

0500

100015002000

2500300035004000

0 5

COMPRESSION (mm )

FOR

CE

( N )

P-Rnf

SF-Rnf

Fig. 5. Compressive strength of URF Composites Tensile Strength:

It has been observed that composites with particle reinforcement showed more tensile strength which was followed by short fiber reinforced composites (Fig.6).

0200400600800

10001200140016001800

0 2 4 6

EXTENSION (mm )

FOR

CE

( N )

P-Rnf

SF-Rnf

Fig. 6. Tensile strength of URF Composites Morphological studies

Morphological analysis of different samples was carried out by studying SEM micrographs as shown in Fig (7-9). These SEM micrographs of

Synthesis and Evaluation of Green Composites Based on Hibiscus Sabdariffa Reinforced 215

the samples give us information about the morphology of the resin and its respective biocomposite. These micrographs clearly show the difference between loaded and unloaded matrix.

Fig. 7. SEM of URF Resin

Fig. 8. SEM of P-Rnf URF Resin

Fig. 9. SEM of SF-Rnf URF Resin

4. Conclusion

In case of mechanical behaviour particle reinforcement of the URF resins has been found to be more effective as compared to short fiber reinforcement. This could be due to more matrix- particle interfacial interaction as compared to short fiber–matrix interfacial interaction. As concluding remarks, the present results suggest that use of Hibiscus Sabdariffa fiber as reinforcement in a natural fiber composite system may be a potential candidate for effectively improving the properties and performances of biodegradable polymer matrix resins References

[1] B. S. Kaith, A. S. Singha, D. K. Dwedi, Sanjeev Kumar, D. Kumar and A. Dhemeniya, International Journal of Plastic Technology, 7(2003) 119-125.

[2] D. K. Dwedi, A. S. Singha, ,Sanjeev Kumar and B. S. Kaith International Journal of Plastic Technology, 8(2004) 299-304.

[3] H.G.B. Premalal, H. Ismail , A. Baharin . Polym Testing 2002; 21(7):833–9.

[4] J.I. Son, H.J. Kim, P.W. Lee. J Applied Polymer Science 2001; 82(11):2709–18.

[5] J.I. Son, H.S. Yang, H.J. Kim. J Thermoplastic Compos Materials 2003.

[6] R.E. Jacobson, M.S. Engineer, D.F. Caulfield, R.M. Rowell, A.R. Sanadi. In: Proceedings of 2nd Biomass Conference of the Americas: Energy, Environment, Agriculture and Industry, August 21–24, Portland OR. Golden, CO: National Renewable Energy Laboratory, 1995. p 1171–1180.

Phonon Properties of Inter-Metallic Compound AuIn2

M. M. Sinha Department of Physics, Sant Longowal Institute of Engineering and Technology,

Longowal, Distt. Sangrur-148106 (Punjab), INDIA E-mail: [email protected]

Abstract

The intermetallic compound AuIn2 exhibit superelasticity such that the intermetallic compound possess

shape memory characteristics when deformed at low stress levels yet also provide sufficient strength to maintain its shape when employed as a gasket or seal. In recent years, the binary intermetallic compound AuIn2, has attracted considerable attention. The main reason for this attention is the discovery of the coexistence of the nuclear ferromagnetism and superconductivity in AuIn2. This discovery has led to significant interest in the structural and dynamical properties of this compound. It is therefore important to perform a true investigation of vibrational properties of AuIn2 to interpret the role of electron and phonon interaction in this compound. A de Launey angular force constant model has been applied, in the present investigation, to study the phonon properties of AuIn2. 1. Introduction Intermetallic alloys which can be made ductile at low temperature and strong at high temperatures are of great value as materials for current and future technology. It is interesting to note that these intermetallics exhibit a striking variety of colors, ranging from purple (AuAl2) to copper (PtIn2) to gold (PtAl2, PtGa2). The binary intermetallic compound AuIn2 that crystallizes in the fluorite structure has received a great deal of attention owing to its many unique properties. The main reason for this attention is the discovery of the coexistence of the nuclear ferro-magnetism and superconductivity in AuIn2 [1, 2]. This discovery has led to significant interest in the structural and electronic properties of these materials [3–13]. Several experimental techniques have been used to investigate the electronic structures of AuX2 (X = Al, Ga, and In), such as x-ray [4] and ultraviolet [5] photoemission spectroscopic experiments, optical reflectivity measurements [6], and angle-resolved photoemission spectroscopy (ARPES) [7, 10, 11]. Besides these experimental studies, all-electron full-potential linear augmented-plane-wave method [10], a mixed-basis band-structure interpolation scheme [12] has been used to calculate the electronic structure of these three materials. Despite much work on structural and electronic properties of these materials, their dynamical properties are relatively poorly known in the literature. Raman-scattering experiment and infrared techniques have been used to

measure the zone-centre phonon modes of AuGa2 and AuIn2 [14]. On the theoretical side, Sinha et al. [15] have studied phonons in AuGa2 and PtGa2 while not much attention has been paid to the lattice dynamical studies of AuIn2. It is important to perform a true investigation of electronic and vibrational properties of these materials because electrons and phonons in metals play an important role in the superconductivity due to their interactions. It is therefore, in the present investigation, zone center phonons and phonon dispersion of AuIn2 has been studied by applying a de Launey angular force constant model [16]. 2. Theory The binary intermetallic compounds AuIn2 has a cubic fluorite (CaF2) structure [7] in which Au atoms form a fcc sublattice and the In atoms occupy the tetrahedral sites located one-quarter of the way up the body diagonals. The Bravais lattice is fcc, the space group is Oh

5 (Fm3m) and the primitive unit cell contains one formula unit. For a one-phonon process, the displacement of the three atoms in the unit cell transforms as

Γ = 2T1u + T2g at the Brillouin zone center. The triply degenerate T2g modes are Raman active modes due to the motion of In atoms against each other, and one of the T1u modes corresponds to three acoustic branches. The remaining triply degenerate T1u modes are infrared active modes. To elucidate the nature of the bonding and other thermal properties in intermetallic

Phonon Properties of Inter-Metallic Compound AuIn2 217

compounds, the lattice dynamics of AuIn2 has been investigated by using a de Launey angular force constant model [16]. In the present analysis three central force constants α1, α2 and α3 existing between Au and In1 or In2, In1 and In2 and In1–In1 or In2–In2, respectively, and the angular force constant α1

’ between Au and In1 or In2 and α2

’ between In1 and In2 are taken into consideration for getting a dynamical matrix of (9×9) of the fluorite structure. The central force constant between Au–Au is assumed to be the same as that between In1–In1 or In2–In2. The long wavelength limit method has been used to establish the analytical relation between the force constants and elastic constant [17]. The expressions for the phonon frequencies at the zone center and the analytical relation between elastic and force constants used in present investigation are given by the following equations; 4/3(α1 + 2α1

’ ) = [mM/(2m+M)]ωTO2 (1)

4/3(α1 + 2α1’ ) + 4 (α2 + α2

’ ) = mωR2 (2)

2aC11 = 4/3(α1 + 2α1’ ) + 4α2 + 12α3 (3)

2aC12 = 4/3(α1 - 4α1’ ) - 4 α2

’ + 6α3 (4) 2aC44 = 4/3(α1 + 2α1

’ ) + 4 α2’ + 6α3−

4/3(α1 - α1’ )2 / 4/3(α1 + 2α1

’ ) (5) Here M being the mass of Au atoms and m the mass of In. ωTO represents the infrared optical frequencies whereas ωR is the Raman active mode at the zone center. In AX2 (X=In, Al), the A-X distance is about 0.2 Å less than the sum of the 12 coordination radii of the constituent atoms suggesting a covalent character of the bond. Being covalent bonding character in AuIn2, the long range electrostatic interaction is not taken into consideration in the present calculation. Table 1. Input data

Zone centre phonon

frequencies [14] (cm-1)

Elastic constant [19] (1011 dynes. cm-2)

ωTO ωR C11 C12 C44 115 124 10.0 7.33 2.933

By using the following results as given in

table 1 for Zone center phonon frequencies [14] and elastic constants [20] to the above equations, the central and angular force constants are calculated and are listed in table 2.

Table 2. Calculated data

Force constant (104dynes cm−1) α1 α1

’ α2 α2’ α3

4.150 -0.514 -0.339 0.959 0.851

Table 3. ZC phonon frequencies (in cm-1)

Mode Present calculation

Experiment [14]

Other calculation

[18] ωTO 115 -- 121 ωR 124 124 115

With the force constants listed in table 2, the

dynamical matrix (9×9) is solved along the three principal symmetry directions. The calculated zone center phonon frequencies are shown in table 3 along with available experimental results. The phonon dispersion curves are plotted along three symmetry directions and are shown in fig.1.

0.0 0.2 0.4 0.6 0.8 1.00

50

100

150

LXΓ Γ

Fig.1 Phonon dispersion curves of AuIn2

[kkk]

0.0 0.1 0.2 0.3 0.4 0.5

wavevector (k)

[kk0]

1.0 0.8 0.6 0.4 0.2 0.0

[k00]

Fr

eque

ncy

(cm

-1)


The elastic constants of AuIn2 are not available in the literature and therefore in the present calculation the value of elastic constants


of AuIn2 are assumed to be that of AuGa2 [3]. The reason of this assumption is the closeness of bulk modulus of AuGa2 (0.96 Mbar) to that of AuIn2 (0.79 Mbar) [3]. The calculated force constants listed in table 2 clearly shows that the interatomic interaction (α1) between Au and In1 or In2 is strongest among all as the Au–In1 or In2 bond length (2.79Å) is smallest among all other bonds [11]. It is obvious from table 3 that the calculated zone center phonons are well in agreement with experiment [14] and differ slightly from other calculation [18]. The calculated phonon dispersion curves along three symmetric directions are shown in figure 1 reveals that there are three acoustic and six optic modes for all wave vectors along [kk0] directions. However, there are two acoustic and four optic modes along the high symmetry directions [k00] (Γ-X) and [kkk] (Γ-L), due to the degeneracy of the transverse modes in both acoustic and optic branches. All the optical modes are found to be quite dispersive along the main symmetric directions. There is clear gap between the acoustic and optic branches for AuIn2 and is different from other calculation [18]. The reason of this gap is due to good difference in mass of Au and In atoms. The present calculations are in good agreement with experiment. References [1] J. H. Wernick, A. Menth, T. H. Geballe, G.

Hull and J. P. Maita J. Phys. Chem.Solids 30 (1969) 1949

[2] S. Rehmann, T. Hermannsdorfer and F. Pobell Phys. Rev. Lett. 78 (1997) 1122

[3] L. R. Testardi Phys. Rev. B 1 (1970) 12 [4] T.K. Sham, M. L. Perlman and R. E. Watson

Phys. Rev. B 19 (1979) 539 [5] I. Perez, B. Qi, G. Liang, F. Lu, M. Croft

and D. Wieliczka Phys. Rev. B 38 (1988)12233

[6] H. R. Philipp Phys. Status Solidi a 69 (1982) 547

[7] J. G. Nelson, W. J .Gignac, S. Kim, J. R. Lince and R. S. Williams Phys. Rev. B 31 (1985) 3469

[8] P.E. Bl¨ochl, O.Jepsen and O.K. Andersen Phys. Rev. B 49 (1994) 16223

[9] L. S. Hsu Phys. Lett. B 8 (1994) 1297 [10] L. S. Hsu, G. Y. Guo, J. D. Denlinger and J.

W. Allen J. Phys. Chem. Solids 62 (2001) 1047

[11] L.S. Hsu, Y. K .Wang, Y. L. Tai and J. F. Lee Phys. Rev. B 72 (2005) 115115

[12] S. Kim, J. G. Nelson and R. S. Willlams Phys. Rev. B 31 (1985) 3460

[13] H..M. Tutuncu, H. Altuntas., G. P. Srivastava and G. Ugur Phys. Status Solidi c 1 (2004) 3027

[14] W. J. Brya Solid State Commun. 9 (1971) 2271

[15] M.M. Sinha, J. S. Kim J. Alloys and Compounds 365 (2004) 49

[16]J. de Launey, Solid State Phys. 2 (1956) 219. [17] H.C. Gupta, M.M. Sinha, B.B. Tripathi,

Solid State Commun. 62 (1987) 777. [18] G. Ugur and F. Soyalp, J. Phys.: Condens.

Matter 18 (2006) 6777 [19] L.R. Testardi, Phys. Rev. B 1 (1970) 4851.

Stress Analysis of a Laminated Composite Plate with Central Hole Under in-Plane Static Loading using Ansys – a Case Study

R. N. Zaware, M. P. Nagarkar and R. R. Navthar.

Department of Mechanical Engineering,, PDVVP College of Engg., Ahmednagar - 414003 (M.S.), INDIA.


Abstract

Stress analysis of mechanical components carries great importance in modeling and prototyping. Here, an attempt is made to study the behavior of rectangular plate under static loading. The rectangular plate is having a central hole. To study the behavior, various experimental as well as numerical and analytical methods are available. The aim of this paper is to analyse the effect in-plane static loading on stress and stress concentration around hole in a laminated plate. Studies were carried out for three different hole diameter to plate width ratio, and five different loading cases. The results are obtained for a laminated plate having 45° fibre orientation. The stress distribution in rectangular composite plate with central hole is studied using FEA package, ANSYS. 1. Introduction

The recent widespread use of CAD has enabled mechanical engineers to use Computer Aided Engineering (CAE), FEM, FEA tools for predicting the performance of components from various aspects in the early stages of its development upto the optimization and later after development stages. CAE, FEA, which is used to be a merely a qualitative prediction means, now allows even quantitative evaluations to be performed.

Laminated composites found widespread applications in various fields of engineering such as aerospace, marine, automobile and mechanical. High stress due to discontinuity or abrupt change in geometry is known as stress concentration and always found at the edges of discontinuity. From the design point of view, it carries great importance to know the stresses and stress concentration at the discontinuity or abrupt change in the geometry.

In this paper a study of rectangular laminated plate with central hole under different in-plane static loading is made. To study the behavior, various experimental as well as numerical and analytical methods are available but the finite element method adopt for whole analysis. The aim of this research work is to investigate the effect of fibre orientations on stress concentration in a single layer laminate with central circular hole.

Plate with Central Hole The study involves modeling of a given composite plate in ANSYS. [1,2] For the study purpose, the dimensions of plate are taken 200mmX100mmX10. The diameter of central hole is taken in the relation D/b = 0.2, 0.25, 0.5 where b = width of the plate and D = diameter of the hole. Fig. 1 shows the test plate.

Fig. 1. A plate with central hole. The plate is subjected to a uniform static load σ = 10, 25, 50, 75, 100N/mm2. Finite Element Analysis

An eight-nodded Linear Layered Structural 3-D Shell Element with six degrees of freedom at each node (specified as Shell99) was selected. Each node has six degrees of freedom, making a total 48 degrees of freedom per element. Figure 2 provides the detail and Geometry of element type[3].


Fig. 2. ANSYS Shell 99 Element. To construct the geometric model of the laminated plate in ANSYS, a quarter plate is modeled using key points, lines and areas. Then a circle is subtracted to get the required plate with hole. Next task is to discretize the laminated plate. Model is meshed with mapped meshing using element size 2.5. Figure 3 shows a meshed model for D/A =0.5. [3]

Fig. 3. ANSYS meshed model of a plate with hole (with Shell99 element type). 2. Results and discussion

Analysis is carried out in ANSYS 10 environment. The diameter of central hole is taken in the relation D/b = 0.2, 0.25, 0.5 where b = width of the plate and D = diameter of the hole. So, the hole diameters are 20mm, 25mm and 50mm. The material used is E glass/epoxy having properties: E1 = 34000 MPa, E2 = 6530 MPa E3 = 6530 Mpa, G12 = 2433 Mpa G23 = 1698 Mpa, G31 = 2433 Mpa µ12 = 0.217, µ23 = 0.366, µ31 = 0.217 Mass Density = 2.6x10-6 Kg/mm3 Fibre orientation = 45° The plate is subjected to a uniform static load σ = 10, 25, 50, 75 and 100N/mm2.

Fig. 4. Von Mises stress plot for the plate (Material -e glass/epoxy, D/b = 0.2, σ = 50 N/mm2)

Fig. 5. Von Mises stress plot for the plate (Full View) (Material -e glass/epoxy, D/b = 0.2, σ = 50/mm2)

Fig. 6. Von Mises stress plot for the plate

Stress Analysis of a Laminated Composite Plate with Central Hole 221

(Material -e glass/epoxy, D/b = 0.25,σ = 75N/mm2)

Fig. 7. Von Mises stress plot for the plate (Full View) (Material-e glass/epoxy, D/b 0.25,σ=75N/mm2)

Fig. 8. Von Mises stress plot for the plate (Material-e glass/epoxy, D/b= 0.5,σ=100N/mm2)

Fig. 9. Von Mises stress plot for the plate (Full View) (Material-e glass/epoxy, D/b = 0.5, σ = 100 N/mm2)

Following table shows maximum von Mises stress obtained for various in-plane static loading and various D/b ratios. Table 1. Maximum von Mises stress (N/mm2) for various σ and D/b ratio. σ D/b

10 25 50 75 100

0.20

2.499

6.248

12.496

18.744

24.99

0.25

2.628

6.569

13.138

19.707

26.276

0.50

3.776

9.439

18.879

28.318

37.758

Fig. 10 Graph of static load Vs Maximum von Mises stress (Material -e glass/epoxy, D/b = 0.2,0.25,0.5) 3. Conclusions

The maximum stress is occurred on the hole boundary (with minimum c/s) for a laminated plate. Here the stress concentration is maximum. It is also observed that as the in-plane static load goes on increasing, the stress and stress concentration also goes on increasing with increasing D/b ratio. Future Scope

While analyzing only one composite is considered with 45° fibre orientation. Various composites can be studied with various fibre orientations to analyse the effect on stress concentration.


References

[1] Henrik Karlsson, Evaluation of FE-software for Mechanical Analysis of composite materials, Lulera University of Technology, 2005.

[2] Losec Jason, Stress Analysis using ANSYS – The composite recurve bow, PPT, 2005.

[3] Documentation for ANSYS Release 10.

Low Temperature Specific Heat of Codoped Mg1-X(AlLi)XB2

Nupinderjeet Kaura, Rajneesh Mohanb, N. K. Gaurb and R. K. Singhc aDepartment of Physics, Indian Institute of Technology Delhi, New Delhi-110016, India

bDepartment of Physics, Barkatullah University, Bhopal- 462026 (M.P.), India cMATS University, MATS Tower Pandri, Raipur – 492002 (CG), India


Abstract

In the present paper, we have formulated the Rigid Ion Model (RIM) which includes long range Coulomb, van der Waals (vdW) interaction and short-range Hafemeister Flygare (HF) type overlap repulsion to investigate the effects of doping on the temperature dependence of the specific heat of codoped Mg1-x(AlLi)xB2 (x = 0.1-0.3) compounds in the wide temperature range 10K ≤ T ≤ 300K and Debye temperature (θD) at room temperature. This model has been successfully studied the cohesive and thermodynamic properties of MgB2, Mg1-xAlxB2 (x = 0.1-0.9), Mg1-xMnxB2 (x = 0.01-0.04) and Mg(B1-xCx)2 (x = 0.0, 0.02, 0.05, 0.075, 0.1, 0.2) in the temperature range 5K ≤ T ≤ 1000K and other diborides [3]. The computed θD is closer to the available experimental value available only at room temperature. θD progressively decreases as the doping increases. This is probably due to the softening of the phonon modes induced by the doping. Besides, we report the temperature dependence of lattice contribution to the specific heat for Mg1-x(AlLi)xB2 at x = 0.1, 0.2, 0.3 at temperature 10K ≤ T ≤ 300K.

1. Introduction

MgB2 is not a new compound, it has been known since the early 1950’s. Only in 2001, it was discovered to be a superconductor at a remarkably high critical temperature of about 39K [1]. This discovery stimulated global interest seeking higher Tc and uncovering the basic physics [2]. MgB2 attracts a lot of attention from condensed matter physicists [1, 2, 3]. Its high Tc comes from the exceptionally high vibrational energies in the graphite-like boron planes and thus MgB2 appears to obey conventional models of superconductivity. This relatively simple view (as compared to HTS) opens up a wide range of practical opportunities [2]. Based on various physical property measurements, important critical parameters of the compound viz., critical superconducting temperature (Tc), coherence length (x), penetration depth (l), critical current (Jc) and lower/upper critical fields Hc1/Hc2 have already been determined and reviewed [4, 5]. Compared to HTS, MgB2 possesses simpler structure, lower anisotropy and larger coherence length. Most interestingly MgB2 has nearly transparent grain boundaries [6], which permit excellent current transport. Higher quality grain boundaries and better superconducting critical parameters provide MgB2 an edge over widely studied HTSC cuprates [7]. MgB2 has a simple hexagonal AlB2-type structure (space group P6/mmm), which is common among borides. It contains graphite-type boron layers, which are

separated by hexagonal close-packed layers of the center of hexagons formed by boron and donate their electrons to the boron planes [1, 2]. Similar to graphite, MgB2 exhibits anisotropy in the B-B length: the distance between the boron planes is significantly larger than that in-plane B-B distance [1, 2].

The chemical substitution has been proved to be an important means in both improving the physical properties and elucidating the mechanism for superconductivity in the high Tc cuprates. However, chemical substitutions in MgB2 are not as easy as in the cuprates. Despite of simple structure and apparently simple chemistry, MgB2 has so far proved very difficult to modify systematically through chemical substitutions. This is in sharp contrast to the situation in high temperature cuprate superconductors, where chemical substitution has played an important role in evolving the nature of superconductivity. Various substitutions have been reported, but very few were successful [8-14]. These are the cases of Al on Mg site [8-10] and of C on B site [11-14]. Nevertheless, Al appears to be an exception since AlB2 has the same crystal structure as MgB2 and Al doping level ranging from zero to one can be made. In this paper, we have investigated the effect of Li doping in MgB2 doped with Al. For this purpose, we have formulated the Rigid Ion Model (RIM) which includes long range Coulomb, van der Waals (vdW) interaction and short-range Hafemeister Flygare (HF) type overlap repulsion to


investigate the effects of Li doping on the temperature dependence of the specific heat of codoped Mg1-x(AlLi)xB2 (x = 0.1-0.3) compounds in the wide temperature range 10K ≤ T ≤ 300K and Debye temperature (θD) at room temperature. Earlier this model has been successfully studied the cohesive and thermodynamic properties of MgB2, Mg1-xAlxB2 (x = 0.1-0.9) [15], Mg1-xMnxB2 (x = 0.01-0.04) [15] and Mg(B1-xCx)2 (x = 0.0, 0.02, 0.05, 0.075, 0.1, 0.2) [16] in the temperature range 5K ≤ T ≤ 1000K and other diborides [17, 18].

The interaction potential is given in the next section. The results and discussion are given in the successive section III. 2. Theory of RIM Approach

We have formulated the Rigid Ion Model (RIM) [19] by including the effects of the long-range Coulomb attraction, the short-range Hafemeister Flygare (HF) type [20] overlap repulsion, and the van der Waals (vdW) interactions. Its model potential (φ) consists of the following interatomic interactions:

Here, k (k′) denote the positive (negative) ions and sum is taken over all the ions (kk′). In the above expression, the first term represents the long-range Coulomb and the second term represents the short-range Hafemeister Flygare interaction [20]. The third and fourth terms are the van der Waals interactions due to dipole-dipole and dipole-quadrapole interactions, respectively. The Pauling coefficients [21] are defined as:

where Zk (Zk′) and Nk (Nk′) are the valence and number of electrons in the outermost orbit. The model parameters, hardness (b) and range (ρ) parameters are determined from the equilibrium condition,

and the Bulk modulus,

where k is the crystal constant and r0 is the

interatomic separation at the temperature T=0 K. The values of the model parameters, as listed in Table 1, have been used to compute the cohesive energy for codoped Mg1-x(AlLi)xB2 (x = 0.1-0.3) from Eq. (1). Other thermal properties e.g. Debye temperature (ΘD) and specific heat at constant volume (Cv) are calculated using their well-known expressions as presented elsewhere [10, 11]. The results thus obtained are presented and discussed below 3. Results and discussion By applying the equilibrium condition and Bulk modulus relation (Eqs. 2 and 3), we have computed the model parameters as a function of temperature using the input data [22] at room temperature and van der Waals coefficients calculated using SKV method [23]. Using the values of these model parameters, we have computed the cohesive energy as given by eq (1) to test the validity of our model. It has been noticed from the values of the cohesive energy that the contribution from the SR interaction is less than 10% of the total cohesive energy.

Fig. 1. Temperature dependence of specific heat of Mg0.9(AlLi)0.1B2. This feature is an indicative of the fact that the major contribution to the cohesion is due to the Coulomb attraction along with the supplementary contribution from the vdW attraction in these materials. θD is the important thermal property of the material, which represents the effective cutoff

)3(00

=⎥⎦⎤

⎢⎣⎡

= rrdrd φ

( ) ( )

)1(

/2exp/2exp

2'

/)exp(2

8'

''

6'

''

2'''''

22

1'''1

1'

''

2

rdrc

rrrr

bn

rrrbn

rZZe

kkkk

kkkkkk

kk

kkkkk

kkkkk

kkkkkk

kkkk

kk

−−

−

∑∑

∑

−−

⎥⎥⎥

⎦

⎤

⎢⎢⎢

⎣

⎡

⎥⎦

⎤⎢⎣

⎡−+

−

+−+

+−=

ρβρβ

ρβ

φ

)2()/()/(1 ''' NZNZ kkkkkk ++=β

)4(9

1

0

2

2

0 rrT dr

dKr

B=

⎥⎦

⎤⎢⎣

⎡=

φ

Low Temperature Specific Heat of Codoped Mg1-X(AlLi)XB2 225

phonon energy of the material. Therefore we have computed θD i.e. 638K, 621K and 604K for Mg1-

x(AlLi)xB2 (x=0.1-0.3) respectively,

Fig. 2. Temperature dependence of specific heat of Mg0.8(AlLi)0.2B2. which is closer to the available experimental value [24] available only at room temperature. θD progressively decreases as the doping increases. This is probably due to the softening of the phonon modes induced by the doping.

Fig. 3. Temperature dependence of specific heat of Mg0.7(AlLi)0.3B2. Besides, we report the temperature dependence of lattice contribution to the specific heat for Mg1-

x(AlLi)xB2 at x = 0.1, 0.2, 0.3 at temperature 10K ≤ T ≤ 300K in Fig. 1, 2 and 3 respectively. On inspection of the respective figures, our calculated values of specific heat shows an excellent agreement with the lattice part of the measured data of Monni et al.[24]. The trend of the specific

heat variation with temperature follows the similar trend as obtained in pure and doped MgB2. 4. Conclusions On the basis of overall results, it may be concluded that RIM is capable of giving satisfactory prediction of cohesive and thermal properties of codoped Mg1-x(AlLi)xB2 (x = 0.1-0.3) in the temperature range 10K ≤ T ≤ 300K. Acknowledgements

The authors are thankful to the University Grants Commission (UGC), New Delhi for providing financial support to this work. One of us (NK) is thankful to the Department of Science and Technology, New Delhi for the award of Fast Track Research Award. References [1] J. Nagamatsu, N. Nakagawa, T. Muranaka,

and J. Akimitsu, Nature 410 (2001) 63. [2] T. Muranaka, Y. Zenetani, J. Shimoyama

and J. Akimitsu, Frontiers in Superconducting Materials: P. 937-981, Ed. by A.V. Narlikar, Springer-Verlag publishing, Germany (2005).

[3] T. Dahm, Frontiers in Superconducting Materials: P. 9831009, Ed. by A.V. Narlikar, Springer-Verlag publishing, Germany (2005).

[4] S.L. Budko, G. Lapertot, C. Petrovic, C.E. Cunninghan, N. Andersen and P.C. Canfield, Phys. Rev. Lett. 86 (2001) 1877.

[5] O.F. de Lima, R.A. Ribeiro, M.A. Avila, C. A. Cardoso, and A.A. Coelho, Phys. Rev. Lett. 86 (2001) 5974.

[6]. S.B. Samanta, H. Narayan, A. Gupta, A.V. Narlikar, T. Muranaka, and J. Akimitsu, Phys. Rev. B. 65 (2002) 92510.

[7]. S.X. Dou, A.V. Pan, M.J. Qin, and T. Silver, Frontiers in Superconducting Materials: P. 1012-1048, Ed. by A.V. Narlikar, Springer-Verlag publishing, Germany (2005).

[8]. D.G. Hinks, H. Claus, J.D. Jorgensen, Nature 411 (2001) 457.

[9]. R.J. Cava, H.W. Zandbergen and K. Inumaru, Physica C 385 (2003) 8.

[10].A. Berenov, A. Serquis, X.Z. Liao, Y.T. Zhu, D.E. Peterson, Y. Bugoslavsky, K.A. Yates, M.G. Blamire, L.F. Cohen and J.L. MacManus-Driscoll, Supercond. Sci. Technol. 17 (2004) 1093; B. Birajdar, T. Wenzel, P. Manfrinetti, A. Palenzona, M. Putti and O. Eibl, Supercond. Sci. Technol. 18 (2005) 572.


[11].M. Paranthaman, J.R. Thompson, D.K. Christaen, Physica C 355 (2001) 1.

[12].A. Bharathi, S.L. Balaselvi, S. Kalavathi, G.L.N. Reddy, V.S. Shastry, Y. Hariharan and T.S. Radhakrishnan, Physica C 370 (2002) 211.

[13].R.A. Ribeiro, S.L. Bud’ko, C. Petrovic and P.C. Canfield, Physica C 384 (2003) 227.

[14].M. Avdeev, J.D. Jorgensen, R.A. Ribeiro, S.L. Bud’ko and P.C. Canfield, Physica C 387 (2003) 301.

[15].N. Kaur, R. Mohan, N.K. Gaur and R.K. Singh, Physica C 451 (2007) 24; N. Kaur, N. K. Gaur and R. K. Singh, Sol. Stat. Phys. (India) 50C (2005) 729.

[16].N. Kaur, N.K. Gaur, R. Mohan and R.K. Singh, J. Phys. Chem. of Solid 68 (2007) 2247.

[17].N. Kaur, N.K. Gaur and R.K. Singh, Mod. Phy. Lett B 21 (2007) 885.

[18].N. Kaur, N. K. Gaur and R. K. Singh, Sol. Stat. Phys.(India) 51 (2006) 707; N. Kaur, N. K. Gaur and R. K. Singh, Sol. Stat. Phys.(India) 51 (2006) 705.

[19].E.W. Kellermann, Phil. Trans. R. Soc. Lond. A 238 (1940) 513.

[20].D.W. Hafemiester and W.H. Flygare, J. Chem. Phys. 43 (1965) 795.

[21].L. Pauling, Z. Kristallogr., 67 (1928) 377; L. Pauling, J.A.C.S., 50 (1928) 1036; L. Pauling, The Nature of the Chemical Bond (Ithaca: Cornell University Press), 1945.

[22].M. Monni, C. Ferdeghini, M. Putti, P. Manfrinetti, A. Palenzona, M. Affronte, P. Postorino, M. Lavagnini, A. Sacchetti, D. Di Castro, F. Sacchetti, C. Petrillo and A. Orecchini, Phys. Rev B 73 (2006) 214508.

[23].J.C. Slater and K.G Kirkwood, Phys. Rev 37 (1931) 682.

[24].M. Monni, C. Ferdeghini, M. Putti, P. Manfrinetti, A. Palenzona, M. Affronte, P. Postorino, M. Lavagnini, A. Sacchetti, D. Di Castro, F. Sacchetti, C. Petrillo and A. Orecchini, Phys. Rev B 73 (2006) 214508.

Restricted Diffusion in Nano- Channels

K. Tankeshwar and Sunita Srivastava*

Computer Center, Department of Computer Science and Applications, Panjab University, Chandigarh-160014, INDIA.

* Department of Physics, Panjab University, Chandigarh-160014, INDIA. E-mail: [email protected], [email protected]

Abstract

In this work we present results for the liquid metal confined to nano width along its liquid vapour

coexistence curve. It has been found that as one moves towards confining walls the self diffusion coefficient decreases, affecting the fluidity of the fluid in a nano -channel. It has also been reaffirmed that the effect of confinement is more for denser fluid than for a dilute fluid. The results are contrasted with that of inert fluids. The macroscopic self- diffusion coefficient has been found to be affected more by the micro/nano-scale confinement in liquid metal than in inert fluid when both are near their triple point.. The denser fluid provides an additional artificial wall which further restricts the flow of fluid. Relevance of the work to the study of flow of fluid like blood in arteries has also been discussed. 1. Introduction

In recent years the study of fluids confined to micro/nano geometries has attracted many researchers due to its multifold applications in science and technology [1-3]. Fabrication [1] of nanofluidics devices with channels down to 10nm has been carried out and could hold fluids in the range of pico- liter. The study of transport coefficients of a fluid in nano devices is one of the major concerns of the researchers. Recently, we have proposed [4] a dynamical model to study the effect of confinement down to nano-dimension on the self- diffusion coefficient. The model was built on the consideration that the confinement affects molecular motion at microscopic level. The model introduces the concept of microscopic (local) self-diffusion coefficient that varies as a function of distance from the walls of the channel. The model has been applied to inert liquids like Ar or Kr. However, it remains to investigate the difference between the flow of fluid which is of more complex than that of inert fluid. Therefore in the present work we present results for the liquid metal along the liquid vapour coexistence curve. The results are also contrasted with that of inert fluids. We consider the two fluids near their triple point. 2. Modeling

We present below the model to study the diffusion of particles in a direction perpendicular to the confining wall. We consider the fluid to be confined to in a one of the direction. The configuration space of many

body systems is divided into number of cells. Each cell is characterized by a fixed configuration associated with the local minima on the potential energy hyper-surface of the system. We assume that (i) Within the cell the liquid configuration executes harmonic motion about local minima which are described by fixed frequency (ii) The system jumps between cells with a certain jump frequency τ -1. (iii) Within the cell the liquid configuration executes harmonic oscillations about local minimum which are described by frequency ω. (iv) The frequency of oscillation is affected due to presence of external factors like compression etc.

To understand the direct effect of confinement on atomic motion at first instance we consider only non-structural walls. The walls first affect the amplitude of the harmonic motion of the particles in a given cell which are closer to the wall in direction of confinement. This in turn affects the motion in z-direction of neighboring particles. Thus when the width of channel is of nano/micro size particles the particles find themselves in compression like situation. This will bring assumption (iv) in force and one has to study the frequency defined locally. To consider the effect of confinement on the frequency and amplitude of single harmonic oscillator executing motion in z direction we consider motion represented by z(t)=Asin(ωt) with A as amplitude when the liquid is not confined. Let particle in a given cell experiences a compression like situation due to confinement which results in decrease in its amplitude by ‘d’. Let at t = t1 particle reaches


confined maxima with z(t1)=Ad=Asin(ωt1) which provides

)1(sin1 11 A

dt −= −

ω. (1)

This results in change in frequency and velocity of the particle. Let the new frequency be represented by Ω and is obtained to be

)/1(sin22 1

1 Adt −==Ω −

πωπ . (2)

From the above expression it is obvious that when compression‘d/A’ becomes zero, frequency remains unchanged and for definite positive compression new frequency is less than old frequency. In a channel the compression will vary with distance from the wall as a function of z and let it be represented by c(z). The expression for the frequency is then modified to

))(1(sin2

)( 1 zcz

−=Ω −

πω. (3)

Considering the waiting time distribution [5] for the cell jump effecting the contents of subvolume given by sech(t/τ), the Velocity auto correlation function is then given by

))(cos()/(sec),( tzthmTkztV B Ω= τ ,(4)

where kB , T and m are Boltzmann constant, temperature of the system and mass of particles of the system. The local diffusion coefficient related to the time integral of velocity auto-correlation function is then obtained to be

)2)((sec2

)( τπτπ zhmTkzD B Ω= .(5)

The self diffusion coefficient is then given by

∫=l

dzzDl

D0

.)(1 (6)

In above equation 2l is width of the cannel in z direction measured in units of atomic diameter. The effect of wall will reduce as one move away from it and will be nearly zero at the center of the channel. If we follow exponential law for fall of effect of wall on atomic motion then c(z) is given by c(z)=exp(-(l-z)). For large value of l the effect

on diffusion coefficient will be very small. However, for small width of channel (few atomic diameters) the diffusion coefficient reduces drastically. The above consideration has provided [4] a simple and useful understanding about the diffusion coefficients in a direction perpendicular to the wall. 3. Results and Discussion

In order to calculate diffusion coefficient we require sum rules of velocity auto correlation as input. We use results obtained by Sharma and Tankeshwar [6] for expanded Rb along its liquid vapour curve. The results obtained for liquid Rb from the above model is shown in Fig.1. In Fig 1. we have plotted ratio of self diffusion coefficient D(z); to the value D(0) versus z.

0 20 40 60 80 1000.0

0.2

0.4

0.6

0.8

1.0

D(z)

/D(0

)

Z

T=350 T=1373

Fig. 1. Variation of D(z)/D(0) with z. Here z (in units of atomic diameter) represents distance (in the direction of confinement) of middle of geometry from the wall confining the fluid. In the present case we consider nano channel has width which is 200 times of atomic diameter and is of the order of 60-70 nm. The results are shown at T=350 K (m.p.) and 1373 K. It is seen that as z is within the 10 atomic diameter of the wall the self diffusion starts decreasing. This is more evident in case of liquid Rb than in case of vapour at high temperature. Thus denser is fluid it has more tendency of showing the freezing near the walls of confinement. If the perpendicular diffusion falls, it affects other directional diffusion coefficients. Hence, on freezing can provide an artificial wall, which decreases the width of a channel and can trigger more restriction on the diffusion coefficient.

Restricted Diffusion in Nano- Channels 229

Since the complexities of the fluid enter in our model through the sum rules, the model can be applied to any complex liquid like blood. Arteries are of nano size if we compare it with the average size of blood particles. Thinner blood provide a smooth flow, however if blood is thick (dense) it has chances of clotting (freezing) near the wall. If this process continued and blood has achieved the critical value of thickness it can even block the arteries. We also compare results with those obtained for liquid Ar at its triple point. The comparison of results is shown in Fig.2 wherein D(z)/D(0) is plotted against z. In Fig.2 solid line represents the diffusion of liquid metal whereas the dotted line that of the liquid Ar. It can be seen from the Fig. 1 that as one moves towards the confining walls the self- diffusion coefficient decreases, affecting the fluidity of the fluid in a nano - channel. It is seen that layers of fluids which are close to wall (at z=100) have diffusion coefficient which is comparatively lesser than at the middle. The macroscopic self- diffusion coefficient has also been found to be affected more by the micro/nano-scale confinement in case of liquid metal than in inert fluid. It has also been reaffirmed that the effect of confinement is more for denser fluid than for a dilute fluid. The denser fluid provides an additional artificial wall which further restricts the flow of fluid. Relevance of the work to the study of flow of fluid like blood in arteries shall also be discussed.

Fig. 2. Variation diffusion coefficient with z. Wall is at z=100. Solid line: Rb metal, dashed line: liquid Ar. References: [1] Anpan Han et al (2006) Nanotechnology

17 2498 [2] Gatimu E.M., Sweedler J.V. and Bohn

P.W. (2006) Analyst, 705 [3] Aggarwal N, Sood J., Tankeshwar

K.,(2007) 18, Nanotechnology, 335707 [4] K. Tankeshwar and S. Srivastava (2007),

18, 485714 Nanotechnology [5] K. Tankeshwar, B.Singla and K.N.Pathak

J.Phys. Cond. Matt. (1991) 3 3173 [6] Sharma Saroj and K.Tankeshwar, (1996):

Condensed Matter 8, 10839.

Mechanical Strength and Phase Transformation of AIN

P. S. Bisht, Virshali Joshi and U. P. Verma SOS in Physics, Jiwaji University, Gwalior - 474011 (M.P), Government Post Graduate College, Guna - 473001 (M.P.)

E-mail: [email protected] : [email protected]

Abstract

Group III-nitrides are of great technological interest in both fundamental sciences and technical application. Most of the common nitrides are well known as hard and wide band gap semiconductor materials. In general they have been studied in zinc blend and wurtzite phases. In this paper we have used first principle method to study the mechanical and electronic properties (band structure) of the AlN in RS (B1) and WZ (B4) phases as a function of pressure. All the calculations are performed using the VIENNA package (Wien2k_07). This is the full potential augmented plane wave (LAPW) within the density functional theory (DFT). The generalized gradient approximation (GGA) is used for the exchange and correlation function. The present results show that AlN in B1 phase is harder than their pre-existing phases (WZ and ZB). The obtained results are compared with the experimental and other theoretical work.

1. Introduction

The III-nitrides are now days widely used by semiconductor industry. Recently their has been increasing scientific and technological interest in wide band gap semiconductors due to several outstanding mechanical properties such as the high melting point, high thermal conductivity, mechanical stability and large Bulk moduli. Furthermore, for their electronic properties, in particular for the large band gap and low dielectric constant, the wide band gap semiconductors are promising materials for technological developments of opto- and microelectronic devices working in the ultra violet spectral region and under high temperature conditions. Among the all III-nitrides, AlN has the largest band gap and most of the work reported so far refers to the stable hexagonal wurtzite (WZ) phase of AlN [1,2]. The metastable cubic ZB modification arises as an advantageous alternative for devices. Despite the technological developments of AlN, few of the theoretical work were made for ZB [3-7] and RS phase [8-10].

In this paper, we focus our attention to structural, electronic, phase transition and elastic properties of AlN in ZB and RS phases as a little work has been reported either theoretically or experimentally on elastic and electronic properties of AlN; especially in RS phase. We have also studied the transition pressure at which ZB phase changes to RS phase. To solve our purpose we use the FP-LAPW approach. The obtained results are compared with the previous theoretical calculations and available experimental finding.

2. Computational details

We use the first-principle total energy calculation for RS (B1) and ZB (B3) under hydrostatic pressure. The calculations are performed using the FP-LAPW approach plus local orbital [11, 12] with in the frame work of density functional theory (DFT) [13] as implemented in the WIEN2K code [14]. The exchange correlation potential was calculated by generalized gradient approximation (GGA) based on the Perdew-Burke-Ernzerhof (PBE) [15].

In this method, the unit cell is partitioned into non-overlapping muffin-tin spheres around the atomic sites, and an interstitial region. Among these two types of regions different basis sets are used. In the ZB structure the muffin-tin (MT) sphere radii of 1.72 and 1.62 Bohr were used for Al and N atoms, respectively, while in the rock-salt structure 1.70 and 1.60 Bohr were used because the lattice constants of the two structures are different. The valance wave functions inside the MT sphere were expended into spherical harmonics up to l=10 and the RmtKmax were taken to be 7.0.

We use 1000 k points in the Brillion zone for the zinc-blende structure and 1200 k points for the rock-salt structure. The self-consistent calculations were considered to be converged only when the total energy and the charge of the system meet the convergence limit of 0.0001 eV and 0.001eV respectively. This method has been

Mechanical Strength and Phase Transformation of AIN 231

proved successful in the determination of the structural and mechanical properties of the semiconductor materials [16]. 3. Results and discussions 3. 1. Structural properties

In this study, the exchange-correlation potential function, the GGA, were employed to characterize the atomic behavior of cubic phase (ZB and RS) AlN at ambient condition. The obtained lattice constants and bulk modulus are listed in table 1. The bulk modulus (B) and lattice constants were evaluated from the Murnaghan equation fitting of the total energies as a function of the unit cell volume. The calculated bulk moduli and lattice constants are in good agreement with other theoretical work listed in table 1. From table it may be observed that bulk modulus of RS AlN is higher than ZB AlN. It implies that the AlN in RS phase is mechanically more stable than ZB phase. Further, this behavior of AlN is admitted in next section 3.2 (elastic properties).

-595.37

-595.36

-595.35

-595.34

-595.33

-595.32

-595.31

-595.305 10 15 20 25

Volume ( Ǻ )

Tota

l ene

rgy

(eV)

RS

ZB

Fig. 1. Plot of Total energy as a function of unit cell volume for AlN in ZB (dark line) and RS (light line) structures. Figure 1 show the calculated energy-volume curves for ZB and RS AlN. From figure, the total energy per formula unit of the ZB phase is about 0.021 Ryd smaller than that of the RS phase. This indicates that the ZB phase is the ground state phase at zero pressure. At a sufficient high pressure the rock salt phase would be favored. The common tangent construction gives a equilibrium transition pressure (Pt) value of about 12.29 GPa. Pt is the pressure at which the enthalpy of the initial ZB phase is equal to that of the final RS phase. Our calculated transition pressure along with lattice constant is in good agreement with earlier reported theoretical works [8, 10] table1.

3. 2. Elastic properties

In this section we investigate the high pressure behavior of the elastic moduli for cubic AlN. Figure 2 shows the pressure dependence of the elastic constants. All constants increase monotonically with increasing pressure. The elastic constant satisfies the generalized elastic stability criteria for cubic crystal under hydrostatic pressure [17], C11+ C12>0, C44 >0 and C11 – C12>0, where C11, C12 and C44 are the relevant elastic stiffness constants for the cubic structure. The first term in the mechanical stability criteria is closely related to the bulk modulus which is obviously necessary for stability. The latter two are thought to be play a more subtle role since C44 involves shearing of the Al-N bonds and C11 - C12 involves stretching (and compression) of the Al-N bonds with a combination of bending and starching of Al-Al bonds. We note that the inequality C11>C44>C12 is satisfied over wide pressure range for ZB and RS AlN.

C44

0100200300

400500600

-5 0 5 10152025303540

Pressure (GPa)

Ela

stic

stif

ness

co

nsta

nt (G

Pa) C11

C12

ZB

0100200300400500600700

-5 0 5 10 15 20 25 30 35 40

Pressure (GPa)

Ela

stic

stif

ness

co

nsta

nt (G

Pa) C11

C44

C12

RS

Fig. 2 Plot of elastic moduli as a function of pressure for ZB and RS AlN.

The elastic stiffness constants were obtained by fitting the total energy of the strained crystal to third order polynomial of the strains. Since there are three independent elastic constants for a cubic phase, three type of


strained, viz, the volume change, volume conserved tetragonal and rhombohedral shear strains were applied to the optimized structures to calculate the elastic constants. The calculated elastic stiffness constants (C11, C12, C44) for RS AlN are greater than ZB AlN for wide range of pressure. At zero pressure the calculated elastic constants (table 1) are in close agreement with the earlier reported theoretical work [8-9] for ZB AlN, but for RS AlN this information is being provided for the first time.

PW Exp. Theoretical

ZB 4.38 4.370* 4.377[8], 4.31-4.376*Lattice

Constant (in Ǻ) RS 4.07 4.05* 3.978*, 4.229*,

4.0302[8]

ZB 195.5 202 [25]

203.2[26] 194.028[8],211-216*

Bulk moduli(B) (GPa)

RS 254.3 295* 236, 370-348 [24] 256.89[8], 297.5*,

272*

Transition pressure (GPa)

ZB

RS 12.29 -- 4.5[8], 5.0[7], 7.1[10], 12.29a, 16.6a

* refs of [8], a refs of [10]

Table 1. Calculated theoretical equilibrium lattice constant, bulk modulus and transition pressure of ZB and RS AlN using murnaghan equation of state.

C11 C12 C44

PW

ZB

RS

315

420

130

165

245

300

Exp. - - -

Theo.

ZB

RS

330.71[8], 294-313*

-

162.199[8], 152-168*

-

156.41[8], 78-202*

-

* refs of [8]

Table 2. Calculated elastic constant, bulk modulus and transition pressure of ZB and RS AlNusing the third order polynomial of the strain.

Fig. 3. Plot of bulk moduli as a function of pressure for ZB and RS AlN.

Aggregate values of the bulk modulus (B) were obtained from the individual elastic constants (figure 2). The bulk modulus increases with the increase in pressure and reaches 300 and 375 Gpa, respectively, for ZB and RS AlN at a pressure of 35 Gpa. For this wide range of pressure the calculated bulk moduli for RS AlN is greater than the bulk moduli of ZB AlN. From the figure it is interesting to note that at 28 Gpa of pressure the bulk modulus of RS AlN reaches to 348 Gpa which corresponds to the bulk modulus of the superconducting phase of NbN [18] and at 30 Gpa of pressure the bulk modulus of RS AlN reaches to 369 GPa the zero pressure value of the bulk modulus of cubic BN [19]. From the literature it is clear that the bulk modulus of BN is higher than the bulk modulus of NbN. The behavior of moduli suggests an increasing hardness in the ZB and RS AlN with pressure.

3.3. Electronic properties

Figure 4 is a plot of energy band gap as a function of pressure. The calculated optical band gaps at zero pressure are 3.31 and 4.40 for ZB and RS AlN, respectively. It is observed that both the structures are indirect in band gap and increases monotonically with pressure. As expected from DFT calculations, the band gap

ZB

0

100

200

300

400

-5 0 5 10 15 20 25 30 35 40Pressure (GPa)

Bul

k M

odul

ii (G

Pa)

RS

0100200300400500

-5 0 5 10 15 20 25 30 35 40

Pressure (GPa)

Bulk

Mod

uli

(GPa

)

Mechanical Strength and Phase Transformation of AIN 233

obtained is somewhat lower than the experimental values of 5.4 eV [20] for ZB AlN. Many papers are reported [16, 21, 22, 23] to estimate the band gap of AlN in ZB phase, but, still RS phase is lacking such data both experimentally as well as theoretically. Our estimated band gap of AlN in ZB phase is in good agreement with the earlier reported band gap of 3.1 eV [23] and 3.24 eV [8].

Fig. 4 Plot of energy bandgap as a function of pressure for ZB and RS AlN

4. Conclusions

We have noticed that AlN in ZB phase has the highest total energy, but elastically it is less stable than RS phase. The behavior of the elastic moduli suggests an increasing hardness in AlN with pressure. The bulk moduli evaluated from the elastic constants at zero pressure were consistent with those obtained from Murnaghan equation for ZB and RS AlN. Our present study shows that AlN in RS phase has high intrinsic strength. Therefore, AlN in RS phase is effective for anti-oxidant properties.

References

[1] S. C. Jain, M. Willander, J. Narayan, and R. van Overstraeten, J. Appl. Phys. 87 (2000) 965.

[2] A. G. Bhuiyan, A. Hashimoto, and A. Yamamoto, J. Appl. Phys. 94 (2003) 2779.

[3] K. Kim, W. R. L Lambrecht, B. Segall, Phys. Rev. B. 53 (16) (1994) 310.

[4] N. E. Christensen, I. Gorczyca, Phys. Rev. B 50, (1994) 4397.

[5] A. F. Wright, J. Appl. Phys. 82 (1997) 2833.

[6] K. Karch, J. M. Wagner, F. Bechstedt, Phys. Rev. B 57 (1998) 7043.

[7] S. Goumiri-said, M. B. Kanoun, A. E. Merad, G. Merad, H. Aourag, Chem. Phys. 302 (2004) 135

[8] Y. Ö. Ciftci, K. Colakoğlu, and E. Deligöz Phy. Stat. sol. ( c ) 4 (2007) 234-237.

[9] C. C Silva, H. W. Leite Alves, L. M. R. Scolfaro, and J. R. Leite Phys. Stat. sol © 2,No. 7 (20005) 2468-2471

[10] Jorge Serrano and Angel Rubio, E. Hernandez, A. Munoz, Phys. Rev. B. 62 (16), (2000) 16612-16622.

[11] Sjostedt, E., Nordstrom, L. & Singh, D. J. Sol. Stat. Commun. 114, (2000) 15-20.

[12] G. K. H. Madsen, P. Blaha, K. Schwarz, E. Sjostedt, & L. Nordstrom, Phys. rev. B 64 (2001) 195134.

[13] P. Hohenberg and W. Kohn, Phys. Rev. 136 (1964) 864

[14] P. Blaha, K. Schwarz, P. I. Sorantin, and S. B. Trickey, Comput. Phys. Commun. 59 (1990) 399; K. Schwarz, P. Blaha, and G. K. H. Madsen, Comp. Phys. Commun. 147 (2002) 71; K. Schwarz and P. Blaha, Comput. Mat. Sci. 28 (2003) 259.

[15] Perdew J.P., Burke S. and Ernzerhof M. 1996, Phys.Rev.Let. 77, 3865.

[16] F. Litimein, B. Bouhafs, Z. Dridi, and P. Ruterana, N. J. Phys 4 (2002) 64.

[17] J. F. Nye, Physical Properties of Crystal (Oxford University Press, Oxford, 1985).

[18] Xiao-Jia Chen, Viktor V. Struzhkin, Zhigang Wu, Ronald E. Cohan and russel J. Hemley PNAS 102, (2005) 9.

3.00

3.10

3.20

3.30

3.40

-5 0 5 10152025303540

Pressure (GPa)

Ene

rgy

band

gap

(eV

)

ZB

3.003.504.004.505.005.506.00

-5 0 5 10 15 20 25 30 35 40

Pressure (GPa)

Ener

gy b

andg

ap (e

V) RS


[19] C. Lee, W. Yang and R. G. Parr, Phys. Rev. B 37 (1988) 785.

[20] I Vurgaftman, J. R. Meyer, and L. R. Ram-Mohan, J. Appl. Phys. 89 (2001) 518; I. Vurgaftman and J. R. Meyer, J. Appl. Phys. 94 (2003) 3675.

[21] S. K. Pugh, D. J. Dugdale, S. Brand, and R. A. Abram, Semicond. Sci. Technol. 14 (1999) 23.

[22] A. E. Merad, M. B. Kanoun, J. Cibert, H. Aourag, and G. Merad, Mat. Chem. Phys. 82 (2003) 471.

[23] P. Jonnard, N. Capron, F. Semond, J. Massies, E. Martinez-Guerrero, H. Mariette European Phys. J. B 42 (2004) 351.

[24] B.Delley, J. Chem. Phys. 113 (2000) 7756.

[25] M. E. Sherwin and T. J. Drummond, J. Appl. Phys. 69 (1991) 8423.

[26] S.Q Wang and H Q Ye, J. Phys. Cond. Matter 14 (2002) 9579-9587.

Novel Feature of Quantum Transport Through Ultra-Thin Quantum Film

Santanu K. Maiti1,2

1Theoretical Condensed Matter Physics Division, Saha Institute of Nuclear Physics 1/AF, Bidhannagar, Kolkata-700 064, India

2Department of Physics, Narasinha Dutt College, 129, Belilious Road, Howrah-711 101, India E-mail: [email protected]

Abstract

A novel transport phenomenon is explored by investigating the disorder effect on electron transport through an ordered-disordered separated ultra-thin quantum film with side attached metallic electrodes. In the strong disorder regime, the current amplitude increases with the increase of the disorder strength, while it decreases in the weak disorder regime. This anomalous behavior is completely opposite to that of a bulk disordered system, where the current amplitude always decreases with the increase of the disorder strength. In this present model, we also discuss the significant effects of the total number of layers and the size of each layer of the film on such transport.

1. Introduction

Much progress in nanoscience and technology has enabled one to study electron transport through nano-structures including atomic wires or different types of organic molecules bridging over electrodes in a very tunable environment. These quantum systems are the recent focus of nanotechnologies since they are taken as the basic units of future nanoelectronic devices, and the transport properties through such systems provide many attractive features as predicted by several works so far. The field of molecular electronics offers a great potential since the electron transport is predominantly coherent1,2 in molecular devices and the study of such systems led recently to numerous achievements on the part of several experiments. The theoretical description of the electron transport through a molecule placed between two metallic electrodes was first studied by Aviram et al.3 in 1974, and later, lot of experiments4,5 have been carried out through different molecular bridge systems. The sensitivity of the electron transport through a molecular bridge significantly depends on the several key factors, like (a) quantum interference of electron waves6,7,8 associated with the geometry that the molecule adopts within the junction, (b) coupling strength9,10 of the molecule to the electrodes, (c) structure of the molecule itself11 and many other important parameters of the Hamiltonian that are needed to describe the

system. The investigation of the current-voltage (I-V) characteristics in the bridge systems reveals several common features. Depending on the molecular coupling strength, the overall (I-V) characteristics show step-like or fairly smooth-like behavior and the amplitude of these currents significantly depends on both the molecular coupling and as well as on the quantum interference of the electron waves12,13. Although all these effects have been pointed out quite extensively in the literatures, but yet the full knowledge of the conduction mechanism through a bridge system is still unclear to us. With the advancement of nanoscience and technology it can be made possible to provide a mesoscopic device where the charge carriers are scattered mainly from its surface boundaries12 and not from the inner core region. This is completely opposite to that of a conventional doped system where the dopant atoms are distributed uniformly along the nanowire. Motivated by such kind of systems, in this article we concentrate our study on the electron transport through a finite size ordered-disordered separated quantum film attached to two metallic electrodes, and explore a novel feature of the electron transport in which the current amplitude increases with the increase of the disorder strength in the strong disorder regime, while it decreases in the weak disorder regime. From our study it is also observed that, the electron transport through the film is significantly influenced by the total number of layers and the


size of each layer of the film which manifest the finite quantum size effects. Here we utilize the simple tight-binding model to describe the system, and adopt the Newns-Anderson chemisorption model13,14 for the description of the electrodes and for the interaction of the electrodes with the film. We organize this article specifically as follows. In Section 2 we describe the model and formulate the theoretical method for our calculations. Section 3 discusses the significant results, and at the end, our conclusions will be available in Section 4. 2. The model and a brief description of the theoretical formulation This section provides the model and the methodology for the calculation of the transmission probability (T), conductance (g)

Fig. 1. A quantum film of two layers attached to two metallic electrodes (source and drain). The front layer (green color) is the disordered layer, where the impurities are given in all the lattice points of this layer, while the other one layer (light blue color) is the ordered layer. The couplings between the two successive layers are described by the red color. and the current (I) through a thin quantum film attached to two semi-infinite one-dimensional metallic electrodes by the use of Green's function technique. The schematic view of such a bridge system is shown in Fig.1. At very low temperature and small applied voltage, the conductance g of the thin film is expressed through the Landauer conductance formula15

Theg

22= (1)

where the transmission probability T can be written as15

][ aFilmD

rFilmS GGTraceT ΓΓ= (2)

In this expression rFilmG and a

FilmG are the retarded and the advanced Green's functions of the thin film, and ΓS and ΓD describe its coupling with the source and the drain, respectively. The Green's function of the film is represented in this form

1)( −Σ−Σ−−= DSFilmFilm HEG (3) where E is the energy of the source electron and HFilm corresponds to the Hamiltonian of the film which can be written in the tight-binding representation within the non-interacting electron picture as

∑∑><

++=ij

ijjiiii

iFilm cccctccH )( †††ε (4)

Here єi’s correspond to the on-site energies and assuming the nearest-neighbor hopping strengths both for the longitudinal and the transverse directions in each layer of the thin film are identical with each other, we set the hopping strength by the parameter t. Similar hopping strength t is also taken for the two nearest-neighbor lattice points in the successive layers of the film, as an approximation. To introduce the impurities in this film, we take the site energies (єi’s) randomly form a “Box” distribution function of width W i.e., the system is subjected to the diagonal disorder only. In this present model, the two electrodes are described by the similar kind of tight-binding Hamiltonian as prescribed in Eq.(4), where the on-site energy and the nearest-neighbor hopping strength are represented by the parameters єi′ and v, respectively. The couplings of these two electrodes to the film are represented by the hopping parameters τS and τD, respectively. In Eq.(3), ΣS and ΣD represent the self-energies due to the coupling of the film with the source and the drain, respectively, and here we use the Newns-Anderson chemisorption model13,14 to incorporate all the informations of this coupling. By utilizing the Newns-Anderson type model, we can express the conductance in terms of the effective film properties multiplied by the effective state densities involving the coupling. This allows us to study directly the conductance as a function of the properties of the electronic structure of the film between the electrodes. The current passing across the film can be assumed as a single electron scattering process between the two reservoirs of charge carriers. The current-voltage (I-V) characteristics can be computed through the expression15

Novel Feature of Quantum Transport Through Ultra-Thin Quantum Film 237

dEETffeVI DS )()()( ∫∞

∞−

−=ηπ

(5)

Where fS(D)=f(E-μS(D)) gives the Fermi distribution function with the electrochemical potential μS(D)=EF-eV/2. For the sake of simplicity, here we assume that the entire voltage is dropped across the film-to-electrode interfaces and this assumption doesn’t significantly affect the qualitative aspects of the current-voltage characteristics. Such an assumption is based on the fact that the electric field inside the film, especially for short films, seems to have a minimal effect on the g-E characteristics. On the other hand, for larger film sizes and higher bias voltage, the electric field inside the film may play a more significant role depending on the size and the structure of the film16, but yet the effect is too small. Our main aim in this article is the determination of the typical current amplitude which can be computed through the relation

VWtyp II ,2 ⟩⟨= (6)

where, W and V correspond to the impurity strength and the applied bias voltage, respectively. Throughout this article we study all the results at absolute zero temperature, but the qualitative behavior of these results are invariant up to some finite temperature (~ 300 K). The reason for such an assumption is that, the broadening of the energy levels of the thin film due to its coupling with the electrodes is much larger than that of the thermal broadening. For simplicity, we take the unit c=e=h=1 in our present description. 3. Results and discussion Here we explore the significant results and describe the strange effect of disorder on the electron transport through an ordered-disordered separated thin film attached to two metallic electrodes. Our study predicts the essential mechanisms and the basic principles for the control of electron transport through any finite size conducting bridge system. To illustrate the size of the thin film we use the parameters Nx, Ny and Nz, where they correspond to the total number of atomic sites along the x, y and z directions, respectively. In the ordered-disordered separated films, we introduce the impurities in the front layer (layers) keeping the back layer (layers) as impurity free, while for the bulk disordered films the impurities are given in

all the layers choosing the site energies randomly from a “Box” distribution function of width W . The two electrodes are connected at the two extreme corners of the lowermost back layer as presented in Fig.1. Throughout our discussion, we set the values of the different parameters as follows: the hopping strengths of the film to the electrodes (source and drain) τS=τD=1.5 and the nearest-neighbor hopping strength t=2. In the electrodes, we take the hopping strength between the nearest-neighbor sites v=4 and, for the sake of simplicity, the on-site energies (єi′’s) of all the atomic sites in these electrodes are taken as zero.

Fig. 2. Typical current amplitudes (ITyp) as a function of the impurity strength (W) for the thin films with two layers (Nz=2), where the system size for each layer is taken as: Nx=4 and Ny=4. The red and the blue curves correspond to the ordered-disordered separated and the bulk disordered films, respectively. In the ordered-disordered separated film, we introduce the impurities only in the front layer. Since the impurities are taken randomly from some “Box” distribution function, we compute the typical current amplitude (Ityp) by averaging over 50 random disordered configurations in each case to get much more accurate results. In Fig.2, the results of the typical current amplitudes as a function of the impurity strength W are shown for the quantum films considering the system size Nx=4, Ny=4 and Nz=2. The red and the blue curves correspond to the ordered-disordered separated and the bulk disordered thin films, respectively. In this two-layer ordered-disordered separated film, we introduce the impurities only in the front layer keeping the back layer as impurity free. For the bulk disordered thin film, the typical current amplitude gradually decreases with the increase of the impurity strength W. This can be very well understood from the theory of Anderson localization, where we get more and more


localization with the increase of the impurity strength. Such a phenomenon is very well established in the transport community from a long back ago. A peculiar behavior is observed in the variation of the typical current amplitude with the impurity strength only when we introduce the impurities in the front layer, keeping the back layer as impurity free. It is observed that, the current amplitude initially decreases with the increase of the impurity strength, but after reaching to a minimum the current amplitude again increases with the impurity strength. This feature is completely opposite to that of the bulk disordered system.

Fig. 3. Typical current amplitudes (ITyp) as a function of the impurity strength (W) for the thin films with three layers (Nz=3), where the system size for each layer is taken as: Nx=3 and Ny=3. The red and the blue curves correspond to the identical meaning as in Fig. 2. In the ordered-disordered separated film, we introduce the impurities only in the topmost front layer keeping the other two layers as impurity free. Such an anomalous signature can be justified in this way. We can assume the ordered-disordered separated film as a coupled system, where the disordered layer is coupled with the impurity free layer. Thus it can be emphasized that, in the absence of any coupling the localized states are obtained in the disordered layer, while the extended states are obtained in the ordered layer. For the coupled film, the coupling between the localized and the extended states is strongly influenced by the strength of the impurity and is inversely proportional to the impurity strength. Accordingly, the coupling effect becomes strong in the weak impurity regime, while the effect is less significant in the limit of strong impurity. Therefore, in the weak disorder regime the electron transport is strongly influenced by the disordered layer in which the electron states are scattered more and accordingly, the current

amplitude decreases. On the other hand, for the strong disorder regime the extended states are less influenced by the disordered layer since the coupling is weak in this regime, and it gradually decrease with the increase of the disorder strength which provides the larger current amplitude. Thus we can say that the electrons moving through the ordered layer encounter weaker resistance from the disordered layer and they achieve higher mobility. For the large enough disorder strength, the extended states of the ordered layer are almost unaffected from the disordered layer, and in that situation we get the current only due to the extended states which is the trivial limit. So the exciting limit is the intermediate limit of W. To emphasize the effects of the system size and the number of layers in the film on such electron transport, we concentrate our study on the results those are given in Fig. 3 and Fig. 4. The typical current amplitudes plotted in Fig. 3 are given for the films with system size Nx=3, Ny=3 and Nz=3, while the results shown in Fig. 4 correspond to the films with system size Nx=5, Ny=5 and Nz=4. The red and the blue curves both for these two figures represent the identical meaning as in Fig. 2. For the ordered-disordered separated film with Nz=3, we introduce the impurities in the topmost front layer, while, for the case of Nz=4 the impurities are given in the topmost two front layers keeping the other layers

Fig. 4. Typical current amplitudes (ITyp) as a function of the impurity strength (W) for the thin films with four layers (Nz=4), where the system size for each layer is taken as: Nx=5 and Ny=5.

The red and the blue curves correspond to the identical meaning as in Fig. 2. In the ordered-disordered separated film, we introduce the impurities only in the topmost two front layers keeping the other two surfaces as impurity free. For all these films, we get the similar kind of behavior both for the ordered-disordered

Novel Feature of Quantum Transport Through Ultra-Thin Quantum Film 239

separated and the bulk disordered cases as predicted in Fig. 2. This is quite obvious since we can map the ordered-disordered separated film with more than two layers into an effective two-layer ordered-disordered separated film. So this is the unique property of any ordered-disordered separated thin film. Now the other significant observation is that the typical current amplitude where it goes to a minimum strongly depends on both the system size and the number of the layers of the film, which reveal the finite quantum size effects in the study of electron transport phenomena. The underlying physics behind the location of the minimum in the current versus disorder curve is quite interesting. The current amplitude is controlled by the two competing mechanisms. One is the random scattering in the ordered region due to the localization in the disordered region which tends to decrease the current, and the other one is the vanishing influence of random scattering in the ordered region due to the strong localization in the disordered region which provides the enhancement of the current. Now depending on the ratio of the atomic sites in the disordered region to the atomic sites in the ordered region, the vanishing effect of random scattering from the ordered states dominates over the non-vanishing effect of random scattering from these states for a particular disorder strength (W=Wc), which provides the location of the minimum in the current versus disorder curve. 4. Conclusion In conclusion of this article, we have focused a strange effect of disorder on quantum transport through a small finite size ordered-disordered separated quantum film attached to two metallic electrodes by the use of the Green’s function technique, based on the tight-binding formulation. Our results have provided an anomalous quantum transport in which the current amplitude increases with the increase of the disorder strength in the strong disorder regime, while the current amplitude decreases in the weak disorder regime. This is the unique property of the electron transport through any ordered-disordered separated quantum film. Such a novel feature is completely opposite to that of the conventional disordered system, where the current amplitude always decreases with the increase of the disorder strength. Lastly, we have observed that for an ordered-disordered separated film, the typical current amplitude where it goes to a minimum strongly depends on

the size and also on the total number of layers of the film which manifest the finite quantum size effects. Throughout our study of the electron transport, we have considered several realistic assumptions by ignoring the effect of electron-electron interaction and the influence of all the inelastic processes. More studies are expected to take into account the Schottky effect, the static Stark effect, etc. References

1. A. Nitzan, Annu. Rev. Phys. Chem. 52 (2001) 681.

2. A. Nitzan, M. A. Ratner, Science 300 (2003) 1384.

3. A. Aviram, M. Ratner, Chem. Phys. Lett. 29 (1974) 277.

4. J. Chen, M. A. Reed, A. M. Rawlett, J. M. Tour, Science 286 (1999) 1550.

5. M. A. Reed, C. Zhou, C. J. Muller, T. P. Burgin, J. M. Tour, Science 278 (1997) 252.

6. R. Baer, D. Neuhauser, Chem. Phys. 281 (2002) 353.

7. M. Magoga, C. Joachim, Phys. Rev. B 59 (1999) 16011.

8. R. Baer, D. Neuhauser, J. Am. Chem. Soc. 124 (2002) 4200.

9. S. K. Maiti, Phys. Lett. A 362 (2007) 225. 10. S. K. Maiti, Phys. Lett. A 366 (2007) 114. 11. M. Ernzerhof, M. Zhuang, P. Rocheleau, J.

Chem. Phys. 123 (2005) 134704. 12. L. P. Kouwenhoven, C. M. Marcus, P. L.

McEuen, S. Tarucha, R. M. Westervelt, N. S. Wingreen, in Mesoscopic Electron Transport: Proc. NATO Advanced Study Institutes (NATO Advanced Study Institute, Series E: Applied Sciences) 345 (1997).

13. V. Mujica, M. Kemp, M. A. Ratner, J. Chem. Phys. 101 (1994) 6849.

14. V. Mujica, M. Kemp, A. E. Roitberg, M. A. Ratner, J. Chem. Phys. 104 (1996) 7296.

15. S. Datta, Electronic transport in mesoscopic systems, Cambridge University Press, Cambridge (1997).

16. W. Tian, S. Datta, S. Hong, R. Reifenberger, J. I. Henderson, C. I. Kubiak, J. Chem. Phys. 109 (1998) 2874.

Electronic Band Structure of Isolated Zigzag Single Wall Carbon Nanotubes

Vikas Thakur, P.S. Bisht, U. P. Verma and P. Raja Ram

School of Studies in Physics, Jiwaji University Gwalior, (M.P), India, 474011 Email: [email protected]

Abstract

The electronic band structure have been investigated for (n,0) isolated single wall carbon nanotubes

(SWCNTs) over a wide range of radius varying from (3,0) to (15,0). The first principle calculations i.e. density functional theory (DFT) were performed using SIESTA ab initio code. The exchange-correlation energy was approximated by the local density approximation (LDA). The nonlocal norm-conserving pseudopotential was used in calculations. The calculated electronic band structure shows that, for n=3,4,5 and 6 zigzag SWCNTs are metallic while for n≥7 zigzag SWCNTs are semi conducting in nature. The large downshift of conduction band minima was observed in lower diameter tubes (or at high curvature) which lead to their overlap with valance band states. This support the σ*-л* band hybridization at high curvature. The calculated values of band gap have been compared with previously reported theoretical and experimental results. Our calculated values of energy band gap for (9,0), (12,0) and (15,0) are in excellent agreement with the scanning tunneling spectroscopy (STS) experimental values. The calculated data showed that (11,0) SWCNT have highest band gap (0.92429 eV) among these tubes and (15,0) tube have least band gap (0.02175 eV).

1. Introduction

The nanostructures such as quantum dots, nanowires and carbon nanotubes (CNTs) possess unique properties that make them promising candidates for future technology applications. However, to truly harness the potential of nanostructures, it is essential to develop a fundamental understanding of the basic physics that governs their behavior in devices. CNTs are amongst the most explored one-dimensional nanostructures and have attracted tremendous interest from fundamental science and technological perspectives. They also exhibit a rich variety of intriguing electronic properties, such as metallic and semiconducting behavior.

CNTs were first discovered by S.Iijima [1] in 1991 in the multiwall form with few tens of nanometers in outer diameter. Two year later single walled carbon nanotubes (SWCNTs) were reported [2,3]. SWCNTs are basically graphite sheets rolled up into a cylinder with diameter of the order of few nanometers, which are characterized by two integers (n,m) defining the rolling vector of graphite [4]. The series of bands in electronic structure of CNT arises from the confinement of electron wave function around the nanotubes circumference.

Zone-folding model [5] provides basic information about electronic structure of CNTs and accordingly they are metallic if n-m is an integer

multiple of three, otherwise semiconducting in nature. However, experimental studies [6] indicate much more complicated structural dependence of electronic properties of CNTs. First principles local density approximation (LDA) calculations [7] showed that the σ*-л* band hybridization becomes significant at small radius (or at high curvature). This leads to modification of electronic properties of CNTs. Also, metallic CNTs may open a small band gap if the bond symmetry is broken due to curvature [8,9].

In this paper we present a consistent ab initio study of electronic band structure of zigzag SWCNT’s over a wide range of radius varying from (3,0) to (15,0). The mixing of conduction band states with valance band states at high curvature have been carefully studied. The calculated band gap data have been reported.

2. Computational Detail

Self-consistent density functional theory (DFT) [10,11] calculations were performed with the SIESTA [12] ab initio code. The exchange-correlation energy was approximated by the LDA [11,13,14] in the Ceperley-Alder formula. The nonlocal norm-conserving pseudopotential [15-18] was used instead of full potential. The valance electrons

Electronic Band Structure of Isolated Zigzag Single Wall Carbon Nanotubes 241

were described by localized pseudoatomic orbitals (PAOs) with a double-ζ singly polarized (DZP) basis set. Since the confinement of the atomic orbitals leads to rise in the energy so it has been accounted for by choosing an energy shift of 200 meV for PAOs that minimize the total energy of the system. The convergence of the total energy in the real space mesh size was investigated and we found that, in all cases, the total energy converges at plane-wave mesh cutoff of 250 Ry.

In present calculations we have chosen tetragonal supercell geometry where tube axis was taken along z-direction. The lattice constants along x and y directions were chosen such that the interaction between nearest neighbor tube is negligible. The lattice constant along z-direction was taken to be equal to the one dimensional lattice parameter of tube along the tube axis. The relaxed atomic positions were used in calculations. To relax atomic positions the conjugate gradients (CG) method in molecular dynamics have been used. Maximum force tolerance in CG coordinate optimization move was set at 0.04 eV/Å.

To fix the number of k points for k-point sampling, Monkhorst-Pack special k-point scheme [19] has been used. The convergence of total energy with respect to number of k-points was investigated and we found that total energy converges at number of k-point equal to 13 which corresponds to 1x1x25 Monkhorst-Pack. Owing to the very large lattice constants of the supercell along x and y direction, k-point sampling has done only along the tube axis i.e. along z-direction. The band energies were calculated along high-symmetry directions Г (gamma) and X. 3. Results and discussion

In the present calculations we have chosen isolated (n,0) SWCNT’s (n ranges from 3 to 15) to study their electronic properties. In present calculations we found that for n=3,4,5 and 6 zigzag SWCNT’s are metallic and tubes having n≥7 are semiconducting in nature. It is in agreement with the results reported earlier [7]. To account for this behavior we have calculated the energies of valance band (VB) maxima, conduction band (CB) minima and Fermi level for all cases. The comparative plot for this has been shown in figure I. From this figure we observe that for tubes having n≥7 the VB maxima and CB minima are equally separated from Fermi level i.e. Fermi level lies at the middle of VB maxima and CB minima. The VB maxima lie below the Fermi level and CB minima lie above the Fermi level. This shows intrinsic semiconducting nature of these tubes.

Because of this behavior, by adding some impurities in these tubes we can make them p or n type semiconductores for nanoelectronic applications.

For lower diameter (n,0) SWCNTs (n≤6) the CB minima face large downshift and go deeper into the valance band. In these cases the VB maxima is nearer to Fermi level as compared to CB minima. Such downshift of CB minima results to the mixing with valance band states (σ*-л* band hybridization) and hence become metallic.

-7

-6.5

-6

-5.5

-5

-4.5

-4

-3.5

-3

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16n

Ener

gy (e

V)

Fig. 1. Energies of VB maxima (marked with star), CB minima (marked with square) and Fermi level (dashed line) with respect to n.

Energy band gap (eV) CNT index (n,0)

Present work (LDA)

Ref.[7] (LDA)

Ref. [20] (Exp.)

(3,0) Metallic (4,0) Metallic (5,0) Metallic (6,0) Metallic Metallic (7,0) 0.21926 0.09 (8,0) 0.58099 0.62 (9,0) 0.07512 0.17 0.080 (10,0) 0.76469 (11,0) 0.92429 (12,0) 0.03513 0.042 (13,0) 0.66991 (14,0) 0.71120 (15,0) 0.02175 0.029

Table 1. Electronic band gap of (n,0) isolated SWCNT’s at Γ-point. Ref. 9 is LDA results, and Ref. 30 is STS experimental results.

The calculated band gaps at Γ-point for isolated zigzag SWCNT’s have been given in Table I. The available theoretical data as well as experimental results available in literature have also been included in Table I. The


calculated data showed that (11,0) SWCNT have highest band gap (0.92429 eV) among these tubes and (15,0) tube have least band gap (0.02175 eV). From the table it is evident that our computed values of energy gaps are much better then that of previously reported LDA results and are in excellent agreement with the scanning tunneling spectroscopy (STS) results for the case of (9,0), (12,0) and (15,0) tubes. 4. Conclusion

In this work we present an extensive first-principle analysis of the electronic band structure of (n,0) isolated zigzag SWCNT’s for n=3 to 15. We found that tubes with n=3 to 6 are metallic due to σ*-л* band hybridization. The large downshift of CB minima at high curvature results in σ*-л* band hybridization. The tubes having n≥7 are intrinsic semiconductors in nature with different values of band gap. The (11,0) tube have highest band gap (0.92429 eV) while (15,0) have least band gap (0.02175 eV) among (n,0) tubes for n=7 to 15. Our LDA results for band gap are better then previously reported LDA results. The calculated values of band gap for (9,0) (12,0) and (15,0) tubes are in excellent agreement with the STS experimental results. References [1] S.Iijima, Nature 354, (1991) 56. [2] S. Iijima and T. Ichihasi, Nature 363, (1993)

603. [3] D. S. Bethune, C. H. Kiang, M. S. deVries, G.

Gorman, R. Savoy, J. Vazques, and R. Beyers, Nature 363, (1993) 605.

[4] M. S. Dresselhaus, and G. Dresselhaus, and P. C. Eklund, Science of Fullerenes and Carbon Nanotubes, Academic Press, San Diego, 1996; R. Saito, G. Dresselhaus, and M. S. Dresselhaus , Physical Properties of Carbon Nanotubes Imperial College Press, London, 1998.

[5] R. A. Jishi, L. Venkataraman, M. S. Dresselhaus, and G. Dresselhaus, Phys. Rev. B

51, (1995) 11176; R. saito, T. Takeya, T. Kimura, G. Dresselhaus, and M. S. Dresselhaus, ibid. 57, (1998) 4145.

[6] M. Ouyang, J. Huang, C. L. Cheung, and C. M. Lieber, Science 292, (2001) 702.

[7] X. Blasé, L. X. Benedict, E. L. Shirley, and S. G. Louie, Phys. Rev. Lett. 72, (1994) 1878.

[8] C. T. White, J. M. Mintmire, R. C. Mowrey, D. W. Brenner, D. H. Robertson, J. A. Harrison, and B. I. Dunlap, in Buckminster-fullerenes, edited by W. Edward Billups and Marco A. Ciufolini VCH, New York, 1993.

[9] J. W. Mintmire, D. H. Robertson, and C. T. White, J. Phys. Chem. Solids 54, (1993) 1835; J. W. Mintmire and C. T. White, Carbon 33, (1995) 893.

[10] W. Kohn and L. J. Sham, Phys. Rev. 140, (1965) A1133.

[11] R. M. Dreizler and E. K. U. Gross, Density Functional Theory, Springer-Verlag, Berlin, 1990.

[12]http://www.uam.es/departamentos/ciencias/fismateriac/siesta/

[13] D. M. Ceperley, B. J. Alder, Phys. Rev. Lett. 45 (1980) 567.

[14] J. P. Perdew, A. Zunger, phys. Rev. B 23 (1981) 5048.

[15] G. B. Bachelet, D. R. Hamann, M. Schluter, Phys. Rev. B 26 (1982) 4199.

[16] D. R. Hamann, Phys. Rev. B 40 (1989) 2980.

[17] N. Troullier, J. L. Martins, Phys. Rev. B 43 (1991) 1993.

[18] X. Gonze, R. Stumpf, M. Scheffler, Phys. Rev. B 44 (1991) 8503.

[19] H. J. Monkhorst and J. D. Pack, Phys. Rev. B 13, 5188 (1976).

[20] M. Ouyang, J. Huang, C. L. Cheung, and C.M. Lieber, Science 292, (1998) 1278.

Modeling of Band-Gap Using Strain Theory for Silicon

V.K. Lamba1, Ankur Gupta2, Munish Verma1

1 Department of Electronics, Haryana College Of Technology & Mang, Kaithal. Haryana. 2Deptt of ECE, College Of Technology & Mang, Kaithal. Haryana.


Abstract

We have modeled theoretically the influence of strain on the electronic structure in Si. The strain calculations were done using a semi empirical approximation which enabled three-dimensional (3-D) strain simulations of the device structures. The strain-induced deformation of the conduction band gives rise to a 3-D potential minimum having a depth of 40 milli-electron volt. Our calculations predict strain-induced channeling of electrons to the edges of the structure. 1. Introduction

In order to assess and compare the performances of nanoscale CMOS devices with novel channel materials (e.g. strained Si, Ge, GaAs or InAs), a simulation approach is undertaken in this work. In Sec. 2, the recent experimental demonstration of novel channel material CMOS devices is reviewed. In Sec. 3, we discuss simulation approaches; In Sec. 4, the issues and challenges associated with the calculations of nanoscale novel channel materials are highlighted.

2. Experimental demonstration

The experimental exploration of novel channel materials for CMOS devices is primarily motivated by their excellent transport properties. Their high room-temperature mobilities and saturation velocities are thought to be the key to the next generation ultra-fast, low power CMOS digital logic technology. For long time, strain has been known to improve the channel transport properties of MOSFETs. Strained Si is the only new channel material which has recently made its way into the commercial integrated circuits. Beginning with the 90 nm technology node devices, released in 2003, leading IC industries have incorporated strained silicon, in some form, to improve the channel transport properties [1-4]. Recently, substantial progress has also been made to incorporate strain in SOI-structures using bond-and-etch-back technique [5]. Modulations in electron and hole mobilities with the scaling of body thicknesses in strained SOI have been reported in [5] and [6]. Also, devices fabricated on Si (110) wafer orientations has shown improved mobility characteristics over (100) devices [7, 8]. Recently, similar results for

(110) strained SOI MOSFETs have also been published [9]. Beyond silicon, germanium is an interesting candidate for nanoscale CMOS technology due to its excellent transport properties -two and four times bulk mobilities for electrons and holes compared to silicon, respectively. Room temperature hole mobility in a 7.5nm thick Ge quantum well has already been reported to exceed 2500 cm2/V-sec [10]. 3. Simulation approaches

In addition to experimental exploration of nanoscale novel channel material MOSFETs, physics based simulation for such devices can offer valuable insight into their operation and can help their design optimization. Numerical simulation not only proves valuable to guide experiments and to explain their results, it also helps to identify the strengths and the weaknesses of different approaches in the emerging field of nanoelectronics. The simulation approach has already proven useful in determining the performance limits of the Si technology by comparing their experimental performances with their ballistic performances [11]. A full 2-D simulation tool, nanoMOS 2.5, developed at Purdue University, quantum mechanically models the Si n-MOSFETs fabricated on (100) wafers.

Device modeling at nanoscale consists of self-consistently solving the Schrödinger equation and the Poisson equation. For a given potential profile, Schrödinger equation quantum mechanically calculates the carrier densities and their transmission probabilities, while the Poisson equation ensures that the charge profile is consistent with the potential profile. The most widely invoked assumption to solve the Schrödinger equation is known as the effective


mass approach, where the slowly varying envelope of the electronic wave function is obtained by solving a Schrödinger-like effective-mass equation (EME). Another full 2-D quantum mechanical tool, QDAME, developed at IBM, solves the effective mass equation to treat the open boundary ballistic quantum transport problem by expanding the electronic wavefunction as a linear combination of wavefunctions which satisfy zero value and zero slope boundary conditions [12]. Both nanoMOS and QDAME employ parabolic E-k relationships for electronic band structure. Scaling study results for UTB nanoscale Si (100) n-MOSFETs has already been published using both nanoMOS [13-15] and using QDAME [16].However, there are two limitations. First, for holes in the valence bands, where the parabolic E-k band structure is not valid and the heavy, light and spin-off valleys are strongly coupled, effective mass approach results in a complicated pk . description of band structure, which is not suitable for quantum simulation of hole transport. Second, even for electrons in the conduction band, if the principal axes of the constant energy ellipsoids are not aligned with the device axes (channel, thickness and width directions), the effective mass equation for the electrons becomes enormously complicated. This becomes a serious issue, limiting the application of effective mass equation to novel channel material devices, since for germanium n-MOSFETs, or silicon n-MOSFETs on wafer orientations other than (100), the device axes and the ellipsoid axes are no longer aligned. Consequently, the usefulness of the mode-space approach vanishes for quantum mechanical treatment of electronic transport in novel-channel material n-MOSFETs. QDAME employs a technique to address this problem by discretizing effective mass equation along the principal axis of the ellipsoid; however, this becomes a real-space approach therefore, is numerically cumbersome. In order to use the efficient techniques, such as mode-space approach and NEGF formalism, the first challenge for simulation of novel channel material n-MOSFET is to develop a generalized effective mass approach where the complicated Hamiltonian arising from the non-alignment of device and ellipsoid axis can be simplified.

As already pointed out, effective-mass-approach is an approximation which disregards the atomic scale fluctuation of the electronic characteristics and describes the band edge

electronic properties in an approximate manner. However, as size goes down, the behavior of the electronic states in nanoscale CMOS devices become increasingly sensitive to all sorts of microscopic phenomena: atomic-scale fluctuations, local bond distortions, alloy effects, structure of the interfaces, quantum tunneling and energy quantization. An improved modeling of such effects is not possible within effective mass approach and a full band atomistic treatment is necessary to address them. Semi-empirical tight-binding approach, a full band technique, proves extremely useful for atomistic treatment of nanoscale devices with any materials, provided that the tight-binding parameters for the material are known in advance. It correctly captures full-band effects, such as valley splitting in a nanostructure, and also treats alloy effects on band structure [17]. 4. Calculations

To derive the tight-binding Hamiltonian, the following conditions are assumed: • Atom-like orbital-localized basis functions

have atomic orbitals symmetry, • Tightly bound-overlap of two orbitals on

different atomic sites is zero, • Orthogonality-overlap of two different

orbitals located on same atomic site is zero, • Nearest neighbor interaction- nonzero

matrix elements for Hamiltonian possible only between orbitals located on nearest neighboring sites. Nearest neighbors of a cation are four anions and vice versa, therefore, matrix elements between cat ions and anions are only possibility.

• Two center integrals-nonzero matrix element for Hamiltonian possible only when the potential is on one of the two atoms on which orbitals are located.

The basis set for Hamiltonian [18] consists of the Bloch sums of the localized orbitals, performed as

( )i

R

VRik RnbeN

knbi

biρρ

∑ += .1

We now assume the variational wave-function as

∑∑ λ==λbn

bnbn

knbcknbknbknbk,

,,,

ρρρρρ

As the crystal Hamiltonian is the sum of kinetic and potential energy operators so H is defined as

[ ]∑ ++∇−=i

ciai UUm

H ,,2

2

2η

Modeling of Band-Gap Using Strain Theory for Silicon 245

Substituting the variational wave-function in the Schrödinger equation we have

( )[ ] ( )[ ] knbkHcknbckHbn

bnbn

bn

ρρρρ∑∑ λε−=λε− λλ

,,,

,,,

Now after left-multiplying by knbρ

and using

Orthogonality of atomic orbitals and tight-binding assumption. Now the matrix elements between orbitals sitting on cation and anion atoms, respectively, are calculated and the four terms correspond to anions at following lattice points, are

[ ]0 0 0; 0 0 0 2 2 2 2 2 2L L L L L L

ji ji ji jia a a a a aR R R R⎡ ⎤ ⎡ ⎤ ⎡ ⎤= = = =⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎣ ⎦ ⎣ ⎦ ⎣ ⎦

r r r r

Thus Hamiltonian can now be expressed in the bloch sum basis as

[ ] [ ][ ] [ ]

a a a c

ca cc

H HH

H H⎛ ⎞

= ⎜ ⎟⎝ ⎠

Once calculating Hamiltonian we can calculate band gap, using Kane’s theory as shown in table1.

Table 1. Fermi energy, minimum and maximum band energy along x, y and z axis.

Units Fermi Energy Minium Band Energy

Maximum Band Energy

-5 8.1717 eV -13.2668 eV 11.4572 eV -2 7.082 eV -12.5529 eV 10.7137 eV 0 6.9149 eV -12.6105 eV 9.7099 eV 2 6.1401 eV -12.0436 eV 9.3431 eV

Applied along X,Y,Z axis

5 4.9697 eV -11.155 eV 8.8637 eV -5 6.8494 eV -12.5173 eV 9.7777 eV -2 6.755 eV -12.4434 eV 9.91532 eV 0 4.9697 eV -11.1552 eV 8.8637 eV

Applied along “XY”

plane 2 6.6161 eV -12.3081 eV 9.9284 eV -5 6.5299 eV -12.2176 eV 10.0671 eV -2 6.7373 eV -12.4314 eV 9.8509 eV 2 6.498 eV -12.1917 eV 10.1543 eV

Applied along Z axis

5 6.6264 eV -12.313 eV 10.0484 eV 5. Conclusions Thus we can tailor the band-gap by applying the strain and fabricate new channel materials. References [1] P. Bai et al. IEEE International Electron

Devices Meeting, 2004. IEDM Technical Digest., 657, Dec. 2004.

[2] T. Ghani et al. IEDM 2003 Technical Digest., pages 11.6.1-11.6.3, 8-10 Dec. 2003.

[3] K. Mistry et al. Digest of Technical Papers, 2004 Symposium on VLSI Technology, pages 50-51, 15-17 June 2004.

[4] S. Thompson et al. IEDM 2002. Digest., pages 61-64, 8-11 Dec. 2002.

[5] S. E. Thompson et al. IEEE Transactions on Electron Devices, 51(11):1790, Nov. 2004.

[6] S. E. Thompson et al. IEEE Electron Device Letters, 25(4):191-193, April 2004.

[7] K. Rim et al. IEDM '03 Technical Digest. IEEE International Electron Devices Meeting, pages 3.1.1-3.1.4, Dec. 2003.

[8] K.Uchida, et al. IEEE International Electron Devices Meeting, 2004. IEDM Technical Digest., pages 229-232, Dec. 2004.

[9] M. Yang et al. IEEE International Electron Devices Meeting 2003, pages 18.7.1, 2003.

[10] T. Mizuno, et al. IEEE Transactions on Electron Devices, 52 367, 2005.

[11] M. Myronov, et al. Journal of Applied Physics, 97 (2005) 1-6.

[12] F. Assad, et al. IEEE Transactions on Electron Devices, 47(2000) 232-240,.

[13] S. E. Laux, et al. Journal of Applied Physics, 95(10):5545-5582, 2004.

[14] Z. Ren, et al.. IEDM Tech. Dig., pages 715-718, 2000.

[15] Z. Ren, et al. IEDM Tech. Dig., pages 5.4.1-5.4.4, 2001.

[16] Sayed Hasan, et al. Solid-State Electronics, 48(6):867-875, June 2004.

[17] A. Kumar, et al. IEEE Transactions on Electron Devices, 52(4) 614-617, April 2005.

[18] T. B. Boykin et al. Physical Review B, 71(11):115215(1-6), 2005.

Pair Potentials of Expanded Fluid Rubidium Using Ab-Initio Pseudopotentials

Avneesh Sharma, Nitu Sharma and Raman Sharma

Department of Physics, Himachal Pradesh University, Shimla-171005, India E-mail: [email protected]

Abstract

The pair potentials of fluid rubidium are calculated for a series of states along the saturated –vapor-

pressure curve, ranging from temperature close to the melting point up to the critical point. The calculation is based on the Troullier Martins’s ab-initio pseudopotential. The results are compared with the model potential developed by G.Kahl & J. Hafner. Here, potential calculated using ab-initio pseudopotential seems to be more efficient and reliable for the study of liquids. 1. Introduction

Modern theories make an effort to understand three states of matter [1-5] from the point of view of interaction between atoms and molecules. In solid state, atoms are arranged in regular geometrical patterns and this greatly simplifies their properties. Because of this, theory of solid state made a very rapid progress. On the other hand, gaseous state of matter is easiest to handle and kinetic theory of gases had a great success in explaining their properties. However, liquids which are condensed state of matter exhibit variety comparable to that found in crystalline solids. From the fundamental point of view, ordinary matter is composed by atomic nuclei and electrons interacting through electromagnetic forces. However, in many situations it is enough to consider just electrostatic coulomb interactions to obtain an accurate description of a physical system. Furthermore, in atoms, molecules, and also in condensed matter phases, the electrons can be divided into two categories: though tightly bound to the nucleus, referred to as core electrons, and the others, not so tightly bound, which will be referred to as valence electrons. When the environment of an atom changes, perhaps because of the presence of other interacting atoms, the state of the valence electrons also changes, in some cases drastically (for instance in a metal), whereas the core electrons are hardly affected and therefore in most cases they can be considered to be chemically inert.Consequently, most properties of matter depend on the behavior of the chemically active valence electrons only. The preceding argument shows that the potential felt by the valence electrons in a given system plays a basic role in the determination of its properties. On the other hand, the valence wavefunctions must oscillate inside the

core region, since they must be orthogonal to the core wavefunctions. This strong and complicated character of the potential in the core region makes it rather difficult to perform accurate calculations. Fortunately, most properties do not depend very strongly on core electrons, because the important interaction among the different atoms of the system takes place where their valence wavefunctions overlap, and is certainly outside the core regions. However, we can replace the actual potential energy in the core region by the pseudopotentials that gives the same wave function outside the core as does the actual potential. It was the startling that in the core region, the pseudopotential was nearly zero. In view of this, pseudopotential theory has enormously contributed to the advances in the properties of pure liquid metals [6].

Ab-initio Pseudopotential In the present work ab initio pseudopotentials [7-8] have been generated using first principle approach of vanderbit. These pseudopotentials are constructed by all electron calculation of free atom in a reference configuration. The construction of pseudopotential is often done separately for each angular momentum quantum number l of the pseudo wave function. These Pseudopotentials are called non-local pseudopotentials which have the form:

∑ <>=

lmlmllmNL YVYV // (1)

Where spherical harmonics Ylm project out each angular moment component. This type of pseudopotential require the projection operator to be used for each plane wave which needs efficient computational effort. A pseudopotential which uses same potential for all the angular momentum

Pair Potentials of Expanded Fluid Rubidium Using Ab-Initio Pseudopotentials 247

components is called the local pseudopotential. In order to generate the pseudopotential for a given metal following steps are used. 1. Valence and core states are selected. 2. A reference configuration is chosen. 3. The matching radii is chosen 4. Pseudopotential are generated.

Inter-ionic pair potential The effective pair potential has the familiar form of the screened coulomb potential given by: ])sin()(21[

0dq

qqrqF

RZZ

V Nij

jiij ∫

∞−=π

(2) Where in above equation,

ji ZZ , are valances and

)(qF Nij

is a normalized energy wave number dependent characteristic that contains total band structure effects in the liquid metals using self consistent electron screening.

⎥⎦

⎤⎢⎣

⎡−

−ΠΩ

−=)(1

1*)(1)()()()

8()( *

*2

qGqqqwqwqqF ji

Nij ε

ε (3)

Here the quantities )(),( qwqw ji are Fourier transforms of self consistent bare ion (unscreened) atomic pseudopotentials for metallic component, obtained through generalized first principle pseudopotential theory. G(q) is the exchange correlation function. )(* qε is modified Hartree dielectric function. Ω is the atomic volume of pure elements. In equation (.3) )(* qε is given by: ( ) ( )( ) ( )( )qGqq −−+=∗ 111 εε (4) Where Hartree dielectric function is given by:

( )⎟⎟⎟⎟⎟

⎠

⎞

⎜⎜⎜⎜⎜

⎝

⎛

−

+

⎟⎟⎠

⎞⎜⎜⎝

⎛

−+

⎟⎟⎠

⎞⎜⎜⎝

⎛+=

F

F

F

F

Fkqkq

kqkq

kq

q2

2ln

4

41

2

112

2

2

πε

(5)

In equation (5) 31

23⎟⎟⎠

⎞⎜⎜⎝

⎛Ω

=ZkF

π is Fermi wave vector.

( )qG is the Shaw exchange and correlation factor [9] which takes into account the correlation of conduction electrons.

2. Application to expanded fluid Rubidium In the following we shall discuss the application of the model described above to expanded fluid rubidium along the saturated-vapor-pressure curve. The calculation have been performed for a set of states and the results are compared with the model potential developed in ref [10].

-10

-5

0

5

10

15R bT = 400 KVolume = 670 (a.u)3

line (ab-initio)Doted line (ref[10])

-10

-5

0

5

10

15R bT = 900 KVolume = 789 (a.u)3

0 10 20 30-10

-5

0

5

1015

Fig. 1. Pair Potential using ab-initio Pseudopotential [Solid line]and Model pair potential [doted line] from ref[10].

r (a.u.)

V(r)

(mR

y)

R bT = 1200 KVolume = 867 (a.u)3

3. Conclusion

The present work consists of theoretical

description of pair potentials of fluid rubidium. The understanding of inter-atomic interactions within liquid metals is quite necessary to study their pair potentials. Till recently for calculating the properties, Model Pseudopotentials have been used. While calculating Model Pseudopotentials adjustable parameters are unavoidable, this causes variation in results.

Here, ab-initio pseudopotentials have been used to include the inter-atomic interactions in the liquid systems. The ab-initio pseudopotentials help in calculating the properties of materials without the use of adjustable parameters with the help of


fundamental quantum mechanical equations. Results are compared with the model potential. Comparison shows that ab-initio pseudopotentials are more reliable and more efficient than model pseudopotentials for the study of liquid systems. References [1] Temperley H.N.V. and Trevena D. H., Liquids

and their properties, (Ellis Horwood: Chichester)1978.

[2] Rice S. A. and Grey P., The Statistical mechanics of simple liquids, (Interscience: New York)1965.

[3] Torren J. E., Interatomic Potentials, (Academic Press: New York)1972.

[4] Ashcroft N. W., Interatomic Potentials and Simulations of Lattice Defects, Plenum: New York)1972.

[5] Kittle C., Introduction to Solid State Physics, (John Wiley: New York)1976.

[6] Yastrebov L. I. and Ktsneison A. A., Foundation of one-electron theory of solids, Mir Publishers Moscow(1987).

[7] Thakur A. and Ahluwalia P.K., Chin. Phys. Lett. 22, 10(2005)2611.

[8] Lee C. Yang W. and Parr R.G., Phys. Rev. B, 37(1988)785.

[9] Shaw R. W., J. Phys. C., 3(1970)1140. [10] G.Kahl and J. Hafner, Phy. Rev. A,

29(1984)3310.

Modelling and Analysis of Interconnect on Wafer

Munish Verma 1, S. S. Gill2, V. K. Lamba1

1 Department of Electronics, Haryana College Of Technology & Mang, Kaithal. Haryana. 2 Deptartment of Electronics, GNDEC, Ludhiana, Punjab.


Abstract

The Electrochemical deposition ECD is one of the major thin film deposition processes in microelectronics as it involves complex surface reactions, surface interactions that affect overall topology of microscale features (interconnects) and transport phenomenon that range over different length range. In this work, we have simulated and developed simulation program for investigating the deposition features and their effects on various topologies for ECD process. The technique and model developed in our work can be used to investigate systematically the topology of Cu deposited by ECD process and helps in design and optimization of processing technology. 1. Introduction As the fabrication of interconnects involves many processing steps, which results in a small overall yield and varied deposition profiles or topologies on a wafer of a given lot. In microelectronics industry, the design of process is typically performed with trial and error based experiments and these experiments are often time consuming and costly. In addition, such an approach may be quick solution to a particular problem but does not provide any understanding of process. Therefore, any systematic issue can not be captured. Our objective of this work is to develop a predictive simulation tool so that it can provide a guideline for design and optimization of ECD process, thereby reducing time and resources. The paper is divided into four sections. I section includes issues in modelling, II section includes working methodology, III section includes comparison of simulation results with experimental data, IV section includes conclusions. Issues in modeling

In ECD process, the predictive simulation tool requires detailed surface reaction mechanisms and rigorous description of transport phenomenon. As the modelling of transport phenomenon in ECD process is complicated by different length scales involved in the system, the individual features and the cell needed to be modelled simultaneously. As HPCVD has been modelled by Nemirovskaya ET al.1, that the two dimensional problem requires Ơ (107) unknowns. Even with a parallel computer that would take a significant amount of time. Furthermore, the features are continuously

growing, requiring remeshing at every time step which is computationally demanding procedure. So we used multiscale simulation2, 3, 11 approaches where the domain is decomposed into sub-domains and different models were used for different sub-domains. Then each is linked together to ensure consistency among them. Here the most suitable governing equation and numerical methods are assigned to each sub-domain model. Working methodology

Our computational process for the Cu ECD is based on the following assumptions: (i) The counter electrode is configured so that the primary current distribution would be uniform in the absence of electrode resistance effects2, (ii) The electrical double layer and the mass transfer diffusion boundary layer attain a steady state rapidly compared with the slow transient process of deposit thickening, (iii) A substantial number of electrical contact points are made around the edge of the wafer so that azimuthal potential variations are negligible, (iv) From a macroscopic viewpoint, the surface of the working electrode remains smooth during the course of deposition3, (v) The ohmic resistance in the solution is negligible with respect to other resistances in the cell, (vi)The electrolyte contains an excess of supporting electrolyte so that migration of copper ions in the solution is negligible4, (vii) The solution is stirred such that the cupric ions are uniformly accessible to the surface at all points. The algorithm4, 12 used for the computation for electrode resistance consists of



……………. Calculation of resistance to the passage of electron through the seed layer/barrier layer.

Calculation of electrodeposition reaction rate which is a function of electronic current in the metal phase i(r, t), also varies with radial position across the wafer.

Local deposition rate which can be obtained from the concentration difference.

………….

2. Model for calculating electric field within electrolyte

We assume a cylindrical plating cell8, 9, 10 with radius R and height h, with anode to be a disc with radius r (r = R) positioned at the bottom of cylindrical cell5, 9. Cathode is a disc with varied position of seeds as shown schematically in Fig. 1. If distance h is small in between cathode and anode, we assume electric field is homogenous in nature. Then, using Laplace equation, we calculated potential and charge density on the resistive electrode. And if h is very very small, the potential is distributed linearly in the cell and transport of ions is governed by the intensity of electric field and in the direction of current density. Here, we have used constant current boundary conditions for process operating at 0.8 Amp/ cm2. The current density had a nominal value based on the projected area. The typical process parameters obtained from the above calculations are: (i) Effects due to cell configurations that includes shape of electrode and scale variation between different points,(ii) Effects due to the seeds, (iii) Effects of process chemistry which includes electrolyte conductivities and additives.

Fig.1. Cell “Generic” (or ‘base’) configuration: wafer radius = 100 mm, distance between wafer and anode =150 mm, rotation = 60 rpm, flow from bottom = 4 GPM, gap between the wafer’s edge and cell’s wall = 10 mm, iaverage = 20 mA/cm2, κ = 0.55 S/cm, seed thickness = 1000 Å.

3. Comparison of simulation results with experimental data Effects due to cell configuration

From Fig. 2, of current density versus distance along the electrode, we conclude that for the maximum deposition rate its value must be around 0.6 Amp/cm2 and the shape of electrode must be elongated convex type. To get uniform topology of the surface, the value of current density must be

Fig. 2. Graph of current density versus distance along the electrode for different shapes of electrodes.

A-parallel electrodes, B-convex electrodes, C-elongated concave electrodes, D-elongated convex electrodes, E-triangular convex electrodes, F-elongated triangular concave around 0.5 Amp /cm2 and the deposition distance from the electrode must be around 75 to 150 mm. If in a process, we want a large area of copper deposition with linear topology, the current density may be around 0.3 Amp /cm2 and topology of electrode is parallel to each other. From Fig. 3, of deposit thickness versus distance along the electrode, we conclude that if we want a thick metallization layer, the current density must be around 0.6 Amps /cm2 and the distance should be 75 to 150 mm along the electrode. From Fig. 4, of current density versus distance along the electrode, we conclude that current density decreases as the distance between two electrode increases. From Fig. 5, of deposit thickness versus distance along the electrode, we conclude that deposition topology remains the same except at the wafer edges where an exponential thickness is observed. From Fig. 6, of current density versus distance along the electrode, we conclude that deposition rate increases at the edges position of seeds and shape of curve is parabolic in nature.

Modelling and Analysis of Interconnect on Wafer 251

Fig. 3. Graph of total deposit thickness versus distance along the electrode for different shapes of electrodes. Shape A-parallel electrodes, shape B-convex electrodes, shape C-elongated concave electrodes, shape D-elongated convex electrodes, shape E-triangular convex electrodes, shape F-elongated triangular concave Effects Due to Additives

A correlation between copper deposition potential and surface concentration of cuprous ions at the electrode has been calculated through simulation and later results are confirmed using experimental results of Vereecken ET al.6, for rotating ring disk experiment that the deposition is faster at cuprous rich surface than at cuprous poor surface for the same applied potential. The polarizing additives inhibit the rate while the depolarizing additives accelerate the rate of formation cuprous ions.

Fig. 4. Graph of current density versus distance along the electrode for varying distance between electrodes from 10 mm to 40 mm Effects Due to Electrolyte Composition

From simulation results we conclude that the electrolyte solution must be low conductive in nature as the proton mobility is about 6 to 7.5 times than the mobility of copper or sulphate ions. The results are confirmed by the experimental results of Uziel ET al.7, where the electrolyte is acid free and the deposition thickness is uniform in nature.

Fig. 5. Graph of total deposit thickness versus distance along the electrode for varying distance between electrodes from 10 mm to 40 mm. 4. Conclusions

The design of electrochemical deposition process for copper deposition involve numerous parameters including shape of electrodes, varying distance between electrodes, position of seed, presence of additives, electrolyte composition, among many others. Simulating the effects of those parameters, we conclude that for depositing a uniform profile, the current density should around 0.3 amp /cm2 with depolarized additives, no acids and shape of electrodes must be low convex type.

Fig. 6. Graph of current density versus distance along the electrode for varying position of seed at electrode.


References [1] M. Nemirovskaya, G.S. Kim, and K.F. Jensen,

in preparation for Journal of Applied Physics (2002).

[2] T.O. Drews et al., IBM J. Res. & Dev., Vol. 49, No. 1, Jan 2005

[3] ‘CELL-DESIGN’, Computer Aided Design and Simulation software for Electrochemical Cells, L-Chem, Inc. 13909 Larchmere Blvd. Shaker Heights, OH44120

[4] R. Alkire, J. Electrochem. Soc. 118, 1935–1941 (1971).

[5] Sergry Chivilikhin et al., in presentation at metallization symposium, AIChe Annual Meeting, Nov 2003.

[6] P. M. Vereecken, R. A. Binstead, H. Deligianni, P.C. Andricacos, IBM J. of Res. and Dev., Vol. 49, No. 1, Jan 2005.

[7] Uziel Landau, The Electrochemical Society Proceedings, Volume 94-9, 1994

[8] P.C. Andricacos, C. Uzoh, J. O. Dukovic, J. Horkans and H. Deligianni, IBM J. of Res. and Dev. 42(5), pp. 567-574, September, 1998

[9] D. C. Edelstein, in Advanced Metallization

Conference Proceedings in 1998, pp. 669-671, Mat. Res. Soc. 1999

[10] J. Dahm and K. Monnig, Ibid, pp. 3-15 [11] ‘CELL-DESIGN’, Computer Aided Design

and Simulation software for Electrochemical Cells, L-Chem, Inc. 13909 Larchmere Blvd. Shaker Heights, OH44120

[12] Yezdi Dordi, Uziel Landau, Jayant Lakshminkanthan, Joe Stevens, Peter Hey and Andrew Lipin, Abstract # 365, The Electrochemical Society Meeting, Toronto, Canada, May 2000

[13] U. Landau ET. al., Abstract No 263, 195th Meeting of the Electrochem. Society, Seattle, WA. May 2-6, 1999.

[14] Uziel Landau, John J. D’Urso, and David R. Rear, US Patent # 6,113,771. Sept. 5, 2000

[15] C. W. Tobias and R. Wijsman, J. Electrochem. Soc. 100, 450 (1953).

The Effect of Germanium Content on Physical Properties of [Se80te20]100-X Gex (X=0,2,4,6)

Mainika and Nagesh Thakur

Electronic Research Laboratory, Department of Physics, H.P.University, Shimla171005. Email: [email protected]

Abstract

The influence of germanium on the physical properties of [Se80Te20]100-xGex (X=0,2,4,6) was theoretically predicted .In present paper various properties viz. lone pair, constraints, mean bond energy, fraction of floppy, and heat of atomization was studied as a function of average co-ordination number <r> for the present system of glass. Using the chemical bond approach of Tichy and Ticha the mean bond energy <E> has been calculated and found to be proportional to the average co-ordination number <r> and number of lone pair electrons of the system.

1. Introduction

Investigation of physical properties in chalcogenide glasses gained much attention because of their potential technological application proposed in civil, medical and military area [1-3] and mostly Se-Te based ternary compound have received particular attention. Many approaches have been proposed to explain the compositional dependence of various physical properties of chalcogenide network [4-6]. Properties of chalcogenide glassy semiconductor are usually affected by the addition of third element. In present paper we have calculated theoretically the average co-ordination number, number of constraints, average heat of atomization, mean bond energy, lone pair, fraction of floppy modes for [Se80Te20]100-xGex at different composition.

Chalcogenide glasses are often called lone pair semiconductor. The chemical bond with lone pair electrons are characterized by flexibility [7].Increasing the number of lone pair electrons decreases the strain energy in the system, and structures with large number of lone pair electrons favor glass formation [8]. Number of constrains and the average co-ordination number for the system

In chalcogenide glasses the co-ordination number of the covalent atom is given by 8-N Rule [9], where N is the number of the outer shell electrons in a given atom. The degree of cross-linking is often describe by the concept of the average co-ordination number <r>, which is

defined as the atom-averaged covalent co-ordination of the constituent. The glassy network are influence by the mechanical constraints (Nc) associated with atomic bonding and average co-ordination number <r> which is also related to Nc. The bonding constraints in the mean field description are 2-body bonding stretching forces (Nα) and the 3-body bond-bending forces (Nβ) need to be counted [10].

If r is the co-ordination number then the number of the two body bond-stretching constraints/atom is (r/2) and number of bond-bending constraints/atom is (2r-3). Knowing the average co-ordination number of the constraints Nc = Nα + Nβ and the average co-ordination number <r> for different composition of [Se80Te20]100-xGex can be calculated [11] by using the formula <r> = (2/5) (Nc +3) (1)

The values of Nα, Nβ, Nc along with <r> for the a- [Se80Te20]100-xGex (x = 0,2,4,6) glassy system are given in table(1). Table 1. The average co-ordination no. <r> & no. of constraints given by Nc = Nα + Nβ for various composition with Ge. Composition

<r> Nα Nβ Nc

X = 0 2 1 1 2

X = 2 2.04 1.02 1.08 2.1

X = 4 2.08 1.04 1.16 2.2

X = 6 2.12 1.06 1.24 2.3


Role of lone-pair electrons and fraction of floppy modes

In order to calculate the number of lone-pair electrons of a chalcogenide glass system, the average co-ordination number proposed by Phillips [12] was introduced. The number of lone-pair electrons in the chalcogenide glass system can be calculated by using the relation L = V - <r> (2)

Where L and V are the lone-pair electrons and valence electrons, respectively. The number of lone-pair electrons and fraction of floppy modes of a given glassy system [Se80Te20]100-xGex were obtained according to equation (2) & (3) and listed in the table (2). It is clear from the fig1 that the lone-pair electrons, decreases continuously with increase in Ge content. For the binary system the value of L must be larger than 2.6 and for a ternary system, it must be larger than 1[13]. M. F. Thorpe [6] pointed out that underco-ordinated network would posses in the absence of the weaker larger range forces, a finite fraction of zero frequency normal vibration modes, floppy modes. The fraction, f, of zero frequency modes given by f = 2 –5/6 <r> (3) Table 2. Values of <r>, valence electron (v), fraction of floppy modes (f) and lone-pair electrons for the system [Se80Te20]100-xGex

The average heat of atomization (HS), mean bond energy <E> and transition temperature (Tg)

It is interested to relate the optical gap with the chemical bond energy. For this purpose we calculate the average heat of atomization (Hs). According to Pauling [14], the heat of atomization Hs (A-B) at standard temperature and pressure of a binary semi-conductor formed from atom A & B is the sum of heat of formation ΔH & the average heats of atomization Hs

A and

HsB that correspond to the average non-

crystalline polar bond energy of the two atoms.

0 1 2 3 4 5 6

3.75

3.80

3.85

3.90

3.95

4.00

lone

pai

r of e

lect

rons

,L

Ge atomic%age

Fig. 1. The number of lone-pair electrons as a function of Ge content. Hs (A-B) = ΔH + 1/2 [Hs

A + HsB] (4)

The first term in above equation is proportional to the square of the difference between the electro negativity χA and χB of the two atoms. B

( )2BAH χχα −Δ (5) In order to extend this idea to the ternary and higher order semi-conductor compounds the average heat of atomization Hs is defined for a compound AαBBβCγ as a direct measure of the bond strength, as

( )( )γβα

γβα++++

=C

sB

sA

ss

HHHH (6)

Fig. 2. The fraction of floppy modes versus the average co-ordination number in the system.

Composition <r> V L = V-

<r>

Fraction of floppy

modes, f X =0 2 6 4 0.334 X=2 2.04 5.96 3.92 0.301

X=4 2.08 5.92 3.84 0.267

X=6 2.12 5.88 3.76 0.234

0 1 2 3 4 5 60.22

0.24

0.26

0.28

0.30

0.32

0.34

frac

tion

of fl

oppy

mod

es,f

co-ordination no.<r>

The Effect of Germanium Content on Physical Properties 255

As seen from the table that value of Hs increases with the addition of Ge. By adding Ge to this glassy system the average bond strength of the compound increases and hence optical gap will increase. Table 3. Values of average co-ordination no. <r>, average heat of atomization (Hs), transition temp. (Tg) mean bond energy and heat of atomization for Ge, Se, Te, element in [Se80Te20]100-xGex (x=0,2,4,6)

<r> Hs(kcal/g-atom)

<E> (ev)

Tg [K]

Element Heat of atomization (kcal/g-atom)

2 48.72 1.85 295.5 Ge 90 2.04 49.55 1.89 307.9 Se 49.4 2.08 50.37 1.93 320.3 Te 46

2.12 51.20 1.97 332.8

The properties of the chalcogenide glasses are closely related to the over all mean bond energy which represent a function of the mean co-ordination no., the type and energy of the chemical bonds between atoms forming glasses. The average co-ordination no. of the glass <r> system is calculated by using the co-ordination no. ZSe=2, ZGe=4, ZTe=2. We determine the over all mean bond energy <E> with Tichy’s equation [15] for the complex chalcogenide systems:

<E> = Ec + Erm (7)

Where Ec is the average energy of cross-linking and given by

Ec =Pr Ehb (8)

Where the average heteropolar bond energy energy is given by

[( )

]GeTe

SeGeGeGeTeTehb bZaz

EbZEaZE++

= −− (9)

Where EGe-Se and ETe-Se are the hetropoalar bond energies of Ge-Se & Te-Se and calculated by using Pauling relation [14]. The bond energies for homopolar bond are 1.63 for Ge-Ge, 1.91 for Se-Se and 1.43 for Te-Te [14]. The average bond energy per atom of the “remaining matrix” Erm is given by

[ ]><

−><= −

rPrEE RSeSe

rm5.02

(10)

Calculated values of mean bond energies are given in table 3.

The glass transition temperature for the present system is predicted theoretically by using the relation given by Tichy and Ticha [15].

[ ]9.0311 −><= ETg (11)

From the table we see that as Ge content increases the mean bond energy of the system increases. 2. Conclusion

It is seen that the average co-ordination number., number of constraints, average heat of atomization and mean bond energy increases with increase in the Ge content in the present system . So we conclude that the Ge addition to the present system of the glass cause a increase in the average bond strength. References [1] A.M. Andriesh, M. S. Lovu, S.D. Shuter J.

Optoelectronic Advance Material 4(b) (2003) 631.

[2] T. Ohta, J. optoelectronics advance Material 3 (2001) 609. [3] A. V. Kolobov, J. Tominaga, J.

Optoelectronics Advance Material 4 (2002) 679.

[4] K. Tanaka, Phy. Rev. B, 39 (1989) 1270. [5] J. C. Phillips, J. Non-Crystalline Solid, 34

(1979) 135. [6] M. F. Thorpe, J. Non-Crystalline Solid, 57

(1983) 355. [7] H. Yinnon, D. R. Uhlmann, J. Non-

Crystalline Solid, 54 (1983) 253. [8] J. Z. Liu, P.C. Taylor, Solid State

Communication, 70 (1989) 81. [9] N. F. Mott and E. A. Davis, electronic

Processes in Non-crystalline-Crystalline Materials (Claredon Press, Oxford, 1979).

[10] S. A. Fayek, A. F. Maged, M.R. Balboul, Vacuum, 53 (1999) 447.

[11] M. Fadel, Vacuum, 48(1) (1997) 73. [12] Achamma George, D. Sushamma, P.

Predeep, Chalcogenide Lett., 3(4) (2006) 33. [13] Liang Zhenhua, J. Non-Crystalline Solid,

127 (1991) 298. [14] L. Pauling, The Nature of Chemical Bond,

Cornell University Press, New York, 1963. [15] L. Tichy, H. Ticha, J. Non-Crystalline Solid,

189 (1995) 141-146.

High Pressure Phase Transition of Zinc-Blende Copper Chloride

Deoshree Baghmara, Sadhna Singha, D. C. Guptab and N. K.Gaura

‘a’Department of Physics, Barkarullah University, Bhopal-462 026. ‘b’ School of Studies in Physics, Jiwaji University, Gwalior- 474 011.

E-mail [email protected]

Abstract

This paper reports an investigation of high pressure, phase transitions of Copper Chloride (CuCl) from zinc blende (ZnS) to sodium chloride (NaCl) structure passing through an intermediate (tetragonal) structure, using an effective interionic potential approach. The phase transition pressure (Pt) and associated volume collapses [∆V (P)/V (O)] obtained by us for the respective phases, show a reasonably good agreement with the experimental data. Also, the variation of Gibbs free energy (∆G) and relative volume collapse as a function of pressure (P) are discussed. 1. Introduction

Copper halides (CuX; X= Cl, Br, I) are members of a family of semiconductors which has received much attention during past few decades [1]. Among these halides copper chloride (CuCl) is the most ionic I-VII semiconductor with four valence electrons per atom, which crystallizes into tetrahedrally co-ordinated zinc-blende (ZB) lattice structure under ambient conditions in [2-4]. Due to the electrostatic interactions [5] the zinc blende structure of the compound becomes unstable and undergoes a structural phase transition. Copper chloride adopts rock-salt structure at a pressure in the region of ~ 10 GPa [6-9]. In addition, copper chloride undergoes a number of structural phase transitions under pressure before attaining the rock salt structure. Due to the dynamic instabilities [11], zinc blende (ZB) CuCl first transforms to an intermediate phase, then goes to a cubic rock salt (RS) phase [10] under elevated pressure.

In the present paper, we have reported the

structural properties of the compound under hydrostatic pressure. The structural properties, phase transition pressure (Pt) and relative volume collapse of copper chloride is being calculated for parent structure zinc blende (ZnS) to intermediate (tetragonal) structure and adopted sodium chloride (NaCl) structure by using an effective interionic potential described in section 2. Results are displayed in section 3 with its comparison to the available experimental data and discussion. 2. Computation Method

It is well known that pressure causes change in the volume of the crystal and consequently it alters

the charge distribution of the electron shells. Therefore, a proper account of phenomenon has to be taken in a theoretical analysis of the crystal under compression. The stability of a lattice is attained at the minimum Gibbs free energy for a particular lattice spacing r, given as:

(1) TSPVUG = + −

Here, U is the internal energy, which at 0 K corresponds to the cohesive energy, S is the vibrational entropy at absolute temperature T and V is the volume at pressure P. At temperatures sufficiently near zero (i.e. T = 0K) one can ignore [12, 13] the entropy term (TS) and thus Gibbs free energy is expressed as:

)()()(333

rPVrUrG BBB = + (2)

)()()( rPVrUrGTTT BBB ′′ ′ (3) = +

)()()( rPVrUrG BBB ′′111

′′= + (4) for the zinc-blende (B3), tetragonal (BT) and rock salt (B1) structure with lattice energies (UB) [14] defined as: B

∑ ⎟⎟⎠

⎞⎜⎜⎝

⎛ −++

⎟⎠⎞

⎜⎝⎛−−=

ij

ijjiij

MB

rrrb

rC

rZe

rU

ρβ

α

exp

)( 6

22

3 (5)

∑ ⎟⎟⎠

⎞⎜⎜⎝

⎛ ′−++

⎟⎠⎞

⎜⎝⎛

′′

−′

′−=

ij

ijjiij

MB

rrrb

rC

rZerU

T

ρβ

α

exp

)( 6

22

(6)

∑ ⎟⎟⎠

⎞⎜⎜⎝

⎛ ′′−++

⎟⎠⎞

⎜⎝⎛

′′′′

−′′

′′−=

ij

ijjiij

MB

rrrb

rC

rZerU

ρβ

α

exp

)( 6

22

1 (7)


High Pressure Phase Transition of Zinc-Blende Copper Chloride 257

Here, first term represents the Coulomb interaction corresponding to the nearest nieghbour separations )/( rrr ′′′ and Madelung constant

)/( MMM ααα ′′′ for ZB (Tetragonal/RS) structures and Ze is the ionic charge [15]. The second term is the van der Waals (vdW) coefficient due to dipole-dipole (d-d) interaction, with )/( CCC ′′′ as their overall coefficients [16] estimated by Slater Kirwood variation (SKV) method. The last term is the Hafemeister-Flgyare (HF) type [17] repulsive interaction operative upto the second neighbour ions. ijβ is the Pauling coefficients with ( )ji rr as the ionic radii of the cations (anions); ρ and b are the range and hardness model parameters determined by the equilibrium condition

0)(=

= orrdrrdφ (8)

where, r is the nearest interionic separation; ro is the equilibrium separation i.e. r = ro and the bulk modulus [18]

orro drrd

KrB

=⎥⎦

⎤⎢⎣

⎡= 2

2 )(9

1 φ (9)

with, K as the crystal structure constant. 3. Results and discussion

The effective potential consists of two model parameters,ρ, b: range and hardness can be evaluated from the knowledge of the second order elastic constants and potential energy at equilibrium condition. The values of input data and model parameter are listed in Table 1.

Table 1. Input data and Model parameters of CuCl

a. ref [14,19], b. ref [14,20]. In order to obtain the structural phase transition pressure, we have minimized the lattice energy using the equations (5), (6) and (7) for the

equilibrium interionic spacing )r(and)r((r), ′′′ corresponding to Zinc

Blende (ZB), intermediate (Tetragonal) and Rock Salt (RS) phases respectively. The cohesive energies

1T3 BBB UandU,U at zero pressure (P=0) are reported and compared with available experimental data [21]. Our calculated cohesive energy are in good agreement with the available experimental data. It is interesting to note that the short-range repulsive energy is 10% of the total energy which indicates that the cohesion in these crystals are due to the Coulomb interaction. Table 2. Cohesive energy of CuCl for different phases in Kcal/mole.

3BU

Cal. Exp. TBU

Cal. Exp. 1BU

Cal. Exp. -287.5

-318.5a

-261.9

-219.3a

-213.5

-

a. ref [21]. Table 3. Structural properties of the compound.

CuCl ZB-

Tetra. Cal xp.

Tetra.-NaCl Cal. Exp.

Pt (in GPa.) [ΔV(P)/V(O)]%

5.5 5.6

5.0a

- 14.25 9.41

≥ 10.0a

- a. ref [9,21,22]. The Gibbs free energy differences ΔG

[= )GG(and)G(G1TT3 BBBB −=− ] have

been obtained at different pressure (P) for different phases, on the lines of our earlier papers [23-26]. The variation of (ΔG) with pressure (P) are shown in fig 1 (a) and (b) value of ΔG is positive for ZB phase and it decreases with pressure and becomes negative for Tetragonal phase crossing to zero fig (a). Also ΔG for tetragonal to rock-salt phase is shown in fig 1(b). The pressure at which ΔG approaches zero corresponds to the phase transition pressure (Pt) as indicated by arrows in the figures for respective phases.

Input Parameters Model Parameters

C11(1011 dynes/cm2) 4.54a

C12(1011 dynes/cm2) 3.63a

C44(1011 dynes/cm2) 1.36a

ro (Ǻ) 2.32b

(ro) (Kcal/mole) 232.00a

ρ(Ǻ) 0.313

b(10-12ergs) 0.366


It is noticed from the fig 1, that our approach

has predicted correctly the relative stability of competitive structures. The values of the relative volume change [ΔV (Pt)/V (O)] associated with the compression (a sudden collapse in volume) has been obtained and plotted to depict the phase diagram in figure 2 for the different phases as indicated by arrows. The values of phase transition pressure and relative volume collapse has been reported and compared with available experimental data in Table 3. The volume changes in our calculations are in accord with those expected when the coordination changes during the phase transition. Our calculated phase transition pressures are in good agreement with the experimental data.

On the basis of above discussion it may be

concluded that our potential model has correctly predicted the phase transition pressure and relative volume collapse. The value of phase transition pressures calculated using the model [Pt =5.5 GPa and 14.25 GPa.] for different phases are closer to the experimental value [Pt = 5.0 GPa and ≥ 10.0 GPa] [9, 21, 22]. The phase diagram depicted in figure 2 shows that the phase transition of copper chloride (CuCl) from zinc blende (BB3) to rock salt (B1) through an intermediate tetragonal (BT) structure is accompanied by volume collapse. The volume collapse [∆V(Pt)/V(O)]% associated with these phase transition are 5.6 and 9.41 for respective phases, these values could not be directly compared with experimental due to a lack of the experimental data and hence our comment on their reliability are restricted, but phase diagram identities follows the same trend as obtained experimentally for compounds of the same group. It is thus obvious from the overall achievements that the present potential is adequately suitable for describing the phase transition studies of superionic solids. References [1] C. H. Park and D. J. Chadi, Phys. Rev.

Lett. 76 (1996) 2314. [2] M. Cardona, Phys. Rev. 129 (1963) 69. [3] C. Ulrich, K. Syassen and M. Cardona,

phys. Rev. B. 60 (1999) 9410. [4] A. Gobel, T. Ruf, C. Lin and M. Cardona

Phys. Rev. B. 56 (1999) 210.

-4

-2

0

2

4

0 5

Pressure (GPa.)

ΔG

(Kca

l/mol

.)

10

Pt

(a)

(b

-4

-2

0

2

4

0 10

Pressure (GPa.)

ΔG

(Kca

l/mol

.)

0.6

0.8

1.0

0 10 20Pressure (GPa.)

Volu

me

chan

ge [V

(Pt)

/V(O

)]

Pt

Pt

ZnS

Tetragonal

NaCl

Fig. 2. Relative volume change as a function of pressure (P) form ZnS-Tetragonal-NaCl structure.

20

Pt

Fig 1. Variation of Gibbs free energy as a function of pressure (P) form (a) ZnS-Tetragonal and (b) Tetragonal-NaCl structure.

High Pressure Phase Transition of Zinc-Blende Copper Chloride 259

[5] J. C. Philips, Rev. Mod. Phys. 42 (1970) 317. [6] V. Meisalo and M. Kalliomaki, High Temp-

High Press. 5 (1973) 663. [7] G. J. Piermarimi, F. A. Mauer, S. Block, A.

Jayaraman, T. H. Geballe and G. W. Hull, J. Solid State Commun: 32 (1979) 275.

[8] N. R. Serebryana, S. V. Popova and A. P. Rusakov, Sov. Phys. Solid State, 17 (1976) 1843.

[9] S. Hull and D. A. Keen, Phys, Rev. B. 50 (1994) 5868.

[10] Y. Ma, J. S. Tse and D. D. Klug, Phys. Rev. B. 69 (2004) 064102-1.

[11] S. Baroni, S. De Gironchi, A. Corso and P. Giannozzi, Rev. Mod. Phys 73 (2001) 515.

[12] J. M. Tranquada and R. Ingalls, Phys. Rev. B. 34 (1986) 4267; Phys. Lett. A 94 (1983) 441.

[13] K. N. Jog , R. K. Singh and S. P. Sanayal, Phys. Rev. B. 35 (1987) 5235; 31 (1985) 6047.

[14] R. K. Singh and D. C. Gupta, 1L Nuovo Cimento D 2 (1987) 1253; R. K. Singh, D. C. Gupta and S. P. Sanayal, Phys. Satus Solidi B 149 (1988) 356.

[15] R. K. Singh, Phys. Rep. 85 (1982) 259.

[16] R. K. Singh, S. Singh, Phys. Rev. B. 39 (1989) 761.

[17] D. W. Hafemeister and Flygare, J. Chem. Phys. 43 (1965) 795.

[18] N. K. Gaur, N. Kaur, M. Manke, J. Galgale and R. K. Singh, Mod. Phys. Lett. B, 17 (2003) 39.

[19] K. Kunc, M. Balanski and M. A. Nusimovice: Phys, Rev. B. 12 (1975) 4346; Phys. Stat. Sol. B, 71 (1975) 341; 72 (1975) 229.

[20] R. K. Singh and P. Khare, J. Phys. Soc. Jpn., 51 (1982) 141.

[21] S. Ves, D. Glotzel and M. Cardona, Phys. Rev. B. 24 (1981) 3073.

[22] A. Blacha, N. E. Christensen and M. Cardona, Phys. Rev. B. 34 (1986) 2413.

[23] R. K. Singh and S. Singh, Phys. Rev. B. 39 (1989) 671.

[24] R. K. Singh and S. Singh, Phase Transit 15 (1989) 127.

[25] R. K. Singh and S. Singh, Phys. Rev. B. 45 (1992) 1019.

[26] R. K. Singh and S. Singh, Phase Transit 654 (1995) 61.

STABILITY OF Na CLUSTERS INSIDE C240 MOLECULE

Harkiran Kaur*, K. Ranjan# and Keya Dharamvir# *University Center of Instrumentation and Microelectronics, Punjab University, Chandigarh

#Deptt. of Physics, Punjab University, Chandigarh E-mail: [email protected]

Abstract

Fullerene is a class of carbon compounds which are closed cage clusters of carbon atoms. The number of carbon atoms in these clusters varies from 20 to 960 and results in a variety of molecules. Three dimensional assembly of these molecules form solids which have very interesting electronic and mechanical properties. A fullerene is a hollow molecule which may accommodate guest atoms inside it. This process is known as endohedral doping. An endohedrally doped fullerene molecule has modified electronic and mechanical properties. This in turn modifies the properties of the solid formed with these molecules. Therefore one can mimic electronic and mechanical properties with different number and type of dopants. Small alkali atoms, alkaline earth metals and rare earth metals may be doped inside fullerene molecules. In the present paper we will address the issue of structural stability of endohedrally Na doped C240 molecule. C240 is a large molecule having mean radius of about 7.11Å. Therefore a number of Na atoms may be doped inside it. In our calculations, the number of the Na atoms doped inside C240 varies from 4 to 9. We have addressed the question of charge transfer to C240, by considering appropriate interaction between Na and C atoms. The binding between Na atoms has been modeled using Gupta potential. We have found that the binding energy of the cluster increases as the no. of Na atoms increase.

1. Introduction

Carbon is one of the commonest substances on Earth and it is widely distributed in nature. All life is carbon-based, and this gives rise to the description of the chemistry of carbon compounds as Organic. Carbon was probably the first element that man was aware of since it is produced in the form for charcoal from burned wood. Charcoal, used in early cave painting and by the artists today, consists of very small crystals of graphite, one of two forms of carbon [1]. The other form is diamond, which is especially famous for its hardness and its high dispersion of light. Until the late twentieth century, graphite and diamond were the only known allotropes of carbon. Fullerenes were first discovered in 1985 in an apparatus designed by Prof. Rick Smalley to produce atomic clusters of the non-volatile element.[2] The discovery of this new form of pure carbon, and the insight that led to the interpretation of the magic number features in the observed mass spectra in terms of hollow carbon molecules, led to the award of the Noble Prize in Chemistry to Curl, Kroto and Smalley in 1996. 1.1 Fullerene

Fullerene is a large molecule composed exclusively of carbon atoms and having shape of

an empty cage with a minimum of 12 pentagons. These Carbon Fullerenes commonly refers to a molecule with 60 carbon atoms, C60, and with an icosahedral symmetry, but also includes larger molecular weight fullerenes Cn (n>60) [3]. Examples of larger fullerenes are C70, C80, C240 and higher mass fullerenes, which possess different geometric structure. Figure 1 shows the structure and geometry of some fullerene molecules. The carbon atoms are arranged in pentagons and hexagons within a fullerene molecule. In C60, 60 carbon atoms are located at the vertices of a regular truncated icosahedron and every carbon site is equivalent to every other site.

Fig.1 Molecule of C60

The average nearest neighbor C-C distant in C60 is 1.44 Ǻ and each carbon atom inside C60 is


trigonally bonded to other carbon atoms. Soon a whole family of related molecules was discovered, including C70, C84 and other “fullerene” –clusters as small as C24 and as large as a postulated C240.

Possible applications of interest to industry include optical devices; chemical sensors and chemical separation devices; production of diamonds and carbides as cutting tools or hardening agents; batteries and other electrochemical applications, including hydrogen storage media; drug delivery systems and other medical applications; polymers, such as new plastics; and catalysts which appear to be natural application for fullerenes. 1.2 Endohedral fullerenes

Fullerene can be doped in several different ways. For example, Endohedral doping- where the dopant is inside the fullerene shell. Substitutonal doping –where the dopant is included in the fullerene shell. Exohedral doping- where the dopant is outside the fullerene shell [ 7].

Fig.2. Endohedral Metal doped C60 (M@C60 )

Therefore, Endohedral fullerene or endofullerene are the endohedrally doped molecular unit where the dopant is inside the hollow core of the fullerene shell. For example La atom might be trapped inside a C60 molecule to form an “endohedral” fullerene [4]. The endohedrally doped configuration denoted by La@C60 for one endohedral lanthanum in C60 or Y2@C82 for two Y atoms inside a C82 fullerene. In principle, many different atomic species can be inserted within fullerenes. The insertion of one, two, and three metal species inside a fullerene cage is common and up to four metal

atoms have thus far been introduced. Examples of endohedrally doped rare earth, alkaline earth, and alkali metal fullerenes are N@C60, Y2@C60, La2@C80, La@C82, and La2@C84.

As the novel form of fullerene-based materials, endohedral fullerenes represent a novel type of nanostructures, which are characterized by a robust fullerene cage with atoms, ions, or clusters trapped in its hollow. Because of the electron transfer from the encaged species to the fullerene cage, this new type of molecules has opened many possibilities for research and has been attracting the wide interest not only in physics and chemistry but also in such interdisciplinary areas as materials and biological sciences. 1.3 Sodium doped C240 A perfect example of an endohedral fullerene is sodium doped C240, Na atoms trapped inside a C240 molecule to form an “endohedral” fullerene. The endohedral doping configuration is denoted by Na@C240 for one endohedral sodium in C240 or Na2@C240 for two Na atoms inside a C240 fullerene and so on. We will study the interaction between the endohedral clusters and the cage by using an approximation in which both the Na and C atoms are frozen at the same positions as they have in the pure NaN clusters and the C240 fullerene.The endohedral cluster (Na)N is placed such that its centre of mass coincides with the center of C240 and the potential is spherically averaged around this point. These molecules are interesting to material scientists as one can play with electronic and optical properties with different no. and type of dopants. To start with structural stability is one of the important aspects to investigate. In this project, we have studied the stability of Na clusters with various No. of atoms. The equilibrium is obtained with minimization with charge and size of the each cluster. 2. Theoretical formalism C240 fullerene molecule is considered as a spherical cage where the carbon atoms are distributed in a similar way as in the corresponding truncated icosahedrons fullerene; each carbon atom is placed on the vertices of slightly distorted pentagonal and hexagonal rings distributed on a spherical hollow cage with radius R=7.11Å [5]. Endohedral doping of Na atoms to C240 molecule may result in charge transfer from Na atoms to C240 molecule. we can

Stability of Na Clusters Inside C240 Molecule 262

expect that one or more electrons may be transferred from the Na cluster to the fullerene cage. In our calculations, the No. of atoms in the cluster varies from 4 to 9. Various interactions between Na atoms and Na and Carbon atoms of the C240 has been considered to calculate total cohesive energy of the system and hence the stability of clusters of Na atoms (cluster) inside C240. The various components of the total cohesive energy are as under. (i) Interaction between sodium atoms in the cluster

Let the number of Na atoms in the cluster is N. To model the metallic bonding in sodium cluster we have used the many-body Gupta potential [6,7], which is based on the second moment approximation of a tight-binding Hamiltonian. It uses exponential functions rather than powers of the separation. Its analytical expression is given by [8],

)21

)]1)0

(2exp[

)10

(exp1

(2

1

−∑≠

−−

−∑≠

−∑=

=

ji r

ijrq

r

ijr

jip

N

iAGPV

ξ

(1)

where rij is the distance between Na atoms of the clusters and r0, A, ξ, p and q are adjustable parameters [9]. For sodium clusters, these parameters have been fitted to band structure calculations [10]. The values are: A = 0.01595 eV, ξ = 0.29113 eV, r0= 6.99 bohr, p = 10.13, q = 1.30 This potential has been already used to study the structure and thermodynamic properties of sodium clusters [6,7]. The Coulomb interaction between Na Atoms is given by

∑

≠

×=

N

ji

ji ijr

jqxiqxKCOLV

,2

1 (2)

where K is the constant, qx is the charge on each Na atom of the cluster and rij is the distance between Na atoms of the clusters. Combining Eqn. 1 and 2, we have

VCLST = VGP + VCOL (3)

(ii) Interaction between Na clusters and C240 Now we will consider the interaction

between Na atoms with the carbon atoms of C240

molecule. Let q be the charge transferred to C240 molecule. We consider that q is uniformly distributed among all carbon atoms on the C240 molecule. Therefore the charge on each C atom is qc = q/240. Coulomb interaction may be superimposed with van der Waals interaction used earlier in exohedral doped NaxC60 compounds [11]. The resultant contribution to the energy is

)exp(6

240

2

1240

ijrB

ijr

A

N

i j ijr

cqikqxNaCV

α−+−

∑ ∑=

(4)

where A=6.95, B=1520, α=3.60, qx=charge on each Na cluster, qc = charge on each C240, rij is the distance between Na atom of the clusters and C atom of C240.

(iii) Interaction due to C240

The interaction between a pair of C atoms on C240 is considered as Coulomb type only, which arises due to charge transfer from the Na cluster.

∑≠

×=

240

,240 2

1

jiji ij

cjci

rqq

kV (5)

where, qc is the charge on each C atom. The cohesive energy may be written as sum

of all the contributions described above. 240240 CNaCCLSTCOH VVVV ++= (6) Apart from this, the electron affinity ofC240 and ionization potential Na needs to be considered to discuss the stability of cluster in these systems. The SAPS calculation [3] of the electron affinity, defined as A(C240)=E(C240)-E(C240), gives the value A(C240)=3.81ev. subsequent addition of electron will modify the higher electron affinity of C240. we have estimated it by adding on shell Coulomb repulsion (φ) to it. It has b the The contribution to the cohesive energy is given by

IPqqqEAqU ×+−+×−= )1(2

1)( ϕ (7)

where q is charge on C240, EA is Electron Affinity, IP ionization potential of Na and equals

to 5.14 eV and 05.7

6021.19 ×=ϕ .


Table1. Structure and energy of various clusters inside C240

S. No No. of Na

atoms

Structure of Na cluster

Size of cluster

Edge length (a) Å

Volume per Na atom (V/N)

Total Cohesive

Energy(E) eV

Energy Per Na

atom(E/N)

1 4 Tetrahedron 2.51 3.549 1.3176 -1.881 -0.47

2 5 Tetrahedron+ centre

3.72 5.260 3.4302 -2.444 -0.49

3 6 Octahedron 4.92 3.478 3.3054 -4.358 -0.73

4 7 Octahedron + centre

6.14 4.341 5.5088 -5.238 -0.75

5 8 Simple Cube 3.24 3.24 4.2515 -7.209 -0.90

6 9 Simple Cube+ centre

3.56 3.56 5.0131 -8.957 -1.00

Now, Total cohesive energy is obtained by adding Eqn. 6 and 7. VTCOH = VCOH + U (8) The Total cohesive energy VTCOH, contains terms which depends upon the shape and size of the cluster of Na atoms (Eqn. 6). We can also vary the fraction of charge q transferred to C240 molecule. Minimization of VTCOH results in equilibrium value of size of the cluster and charge on it. The C240 molecule is assumed to be rigid. 3. Results and discussion

Table 1shows the minima of VTCOH w. r. t. size of the cluster and charge q for various clusters at equilibrium. From fig.3 it is clear that VTCOH is minimum for no transfer of charge i. e. q=0 in a cluster of 4 atoms and similar trend is seen in all remaining clusters considered here.

-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

-2.0

-1.8

-1.6

-1.4

-1.2

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

Tot

al c

ohes

ive

ener

gy

Charge on cluster

4 Na atoms Cluster

Fig. 3. VTCOH vs charge on the clusters The details of the structure have been summarized in the table 1 given above. From Table 1 we observe that size of the cluster increases with the increase in no. of Na atoms. The binding energy per atom also shows gradual rise with increase in no. atoms in the cluster.

Stability of Na Clusters Inside C240 Molecule 264

4 5 6 7 8 9-10

-9

-8

-7

-6

-5

-4

-3

-2

-1

Tota

l coh

esiv

e en

ergy

No. of Na atoms

Fig. 4. VTCOH vs No. of Na atoms in the Clusters

Therefore there is a tendency of the clusters to grow in size inside C240. Fig. 4 shows the variation of VTCOH with no. of Na atoms in the cluster. The VTCOH increases montonically with no. of atoms. 4. Conclusions

We conclude that as the molecule C240 is large in size therefore the interaction between cluster and C240 is very small while the cluster is neutral. But it becomes repulsive with more and more transfer of charge and results in lowering the binding energy which ultimately makes the system unstable. But the choice of the type of dopants is determinantal to predict the possibility of charged cluster formation. In case of Na ionization potential is quite large say 5.14 eV. It

lowers the binding energy with the same magnitude. If we have a dopant with smaller ionization potential, the fractional charge transfer may be energetically favorable. Further the smaller size of fullerene may also be favorable for cluster stability inside fullerene molecule. References 1. Taylor, R. Lecture Notes on Fullerene

Chemistry: A handbook for chemists; Imperial College: London, 1999

2. Kroto, H. W.; Heath, J. R.; O’Brien, S. C.; Curl, R. F.; Smalley, R. E. Nature 318 (1985) 162

3. "http://en.wikipedia.org/wiki/Endohedral_fullerenes"

4. Science of Fullerenes and Carbon Nanotubes by M. S Dresselhaus, G.Dresselhaus, P.C.Eklund.

5. Jian ping Lu and Weitao yang, Phys Rev B 49 (1994) 11421

6. R. P. Gupta, Phys Rev B 23 (1981) 6265 7. J. M. Cabrera-Trujillo et al., Phys Rev B 53

(1996)16059 8. Juan A. Reyes-Nava, Ignacio L.Garzon, and

Karo Michaelian Phys. Rev. B 67 (2003) 165401

9. F. Cleri and V.Rosato, Phys. Rev. B 48 (1993)22

10. Y. Li, E. Blaisten-Barojas, and D.A. Papaconstantopoulos, Phys. Rev. B 57(1998)15 519

11. K. Ranjan, Keya Dharamvir and V. K. Jindal, Physica B 371(2006) 232

Phase Transition in ASe (A= Ca, Eu and Th)

D.C. Gupta†, K.C. Singh†, Subhra Kulshrestha† and Sonia Mehra† † School of Studies in Physics, Jiwaji University, Gwalior – 474 011


Abstract

The properties of ASe(A=Ca, Eu and Th)have been investigated under high pressure by modified charge transfer potential. This model based is based on long-range attractions, charge-transfer mechanism, effects of covalency and short-range overlap repulsive interactions. Selenium compounds undergo transition from NaCl to CsCl phase under high pressure. The values of the phase-transition pressure (PT= 36.30, 14.15 and 14.34 GPa) computed from the present model shows good agreement with their corresponding experimental data for CaSe, EuSe and ThSe respectively and compression of 8.20 %, 12.10% and 10.34% respectively.

1. Introduction Selenium of Calcium, Europium and Thorium belong to the family of rare-earth compounds. They form a closed-shell ionic system which crystallizes in rock-salt (B1) structure at ambient conditions. In these Selenium Ca belongs to 2nd group, Eu is from lanthanide family while Th is a member of actinide series. The high pressure studies of CaSe has been reported to have a structural transformation from B1 to B2 phase at 38 GPa [1], EuSe transforms at comparatively lower pressures, i.e., at 14.50 GPa [2] while ThSe have been reported to transformation from B1 phase to B2 phase at further lower pressure, i.e., at 15.00 GPa [3]. These rare-earth materials lead to a mixed valency state under pressure. CaSe does not have d-electrons in the valency state while EuSe and ThSe are f-band materials. The structural and electronic properties of CaSe under high pressure have been calculated by GGA and LDA [4]. The cohesive properties of CaSe and EuSe at high pressure are done by [5,6] using ab-initio method. The three-body potential (TBP) model has been found to predicted successfully the various lattice static, dynamic, anharmonic, elastic, photo-elastic and high pressure phase transition properties of a variety of materials with cubic structure [7]. In the present paper we have developed a modified charge-transfer potential (MCTP) model, which includes covalency effects [8] along with the long-range Coulomb and many-body effects and short-range repulsion. 2. Methodology

The calculations are performed by modified charge-transfer potential (MCTP) model consists

of the long-range coulomb attraction, charge transfer mechanism, effect of covalency and short-range overlap repulsive effects. With the application of increases pressure the crystal leads to decrease in its volume. In close-shall materials, causes a charge transfer or three-body interaction (TBI) due to the increased overlap of electron shells of the adjacent ions. The stability of a lattice is attained at the minimum Gigs free energy G (=U+PV-TS) here, U is the internal energy equivalent to the lattice energy. S is the vibration entropy at absolute temperature T and V is the volume at pressure P. At temperature sufficiently near zero (i.e. T=0) one can ignore the entropy terms (TS) and represses it as G=U+PV [8]. The Gibbs free energies for the two phases are expressed as

(for B1 phase)

and

(for B2 phase)

Here UB1 and UB2 are internal energies of B1 and B2 phases respectively. Then the internal energies expression for the modified charge-transfer potential (MCTP) model can be written as

U(r) =UCOL +UCTP +UHF

The first term represents the coulomb energy [8] due to attraction force, this can write as

rZe

rU MCOL

22

)(α

−=

The second term UCTP corresponds to energy due to charge transfer potential (CTP) and covalency.

111 )()( BBB PVrUrG +=

222 )'()'( BBB PVrUrG +=


rrnfZe

rU mMCTP

)(2)(

2α−=

Where, n is the number of nearest neighbour and fm(r) is the charge transfer potential parameter, αM is Madelung constants and Ze is ionic charge.

The last term is the Hafemeister and Flygare (HF) type short range overlap repulsive potential, the Short range (SR) repulsive potentials have been repotted by Hafemeister-flyger (HF) [8] and Hafemeister-Zarht (HZ)[8] with some refinement for better description of the overlap repulsion. The expression for short range potential proposed by HF is given as

1( ) exp(( ) / ) ' exp((2 ) / )2

1 ' exp((2 ) / )2

HF ij i j ij ii i ii

jj j jj

U r nb r r r n b r r

n b r r

β ρ β ρ

β ρ

= + − + −

+ −

Where, n is the number of nearest neighbours (nn), n’ is next nearest neighbours (nnn) and βij are Pauling coefficients define as

j

j

i

iij n

ZnZ

++=1β

Where, Z and n are valence and the number of outermost electrons respectively.

The values of range (ρ) and hardness (b) parameters and modified many-body interaction parameter ƒm(r) have been determined from the equilibrium condition as

T

or=ror=r

KrBdr

Uddr

dU 9 and 0 2

2

==

Where, U is modified as charge-transfer potential. Where K=V/r3 as the structure dependent constant.

In second term in lattice energy the CTP parameter modified with covalency effects can be calculated from the relation as given below

The value of CTP parameter can be calculated from Cochran relation as.

The covalency term is defined as

2 2 *

3

4 1SP sCOV

o g

V e efr E e

σ ⎡ ⎤= −⎢ ⎥

⎣ ⎦

Here*

1 se nce

⎡ ⎤− =⎢ ⎥

⎣ ⎦

Where, nc is the number of electrons transferred to the unoccupied orbital of a cation from its surrounding anion at the nearest neighbour.

We can determine V2spσ/Eg2 using a hypothesis which was derived from correlations between the hyperfine coupling constants of the transition metal impurity ions and the Szigeti effective charge e*s of the host crystal.

So

*

2

2

1

1 2

s

S P

g

eV e

Eσ

−=

This relation is obtain by using the value of nc and nc/12 =V2spσ / Eg2 in the LCAO approximation.

The value of Eg, i.e., the transfer energy of electron from anion to cation is determined by the same approximation as

Eg = 0

2)12(r

eIE M −+−

α

Here, E denotes the electron affinity of the cation and I denote the ionization potential of the AEC atom.

εo(ε∞) are denoting the static and optical dielectric constant and the transverse optical phonon frequency at zone centre by ωt ,e*s is given by

(e*)2 = 2

02

)2(4)(9

+−

∞

∞

επεεμω

Nt

For calculating ωt, we use LST relation

20

2LO

TO

ω εω ε∞

=


In order to check the relative stability of the two phases, we have minimized the lattice energies in both the real and hypothetical phases

( ) ( ) ( )m CTP COVf r f r f r= +

( ) exp( / )CTP o of r f r ρ= −

Phase Transition in ASe (A= Ca, Eu and Th) 267

at zero pressure using the model parameters. At transition pressure both the phases are in equilibrium. Under high pressure, the Gibbs free energy changes and the crystal structure is stabilized at a new value of equilibrium separation. This variation of interionic separation causes corresponding changes in the physical properties of the solids. In Fig.-1, we have depicted the values of the calculated phase transition pressure for ASe (A=Ca, Eu and Th) compounds. We can also find from Table 1 that the calculated values of the phase their transition pressures are in good agreement with experimental values. It is clear form figure, as the atomic wait of A (=Ca, Eu and Th) increases then phase transition decreases. The equilibrium separation under pressure has been used to calculate the relative volume change (V(P)/V(0)) with pressure and the equation of state (EOS) for present material. The graphs of EOS for these compounds are depicted in Fig- 2. The value of (∆V(Pt)/V(0)) have also been obtained from these figures and reported in Table-1. The calculated values of the relative volume change (∆V(P)/V(0)) are 8.20%, 12.10 % and 10.34 % for CaSe, EuSe and ThSe, respectively.

Table 1. Calculated lattice parameter, phase transition pressure, relative volume collapse, and transition volumes for AS (A=Ca, Eu and Th) compounds.

Fig. 1. The value of phase transition pressure (PT) for CaSe, EuSe and ThSe

0

20

40

CaSe EuSe ThSe

P T (G

Pa)

Fig. 2. Variation of relative volume (V(P)/V(0) with pressure (P) for CaSe, EuSe and ThSe

0.6

0.8

1

0 10 20 30 40Pressure (GPa)

V(P

) / V

(0)

PT = 36.30 GPaCaSe

EuSe

ThSe

PT = 14.15 GPa

PT = 14.34 GPa

Acknowledgements The authors, (particularly DCG) are thankful

to the University Grant Commission, New Delhi for financial support through grant No. F.31-9/2005. References [1.] H. Luo, et.al., Phys. Rev. B50 (1994) 16232. [2.] M. Ananyas, Ph.D. Thesis, Barkatullah

University,y, Bhopal (2001). [3.] U. Benedict, et.al., J. Less-Common Mat. 98

(1984) 301. [4.] Z. Charifi, et.al., J.Phys.: Cond. Matter 17

(2005) 4083. [5.] P. Cortona and P. Masri, J.Phys.: Cond.

Matter 10 (1998) 8947. [6.] M. Horne, et.al., J. Phys.: Cond. Matter 16

(2004) 5061. [7.] R.K. Singh and D.C. Gupta, IL Nuovo

Cimento D9 (1987) 1253; Phys. Rev. B40 (1989) 11278; ibdi, 43 (1990) 11185; 45 (1991) 7031.

[8.] K. Motida, J. Phys. Soc. Japan 49 (1980) 213 and 218;; ibdi, 50 (1981) 1247; 55 (1986) 1636; 56 (1987) 1785.

[9.] A.S. Mischehenko and K.A. Kikoin, J. Phys.: Cond.d. Matter 3 (1991) 5937. Jayaraman et.al., Phys. Rev. B9 (1974) 2513.

[10.] R.W.M. D’Eye., P.G. Sellman, J.R. Murray, J. Chem. Soc. 2 (1952) 555.

Solids CaSe

EuSe

ThSe

Reference

2.95 3.09 2.94 Present Study

a (B1)

2.96a 3.09b 2.94c Exp[1a,9b,3c]

1.79 3.25 2.95 Present Study

Equilibrium Lattice Constant (in 10-10 m)

a (B2)

1.81 - - Exp [1] 36.30

14.15

14.34

Present Study

Phase Transition Pressure (PT) 38.0

0a 14.50b

15.00c

Exp[1a,10b,11c]

8.20 12.10

10.34

Present Study

Relative Volume collapse (in %) 7.70a 12.8

0b 9.00c Exp[1a,10b,11c]

0.73 0.61 - Present Study V(PT)/V(0)

(in B1) 0.73a 0.59b - Exp[1a,10b]

Parametric Analysis and Optimization of Cutting Parameters for Turning Operations Based on Taguchi Method

Amar Patnaika , S.K.Pradhana and Maheshwar Dwivedyb

aDepartment of Mechanical Engineering, National Institute of Technology, Hamirpur-177 005 (HP), India. bB.I.TS Pilani.


Abstract

Surface quality is one of the specified customer requirements for machined parts. There are many parameters that have an effect on surface roughness, but those are difficult to quantify adequately. In finish turning operation many parameters such as cutting speed, feed rate, and depth of cut are known to have a large impact on surface quality. In order to enable manufacturers to maximize their gains from utilizing hard turning, an accurate model of the process must be constructed. Several statistical modeling techniques have been used to generate models including regression and Taguchi methods. In this study, an attempt has been made to generate a surface roughness prediction model and optimize the process parameters Genetic algorithms (GA). Future directions and implications for manufacturers in regard to generation of an robust and efficient machining process model is discussed.

1. Introduction

Surface roughness has received serious attention for many years. It has formulated an important design feature in many situations such as parts subject to fatigue loads, precision fits, fastener holes, and aesthetic requirements. In addition to tolerances, surface roughness imposes one of the most critical constraints for the selection of machines and cutting parameters in process planning. A considerable number of studies have investigated the general effects of the speed, feed, and depth of cut on the surface roughness. Process modelling and optimization are the two important issues in manufacturing products. The manufacturing processes are characterized by a multiplicity of dynamically interacting process variables [1, 2]. A greater attention is given to accuracy and surface roughness of product by the industry these days. Surface finish has been one of the most important considerations in determining the machinability of materials. Surface roughness and dimensional accuracy are the important factors required to predict machining performances of any machining operations [3]. The predictive modeling of machining operations requires detailed prediction of the boundary conditions for stable machining [4,5]. The number of surface roughness prediction models available in literature is very limited [3,5]. Most surface roughness prediction models are empirical and are generally based on experiments in the laboratory. In addition it is

very difficult in practice, to keep all factors under control as required to obtain reproducible results [5]. Generally these models have a complex relationship between surface roughness and operational parameters, work materials and chip-breaker types. Optimizations of machining parameters are not only increases the utility for machining economics, but also the product quality increases to a great extent [1]. In this context, an effort has been made to estimate the surface roughness using experimental data. It has also been made an attempted to optimize the surface roughness prediction model using a Genetic Algorithmic approach. Since turning is the primary operation in most of the production processes in the industry, surface finish of turned components has greater influence on the quality of the product. Surface finish in turning has been found to be influenced in varying amounts by a number of factors such as feed rate, work material characteristics, work hardness, unstable built-up edge, cutting speed, depth of cut, cutting time, tool nose radius and tool cutting edge angles, stability of machine tool and workpiece setup, chatter, and use of cutting fluids.

Taraman [6] used Response Surface Methodology (RSM) for predicting surface roughness of different materials. A family of mathematical models for tool life, surface roughness and cutting forces were developed in terms of cutting speed, feed, and depth of cut. Hasegawa et al., [7] conducted 34 factorial designs to conduct experiments for the surface roughness prediction model. They found that the


surface rough increased with an increase in cutting speed. Sundaram and Lambert [8, 9] considered six variables i.e speed, feed, and depth of cut, time of cut, nose radius and type of tool to monitor surface roughness.

To improve the efficiency of these turning processes, it is necessary to have a complete process understanding and model. To this end, a great deal of research has been performed in order to quantify the effect of various hard turning process parameters to surface quality. These factors can be divided into a) setup variables, b) tool variables, and c) workpiece variables. In order to gain a greater understanding of the turning process it is necessary to understand the impact of the each of the variables, but also the interactions between them. It is impossible to find all the variables that impact surface roughness in turning operations. In addition, it is costly and time consuming to discern the effect of the every variable on the out put. In order to simplify the problem, one needs to eliminate or select specific variables that correspond to practical applications. Taguchi method [10] consist of a plan of experiments with the objective of acquiring data in a controlled way, executing these experiments and analyzing data, in order to obtain information about the behaviour of a given process. It uses orthogonal arrays to define the experimental plans and the treatment of the experimental results is based on the analysis of variance (ANOVA) [2]. interactions [11-12]. The outputs to be studied are surface roughness (Ra) and tool life (T). 2. Plan of experiment based on the Taguchi method

For the experimental plan, the Taguchi method for three levels was used with careful understanding of the levels taken by the factors. Table 1 indicates the factors to be studied and the assignment of the corresponding levels. According to the Taguchi design concept, a L27 orthogonal array was chosen for the experiments (Table 2). The analysis was made using the popular software, column to the depth of cut (d) and the remaining were assigned to the specifically used for design of experiment applications, known as MINITAB 14. The plan is made of 27 tests (array rows) in which the

first column was assigned to the cutting velocity (Vc),the second column to the feed rate (f) and the fifth Level Cutting

velocity Vc (m/min)

feed rate f (mm/rev)

Depth of Cut d (mm)

1 2 3

135 210 285

0.08 0.20 0.32

0.60 1.00 1.60

Table 1. Cutting parameters and their levels 3. Results and discussion

The test plan was developed with the aim of relating the influence of the cutting velocity (Vc), feed rate (f ) and depth of cut (d) on surface roughness (Ra) and tool life (T). We should mention that only one observation for a treatment is noted. The statistical treatment of the data was made in three phases. The first phase was concerned with the ANOVA and the effect of factors and the interactions. The second phase allows us to obtain the correlations betweens the parameters. Afterwards, the results were obtained through confirmation tests. In the final stage, optimization of turning parameters was carried out by using a Genetic Algorithm 3.1 ANOVA and the effects of factors

Analysis of Variance of the data with the surface roughness (Ra), and tool life (T) with the objective of the analyzing the influence of cutting velocity (Vc), feed rate (f ) and depth of cut (d) on the total variance of the results is performed. The experiments were conducted for each combination of factors (columns) as per selected orthogonal array. The number of observations under each combination of factors is one, i.e. the number of replications is one. The experimental results are shown in Table 3. Tables 4-5 show the results of the ANOVA with the surface roughness (Ra), and tool life (T) respectively. This analysis was undertaken for a level of significance of 5%, that is, for a level of confidence of 95%. The last column of the tables indicates that the main effects are highly significant (all have very small P-value)


Table 3. Experimental design using L27

From Table 4, one can observe that the cutting velocity (p= 0.001) and feed rate (p=0.000) have great influence on surface roughness. The interactions of cutting velocity/feed rate (p=0.000) and cutting velocity/depth of cut (p=0.001). But The factor depth of cut (p=0.028) and the feed rate/ depth of cut (p=0.300) have present less significant contribution on the surface roughness. From Table 5, one can observe that the cutting velocity (p=0.000) and feed rate (p=0.000) have great influence on the tool life. The interactions cutting velocity/ feed rate (p=0.000) and cutting velocity/ depth of cut (p=0.000), shows significance of contribution on the tool life. Whereas, depth of cut (p=0.030) and the interaction feed rate/depth of cut (p=0.003) have less significant effect as compare to the other factors and interactions on tool life.

Table 4. ANOVA table for surface roughness (Ra) Source DF SS MS F P A 2 17.6570 8.8285 41.10 0.001 B 2 92.1785 46.0892 247.64 0.000 C 2 0.6124 0.3062 9.40 0.028 A*B 4 0.7407 0.1852 91.66 0.000 B*C 4 0.0118 0.0030 1.46 0.300 A*C 4 0.1266 0.0317 15.67 0.001 Error 8 0.0162 0.0020 Total 26 111.3432 Table 5. ANOVA table for the tool life (T) Source DF SS MS F P A 2 671614 335807 68.71 0.000 B 2 3058621 1529311 437.32 0.000 C 2 26618 13309 9.37 0.030 A*B 4 13942 3486 512.40 0.000 B*C 4 73 18 2.69 0.109 A*C 4 5633 1408 207.04 0.000 Error 8 54 7 Total 26 3776556

Test Cutting Speed Vc (m/min)

Feed Rate f (mm/rev)

Depth of Cut d (mm)

Surface Roughness Ra (µm)

S/N ratio (db)

Tool Life T (Sec)

S/N ratio (db)

1 135 0.08 0.60 2.0855 -6.3842 1713.96 64.6800 2 135 0.08 1.00 2.3377 -7.3758 1650.48 64.3522 3 135 0.08 1.60 2.5220 -8.0349 1597.58 64.0693 4 135 0.20 0.60 4.3262 -12.7221 1272.25 62.0914 5 135 0.20 1.00 4.7142 -13.4682 1200.83 61.5896 6 135 0.20 1.60 5.0440 -14.0555 1139.99 61.1380 7 135 0.32 0.60 6.8870 -16.7606 830.53 58.3871 8 135 0.32 1.00 7.2362 -17.1902 761.76 57.6364 9 135 0.32 1.60 7.4884 -17.4878 708.86 57.0112

10 210 0.08 0.60 3.4144 -10.6663 1425.66 63.0803 11 210 0.08 1.00 3.6181 -11.1696 1385.98 62.8351 12 210 0.08 1.60 3.7733 -11.5344 1354.24 62.6339 13 210 0.20 0.60 5.9655 -15.5129 981.29 59.8359 14 210 0.20 1.00 6.1983 -15.8455 938.98 59.4531 15 210 0.20 1.60 6.3632 -16.0735 907.24 59.1544 16 210 0.32 0.60 8.0413 -18.1065 595.13 55.4922 17 210 0.32 1.00 8.1965 -18.2726 560.74 54.9752 18 210 0.32 1.60 8.3032 -18.3849 534.29 54.5555 19 285 0.08 0.60 4.3941 -12.8574 1243.15 61.8905 20 285 0.08 1.00 4.5202 -13.1032 1219.35 61.7226 21 285 0.08 1.60 4.6075 -13.2693 1203.48 61.6088 22 285 0.20 0.60 6.8676 -16.7361 817.31 58.2477 23 285 0.20 1.00 6.9937 -16.8941 793.50 57.9909 24 285 0.20 1.60 7.0713 -16.9900 777.63 57.8155 25 285 0.32 0.60 8.5360 -18.6251 481.39 53.6499 26 285 0.32 1.00 8.6039 -18.6939 460.23 53.2595 27 285 0.32 1.60 8.6524 -18.7427 447.01 53.0063


3. 2. Correlations

The correlations between the factors (cutting velocity, feed rate and depth of cut) and the measured surface roughness, and tool life were obtained by multiple linear regression analysis. The mathematical model suggested was in the following form. Y=P0 + P1*Vc + P2 * f + P3 *d+P4*Vc*f+P5*f*d +P6*Vc*d (1)

Here, Y is the performance output terms and Pi (i=0, 1, 6) are the model constants. The constants were calculated using linear regression analysis with the help of SYSTAT 7 software, and the following relations were obtained. The calculated coefficients from SYSTAT 7 software were substituted in Eq. (1). Ra= -0.309+ 0.675Vc + 0.870f + 0.175d-0.234Vc.f – 0.002f.d – 0.143Vcd r2=0.99 (2) T = 1.511- 0.646Vc – 0.783f – 0.186d + 0.189Vc.f + 0.004 f.d + 0.152Vcd r2=0.99 (3)

The higher correlation coefficients (r2) confirm the suitability of the models and the correctness of the calculated constants. In this study, a weighting method is used for the optimization of the process with multi-machining performance outputs. Since surface roughness (Ra) and tool life (T) are the two objectives, in order to overcome the large differences in numerical values between the objectives, the function corresponding to each machining performance output is normalized first.

A weighting method is adopted to formulate a single objective function involving surface roughness (Ra) and tool life (T). Table 6 shows the cutting conditions and cutting time used in turning operations during confirmation experiments, a new set of data was taken out, and conducted a new set of experiments. In Table 7, a comparison was made between the values obtained from the models developed in the present work, Eqs.2-3, with the values obtained experimentally. From the analysis of the table, we can observe that the estimated error is greater especially for surface roughness (Ra) (maximum value 7.0% and minimum 3.33%) and for tool life (T) (maximum value 3.71% and minimum 3.00%). Therefore, it can be concluded that the evolution of correlation equations for the surface roughness and tool life with the cutting conditions (cutting velocity, feed rate and depth of cut) satisfies a reasonable degree of approximation. Table 6. Cutting conditions in confirmation tests.

Test Vc (m/min) f (mm/rev) d (mm) 1c 140 0.16 1.3 2c 220 0.12 1.5 3c 300 0.18 0.9

Table 7. Experimental plan confirmation drilling tests and their comparison with the results.

4. Conclusions

The results outlined in this study lead to conclusions for turning of S45C after conducting the experiments and analyzing the resulting data.

1. Cutting velocity (0.001) and feed rate (0.000) have greater influence on the surface roughness followed by feed rate.

2. Cutting velocity (0.000) and Feed rate (0.000) have greater influence on the tool life.

3. The interaction between cutting velocity / feed rate (0.000) has a significant effect on surface roughness.

4. Similarly, for tool life, the interaction between cutting velocity / feed rate and cutting velocity / depth of cut has greater significant effect.

Test Surface Roughness Tool life Ra (µm) T (Sec) Expt. Model(Eq.(3)) Error(%) Expt. Model(Eq.(4)) Error(%)

1c 7.56322 6.66851 10.448 985.45 858.95 12.83

2c 6.78941 6.19235 8.793 1043.25 924.68 11.36

3c 9.1258 8.82544 3.291 849.38 756.61 10.92


5. The confirmation tests showed that the error associated with surface roughness (maximum value 10.448 % and minimum 3.291%) is lesser than the error associated with tool life (maximum value 12.83 % and minimum 10.92 %).

6. Using experimental data, a multiple linear regression model is developed and proves to be effective in optimizing the cutting conditions in turning operations.

7. The search for optimal turning conditions is based on mathematical formulation of the multi-objective optimization problem, and the contribution of each machining parameter is studied. The algorithm is tested to find optimal values of parameters with varying weighting factors for different objectives.

8. It can be concluded from this preliminary study that the turning were carried out on an engine lathe using tungsten carbide with the grade of P-10 for the machining of S45C steel bar is a very difficult operation, and much more work remains to be done to establish effective turning operation for such materials through improvement of tooling and process parameters.

9. In this study, the Taguchi method gives effective methodology in order to find out the effective performance out put and machining conditions.

References [1] R. Azouzi, M. Guillot, On-line optimization

of the turning using an inverse process neurocontroller, Transactions of ASME, Journal of Manufacturing Science and Engineering 120 (February) (1998) 101–107.

[2] T. Warren Liao, L.J. Chen, Manufacturing Process modeling and optimization based on

multi-layer perceptron network, Transactions of ASME, Journal of Manufacturing Science and Engineering 120 (February) (1998) 109–119.

[3] A. Mital, M. Mehta, Surface roughness prediction models for fine turning, International Journal of Production Research 26 (1988) 1861–1876.

[4] P.R. Motghare, Monitoring of cutting tools by the estimation of tool wear. Unpublished Masters Thesis, Dept of Mech. Engg., (1998). Indian Institute of Technology, Delhi, India.

[5] C.A. Van Luttervelt, T.H.C. Childs, I.S. Jawahir, F. Klocke, P.K.Venuvinod. Present situation and future trends in modelling of machining operations. Progress Report of the CIRP working group on ‘Modelling of machining operations’, Annals of the CIRP, 47/2 (1998) 587–626.

[6] K. Taraman, Multi machining output—multi independent variable turning research by response surface methodology, International Journal of Production Research 12 (1974) 233–245.

[7] M. Hasegawa, A. Seireg, R.A. Lindberg, Surface roughness model for turning, Tribology International December (1976) 285–289.

[8] R.M. Sundaram, B.K. Lambert, Mathematical models to predict surface finish in fine turning of steel, Part I, International Journal of Production Research 19 (1981) 547–556.

[9] R.M. Sundaram, B.K. Lambert, Mathematical models to predict surface finish in fine turning of steel, Part II, International Journal of Production Research 19 (1981) 557–564.

[10] U. Tetsutaro, M. Naotake, Prediction and detection of cutting tool failure by modified group method of data handling, International Journal of Machine Tools and Manufacture 26 (1986) 69–110.

Study of Formation of Diamond-Like Carbon Coatings and Theoretical Calculations of Stiffness of the Films

Avnish Narula

Department of Electrical & Electronics Engineering, University Institute of Engineering and Technology

Panjab University, Chandigarh- 160 025, India E-mail: [email protected]

Abstract

This paper reviews the formation of Diamond-like Carbon (DLC) coatings by chemical vapor deposition techniques (CVD). The method also summarizes the role of atomic Hydrogen and Oxygen species in the deposition of the films. Nano-indentation techniques have been used to calculate the stiffness of the films having average thickness of 1.5 μm. Typical nano-indentation curves and the results from the previous works have been used to calculate the stiffness which comes out to be 1.744 MN/m. A Berkovich indenter having a total included angle of 142.3 degrees was considered to calculate the theoretical values. 1. Introduction

Diamond-like carbon is an umbrella term that refers to 7 forms of amorphous carbon materials that display some of the unique properties of natural diamond, all seven containing significant amounts of sp3 hybridized carbon atoms. They are usually applied as coatings to other materials that could benefit from some of these properties. As implied by the name, Diamond-like carbon, the value of such coatings accrues from some of the properties of diamond to surface of almost any material. DLC coatings made at the same time are amorphous, flexible and yet purely sp3 bonded “diamond”.

This paper reviews the formation of Diamond-like carbon (DLC) coatings by chemical vapor deposition techniques (CVD) [6]. The method also summarizes the role of atomic hydrogen and oxygen species in the deposition of the film [7].

Nano-indentation techniques have been used to calculate the stiffness of the films having average thickness of 1.5 µm [15]. Typical nano-indentation curves and the results from the previous works [14] have been used to calculate the stiffness which comes out to be 1.744 MN/m. A Berkovich indenter having a total included angle of 142.3 degrees was considered to calculate the theoretical values. Formation of DLC films by chemical vapor deposition

Diamond is a thermodynamically Meta stable phase at room temperatures [1]. So synthetic diamonds are made at high temperatures under

high pressures with the aid of transition metal catalysts such as Ni, Fe and Co [2,3]. The growth of diamond films under low pressure (equal to or greater than 1 atm.) and low temperatures (~800˚C) is not a thermodynamically equilibrium process, and differs from other CVD processes. The formation of diamond from gas phase at low pressure was initially reported in late 1960s [4,5]. The typical process of diamond films is illustrated schematically in figure [6].

A gaseous mixture of hydrocarbon (typically methane) and hydrogen is fed into the activation zone of decomposition chamber where activation energy is introduced to the mixture and causes the dissociation of both Hydrocarbon and hydrogen molecules to form hydrocarbon free radicals and atomic hydrogen. Many different activation schemes have been found effective in depositing diamond films and include hot-filament, RF and microwave plasma and flames. Upon arrival on the growth surface, a generic reaction of surface reactions would occur [6];

C H H C H

C CH C CHC C H C C H

D D

D D

D x y D X y

+ → +

+ → −

+ → −

* *

* *

*

........( )

....( )...( )

2

3 3

1

23

Reaction 1 is to activate a surface site by removal of surface site by removal of surface Hydrogen atom linked to carbon atom as the diamond surface. An activated surface site readily combines with either a hydrocarbon


radical (reaction 2) or an unsaturated hydrocarbon molecule (eg. C H2 2 , reaction 3). Reactants

↓

↓

H CH2 4+

↓ Activation

H H

CH H CH H

e heat2

4 3 2

2−

⎯ →⎯⎯⎯

+ ⎯ →⎯ +

,

Schematics showing principal elements in the complex diamond CVD process

Flow of reactants into the reactor, activation of the reactants by the thermal and plasma processes, reaction and

Fig. 1. Flow and Reaction transport the species to the growing surface, and surface chemical processes depositing diamond and other forms of carbon.

A high concentration of atomic hydrogen has proven a key factor in the successful growth of diamond films and atomic hydrogen is believed to constantly remove the graphite deposits on the diamond growth surface so as to ensure continued deposition of diamond [5]. Oxygen species have also proven to be important in the deposition of diamond films by atmospheric combustion flames using oxygen and acetylene [7,8].

Background The measurement of surface mechanical properties on a micro and nanometer scale has been assuming more and more importance for both the research and the industrial sectors. As components to be analyzed become smaller and coatings become thinner, the apparatus required to measure their properties becomes more

complicated and the limits of resolution are continuously being pushed back. Micro / nano-tribological studies are crucial to developing a fundamental understanding of interfacial phenomena occurring at such small scales as well as of the structural and mechanical properties of the materials. Indentation tests, sometimes called hardness tests, are perhaps the most commonly applied means of testing the mechanical properties of materials. The technique has its origin in the Mohs scale of mineral hardness in which materials are ranked according to what they can scratch. The characterization of solids in this way takes place on an essentially discrete scale, so much effort has been expended in order to develop techniques for evaluating material hardness over a continuous range. More recently the nano-indentation technique has been established as the primary tool for investigating the hardness of small volumes of material.

In a traditional indentation test (macro and micro indentation), an indenter tip (frequently made of a very hard material like diamond) with a known geometry and mechanical properties is driven into a specific site of the material surface whose properties are unknown. The load placed on the indenter tip is increased as the tip penetrates into the specimen and soon reaches a user-defined value. At this point, the load may be held constant for a period or removed. The area of the residual indentation in the sample is measured and the hardness, H, is defined as the maximum load, P(max) divided by residual

indentation area. rA

PH max=

For most techniques, the projected area may be measured directly using light microscopy. As can be seen from this equation, a given load will give a smaller indent in a “hard” material than a “soft” one. Nano-indentation (A) Foundering of the traditional techniques

The traditional indentation technique is limited due to large and varied tip shapes, with indenter rigs which do not have a very good spatial resolution (the location of the area to be indented is very hard to specify accurately). Comparison across laboratories is difficult and

Study of Formation of Diamond-Like Carbon Coatings and Theoretical Calculations 275

often meaningless. Nanoindentation improves on these macro and micro indentation tests by indenting on the nano-scale with a very precise tip shape and high spatial resolution to place the indent.

The term “nanoindentation” refers to depth sensing indentation (DSI) techniques used to obtain mechanical properties from very small volumes of material [10]. The Nanoindentation tests can be used in the analysis of organic and inorganic soft and hard coverings. Examples are thin and multilayer PVD, CVD, PECVD, photo resistors, paints, lacquers; and many other types of films, covering optical, micro-electronic, protective, decorative and other applications, Substrates can be hard or soft, including metal alloys, semiconductors, glass, refractive and organic materials. (B) Three-sided Pyramidal Tips

In nanoindentation, small loads and tip sizes are used, so the indentation area may be only a few square micrometers or even nanometers. This presents problems in determining the hardness as the contact area is difficult to find. Atomic force microscopy or scanning electron microscopy may be utilized to image the indentation, but can be quite cumbersome. Instead, an indenter with a geometry known to a high precision is employed for this purpose [10].

The Berkovich tip is the standard nanoindentation tip. It is a three sided pyramid which is geometrically self similar. It has a very flat profile, with a total included angle of 142.35 degrees. The half-angle, or the angle from the perpendicular to one face, is 65.35 degrees. The aspect ratio of the tip is 1:8 [9]. The Berkovich tip has the same projected area to depth ratio as a Vicker indenter [10]. A typical radius of curvature for a standard Berkovich tip for thinner films would be approximately 150 nm. Sharper Berkovich tips with a radius of less than 50 nm are available [9]. As it is three-sided it is easier to grind these tips to a sharp point and so is more easily employed for nanoidentation type tests. It is typically used for bulk materials and films greater than 200 nm thick [10]. Because the Berkovich tip is the standard tip for indentation, it should be used whenever possible. The well-accepted models models for nanoindentation use this geometry. This tip is best tip for most bulk samples, unless the roughness is more than 50 nm RMS. The tip is manufactured from diamond; it therefore experiences negligible deformation during the indentation process. This can be assumed because

the Young’s modulus is approximately ten times than that of the samples under testing. Berkovich tips will generally provide good images. The tips are not so sharp that they wear on materials with modulus greater than 1 GPa. If softer samples are used, Berkovich tips may not image them well.

Typical indentation applications for Berkovich tips are:

• Bulk Ceramics and glasses. • Bulk metals and steels. • Thin hard films and coating greater

than 100 mm thick. • Hard, smooth biomaterials (Polished

bone and tooth samples) • Hard Polymers (Modulus greater

than 1 GPa) [9]. (C) Methodology

During the course of instrumented indentation process, a record of the depth of penetration is made and then the area of the indent is determined using the known geometry of the indentation tip. While indenting various parameters, such as load and depth of penetration can be measured. A record of these values can be plotted on a graph to create a load-displacement curve. These curves can be used to extract mechanical properties of the material [10]. 2. Experimental Set up

The Nanoindenter applies a force P on the test sample by passing a current through a coil that sits within a circular magnet.

Fig. 2. Schematic of Nanoindenter set up

The force on the indenter shaft is directly proportional to the current passing through the coil. The indentator simultaneously measures the displacement h1 of the shaft, measured at the capacitance cauge and the force. The


Nano-indentation can also operate in AC mode; this is where a small force nanoindentation is superimposed onto the primary loading signal acting on the indenter shaft. The oscillating tips resulting response are measured so that the contact stiffness S during loading can be calculated. This AC mode of testing is particularly useful to evaluate thin films on substrates where the mechanical properties change as a function of surface penetration [11]. Oliver-Pharr method

The Oliver-Pharr method is based on elastic solutions by Sneddon, who derived general relationships among the load, displacement and contact area for any punch that can be described as a solid revolution of a smooth function. The Oliver-Pharr method builds on solution by Doerner and Nix who suggested that unloading stiffness could be calculated from a linear fit of the unloading curve [12].

Fig. 3. is a typical load displacement hysteresis curve obtained from an elastic/plastic material and Figure 4 shows the schematic representation of the indent under load and in the unloaded conditions [12].

Fig. 3. Typical indentation Curve

Fig. 4. Schematic representation of indenter sample contact

Oliver-Pharr found that unloading data is usually not linear but better described with a power law. P=A (h-hr) m The modulus from unloading is then calculated using the following equations:

EEE

AEdHdPS

ir

r

)1()1(

2

22 υυ

βπ

−+

−=

==

Where S= Measured stiffness of rhe upper portion of

the unloading curve A= Projected contact area. E(r) = Reduced modulus β = Correction factor which is 1.034 for

Berkovich indenter E and υ = Young’s modulus and Poisson’s

ratio for specimen Ei and iυ = Young’s modulus and Poisson’s

ratio for indenter One major advantage to the Oliver-Pharr method is that a direct measurement of the model is not necessary [17]. Berkovich indenters use an area function based on the geometry of the tip, compensating for elastic load during the test. Use of this area function provides a method for gaining real-time nano-hardness values from the load-displacement graph. An area function A(h) typically describes the projected area of an indent as a 2nd order polynomial function of the indenter depth h. The area function of an ideal Berkovich indenter is the following: A(hc)= 24.5 2

ch C(0)= 24.5 C(1) – C(5) = Fitted Parameters Other terms following the first one describe deviations in geometry due to blunting at the tip.

S

PHHcmax

max ∈−=

Where Pmax = Peak indentation load ∈= 0.75 for Berkovich tip S = Unloading stiffness Hmax = maximum depth of indentation

Study of Formation of Diamond-Like Carbon Coatings and Theoretical Calculations 277

The hardness from unloading is calculated using the following equation

APH max=

Pmax = Peak indentation load A=Projected area of hardness impression

This hardness definition may be different from conventional hardness definition. The observed hardness impression may be less than that at peak load if a portion of the area did not plastically deform [13]. Taking iυ =0.07 Ei=1140 Gpa And using the results from previous works [14] and taking E = 759 Gpa υ = 0.17 E(r) comes out to be 464.6 Gpa And for hc = 0.65 μm For β = 1.034 for Berkovich indenter Unloading Stiffness comes out to be 1.744 MN/m (D) Limitations

Since the analysis is based on elastic constant solution, there is some concern on how well this method works in elastic/plastic situations. Problems associated with the “pile-up” or “skin-in” of the material n the edges of the indent during the indentation process remains a problem that is still under discussion. The Oliver-Pharr method works well for hard ceramics, when sink in predominates but fails with the soft metals which display extensive pile-up. It has been shown that pile up is significant for materials that do not work harden and for which h (f)/h (max) is close to one. When pile-up is significant, the Oliver-Pharr approach underestimates the true contact area by as much as 60% [13].

References [1] R. Bermen, in Physical properties of

Diamond, ed, R. Bermen, Clarendron Press, Oxford, pp. 371, 1965.

[2] H.P. Bovenkerk, F.P. Bundy, H.T Hall, H.M. Strong, and R.H. Wentorf, Nature 184, (1959) 1094

[3] J. Wilks and E. Wilks, Properties and application of diamond; Butterworth-Heinamann, Oxford; 1991.

[4] B.V. Derjaguin & D.V Fedoseev, Sci. Am. 233, (1975) 102

[5] J.C Angus, H.A. Will and W.S. Stanko, J. Appl. Phys. 39, (1968) 2915.

[6] J.E. Butler and D.G. Goodwin in Prperties and growth and Application of Diamond, eds. M.H. Nazare and A. J. Neves, INSPEC, London, PP. 262, 2001.

[7] L.M. Hansen, W.A. Carrington, J.E. Butler and K.A. Snail, Mater. Lett. 7 (1988)

[8] D.E. Rosner, Ann. Rev. Mater. Sci. 2, (1972), 573

[9] http://www.hysitron.com/old/PDF/ Tip%20Selection%20Guide.pdf

[10] Y.T. Cheng, C.M. Cheng, Scaling, Dimensional analyssis and indentation measurement, Mater. Sci. Eng. R, 44(2004) 91

[11] Fischer-Cripps, A.C. Nanoindentation (Springer: NewYork), 2004

[12] W.C. Oliver, G.M. Pharr J. Mater. Res. 7 (1992) 1564

[13]http://www.mse.gatech.edu/Research/Equipment_Facilities/Nanoindentation/nanoindentation.html

[14] Sungwoo Choo et al, Micromech. Microeng., 15, pp. 728-33 (2005).

A Study on Acoustic-Diffusive Wave Propagation Phenomenon in Semiconductor Materials in Contact with Fluid

Indu Sharma, J. N. Sharma and Subash Chand

Department of Applied Sciences, National Institute of Technology, Hamirpur 177 005 India Email: [email protected], [email protected]

Abstract

The paper is aimed at to investigate the propagation of thermoelasto-diffusive (ETP) surface waves in homogenous isotropic, thermally conducting, semiconductor materials in contact with fluid half space or layer of thickness (d). The concept of relaxation of heat and charge carrier fields is also taken. Secular equations, in isolated mathematical conditions and compact form, for the thermoelastic diffusive surface waves in semiconducting material half space or a layer with thickness d are derived. Numerical solution of various secular equations and other relevant relations is carried out for Germanium (Ge) semiconductor material with the help of functional iteration numerical technique. The dispersion curves, attenuation coefficient of the waves are computed and presented graphically, in order to illustrate and compare the analytical results. 1. Introduction Surface acoustic waves (SAW) are one of a broad class of acoustic techniques that have been applied to the study of physical changes at the solid-liquid interfaces. The generation of acoustic or elastic wave accompany the transient, short duration thermal heating of a material is rapidly becoming a powerful tool for characterization of a materials impinging of their microstructure. The vast majority of problems connected with coupled field theory in continua concern such media as deformable dielectric, magnets, isolators or conductors. But there are not so many studies on deformable semiconductors. Most of these are based on classical irreversible thermodynamics. Maruszewski [1-5] presented theoretical considerations of the simultaneous interactions of elastic, thermal and diffusive of charge carrier fields in semiconductors. Sharma and Thakur [6] studied the plane harmonic elastodiffusive surface wave in semiconductor materials. Sharma, et al. [7] have also studied about the propagation characteristics of thermo-elastodiffusive surface acoustic waves in semiconductor materials. Sharma and Thakur [8] studied the propagation characteristics of elasto-thermodiffusive waves in semiconductor materials half-space. The present article deals with the so-called second sound effect occurring during diffusion of charge carrier and with ones concerning conduction of heat in extrinsic semiconductors with relaxation. Here we consider the problem of leaky and non leaky Rayleigh type surface waves and present a systematic analysis of surface wave propagation in thermoelastic p-type semiconductor materials in contact with fluid half space or a layer of finite thickness d based on the governing equations derived by Maruszewski [4] and non-dimensionalzed by Sharma and Thakur [8]. After deriving the secular equations for surface acoustic

wave (SAW) generation in semiconductor materials in contact with fluid the various characteristics of elasto-thermodiffusive (ETP) waves have been investigated in the light of relaxation and life time of charge carrier fields. These waves are coupled with each other and get modified due to thermal variations, thermal relaxation time and life/relaxation time of charge carrier fields. The analytical results so obtained have been verified numerically and are illustrated graphically in case of Germanium (Ge) semiconductor materials half space (or a layer). 2. Problem and basic equations We consider an extrinsic, homogeneous, isotropic, thermally conducting, elastic (p-type) semiconductor materials, initially under undeformed state at uniform temperature 0T . The surface of the semiconductor is in contact with fluid half space or a layer of thickness d. We take the origin of coordinate system oxyz on the top plane interface surface and z-axis pointing normally into the half-space, which is thus represented by 0≥z . We take x-z plane as the plane of incidence and assume that all the particles on a line parallel to y-axis are equally displaced so that all the field quantities are independent of y-coordinate. The x-axis is taken along the direction of acoustic wave propagation in the semiconductor half-space and the densities of the charge carriers at doping level are assumed to be of such values that the life time +

pt and the diffusion coefficients pD are independent of them. Further the disturbance is assumed to be confined to the neighbourhood of the free surface and hence vanishes as .∞→z In linear theory of thermoelasticity in semiconductors, the non-dimensional governing field equations for temperature ),,,( tzxT displacement vector

),,0,(),,( wutzxu = holes diffusion fields

A Study on Acoustic-Diffusive Wave Propagation Phenomenon in Semiconductor Materials 279

),,( tzxP respectively; in the absence of body forces and heat sources; are given by

)1.1(0

.)1(..

222

=∇−

∇−−∇∇−+∇→→→

T

Puuu pλδδ

,

( ) )2.1(0

)()( 0000

22

=+∇∈−

⎟⎟⎠

⎞⎜⎜⎝

⎛+++−∇++−∇ +

utu

PtaPtPaPTtTT

QT

p

ppQnnpqQ

&&ρ&ρ

&&&&&& ααε

)3.1(0

11

2

02

=⋅∇∈∈−∇+∈−

⎥⎥⎦

⎤

⎢⎢⎣

⎡−⎟

⎟⎠

⎞⎜⎜⎝

⎛−

∈−−+∇

++

uTT

PtPttD

PtD

kP

Tpqp

p

p

p

pppp

pp

&ρ&

&&&

ε

χα

Also the governing equations for fluid medium are given by

( )0

341

12

2 =⎟⎠⎞

⎜⎝⎛

∂∂

++−∇ φ

νεδφ &&

tLLL

L (2.1)

012 =−∇ LL

L ψν

ψ & (2.2)

LLLL

LT φρβδρε 2

2

∇= (2.3)

where we have used the quantities

1cxx

∗

=′ω ,

1czz

∗

=′ω , ,' tt ∗=ω

0

'

TTT =

0

'

pPP = , u

Tc

u T0

1'

λωρ ∗

= , wT

cw T

0

1

λωρ ∗

=′

∗= ωτ QQ t , ∗= ωpp tt ', ∗= ++ ωpp tt

'

21

222

c

c=δ ,

)2(0

2

μλρλ

+=∈

e

T

T CT ,

KC e )2( μλ

ω+

=∗

ρμλ 22

1+

=c ,ρμ

=22c ,

eCK

ρχ =

0

0

Tp

T

p

p λλ

λ = ,LvL

L CTλρ

βε ∗

∗

= 0

2

,p

p

p DpKTa

0

02

ρ=∈ ,

0

0

TKpm pq

pq =ε ,0

010 TC

paa

e

pp = ,

e

pp

Cp 0

0α

α =

,2

2

2

22

zx ∂∂

+∂∂

=∇ , ,1 Q

pQp

aaa = ,2 p

pQp

aaa =

,0ppP −= ,)23( TT αμλλ += 01 TTT −=

21

22

1

22 ,,

cc

cc

L

LL

L

LL

LL

ρωμ

νρλ

δ∗

===

** '

0

, 3 , LL L

TT

Tββ β λ αβ

∗= = = (2.4)

Here μλ, are Lame parameters ρ; is the density of

the semiconductor; pλ are the elastodiffusive constants of holes, Tα is the coefficient of linear thermal expansion of the material; K is the thermal

conductivity, pα are thermo diffusive constants of

holes; areaaa pQQp ,, the flux like constant; ,pD are the diffusion coefficients of holes. The quantities

,, qppq mm are the Peltier-Seebeck-Dufour-Soret like

constants; ,, pQ tt are the relaxation times of heat and

hole fields, eC is the specific heat, ,+pt denotes the life

times of the charge carriers’ fields; p are the non-

equilibrium, Lc is the velocity of sound in the liquid,

Lλ is the bulk modulus, Lρ and Lμ are

respectively the density and viscosity of the liquid; *α is the coefficient of volume thermal expansion and LT is the temperature deviation of liquid medium from ambient temperature 0T . 3. Boundary conditions The boundary conditions to be satisfied at the solid-liquid interface are given below: (i) Mechanical conditions:

( ) ( ) LLLzxzxLzzzz wwuu ==== ,,, σσσσ (3.1) (ii) Thermal conditions:

( ) 01,, =+−++ PsaTTKNmKT ppL

Sz

pqz ρ (3.2)

(iii) Holes charge carrier diffusion condition: ( ) 012,, =⎟

⎠⎞

⎜⎝⎛

∂∂

++−++ Pt

tsTTKaNDTm ppL

Spz

pz

qp ρρ (3.3)

where pS sK , are respectively the surface heat conduction coefficient and surface recombination velocity, LT is the temperature deviation of liquid medium, Tε is thermo elastic coupling parameter and χ is the thermal diffusivity. 4. Solution of the problem In order to solve the problem we introduce the scalar point potential function φ and vector point potential function )0,,0( ψψ −=

ρ through the

relation

xzw

zx ∂∂

−∂∂

=∂∂

+∂∂

=ψφψφ ,u (4)

In the case of fluid medium, we have

xzw

zxu LL

LLL

L ∂∂

−∂∂

=∂∂

+∂∂

=ψφψφ , (5)

Where Lφ and Lψ are velocity potentials. We consider the case of time harmonic waves ( )

( ) ( ) ( ) ( )( ) ( ) ctxikzzPzTz

PT

LL

LL

−

=

exp,,,,,

,,,,,

ψφψφ

ψφψφ(6)


where kc /ω= , is the phase velocity, k and ω are wave number and angular frequency respectively of the waves. Substitution of solutions (2.4) - (6) in equations in basics leads to a coupled system of three ordinary differential and hence algebraic equations in terms of ( )LLPT ψφφ ,,,, . The requirement of the existence of non-trivial solution of this system provides us a cubic polynomial characteristic equation in 2m , which gives us the characteristic roots .3,2,1, =± imi In general, the characteristic

roots )3,2,1( =imi are complex and as we are considering surface waves only, so without loss of generality we assume that ( ) 0≥βie mR and choose

only that form of im , which satisfies the radiation condition. Hence the solution is a superposition of the plane waves attenuating with depth. Here 3,2,1,2 =iai are the roots of the complex cubic equation

0246 =−+− CaBAaa (7) where BA, and C are given in appendix 5. Derivation of secular equations We consider the situation in which semiconductor half-space is in contact or loaded with viscous fluid. Upon applying the required interface boundary conditions (3) at the solid-liquid interface ( 0=z ) and subsequently requiring non-trivial solution, after lengthy but straight forward algebraic reductions and manipulations, the secular dispersion relation for Rayleigh waves is obtained.

( )( ) ( ) ( )

( ) ( ) ( )

( )( ) ( )( ) ( ) )8(33221143322113

332211233221110

3211

'1

42

32132

3211

'1

22

32110

RmRmRmBNmNmNmBMmMmMmBLmLmLmaBB

RRRTshaApNNNAp

MMMTshaApLLLaAA

nLT

LT

+−++−++−++−+=

+−+++−+

+−+++−+

γε

γ

Where 3,2,123321 =−= iQPQPL

3,2,132231 =−= imQmQM 3,2,123321 =−= imPmPR

( ) ( ) 3,2,1321

'1

231

'1

1 =−+−= iPPTshQQTshN nLTLT

γε

γ

32323232 ,,,,,,, RRNNMMLL can be

obtained from1111 ,,, RNML in equation (8) by

replacing the subscripts permutation (2, 3) with (1, 3), (1, 2), respectively.

iii PBA ,, and iQ are

defined in the appendix. The secular equation (8) governs the motion of non-leaky Rayleigh (Stoneley) waves in the instant analysis and contains complete information regarding phase

velocity, attenuation coefficient and other characteristics of these waves 6. Numerical results and discussion In this section, we present some results in order to illustrate the analytical developments carried out in the previous sections. To understand the interactions of various fields in thermo elastic semiconductors, the non-dimensional phase velocity and attenuation of ETP-surface wave propagation modes under different situation have been obtained and computed numerically for Germanium (Ge) semiconductor material and their profiles are plotted on linear scale against non-dimensional wave number (R). The physical data of Germanium semiconductor material is given below.

fluidviscousforL 0.1=μ

161623

111116

200

1225

33211211

10004.0108.5/103.1

3106010004.0

10105.0,10

103.51053.01048.0

−−−−−

−−−−−−

−−−−+

−−−

×−=×=×=

==×−=

=×=<=

×=×=×=

vkmKsm

KjKgCKmKvkm

psmDst

kgmNmNm

pqT

pe

qp

pp

αα

ω

ρμλ

In this paper, we present the analysis of thermal relaxation and life times as well as relaxations of hole field on the phase velocity and attenuations. The corresponding profiles are plotted on linear scales against non-dimensional wave number (R) and thermal relaxation time in Figures. 1to4. Here the considered non-dimensional values of lifetimes of charge carrier

,796.0=+nt respectively, corresponding to their

dimension values ,10 12 stn−+ = Fig. (1) shows the

variations of phase velocity with wave number at different life times for leaky Rayleigh waves (LRW) and non-leaky Rayleigh waves (NLRW). We find that considerable changes of phase velocity that occur are within relatively narrow interval of wavelength, i.e. it is found to be decrease in the wave number interval

10 ≤≤ R and after 1≥R , the phase velocity profiles propagate with uniform phase velocity for all values of life times. Fig. 2 show the variations of attenuation with wave number for single values of life times in the case of viscous fluid in contact with semiconductor half space. It is observed that the attenuation increases monotonically in a linear fashion with increasing wave number. Attenuation is caused by both scattering and absorption. Absorption is generally caused by media. This is due to the energy loss by both viscosity and heat conduction. Whenever there is a relative motion between the particles in a media, such as in wave propagation, energy loss occurs. This is due to the stress from viscous forces between particles of the medium. The energy lost is converted into heat. Because of this, the intensity of a wave decreases more rapidly than the inverse square of the distance. A special type of attenuation occurs when a wave travels over a boundary, such as a fluid is in contact with solid surface. In such a situation, the fluid in immediate contact with the surface must be at rest. Subsequent

A Study on Acoustic-Diffusive Wave Propagation Phenomenon in Semiconductor Materials 281

layers of fluid will have a velocity that increase as the distance from the solid surface increases. The velocity gradient causes an internal stress associated with viscosity that leads to loss of momentum. This loss of momentum leads to a decreases in amplitude of the wave close to the surface. Fig.3.shows the variation of phase velocity with wave number at relaxations time 6.0,4.0,2.0,0.0=Qt and 8.0 for NLRW and LRW waves when semiconductor half space is in contact with viscous fluid. For NLRW the phase velocity shows the increases within the interval

2.00 ≤≤ Qt and after 2.0≥Qt , the phase velocity profiles propagate with uniform phase velocity for all values of relaxations time. In the case of LRW waves the phase velocity increases monotonically with thermal relaxations time. Fig. 4 shows the variation of attenuation with wave number for LRW and NLRW at different values of thermal relaxations times.

0.75

0.77

0.79

0.81

0.83

0.85

0.87

0.89

0 0.5 1 1.5 2 2.5 3Wave number (R)

Phas

e ve

loci

ty (V

)

NLRW

LRW

Fig. 1. Variations of phase velocity profiles with wave number at 0.796 life time of holes charge carrier for LRW and NLRW waves in the case of viscous fluid in contact with semiconductor half-space.

0

0.01

0.02

0.03

0.04

0.05

0.06

0 1 2 3 4Wave number (R)

Atte

nuat

ion

(Q)

NLRW

LRW

Fig. 2. Variations of attenuation profiles with wave number at 0.796 life time of holes charge carrier for LRW and NLRW waves in the case of viscous fluid in contact with semiconductor half-space.

0.78

0.785

0.79

0.795

0.8

0.805

0.81

0.815

0.82

0 0.2 0.4 0.6 0.8

Thermal relaxiation time (tQ)

Phas

e ve

loci

ty (V

)

NLRW

LRW

Fig. 3. Variations of phase velocity profiles with thermal relaxation for LRW and NLRW waves in the case of viscous fluid in contact with semiconductor half-space.

0

0.00050.001

0.0015

0.0020.0025

0.003

0.0035

0.004

0.0045

0.005

0 0.1 0.2 0.3 0.4

Thermal relaxiation time (tQ)

Atte

nuat

ion

(Q)

NLRW

LRW

Fig. 4. Variations of attenuation profiles with thermal relaxation time for LRW and NLRW waves in the case of viscous fluid in contact with semiconductor half space.

For NLRW the attenuation profiles increases linearly within the interval 2.00 ≤≤ Qt and after

2.0≥Qt the attenuation shows minima at 3.0≥Qt and after 3.0≥Qt it shows the increasing behavior. While in the case LRW the attenuation first increases monotonically with thermal relaxations time. Attenuation can also occur by a process called relaxation. One of the basic assumptions prior to this discussion on attenuation was that when the pressure or density of a fluid or media dependent only on the instaneous values of density and temperature and not on the rate of change in these variables. However whenever a change occur, equilibrium upset and media adjusts until a new local equilibrium is achieved. This does not occur instaneousaly, and pressure and density will vary in the media. The time to take to achieve this new equilibrium is called relaxation time; As a consequence the velocity of the wave will increase from an initial value to that of a maximum as frequency increases. Again the losses associated with relaxation


are due to mechanical energy being transformed into heat. 7. Conclusions It is found that the thin layer coating of fluid results in smoothening of profiles of various wave characteristics by chopping of their fluctuating behaviour. Moreover, the presence of velocity gradient causes an internal pressure associated with viscosity which leads to loss of momentum. This loss of momentum results in decrease of amplitude of the wave close to the surface. The variation in attenuation profiles are attributed to the fact that the viscosity of liquid affects the attenuation by causing a leakage of energy via shear waves into the fluid and also the longitudinal wave attenuation in fluid dominates the longitudinal component of the wave resulting in energy dissipation of the propagating evanescent wave in the elastic semiconductor. The life time and relaxation time parameters of holes charge carrier fields are noticed to have significant effects on various wave characteristics in addition to holes concentration. The shifting of maxima and minima towards origin with increasing values of life times and changes in the sign of velocity and attenuation profiles clearly define the transition from a damped wave to a growing wave or wave amplification as evidenced from the computer simulated results. References [1] B. Maruszewski, Arch. Mech. 38 (1986) 71. [2] B. Maruszewski, Arch. Mech. 38 (1986) 83. [3] B. Maruszewski, Int. J.Engng. Sci. 25 (1987)

145 [4] B. Maruszewski, J. Acoust Soc. Am. 85 (1989)

1967. [5] B. Maruszewski, Electro-magneto-mechanical

Interactions in Deformable Solids and Structures, edited by Y. Yamamoto and K. Miya (Northland, Amsterdam,(1987).

[6] J. N. Sharma, N. Thakur, Journal of Materials and Structures. 1 (2006) 813-835.

[7] J. N. Sharma, N. Thakur, V Walia, Asian Journal of Chemistry. 18 (2006) 3329-3334.

[8] J. N. Sharma, N. Thakur, Journal of Thermal Stresses. 30 (2007) 357-380.

[9] J. R. Wu, Z.M. Zhu, J. Acoust. Soc. Am. 91(1992) 861-867.

[10] F. Graff, Wave motion in elastic solids, Dover, New York (1991).

Appendix The functions used in equations (7) are given

( ) ( ) qppq

Tpq

pTppQ

Tpqp

TQ

ppqqp i

Aεε

εελεεωλτετεεττεε−

+−++−+++−=

−

11)1(1 1'*

( ) ( )( )qppq

Tppq

ppqp

TQ

pQ i

Bεε

ετεεωτεετττ−

++−−+++=

−

1111 '1'*

qppqppp

Q iC

εετεωττ

−

−=

−

1

'1*

+−− −++= pppppQ

p taait /)( 2000

10

' ωαωατ ,

⎥⎥⎦

⎤

⎢⎢⎣

⎡+

⎟⎟⎠

⎞⎜⎜⎝

⎛−

∈−+= ++

−∗

pp

ppppp

pe

p tttD

itDC

K2

01 11ωχ

αω

ρτ

,)1( 2*22* ck pp τα −= 1−+= ωτ it QQ ,

,)1(21

221 qp

p cik

εεω

β−

−= ,)1( 222 ck −=α

The coefficients used in equation (8) are defined as 22 β+= kp ,

21

2

ρδγρω

ω LL =

( ) ( ) 22'

21'

1 ,tanh,tanhρδ

ρωνγγ LLiadTdT ===

,12

'2

2

21

'2

'1

22

0 γω

γγTpkTTkpA L+⎟⎟

⎠

⎞⎜⎜⎝

⎛−=

( ) ( )( )

( ) ( ) ( )( )( )[ ]222

2211

21

2

21

'2

'1

2

222

22

2

'222

1

2222

4

kkapkpTTk

kkTpakkA

L +−+−+−+

−−+−=

γγωγγγγ

γβγβ

( )⎭⎬⎫

⎩⎨⎧

−+−= 2

2

'222

222

2 2)( pkT

kkAγ

βγβ

⎭⎬⎫

⎩⎨⎧

+⎟⎠⎞

⎜⎝⎛ −+−=

βγβaTpkpA

2

'22

3 22

( )⎭⎬⎫

⎩⎨⎧

−+−= 2

2

'222

222

4 2)( pkTkkAγ

βγβ

⎟⎟⎠

⎞⎜⎜⎝

⎛−−−⎟⎟

⎠

⎞⎜⎜⎝

⎛−=

2

'2

222

21

'2

'1

22

0214γβ

βωγγ

βTkkTTkkB L

( ) ( )

( ) ( )( )

( ) ⎥⎥⎥⎥

⎦

⎤

⎢⎢⎢⎢

⎣

⎡

+

⎟⎟⎠

⎞⎜⎜⎝

⎛

−+

−+−

+

−+−=

222

221

12

21

22

21

'2

'1

22

1

'12

1

22

224

224

k

ka

kkkTT

kTakB

L

γ

γ

ωγγ

γγβ

βγ

β

1

'12

2 )1(4γ

βTSh

akB LT−=

( )⎪⎭

⎪⎬⎫

⎪⎩

⎪⎨⎧

+⎟⎟⎠

⎞⎜⎜⎝

⎛−−= ap

TkkB 42

2

'2222

3 γβ

β

1

'12

4 )1(4γε

βTSh

akB nLT−=

( )( ) ( )[ ]iipnq

iiLTi SmhWmShP −+−−= ε1

( )( ) ( )[ ]iipqniiqn

pLTi SmhWmShQ −+−−= '1 εεε

Plane Harmonic Waves in Generalized Thermoelastic Materials with Voids

D. Kaur1 and J. N. Sharma2

1Department of Applied Mathematics, National Institute of Technology, Jalandhar 144001 India 2Department of Applied Sciences, National Institute of Technology, Hamirpur 177 005 India

Email: [email protected], [email protected] Abstract

The aim of the present paper is to give a detail account of the plane harmonic generalized thermoelastic waves in heat conducting solid with voids. The general characteristic equation being quartic suggests that there are four longitudinal waves namely, quasi-elastic (QL-mode), quasi-thermal (T-mode), volume fraction (φ -mode) and micro-thermal (MT-mode), in addition to transverse waves which can propagate in such solids. The transverse waves get decoupled and remain unaffected due to temperature variation and porosity effects. These waves travel without attenuation and dispersion. The other generalized thermoelastic waves are significantly influenced by the interacting fields and hence suffer both attenuation and dispersion. At quite low frequency, volume fraction and thermal waves do not exist but at high-frequency limits these waves do propagate and thermal wave have finite, though quite, large velocity. Also in the low-frequency regimes, the disturbance is mainly dominant by mechanical process of transportation of energy and at high-frequency regimes; it is significantly dominated by close to diffusive process (heat conduction or volume fraction). The general complex characteristic equation can be solved by using DesCartes algorithm along with irreducible case of Cardano’s method with the help of DeMoivre’s theorem in order to obtain phase speeds and attenuation coefficients. Finally, the numerical solution is carried out and the obtained phase velocities of thermoelastic waves are presented graphically. 1. Introduction

Nunziato and Cowin [1] have presented a nonlinear theory for the behaviour of elastic materials with voids which was, for many researchers, the basis for the study of a porous elastic solids with lacunae finally dispersed in the matrix. Cowin and Nunziato [2] presented a linear theory of elastic materials with voids. This theory is of practical utility in investigating various types of geological, biological and synthetic porous materials for which classical theory is inadequate. Puri and Cowin [3] investigated the behaviour of plane harmonic waves in linear elastic materials with voids. Chandrasekharaiah [4] studied the effect of voids on the propagation of surface waves in elastic half space with voids. Chandrasekharaiah [5] investigated the propagation of Rayleigh-Lamb waves in a homogeneous isotropic elastic plate containing a distribution of vacuous pores (voids). Giovine [6] proposed appropriate constitutive relations to specialize the basic balance equations of the linear theory. Jaric’ and Rankovic’ [7] have studied the nonlinear theory of thermoelastic materials with voids. Iesan [8] developed the linear theory of thermoelastic materials with voids. Kumar and Rani [9] studied the problem of thermal loads in thermoelastic half spaces with voids. Lord and Shulman [10], and Green and Lindsay [11] extended the conventional coupled thermoelasticity to generalized thermoelasticity by including thermal relaxation time as a time lag needed for the establishment of steady state in order to remove the drawback of infinite velocity of propagation of a part of disturbance. Chandrasekharaiah [12] and Hetnarski and

Ignaczak [13] brought out reviews and comparisons of various theories of generalized thermoelasticity.

Keeping in view the recent interest in a theory of thermo-mechanics which allows for “second sound” effect, the present paper is aimed at to give a brief account of the effects of thermal relaxation time and presence of voids on the propagation characteristics of plane harmonic waves in heat conducting elastic materials. In general, four longitudinal waves are found possible to propagate in the considered materials, in addition to decoupled transverse waves. The low and high-frequency behaviour of harmonic waves is also investigated.

2. Problem and Basic Equations

We consider a homogeneous isotropic thermoelastic solid with voids at uniform temperature

0T in the undisturbed state. The basic governing field equations of linear generalized thermoelasticity with voids, in the absence of body forces, equilibrated forces and heat sources, are given by

( )⋅⋅

=∇−∇+∇∇++∇ uTbuu ρρρ ρβφμλμ .2 (1) ⋅⋅

=+⋅∇−−−∇ φρχφξφξφα mTubρ&

212 (2)

⎟⎠⎞

⎜⎝⎛ ++⎟

⎠⎞

⎜⎝⎛ +∇=

⎟⎠⎞

⎜⎝⎛ +−∇

⋅⋅⋅⋅

⋅⋅

φφβ

ρ

0000

02

. tmTutuT

TtTCTK e

&ρ&ρ

& (3)


where ),,( 321 uuuu =ρ

is the displacement vector,

),,,( 321 txxxT is the temperature change, μλ, are Lame′ parameters; K is thermal conductivity;

eC and ρ are respectively the density and specific heat at constant strain; φ is change in volume fraction,

tt ααμλβ ,)23( += is linear thermal expansion;

,,,, 21 mb ξξα and χ are material constants due to the presence of voids. We define the quantities

φχωφω

βρω

ωω

21

000

0

1i

1

c ,* ,

*u ,* ,

*

2∗

=′=′=′

=′=′=′

ttTTT

Tuc

ttc

xx ii

i

( )

21

232

121

02

0

21

10

2

cc

,

,2 2

==

=+

=∗

δαρχβ

χωβμλρβ

ε

cTb

a

Tbc

aC

T

eT

1

24

15

04

211

321

222

,

,c,cc

3

2

ξωξξ

χω

αχ

αωξδ

∗

∗

∗

==

===

Kmca

mTaa (4)

where

ρχα

ρμ

ρμλ

==+

= 23

22

21 c,c,2c , ( ) ,2*

KCe μλω +

= (5)

Upon introducing quantities (4) in equations (1) to (3) and suppressing the primes for convenience, we obtain

( ) ⋅⋅=∇−∇+∇∇−+∇ uTauuρρρ

φδδ 1222 .1 (6)

( ) ( )2

1432

2

δφφξφφ⋅⋅

=++−⋅∇−∇ Taaua &ρ (7)

0. 05002 =⎟⎟

⎠

⎞⎜⎜⎝

⎛+−⎟⎟

⎠

⎞⎜⎜⎝

⎛+∇−⎟⎟

⎠

⎞⎜⎜⎝

⎛+−∇

⋅⋅⋅⋅⋅⋅φφε tautuTtTT T

&ρ&ρ& (8)

A plane displacement wave of harmonic time dependence propagating in a direction defined by the propagation vector p

ρ is represented as

).(exp ctprikdAu −=→→ρρ

(9) where

),(1cccc =′ ),

*( 1

ωkc

kk =′ )( *ωωωω =′ are

respectively, the non-dimensional phase velocity , wave number and angular frequency with suppressed

primes and are related by .k

c ω= Here A is unknown

amplitude of the wave and a unit vector dρ

defines the direction of motion. If deformation affects the thermal state and volume fraction of the medium, a

displacement wave is also accompanied by thermal and volume fraction waves, which are scalar quantities, and may be assumed of the form

).(exp ctprikBT −=→→

(10)

).(exp ctprikC −=→→

φ (11) where B and C are unknown amplitudes. Substitution of solutions (9) to (11) in equations (6) to (8) leads to ( ) ( ) ( ) 0.1 1

11222 =−+−+− −− CpaikBpikAdppAdcρρρρρρ

δδ (12)

( ) 01.21

2

203

42

21 =⎟

⎟⎠

⎞⎜⎜⎝

⎛−++− −− Cc

ka

BakAdpaikδ

ξρρ (13)

( ) 01).( 205

20

20 =+−− CcaBcAdpcik T ττετ

ρρ (14)

where 100

−+= ωτ it . The equations (12) to (14) constitute a coupled system of homogeneous algebraic equations in the unknowns A, B, and C. This system will have a non-trivial solution if and only if the determinant of the coefficients of A, B, and C vanishes. This requirement provides us ( ) ( ) 0.22 =+− dppSdc

ρρρρδ (15)

where

Tcc

S T −⎟⎟⎠

⎞⎜⎜⎝

⎛−

−−= 20

202

11

ττε

δ

( )

⎥⎥⎦

⎤

⎢⎢⎣

⎡

⎟⎟⎠

⎞⎜⎜⎝

⎛⎟⎟⎠

⎞⎜⎜⎝

⎛

−−−−

⎥⎦

⎤⎢⎣

⎡−

−

=2

20

20

023

21

2

2

20

202

3

111

.1

ccca

dppcc

caT

T

Tb

ττε

ξωδ

ωε

ττε

εε

φ

φ

ρρρ

4

2

aa

b =ε ,3

54

aaa

=φε , ξωξ i−=10 ,T

b

aaa

εεεφ=

3

41 (16)

The equation (15) is a vector equation consisting of a set of three equations which must be solved for unknown vector d

ρ. If we take ),,( 321 dddd =

ρ

and ),,( 321 pppp =ρ

, the equation (15) is equivalent to the following system of equations 0)( =jij da (17)

where the matrix )( ija is given by

⎥⎥⎥

⎦

⎤

⎢⎢⎢

⎣

⎡

+−

+−

+−

=23

223231

3222

2221

312121

22

)(

pScppSppS

ppSpScppS

ppSppSpSc

aij

δ

δ

δ

For a nontrivial solution of equation (17) the determinant of matrix )( ija must be zero. This gives us a characteristic equation which can be solved for 2c . In the absence of voids )0( φεε ==b and

thermo-mechanical coupling )0( =Tε , the matrix in

Plane Harmonic Waves in Generalized Thermoelastic Materials with Voids 285

(17) is Hermitian (symmetric) and the three eigenvalues are real, corresponding to three elastic waves propagating in any fixed direction. However, in the presence of thermal coupling )0( ≠Tε and absence of

voids )0( φεε ==b the matrix )( ija is neither symmetric nor Hermitian. The characteristic equation in this case, after rationalization and simplification, is of fourth degree in 2c having in general, complex roots. Because the thermal coupling is usually small, three of the roots will have small imaginary parts and have real parts approximately equal to the corresponding eigenvalues of )( ija with 0=Tε . These three roots give three attenuated elastic waves and the fourth root corresponds to the thermal wave which has a large attenuation coefficient. Again in the presence of voids and thermoelastic coupling

)0,0,0( ≠≠≠ φεεε bT , the matrix )( ija is not Hermitian (symmetric). The characteristic equation 0=ija , after rationalization and

simplification, is of sixth degree in 2c having in general, complex roots. Here four of the six roots correspond to attenuated elastic and thermal waves and the fifth and sixth will be associated to the volume fraction wave ( −φ mode) and micro-thermal wave (MT-mode), respectively. These waves also have large attenuation coefficients.

2. 1. Transverse waves For transverse motion pd

ρρ±≠ , 0. =dp

ρρ and hence

the secular equation (15) implies that δ±=c (18) This equation (18) defines transverse wave which do not interact with temperature and volume fraction fields and hence travels with nondimensional velocity δ without dispersion, attenuation and dissipation.

2. 2.Longitudinal waves For longitudinal motion ,pd

ρρ±= ,0. ≠dp

ρρ the secular

equation (15) in this case leads to

01

1 20

202 =−⎟⎟

⎠

⎞⎜⎜⎝

⎛

−−− T

cc

c T

ττε (19)

This complicated equation (19) shows that the phase velocity depends on ω or k , meaning that thermoelastic and volume fraction waves are dispersive in character. Because the solution of equation (19) for c is generally complex valued, so generalized thermoelastic and volume fraction waves also suffer attenuation. 3. Low and high frequency behaviour In low and high-frequency regimes, characterized by

1<<ω and 1>>ω , the wave like modes are determined by isentropic (constant entropy) and isothermal material parameters, respectively. We now inspect the low and high- frequency behaviour of waves.

When 0→ω , the secular equation (19) provides us

( )( ) ,1

1,0,02

φ

φ

εεεεε

ε+

+−+±=

T

TbTc

(20) The first two value of c corresponds to (T-mode) and volume fraction (φ -modes) wave and the third value referred to modified longitudinal elastic (QL) wave. Clearly there is no damping in either of these modes. The close inspection of last value of c in (20) reveals that the phase velocity of QL-wave is increased by a small amount due to the thermomechanical coupling but decreased because of porosity effects at adiabatic (isentropic) conditions. In the absence of voids

bm == 0 which implies that bεεφ == 0 and consequently, the above equation (20) provides us

Tcc ε+±== 1,0,0 (21) For ∞→ω , the secular equation can be written from (19) in a straight forward manner by replacing

0τ with 0t and we obtain

( )[ ] 01111 40

202

1

2

=+++−⎟⎟⎠

⎞⎜⎜⎝

⎛− ctctc

Tεδ(22)

This reduced frequency equation (22) provides us three real values of 2c for non-zero 0t given by

( ) ( )

0

20

2000

21

2

2

4111,

t

ttttc

TTT ⎥⎦⎤

⎢⎣⎡ ++−±++

=εεε

δ (23)

These correspond to three modes of wave propagation namely, volume fraction wave (φ -mode), quasi-longitudinal elastic (QL) wave and thermal wave (T-mode), respectively. Clearly the last two waves are influenced by thermal relaxation effects and thermomechanical coupling. The volume fraction wave travels with velocity 1δ is not accompanied by thermal and elastic field and vice versa at extremely high frequencies. In case thermal and elastic fields are not coupled with each other ( )0=Tε , the equation (23) implies that

01

1,1,t

c δ= . This shows that thermal mode has

a finite, though quite large, speed. In the absence of thermal relaxation time )0( 0 =t , the above secular

equation leads to ∞= ,,1 1δc for longitudinal elastic wave, volume fraction wave and thermal modes, respectively. The third mode (T-mode) has infinite speed of propagation being diffusive in character in this case. In the general case, ∞<<ω0 the secular equation (19), after lengthy but straight forward algebraic reduction and manipulation, can be rewritten as

( ) 01 224

1=−Π

=cii

ζ (24)


where 3,2,1,2 =iiζ are the roots of the quadratic equation

0234 =+−+− PNML ζζζζ (25) Here

( ) ⎟⎟⎠

⎞⎜⎜⎝

⎛−−++++=

T

bT

aL

εεε

ξωδ

τετ φ2

023

21

00111 (26)

( ) ( )[ ]

( )2

43102

032

10

0203

21

0

21

1111

ωε

εεεετ

ωξ

δτ

ετω

εξ

δτ

φφ

φ

T

Tb

T

aaaa

aM

++⎥⎦

⎤⎢⎣

⎡−+

+++⎥⎥⎦

⎤

⎢⎢⎣

⎡ +−+=

(27)

( )[ ]

( ) ( )22

032

032

1

0203

21

00

1

111

TbT

T

aa

aN

εεωετε

ωεξ

δ

ετωξ

δττ

φφ ++⎥⎥⎦

⎤

⎢⎢⎣

⎡ +−

⎥⎦

⎤⎢⎣

⎡+++−+=

(28)

( )ξωξ

ω

εξ

δτ φ i

aP −=

⎥⎥⎦

⎤

⎢⎢⎣

⎡ +−= 1,1

0203

21

20 (29)

The secular equation (24) gives us four pairs of values of complex phase velocity c , being eighth degree equation and hence four distinct types of waves are possible to propagate in such materials for ∞<<ω0 . Thus there are, quasi-longitudinal elastic (QL-mode), quasi-thermal (T-mode), volume fraction ( −φ mode) and micro thermal ( −MT mode) waves which propagate in thermoelastic materials with voids. The equation (24) being quartic with complex coefficients can be solved by Descartes algorithm along with irreducible case of Cardano’s method with the help of DeMoivre’s theorem. Upon obtaining the complex roots 4,3,2,1,2 =iiζ of secular equation (24), we get

.4,3,2,1,1=±= jc

jj ζ

(30)

We define QiVc 111 −−− += ω (31)

so that V

RiQRk ω=+= , , V and Q are real. This

shows that V is the propagation speed and Q is the attenuation coefficient of the waves. Upon using representation (31) in equation (30), we obtain

)(Im,)(Re

1jj

jj QV ζω

ζ== , 1,2,3,4j = (32)

The results in the context of conventional coupled thermoelasticity can be obtained from the above analysis by setting 00 =t . 4. Numerical Results and Discussion

In order to illustrate the analytical results obtained in the previous sections we present some numerical simulation results for magnesium crystal like material with physical data given below [9]

2101017.2 −×= Nmλ 21010278.3 −×= Nmμ ,331074.1 −×= mKgρ , ,250 CT o=

113 deg1004.1 −−×= KgJCe

5.0,05.0,0.00 =t 111

1 1034.5 −∗ ×= sω 112 deg107.1 −−×= WmK ,

126126 deg100.2,deg1068.2 −−−− ×=×= NmmNmβ ,21510753.1 −−×= mχ ,10475.1 210

1−×= Nmξ

2105 1013849.1,10688.3 −− ×=×= NmbNa

The computed values of phase velocities of four possible modes of wave propagation namely; QL wave, T-mode, φ -mode and MT- mode are plotted in Figs. 1, 2, 3 and 4 respectively with frequency on log-linear scales. From Fig.1, it is noticed that the phase velocity of quasi-elastic (QL) wave first decreases from its value

8871.0=V at 0=ω in the region 003.00 ≤< ω , and then increases in

10003.0 ≤≤ ω to attain its maximum value at 10=ω . The effect of thermal relaxation time is quite significant at high frequencies but negligibly small at low frequencies. The phase velocity profiles of thermal wave (T-mode) are presented in Fig. 2 on log-linear scales. It is observed that the phase velocities in the range 1.00 ≤≤ ω is almost zero and then increases approximately in Gaussian manner in the ranges 0.11.0 ≤≤ ω for all considered values of thermal relaxation time, and

1001 ≤≤ ω for 05.00 =t , as well as

3001 ≤≤ ω for 5.00 =t and 0.00 =t . The Fig.3 shows the variation trends of phase velocity of volume fractional wave ( −φ mode) in the considered range of frequency on log-linear scales. The phase velocity of volume fraction field remains between its isentropic value 0.0=V at low-frequency and isothermal values 4910.0=V at high-frequency, although it is quite dispersive in the range ∞<< ω0 as can be noticed from Fig. 3. The phase velocity profiles of micro-thermal wave (MT-mode) which is due to the consequent joint effects of the interaction of volume fraction, elastic and thermal effects are plotted in Fig. 4. It is observed from Fig. 4 that this wave appears at low-frequencies ( )3003.0 ≤≤ ω only and disappears

at high-frequencies ( )300≥ω . However, it varies

normally in the range ( )303.0 ≤≤ ω for

0.00 =t but fluctuates in case of other two values of thermal relaxation time. The effect of thermal relaxation time is quite pertinent on phase velocity of these modes.

Plane Harmonic Waves in Generalized Thermoelastic Materials with Voids 287

5. Conclusions

There are four longitudinal partial waves namely; quasi-longitudinal (QL), quasi-thermal (T-mode), volume fraction ( −φ mode) and micro-thermal (MT-mode), in addition to transverse wave modes in the considered solids. The decoupled transverse waves remain unaffected due to thermal relaxation and porosity effects. The longitudinal wave modes are dispersive and attenuated in space. The thermal mode has finite, though quite large, velocity of propagation in non-classical (generalized) thermoelasticity )0( 0 ≠t but it is observed to be diffusive in character in case of classical (coupled) thermoelasticity )0( 0 =t . The thermal relaxation effects are quite significant at high frequency which supports the fact that ‘second sound’ effects are short lived. It is observed that the micro thermal (MT- mode) wave does not exist at extremely high and low frequencies for 00 ≠t and at all values

of the frequency for 00 =t . Thus this mode is a consequence of thermal relaxation effects (phonon-phonon collision) in addition to its association with volume fraction field. The presence of MT-mode is attributed to the fact that core material and pores exhibit different characteristics and effects, though small, due to thermal variations as the rate of expansion and contraction of material particles and pores due to thermal variations is at small departure.

0

0 . 2

0 . 4

0 . 6

0 . 8

1

1 . 2

0 . 0 0 0 0 1 0 . 0 0 1 0 . 1 1 0 1 0 0 0

F r e q u e n c y

Phas

e ve

loci

ty (V

1)

T A O = . 0 5T A O = 0 . 0T A O = 0 . 5

Fig. 1. Variation of phase velocity of QL – mode.

01 02 03 04 05 06 07 08 09 0

1 0 0

0 . 0 0 0 1 0 . 0 1 1 1 0 0 1 0 0 0 0

F r e q u e n c y

Phas

e ve

loci

ty (V

2)

T A O = . 0 5T A O = 0 . 0T A O = 0 . 5

Fig. 2. Variation of phase velocity of T – mode.

0

0 . 2

0 . 4

0 . 6

0 . 8

1

1 . 2

0 . 0 0 0 1 0 . 0 1 1 1 0 0 1 0 0 0 0

F r e q u e n c y

Phas

e ve

loci

ty (V

3)

T A O = . 0 5T A O = 0 . 0T A O = 0 . 5

Fig. 3. Variation of phase velocity of φ - mode

0

1 0

2 0

3 0

4 0

5 0

6 0

0 . 0 0 1 0 . 1 1 0 1 0 0 0

F r e q u e n c y

Phas

e ve

loci

ty (V

4)

T A O = . 0 5T A O = 0 . 0T A O = 0 . 5

Fig. 4. Variation of phase velocity of MT – mode. References 1. J.W. Nunziato and S.C. Cowin, Arch. Rational

Mech. Anal. 72(1979), pp.175-201. 2. S.C. Cowin and J.W. Nunziato, J. Elasticity 13

(1983), pp. 125-147. 3. P. Puri and S.C. Cowin, J. Elasticity 15 (1985),

pp.167-183. 4. D. S. Chandrasekharaiah, Acta Mech. 62

(1986), pp. 77-85. 5. D. S. Chandrasekharaiah, ASME J. Applied

Mech. 54 (1987), pp. 509-512. 6. P. Giovine, Transport in Porous Media 34

(1999), pp. 305-318. 7. J. Jaric’ and S. Rankovic’, Rev. Roum. Sci.

Techn. Mech. Appl. 24 (1979), pp. 793-805. 8. D. Iesan, Acta Mech. 60 (1986), pp. 67-89. 9. R. Kumar and L. Rani, J. Vib. Cont. 11 (2005),

pp. 499-517. 10. H. W. Lord and Y. Shulman, J. Mech. Phys.

Solids 15 (1967), pp. 299-309. 11. A. E. Green and K.A. Lindsay, J. Elasticity 2

(1972), pp.1-7. 12. D. S. Chandrasekharaiah, Appl. Mech. Rev. 39

(1986), pp. 355-376. 13. R. B. Hetnarski and J. Ignaczak, J. Thermal

Stresses 22 (1999), pp. 451-476.

Free Vibrations in a Cylindrical Panel of Heat Conducting Viscoelastic Material

P. K. Sharma, V. Walia and J.N. Sharma

Department of Applied sciences and Humanities, National Institute of Technology, Hamirpur (H.P.)-177 005, India

E-mail: [email protected], [email protected] , [email protected]

Abstract

In the present paper, we choose three displacement functions to represent three displacement components on the basis of three-dimensional thermo viscoelasticity for isotropic media. The fundamental equations are then simplified and free vibration solution for a simply supported cylindrical panel is obtained by using modified Bessel functions with complex arguments. 1. Introduction

The effect of internal friction on the propagation of plane waves in an elastic medium may also be considered owing to the fact that dissipation accompanies vibrations in solid media due to the conversion of elastic energy to heat energy Ewing et al. [8]. Several mathematical models have been used by authors [9, 10, 11] to accommodate the energy dissipation in vibrating solids. where it is observed that internal friction produces attenuation and dispersion and hence the effect of the viscoelastic nature of material medium in the process of wave propagation can not be neglected. The viscoelastic nature of a medium has special significance in wave propagation in a solid medium. The cylindrical panels are frequently used as structural components and their vibration characteristics are obviously important for practical design. Soldatos and Hadhgeorgiou [1] used an iterative approach to predict the frequencies of isotropic cylindrical shells and panels based on the governing equations of three-dimensional elasticity. Leissa [2] studied free vibrations of thick hollow cylinders by Ritz method. Jing [3 employed the perturbation method to study three-dimensional vibrations of fiber reinforced composite laminated cylindrical shells, while Fan and Ding [4] and Ye and Soldatos [5] used state space method to analyze laminated orthotropic cylindrical shells and cross-ply cylindrical panel, respectively. Thick composite cylindrical shells can be used in applications involving aerospace, offshore and submarine structures, pressure vessels, civil engineering structures, chemical pipes and even automotive suspension components. These structures can be easily exposed to a variety of temperature fields in different environments. In the present paper, we choose three displacement functions to represent three displacement components on the basis of three-dimensional thermo viscoelasticity for isotropic media. The fundamental equations are then simplified and free vibration solution for a simply supported cylindrical panel is obtained by using modified Bessel functions with complex arguments.

It can be verified that the resulting frequency equations and expression of stresses, temperature change and displacements are all in real forms. Numerical examples are presented and compared to relevant publications.

2. Formulations and solution

The governing field equations of motion and heat conduction in the absence of body forces and heat sources are as given in [7] and the constitutive relations are as

( )TtTeee kzzrrrr&

21)2( δβλλμλσ θθ +−+++= ,

( )TtTeee kzzrr&

21)2( δ+β−λ+μ+λ+λ=σ θθθθ

( )TtTeee kzzrrzz&

21)2( δ+β−μ+λ+λ+λ=σ θθ

rzrzzzrr eee μσμσμσ θθθθ === , , 2 (1)

zu

ruu

rrue zrr

rr ∂∂

=+∂∂

=∂∂

= zze ,1e ,θθ

θθ

θ

θθ

θ

θθθ

∂∂

+∂∂

=

∂∂

+∂∂

=−∂∂

+∂∂

=

zz

zrrr

urz

ue

ru

zu

ru

ruu

re

1

,e ,1rz

(2)

Where

⎟⎠⎞

⎜⎝⎛

∂∂

+=

⎟⎠⎞

⎜⎝⎛

∂∂

+=⎟⎠⎞

⎜⎝⎛

∂∂

+=

t

tt

e

e

0

1e0

1

,1 ,1

βββ

αμμαλλ (3)

( ) ( ) βαμα+λα=βαμ+λ=β TTe 100 23 ,23

Here ),,( zr uuuu θ=ρ

is the displacement vector,

),,,( tzrT θ is the temperature change, ee μλ , are

Free Vibrations in a Cylindrical Panel of Heat Conducting Viscoelastic Material 289

independent isothermal elasticities, K is coefficient of thermal conductivity, e is the dilatation, eC and ρ are

the density and specific heat at constant strain, 10 αα and

Tα are visco elastic relaxation time and coefficient of

linear thermal expansion respectively, 1 0 and tt are the

thermal relaxation times respectively. ikδ is the Kronecker delta. Here k=1 for Lord-Shulman (LS) theory and k=2 for Green-Lindsay (GL) theory. The comma notation is used for spatial derivatives and superposed dot represents time differentiation. We define the quantities

1*

1*

1

*

1

*00

1*

0

1*

0

1*

t , t

, z ,

,TT,

, ,

tt

zc

rc

r

Tu

Tc

u

uT

cuu

Tc

u

ze

z

er

er

ωω

ωω

βρω

βρω

βρω

θθ

=′=′

=′=′

=′=′

=′=′

( )( )eee

eeee

CT

KC

ct

μλρβμλ

ω

ωωααωααω

2 ,

2

,rr

, , , t

02

*

1

**

00*

110*

0

+∈=

+=

′∂∂

=∂∂

=′=′=′

212

1

2*2

122

21 ,c ,2c ∇

ω=∇′

ρμ

=ρμ+λ

=c

ee (4)

In order to solve equation (1), we take

z

rrr rru

,z

,,,,

u

, 1u , 1

χ

ψφφψ θθθ

=

−−=−= (5)

Assume that the disturbance is time harmonic through the factor tie ω and then seeking the solution in the form .

tienzmr ω⎟⎠⎞

⎜⎝⎛απθ

πφ=φ sin)sin()( ,

tienzmr ω⎟⎠⎞

⎜⎝⎛απθ

πψ=ψ cos)sin()(

tienzmr ω⎟⎠⎞

⎜⎝⎛απθ

πχ=χ sin)sin()( ,

tienzmrTT ω⎟⎠⎞

⎜⎝⎛απθ

π= sin)sin()( (6)

The functions ψ , T and ,χφ are obtained as

0k , )()( 211414 <′+′= rkKBrkIA ββψ ,

[ ]∑=

ββ +=φ3

1)()()(

iiiii rmKBrmIAr

[ ]∑=

ββ +=χ3

1)()()(

iiiiii rmKBrmIAar ,

[ ]∑=

ββ +=3

1

)()()(i

iiiii rmKBrmIAbrT (7)

Where( )

( ) 223

2*1

21

2*0

22

Li

iii tggm

gmmga−−αδ−δ+

= ,

⎟⎟⎠

⎞⎜⎜⎝

⎛δα

−−

⎟⎟⎠

⎞⎜⎜⎝

⎛+

⎭⎬⎫

⎩⎨⎧

δδα−+−δ

−=

2*1

22

32

6

312

2*0

*1

222

314*

0

Li

iL

i

i tggmg

ggmtgggmb , 1,2,3 i =

And im are the roots of

,0246 =−+− CBmAmm (9) bygiven are C and BA, tscoefficienin which

*0

102*

02*

0*1

222

431gAδ

ττ′β∈ω−

δδα−++=

itggg L ,

2*0

*1

102*

03

2*0

*1

24

22

434131 ggδδα

ττ′β∈ω−

δδα−++=

itggggggB L ,

( )22*1

22*

0*1

210

2*0

431g ω−δαδδαττ′β∈ω

+= LL ttiggC (10)

( )( )

( ) 02

4*1

222*03

12

02

2

*0

2*1

221

,g

, 1g

,

ωταδωδ

ωαδδδ

δωαδ

itgt

i

tg

LL

L

+=−=

−+−=

−=

παδω

τββτω

mtt

i

LL =−=

=′∈=

, k

, g , g

2*1

2

221

1*06

*005

(11)

( )( )

1 ,1

,2

, 2

-1 , 2

0*00

*01

200

22

ωααωδδ

ααδαδ

μλμλ

δμλ

μδ

iio

ee

ee

ee

e

+=+=

−+=

++

=+

=

k

k

tititi

ii

10000

211

0*01

*1

1 , 1 , 1

,1 ,1

δωτωτδωτ

ωββωαα

+=′+=+=

+=+=


3. Frequency equation

The displacements, temperature and stresses can be obtained by using (5), (7) and constitutive relations (1) along with non dimensional quantities. Considering the traction-free insulated boundary conditions at the lower and upper surfaces r = a, b and making use of expressions obtained for stresses and temperature one can get the free vibration equation as:

( )8,.......,2,1,,0 == jiEij (12)

( ) ( )

)()()21(

)(21

1111*02

1

22

1*0

2

111

1*0

211

*0

2111

tmIbt

ta

tmItmtmImE

L β

ββ

τββαδ

αδδ

⎟⎟⎠

⎞⎜⎜⎝

⎛+−−+

′−+′′=

( ) ( )

)()()21(

)(21

1221*02

1

22

2*0

2

121

2*0

212

*0

2213

tmIbt

ta

tmItm

tmImE

L β

ββ

τββαδ

αδδ

⎟⎟⎠

⎞⎜⎜⎝

⎛+−−+

′−+′′=

( ) ( )

)()()21(

)(21

1331*02

1

22

3*0

2

131

3*0

213

*0

2315

tmIbt

ta

tmItm

tmImE

L β

ββ

τββαδ

αδδ

⎟⎟⎠

⎞⎜⎜⎝

⎛+−−+

′−+′′=

( )( ) ⎥⎦

⎤⎢⎣

⎡ ′−′′

′βδ−αδ−= β

β 21

1111

1

1*0

*0

217

)()(21

ttkI

tkItkE

)()1( 111121 tmIamE β′−= ,

)()1( 122223 tmIamE β′−= ,

)()1( 133325 tmIamE β′−= , )( 111

27 tkIt

E ′β= β

)(2)(2112

1111

131 tmI

ttmIm

tE ββ′

β+′β

−= ,

)(2)(2122

1122

133 tmI

ttmIm

tE ββ′

β+′β

−=

)(2)(2132

1133

135 tmI

ttmIm

tE ββ′

β+′β

−= ,

)()()( 1121

2

111

111

2137 tkI

ttkI

tktkIkE ′β

−′′′

+′′′′−= βββ

)( 111141 tmImbE β′= , )( 122243 tmImbE β′= ,

)( 133345 tmImbE β′= , 047 =E

Here ( )8,6,4,2=jEij can be obtained by just replacing a modified Bessel function of first kind in

( )7,5,3,1== iEij with the ones of the second kind

respectively, while ( )8,7,6,5=jEij can be obtained

by just replacing 1t in ( )4,3,2,1=jEij with 2t

respectively, where 2/*1/1 tRat −== and

2/*1/2 tRbt +== and Rabt /)(* −= is the thickness to mean radius ratio of the panel. For uncoupled thermo elasticity )0(∈= , the above analysis reduces to one as obtained and discussed by Chen et al.22 in the case that elastokinetics and thermal effects get decoupled. 4. Numerical results and discussions

In order to illustrate and verify the analytical results obtained in the previous sections we present some numerical simulation results. For the purpose of numerical computations we have considered magnesium crystal-like material whose physical data is given below [12]:

2101017.2 −×= Nmλ ,

21010278.3 −×= Nmμ , 331074.1 −×= mKgρ ,,250 CT o=

113 deg1004.1 −−×= KgJCe ,111

1 1034.5 −∗ ×= sω 112 deg107.1 −−×= WmK ,

126

126

deg100.2,deg1068.2

−−

−−

×=

×=

NmmNmβ

,

21510753.1 −−×= mχ

210

5

21021

1013849.1,10688.3

,10475.1

−

−

−

×=

×=

×==

NmbNa

Nmξξ .

For closed cylindrical shell, the central angle πα 2= and the integer ‘n’ must be even, since

shell vibrates in the circumferential full wave. Therefore the frequency equation for closed cylindrical shell can be written by

setting ...)3,2,1(, == ναπν n , where ν is

circumferential wave number. The computations have been done for viscothermoelastic panel (VTE) and thermoelastic panel without voids (TE) in order to make comparative study.

Free Vibrations in a Cylindrical Panel of Heat Conducting Viscoelastic Material 291

For two different values ( 2,1=ν ) of the circumferential wave number, the variations of lowest frequency

2cRω

=Ω with respect to the parameter L

mRtL = of a

simply supported stress free thermally insulated cylindrical shell of magnesium crystal-like material are respectively shown in Figs. 1 and 2 for different values ( 5.0,25.0,1.0,01.0* =t ) of thickness to mean radius of the panel. The effect of viscosity is noticed to be quite prominent with increasing values of Lt and *t as can be seen from lowest frequency profile in Fig. 1.

Figure 1

00.5

11.5

22.5

33.5

44.5

0.1 0.48 0.86 1.24 1.62 2

Parameter (tL)

Freq

uenc

y .

t* = 0.01t* = 0.1t* = 0.25t* = 0.5t* = 0.01t* = 0.1t* = 0.25t* = 0.5

With balls - TE Without balls - TVE

Fig. 1. Variations of lowest frequency (Ω) with respect to parameter ( lt ) for different values of of thickness to

mean radius ratio ( 5.0,25.0,1.0,01.0* =t ) and 1=ν .

Figure 2

00.5

11.5

22.5

33.5

44.5

5

0.1 0.48 0.86 1.24 1.62 2

Parameter (tL)

Freq

uenc

y .

t* = 0.01t* = 0.1t* = 0.25t* = 0.5t* = 0.01t* = 0.1t* = 0.25t* = 0.5

With balls - TE Without balls - TVE

Fig. 2. Variations of lowest frequency (Ω) with respect to parameter ( lt ) for different values of thickness to mean

radius ratio ( 5.0,25.0,1.0,01.0* =t ) and 2=ν .

From Fig. 2, it is observed that the lowest frequency is significantly affected due to viscosity in the range

5.15.0 ≤≤ Lt for all values of *t . The lowest frequency increases monotonically with increasing values of Lt for 1=ν , and 2=ν , which becomes stable and

smooth at higher values of Lt and small values

of *t . The comparison of Figs. 2 and 3 shows that the lowest frequency is significantly affected thickness to mean radius ratio )( *t for 5.0≥Lt ,

1=ν and for 5.1≤Lt , 2=ν both in the presence and in the absence of viscosity.

Reference 1. Soldatos, K.P. and Hadhgeorgian, V.P., Three

dimensional solution of the free vibrations problem of homogeneous isotropic cylindrical shells and panels, J. Sound and Vibration, Vol.-137 pp. 369-384, 1990.

2. Leissar, A.W., Free vibrations of thick hollow circular cylinders from three dimensional Analysis, ASME J. Vib. Acoust.,Vol.-119, pp.89-95,1997.

3. Jiaug, X.Y., 3-D vibration analysis of fiber reinforced composite laminated cylindrical shells, ASME J. Vib. Acoust., Vol.-119, pp. 46-51, 1997.

4. Fam, J.R. and Duag, K.W., Analytical solution for thick closed laminated cylindrical shells, Int. J. Mech. Sci., Vol.-35, pp. 657-668, 1997.

5 Ye, J.Q. and Soldatos, K.P., 3-D vibrations of laminated cylinders and cylindrical panels with symmetric and antisymmetric cross-ply layup, Composite Engineering, Vol.-4, pp 429-444, 1994.

6. J.N. Sharma and R.S. Sidhu, On the propagation of plane harmonic waves in anisotropic generalized thermoelasticity, Int. J. Engng. Sci., Vol.-24, pp. 1511-1516, 1986.

7. Sharma J. N. Free vibration abalysis of homogeneous transversely isotropic thermoelastic cyclinderical panel, J. Acoust. Soc Am. 109, 2000.

8. V.P. Buchwald, Rayleigh waves in transversely isotropic media, Q.J. Mech. Appl. Math, Vol. 14, pp. 193-304(1961)

9. Ewing, W. S. Jardetzky and F. press, Elastic waves in layered media, McGraw Hill New York, 272-280, 1957.

10. C. Hunter, Visco-elastic waves: Progress in solid mechanics (Eds.) Ian Sneddon and R. Hill, North

Interscience, Amsterdam, New York, 1960. 11. W. Flugge, Viscoelasticity, Blasdell, London,

1967. 12. Kumar, R. and Rani, L., Interaction due to

mechanical and thermal sources in thermo elastic half space with voids, Journal of Vibration and Controls, Vol. 11, pp 499-517, 2005.

Transient Waves Due to Continuous Thermal Loads in Thermoviscoelastic Materials

Rattan Chand1, Nisha Sharma2 and J. N. Sharma2

1Department of Physics, Govt. Sen. Secondary School, Sujanpur Tihra, (HP) 176 110 (HP) 2Department of Applied Sciences, National Institute of Technology, Hamirpur 177 005 (HP)

Email: [email protected], [email protected] Abstract

The Kelvin-Voigt model of linear viscoelasticity is used to investigate the transient waves due to continuous thermal loads acting on the boundary of the thermoviscoelastic media. The boundary value problem in the transform domain is solved by employing the Laplace and Hankel transform techniques in the context of various theories of generalized thermoelasticity. The Laplace transforms have been inverted by using a numerical technique and then the inverse Hankel transform integrals are evaluated by using Romberg integration in order to obtain the results in the physical domain. The temperature and stresses are computed numerically and presented graphically in different situations for copper material.

1. Introduction

The theory of thermoelasticity deals with the effects of mechanical and thermal disturbances on the elastic body. There are two defects in uncoupled theory of thermoelasticity (UCT). The theory of coupling of thermal and strain fields gives rise to Coupled theory of thermoelasticity (CT) and was first postulated by Duhamel [1] in 1837, shortly after the theory of elasticity. Neumann [2] in 1855, made attempt at thermodynamical justification of equations of Duhamel’s theory. The work of Biot [3] in 1956 removed the first defect of the uncoupled theory of thermoelasticity. In 1967, Lord and Shulman [4] incorporated a flux-rate term into Fourier’s law to formulate a generalized theory that admits finite speed for thermal signals. This theory also predicts a finite speed of heat propagation. A wave like thermal disturbance is referred to as “second sound” by Chandrasekharaiah [5]. Sharma and Chauhan [6] investigated the disturbance due to normal point load and thermal source acting on the surface of half space by applying Laplace and the Hankel transform techniques. The effect of internal friction on the propagation of plane waves in an elastic medium may also be considered owing to the fact that dissipation accompanies vibrations in solid media due to the conversion of elastic energy to heat energy Ewing, Jardetzky and Press [7]. As pointed out by Freudenthal [8], most of the solids when subjected to dynamic loading, exhibit viscous effects. The Kelvin-Voigt model is one of the macroscopic mechanical models often used to describe the visco-elastic behaviour of a material. This model represents the delayed elastic

response subjected to stress when the deformation is time dependent but recoverable. The dynamical interaction of thermal and mechanical fields in solids has great practical applications in modern aeronautics, astronautics, nuclear reactors, and high-energy particle accelerators, for example. Mukhopadhyay [9] studied the thermal relaxation effects and compared the various theories of generalized thermoelasticity for the thermoviscoelastic interactions in an infinite viscoelastic solid of Kelvin-Voigt type with a spherical cavity. Sharma and Chand, Sharma et al. [10, 11] discussed the transient thermoviscoelastic waves and forced vibrations in a half space due to thermal and mechanical loads.

In the present article the Kelvin-Voigt model of linear viscoelasticity describes the viscoelastic nature of the material. We investigate the transient waves due to thermal sources in a thermoviscoelastic continuum by applying the Laplace and Hankel transform techniques in the context of various theories of generalized thermoelasticity. The results obtained theoretically have been computed numerically and are presented graphically for copper material. A complete and comprehensive analysis and comparison of results in various theories are presented. 2. Formulation of the problem

We consider a homogenous isotropic thermoviscoelastic half space initially undisturbed and at uniform temperature 0T . The Kelvin-Voigt model of linear viscoelasticity that describes the viscoelastic nature of the material has been employed to study the problem. We take the origin of cylindrical coordinate system ( )zr ,, θ as any

Transient Waves Due to Continuous Thermal Loads in Thermoviscoelastic Materials 293

point on the surface 0=z and z -axis pointing vertically downward into the medium so that the half-space occupies the region 0≥z . It is assumed that a continuous heat source is acting at a point on the surface 0=z of the medium and hence all the quantities are independent of the θ co-ordinate. The basic governing equations of motion and heat conduction in the context of the generalized theory of thermoelasticity, in the absence of body forces and heat sources are given by

→→→

=+∇−∇∇++∇ uTtTuu k &&& ρδβμλμ ) ( .)( 122 (1)

→

∇∂∂

+∂∂

++=∇ ut

tt

TTtTCTK ke .)()( 2

2

010

..

0

.2 δβρ (2)

where )1( 0 te ∂∂

+= αλλ , )1( 1 te ∂∂

+= αμμ ,

⎟⎠⎞

⎜⎝⎛

∂∂

+=+=teT 01)23( ββαμλβ

( ) Teee αμλβ 23 += , ( ) eTee βααμαλβ /23 100 += (3)

Here ),0,(),,( wutzru =ρ

is the displacement vector; ),,( tzrT be the temperature change;

ee μλ , are the Lame s, parameters; eC,ρ and

Tα are respectively the density, specific heat at constant strain and coefficient of linear expansion; K is the thermal conductivity;

10 , αα are viscoelastic relaxation times and

10 , tt are thermal relaxation times; k1δ is the

Kronecker’s delta in which 1=k for Lord –Shulman (LS) theory and 2=k in the case of Green-Lindsay (GL) theory. The thermal relaxation times 0t and 1t satisfies the inequalities as

010 ≥≥ tt (4)

in case of GL theory only. However, it has been proved by Strunin recently that the inequalities (4) are not necessary to be satisfied. The initial and regularity conditions are given by ( ) ( ) ( ) ( )0,,00,, ,0,,00,, zrwzrwzruzru && ==== ,

( ) ( )0,,00,, zrTzrT &== ,

for 0,0 ≥≥ rz and

( ) ( ) ( ) ,00,, ,00,, ,00,, === zrTzrwzru (5)

for ∞→> zt ,0 .

3. Boundary conditions

The surface 0=z of the thermoviscoelastic solid is subjected to the action of continuous thermal point load at the origin. Therefore, the corresponding boundary conditions are given as

,2

) ()( ,0 ,0 0

rtfrThT

zT

rzzz πδττ −=+

∂∂

== (6)

where )(rδ denotes the Dirac delta function; )(tf is a well behaved function (Heaviside’s unit function) of time and h is the coefficient of surface heat transfer. Here 0→h corresponds to temperature gradient and ∞→h refers to temperature input acting at the boundary of the half space. We define the quantities

, ,/ ,/ *1

*1

* ttczzcrr ωωω =′=′=′

,/ , / 001* TTTTucu eii =′=′ βωρ

, , 1*

10*

0 tttt ωω =′=′ 0/ Teijij βττ =′

0*

00*

01*

1 , , βωβαωααωα =′=′=′

( )eeeCT

μλρβ

2 0

2

+∈= ,

10e** P/ cTP βω=

KC eee /)2(* μλω += , 21

22

2 / cc=δ

*1

22

21 / , ,2 ω

ρμ

ρμλ hchcc eee =′=

+= (7)

Using quantities (6) in equations (1) to (3) and suppressing dashes, we obtain

( ) →

→→

=+∇⎟⎠⎞

⎜⎝⎛

∂∂

+−

⋅∇∇⎥⎦⎤

⎢⎣⎡

∂∂

+−+−+∇⎟⎠⎞

⎜⎝⎛

∂∂

+

uTtTt

ut

ut

k &&&210

21

20

2221

1

)21(11

δβ

δαδαδδα (8)

( ) →⋅∇⎟

⎟⎠

⎞⎜⎜⎝

⎛

∂

∂+

∂∂

⎟⎠⎞

⎜⎝⎛

∂∂

+=∈+−∇ ut

ttt

TtTT k 2

2

10002 1 δβ&&& (9)

The non-dimensional form of boundary conditions (6) on the surface 0=z is given as

rtfr

hTzT

rzzz πδ

ττ2

) ()( ,0 ,0 −=+∂∂

== (10)

4. Solution of the problem

We introduce the potential functions φ and ψ through the relations

rrzwv

zru ψψφψφ

−∂∂

−∂∂

==∂∂

+∂∂

= , 0 , (11)

Using equations (11) in equations (8) and (9) and applying Laplace transform followed by Hankel transform provide us


0ˆˆ2*

1

22 =⎟⎟⎠

⎞⎜⎜⎝

⎛+− ψ

δαψ pqD (12)

( )( )( ) 0ˆ,ˆ 22

2221

22 =−−−− TmqDmqD φ (13) (19) where

*

0032

221

*0

10*0

4

02

*0

22

21

/

, 2

δτ

δ

ττβτ

δ

pmm

pppmm

=

′∈++=+

(14)

(20) The non-vanishing solution of equations (12) and (13) which satisfies the radiation condition (viz. the disturbance is assumed to be confined to the surface 0=z ) is obtained for ψφ ˆ,ˆ and T respectively. We take )()( tHtf = and apply the Laplace transform followed by Hankel transform to boundary conditions (10). Then, upon using the expressions for ψφ ˆ,ˆ and T after lengthy but straight forward calculation, the displacements, temperature change and stresses in the transformed domain are obtained as

( )[ ] Δ++−= −−− / ˆ 3213321

zzz eMeMeMqu ξξξ ξ (15) [ ] Δ++−= −−− /ˆ 321

32211zzz qeMeMeMw ξξξ ξξ (16)

[ ] Δ++= −−− / )(ˆ 3211321

*1

zzzzz eFMeMeMFp ξξξατ (17)

( ) ( )[ ] Δ+++= −−− /2 3213

23

22211

*1

2 zzzrz eMqeMeMqp ξξξ ξξξαδτ (18)

[ ] /ˆ 212211

1*0

*0 Δ+= −− zz eQMeQM

pT ξξ

τβ

δ (19)

where

[ ( ) ]122

32

*0

*01

1 2 2

FqqFM ξξδπβτ

−+−=

[ ( ) ]1123

2*0

*01

2 2 2

FqqFM ξξπδβτ

−+=

[ ]12*0

*01

3 ξξπδβτ

−−=Fq

M (20)

22*1

2 qpF δα

+=

231 2 δξqF = (21)

( )[ ]( )( )[ ]( )211

23

22

11223

21

2

2

ξξξ

ξξξ

−−+−

−−+=Δ

hFqFqQ

hFqFqQ (22)

The results for the LS theory and GL theory can be obtained by setting k=1 and k=2 respectively, in the values of 100 , τττ and′ and those for coupled thermoelasticity (CT) can be obtained by taking

10 0 tt == in the foregoing analysis. Here 0→h corresponds to temperature gradient and ∞→h refers to temperature input acting at the boundary of the half space. The results for uncoupled thermoviscoelasticity can be obtained by setting thermal coupling parameter and ,0∈= 10 0 tt == in the above analysis. The results for non-viscous thermoelastic continuum can be deducted from the above-obtained results by taking 010 ==αα in the appropriate results and functions. 5. Numerical results and discussion

The physical data for a copper material is given as Mukhopadhyay [9]

,/10950.8 ,/102.4 ,/102.8 33210210 mkgmNmN ee ×=×=×= ρμλ ,///1013.1,05.0,/100.1 28 KsmCalKKT ×=∈=×= −α

111* 1011.1 −×= sω , st 130 10131.6 −×=

sst 1310

131 108831.6,107565.8 −− ×==×= αα

Due to the closeness of results and to avoid clustering of different curves, the variations of temperature change and stresses are presented graphically for one value of time viz. at 25.0=t only in the context of viscous and non-viscous; Lord-Shulman (LS), coupled thermoelasticity (CT), uncoupled thermoelasticity (UCT), Green – Lindsay (GL); theories of thermoelasticity. 5. 1. Continuous temperature input

The computed temperature change, vertical and shear stress functions in case of applied temperature input at the stress free surface of the viscous and non-viscous solid half space are given in Figs. 1 to 3. From Fig. 1, it is noted that the variation of temperature change with radial distance in the context of CT, UCT, CT (NV) and UCT (NV) theories of thermoelasticity for viscous and non-viscous solid half space, first attains maximum value in the vicinity of the load because of large internal friction among the molecules at the time of application of continuous temperature input. After that the temperature change starts decreasing to achieve minimum value at 75.0=r due to small value of internal friction among the rest of the molecules of the solid and then remain close to zero in an oscillating manner. The trend and behavior of temperature change in LS and GL theories of thermoelasticity for viscous solid half space is observed to be opposite to that in the context of CT, UCT, CT (NV) and UCT (NV) theories in the considered domain of radial distance.

Transient Waves Due to Continuous Thermal Loads in Thermoviscoelastic Materials 295

-60

-40

-20

0

20

40

0 1 2

Radial distance

Tem

pera

ture

cha

nge

LS

CT

UCT

Fig. 1. Temperature Change [Temperature input at ( )25.0=t ]

It is found from Fig. 2 that the variations of vertical stress in case of CT, UCT, GL, CT (NV) and UCT (NV) theories for viscous and non-viscous solids decrease from zero to its minimum value in the domain 05.00 ≤≤ r , increase rapidly to the maximum value at 75.0=r and then sinusoidally tends to zero in a decreasing manner afterwards. In LS, theory the variation of this function remains almost close to zero in the considered domain of the radial distance except near the vicinity of the load where slight variations are noticed due to the continuous temperature input at the stress free surface of the viscous solid half space.

-125

-105

-85

-65

-45

-25

-50 0.5 1 1.5 2 2.5

Radial distance

Ver

tical

stre

ss

LSCT X 10UCTGL X 100CT(NV)UCT(NV)

Fig. 2.Vertical stress [Temperature input at ( )25.0=t ]

From Fig. 3, it is seen that the variation of shear stress function in case of GL theory for viscous solid has an opposite trend and behavior to that in the context of CT and UCT theories. The shear stress development in case of CT (NV) and UCT (NV) theories for non-viscous solid half space and LS theory in case of viscous solid is observed to be significantly small in magnitude as compared to other counter part theories of thermoelasticity.

-75

-60

-45

-30

-15

0

15

30

45

60

75

0 1 2

Radial distance

Shea

r stre

ss

LSCTUCTGL X 10CT(NV)UCT(NV)

Fig. 3.Shear stress [Temperature input at ( )25.0=t ]

5. 2. Continuous Temperature Gradient

The computed results for temperature change, vertical and shear stress functions in case of viscous and non-viscous solids under the action of continuous temperature gradient are plotted in Figs. 4 to 6. It is noticed from Fig. 4 that the variations of temperature change in the context of LS, GL, CT, UCT, CT (NV) and UCT (NV) theories first attain maximum values just in the vicinity of the load, decrease rapidly in the domain 3.00 ≤≤ r ; slowly and steadily in the domain 5.23.0 ≤≤ r , before it creeps along the stress free surface in case of CT and UCT theories for both viscous and non-viscous solid half spaces but increase as a hump for LS and GL theories in the domain 7.03.0 ≤≤ r and finally approaches to zero in an oscillating manner for viscous media because of thermal relaxation times.

-0.050.450.951.451.952.452.953.453.954.454.95

0 1 2

Radial distance

Tem

pera

ture

cha

nge

LSCTUCTGLCT(NV)UCT(NV)

Fig. 4. Temperature Change [Temperature gradient at ( )25.0=t ]

It is revealed from Fig. 5 that the variations of vertical stress, in the context of CT, UCT, CT (NV) and UCT (NV) theories for viscous and non-viscous solids show opposite trends and behavior to that in case of LS and GL theories viscous solids.


-80

-60

-40

-20

0

20

40

60

80

100

120

0 1 2Radial distance

Ver

tical

stre

ssLSCTUCTGLCT(NV)UCT(NV)

Fig. 5.Vertical stress [Temperature gradient at ( )25.0=t ]

However, the magnitude of vertical stress in considered theories for non-viscous solid is found to be greater than that of the shear stress for the same theories under similar conditions as can be seen from Fig. 6. The magnitude of variation of these functions in the CT and UCT theories is higher in viscous solid than that of non-viscous one.

-2.5

-1.5

-0.5

0.5

1.5

2.5

3.5

0 1 2

Radial distanceShea

r stre

ss

LSCTUCTGLCT(NV)UCT(NV)

Fig. 6.Shear stress [Temperature gradient at ( )25.0=t ]

6. Conclusions In conclusion, all the considered functions are noticed to vanish at certain values of radial

distance approximately near 5.2=r , which show the existence of wave fronts. . The effect of mechanical relaxation time on various considered functions is noticed to be quite significant and due to which the amplitude of vibrations is subject to suppression along with conversion of some part of thermal energy, though small, to mechanical energy due to internal friction of the material particles. In case of temperature input at the stress free surface of the solid half space, the most of the energy is observed to be carried in the form of thermal wave, significant amount propagate in the form of vertical stress and shear stress, which is quite in agreement with the boundary conditions. However, in case of temperature gradient, the significant amount of energy is carried in the form of vertical stress and shear stress waves in addition to thermal wave which carries major portion of the energy. References [1] J. Duhamel, Thermo-Mechanique, J. de L ‘Ecole

Polytechnique 15, 1837. [2] F. Neumann, Vorlesungen Uber die Theorie

der elasticitat, Meyer, Brestau, 1885. [3] M. Biot, J. Appl. Physics. 27 (1956) 240-253. [4] H. W. Lord and Y. Shulman, J. Mech. Phys

Solids. 15 (1967) 299-309. [5] D. S. Chandrasekharaiah, , Appl. Mech. Rev.

39 (1986) 355- 376. [6] J. N. Sharma, R. S. Chauhan, J. Thermal

Stresses. 24 (2001) 651-675. [7] M. Ewing, W. S. Jardetzky and F. Press,

Elastic waves in layered media, McGraw Hill, New York. (1957) 272-280.

[8] A. M. Fredudenthal, J. Appl. Phys. 25 (1954) 1-10.

[9] S. Mukhopadhyay, J. Thermal Stresses. 23 (2000) 675-684.

[10] J. N. Sharma, R. Chand, J. Thermal Stresses. 28 (2005) 233-252.

[11] J. N. Sharma, R. Chand, D. Chand, Int. J. Appl. Mech. and Engg. 11 (2006) 105-121.

Forced Vibrations of Solid Thick Plate of Thermoviscoelastic Material

P.K. Sharma, K. K. Sharma, V. Walia and A. Kumar Department of Mathematics, National Institute of Technology, Hamirpur – 177 005 INDIA

Email: [email protected]

Abstract

The present paper is aimed at to study the thermoelastic interaction in an infinite Kelvin –Voigt type viscoelastic, thermally conducting thick homogeneous isotropic axisymmetric visco-thermoelastic solid plate subjected to sudden lateral mechanical and thermal loads in the context of coupled (CT), Lord-Shulman (LS) and Green-Lindsay (GL) theories of thermoelasticity. Using Laplace and Fourier transform method; the governing equations of motion and heat conduction have been solved to predict the response of the plate in the physical time domain. A modified Bessel function solution with complex argument is directly used. In the absence of mechanical relaxations (viscous effect), the results for generalized, and coupled theories of thermoelasticity, can be obtained as particular cases from the derived results. In the absence of thermomechanical coupling the analysis for a viscoelastic plate can also be deduced from the present one.

1. Introduction

During the second half of twentieth century non-isothermal problems of the theory of elasticity became increasingly important. This is due to their wide application in diverse fields. The high velocities of modern aircraft give rise to aerodynamic heating, which produce intense thermal stresses, reducing the strength of aircraft structure. In uncoupled thermoelasticity, the temperature is governed by parabolic differential equation that does not contain any elastic term. Biot [1] formulated the theory of coupled thermoelasticity to eliminate the paradox inherent in classical uncoupled theory that elastic change has no effect on temperature. The heat equations in both coupled and uncoupled theories of thermoelasticity are of diffusion type, predicting infinite speed of heat transportation, a physically impossible phenomenon. During the last four decades non-classical theories of thermoelasticity have been developed to remove this paradox. The generalized theory of thermo-elasticity has drawn wide spread attention because it removes the physically unacceptable situation of classical theory of thermo-elasticity, that is the thermal disturbance propagates with infinite velocity. Lord and Shulman [2], and Green and Lindsay [3] are two important generalized theories of thermoelasticity that have become the center of interest of recent research in this area. Lord and Shulman [2] incorporated a flux-rate term into the Fourier’s law to formulate a generalized theory, which includes one thermal relaxation time whereas Green and Lindsay [3] by including temperature rate violate the classical Fourier’s Law of heat conduction when the body under consideration has a centre of symmetry.

The heat conduction equation in both of these theories is hyperbolic and hence predicts finite speed for thermal signals. Dhaliwal and Sherief [4] extended LS - theory to anisotropic thermoelastic body. GL theory involves two relaxation times constrained by inequality t1 > t0 > 0. Recently Strunin [5] proved that no such constraint on relaxation times arises. In both these theories, introduction of the thermal relaxation parameters modifies the basic equations of thermoelasticity. Chandrasekharaiah [6] referred a wave like thermal disturbance as ‘Second Sound’. Some researchers such as Ackerman et al. [7], Guyer and Krumhansl [8], and Ackerman and Overtone [9] proved experimentally for solid Helium that finite, though quite large speed also exists.

Several mathematical models have been used by authors [10, 11] to accommodate the energy dissipation in vibrating solids where it is observed that internal friction produces attenuation and dispersion and hence the effect of the viscoelastic nature of material medium in the process of wave propagation and disturbance due to applied loads can not be neglected. The viscoelastic nature of a medium has special significance in wave propagation in a solid medium. EI- Karamany [12] formulated the boundary integral equation method for generalized linear micro-polar visco-thermoelasticity and obtains the solution of the corresponding equations in the Laplace transform domain. He pointed out that the cases of generalized linear micro-polar thermoviscoelasticity of Kelvin-Voigt model, generalized linear thermo-viscoelasticity and generalized thermoelasticity [2 & 3] theories can be obtained from the given results. Mukhopdhyay and Mukhopdhyay [13] studied the effect of rotation and relaxation time on plane waves in an infinite generalized thermoviscoelastic solid of Kelvin- Voigt


type when the entire medium rotates with a uniform angular velocity. Mukhopdhyay [14] explored the effect of thermal relaxation time parameters on thermoviscoelastic interactions in an infinite body with a spherical cavity subjected to periodic loadings on the boundary of the cavity. Sharma and Sharma [16] have studied the distribution due lateral loads in a homogeneous isotropic thermoelastic solid plate by using a combination of Laplace and Fourier transform technique in the context of generalized theories of thermoelasticity.

The Kelvin-Voigt model is one of the macroscopic mechanical models often used to describe the visco-elastic behaviour of a material. The model represents the delayed elastic response subjected to stress when the deformation is time dependent but recoverable. In the present paper the response of a homogeneous isotropic finitely stretched thermo-viscoelastic plate due to lateral mechanical and thermal load, in the context of generalized theories of thermoelasticity [2, 3] has been investigated by using Laplace and Fourier transforms technique. A relatively straightforward and efficient approach has been presented to solve dynamical generalized thermoelastic coupled problems subjected to impact loads. In the absence of thermomechanical coupling the analysis for a viscoelastic plate can also be deduced from the present one. 2. Formulation and solution of the problem

We consider a finite homogeneous isotropic thermoelastic circular plate of constant thickness 2h and radius R in an undisturbed state. The plate is axisymmetric with z-axis as the axis of symmetry. The origin of cylindrical co-ordinate system ),,( zr θ is taken at the upper surface of the plate given by z = 0 and z-axis normal to it along the thickness as shown in figure. Since z-axis is the axis of symmetry, hence all the quantities are independent of θ -co-ordinate and the displacement vector is given by

),,(),,( woutzxu =→

. Let ),,( tzxTT = be the change in temperature of plate at any time. For linear generalized theories of thermoelasticity the basic governing field equations of motion and heat conduction in the absence of body forces and heat sources are given as given in [20]. The parameters μλ , and β are defined as

)1(1

, 1 ,1

0

10

⎟⎠⎞

⎜⎝⎛

∂∂

+=

⎟⎠⎞

⎜⎝⎛

∂∂

+=⎟⎠⎞

⎜⎝⎛

∂∂

+=

t

tt

e

ee

βββ

αμμαλλwh

ere( ) ( ) eTeeTeee βααμαλβαμλβ /23 , 23 100 +=+= ,

ee μλ , are Lame` parameters, 10 ,αα are visco-

thermoelastic relaxation times and Tα is the coefficient of linear thermal expansion. The constitutive relations are given by

ijkijijkkij TtTee δδβμδλσ )(2 21&+−+= (2)

where 3,2,1,,2/)( ,, =+= jiuue ijjiij is the strain tensor. The plate is assumed to be clamped around the entire perimeter, at rest and undisturbed initially. Therefore, the initial and regularity conditions are given by

auu >======= r 0,at t T0T ,w0 w,0 &&&

atu

==∂∂ rat 0 (3)

when 0,for t w0 >==u Rr → 2. 1. Boundary conditions

The plate is subjected following two types of boundary conditions: (i) Mechanical lateral load atHzz ==== ron 0,T 0,u ),(0σσ ,

RTwu ==== ron 0 (4.1) (ii) Thermal lateral load atHTzz ====σ ron ,)(T 0,u ,0 0 ,

RTwu ==== ron 0 (4.2) where H(t) is Heaviside unit function. We define non-dimensional quantities

0ij

00

1*

1

*

1

* , ,u ,z ,

TTTTu

Tc

cz

crr

e

ij

e β

σσ

βρωωω

=′=′=′=′=′ρρ

i, j =1, 2, 3 (5)

t , , t,t , 0*

01*

10*

0**

11 ttt ωωαωαωωαα =′=′=′=′=′

0*

021

22

21

*21

*1 ,/ ,/ ,/ βωβδωηωη =′=== cccRca

where ( )

( ) ρμ

ρμλ

μλρβμλ

ω eee

eee

eeee ccCK

C=

+=

+∈=

+= 2

221

20* ,

2 ,

2T

,2

Here ∈ is the thermoelastic coupling constant, *ω is the characteristic frequency of the material, and

21 ,cc are respectively the longitudinal and shear wave

Forced Vibrations of Solid Thick Plate of Thermoviscoelastic Material 299

velocities in the medium. Upon using the quantities (5), along with the relations

rrzwv

zru ψψφψφ

−∂∂

−∂∂

==∂∂

+∂∂

= ,0 , (6)

in governing equations and then applying Laplace and Fourier transform with respect to time variable‘t’ defined by

( )∫∫∞

+−∞

∞−

==0

)(,,),,( dtdzetzrfpzrf iqzpt (7)

and simplifying the resulting equations obtained, the potential functions and temperature change in transformed domain is obtained as

( ) ( )[ ]∑=

+=2

100

ˆi

iiii rmKBrmIAφ

)()(ˆ 313313 rmKBrmIA +=ψ (8)

( ) ( )[ ]∑=

+=2

100

ˆi

iiiii rmKBrmIAbT

where 2222ii apqm += , i = 1,2; 2*

122

3 / δαpqm +=

*02

221*

012*

02

*022

21 a , 1

ατ

αττβ

ατ

pa

pp

aa =′∈

++=+ (9) (11)

01*

001*

01*

1 , , ββδααα +=+=+= −−− ppp

( ) 2,1 ,/ 1*0

*22 =−−= ippqmb ii τβα

Here nI and nK , n = 1, 2 are modified Bessel

functions and 2*1

223 / δαpqm += .

The transformed displacement components and stresses are obtained as

ψφ )iqr

u +∂∂

=ˆ

ˆ , ψφ ˆ1ˆˆ ⎟⎠⎞

⎜⎝⎛ +∂∂

−=rr

iqw

ψαδφδαδσ ˆ2ˆ22ˆ *1

22

2*1

22

rpiq

rrppqrr ∂

∂+⎟

⎟⎠

⎞⎜⎜⎝

⎛

∂∂

−+= ,

⎟⎟⎠

⎞⎜⎜⎝

⎛−+−

∂∂

= ψδα

φδασ ˆ12ˆ

2ˆ22*

1

22*1

rpq

riqprz (10)

ψδα

φαδσ

ˆ12

ˆ)1(2

2*1

2

2*1

22

⎟⎠⎞

⎜⎝⎛ +∂∂

−

⎟⎟⎠

⎞⎜⎜⎝

⎛∂∂

+∂∂

−=

rriqp

rrrppzz

Invoking the boundary conditions (4), we obtain

ΔΔ

=ΔΔ

= +3iB , ii

iA ; i = 1, 2, 3 (11)

where ijE=Δ , i ,j = 1,2,3,4,5,6 and

( ) 1,2;i ),(2 102*

122

1 =−= ηαδ iii mImppE

)](1)([2E 13013

1303*1

213 η

ηηαδ mI

mmImpiq −′−=

)(E 1,2;i ),( 13123102 ηη miqImImE iii ==′=

0E 1,2;i ),( 33103 === ηiii mIbE

1,2;i ),( 204 == ηii miqIE

)](1)([E 23023

230343 ηη

η mIm

mIm −′−=

Here ijE , j = 4,5,6 can be obtained by just replacing

modified Bessel function of first kind in ijE , j = 1,2,3

with the ones of second kind respectively while ijE , i

= 5,6 can be obtained by replacing 21 ηη with in

ijE , i = 2,3 and iΔ , i = 1,2,3,4,5,6 can be obtained

by replacing corresponding column of Δ by Tp )0,0,0,0,0,( 1*

0−σ when mechanical load is acting

and by Tp )0,0,0,,0,0( 1− for lateral thermal load acting on the boundary of the plate. The results in the context of coupled visco-thermoelasticity (CVT) can be obtained by setting to = 0 = t1 and those for uncoupled viscothermoelasticity (UCVT) can be obtained by taking coupling constant ∈ = 0 and relaxation times to = 0 = t1 in the above analysis. 3. Inversion of the transforms

Due to existence of damping term in the temperature field equation (1.2), the dependence of roots )3,2,1( =imi on p and q is complicated; hence the inversion of the integral transforms is difficult because the isolation of p is impossible. These difficulties, however, are reduced if we use some approximate or numerical methods. Therefore, in order to obtain the solution of the problem in the physical domain, we must invert the transforms is equations (19) to (22). These expressions can be formally expressed as functions of r, q and p (the Fourier and Laplace transforms parameters respectively) of the form ),,(ˆ pqrf . To obtain the various functions in the physical domain, integral transforms has been inverted by adopting the procedure outlined in [16].


4. Conclusions

In this paper potential functions and modified Bessel function with complex argument has been directly used to study the disturbance of homogeneous isotropic thick plate due to (i) mechanical load and (ii) thermal line load on its free surface using integral transform technique. In order to obtain the solutions in physical domain the integral transforms have been inverted by using numerical technique. The problem is also formulated and solved by finite element method (FEM). The results obtained have been presented graphically for Lord Shulman (LS), Green Lindsay (GL), Green Nagdhi (GN), and coupled (CT) theories thermoelasticity. The results obtained by finite element method are also represented graphically for coupled thermoelasticity. The results obtained by two different methods are found to be quite close and are in agreement to each other. In case of thermal load the disturbance can be experienced near fixed boundary however in mechanical load it decays midway, which implies that disturbance due to thermal load can be experienced at more distant points than that in case of disturbance due to mechanical loads. Numerical result reveals that the effect of relaxation times so called “second sound effects’ are short lived but are significant in the vicinity of load applied. This development can be used to ascertain the validity of finite element results and the application of a particular theory of thermoelasticity in different situations. The existence of ‘second sound’ phenomenon in case of Lord-Schulman (LS) and Green-Lindsay (GL) models has experimental validation and hence these models are recommended for such applications in various situations in order to avoid the physically impossible and unrealistic paradox of infinite velocity of a part of disturbance.

References 1. M. Boit, Thermoelasticity and irreversible

thermodynamics, J. Appl. Phys., vol. 27, pp. 240 – 253, 1956.

2. H. W. Lord and Y. Shulman, The generalized dynamical theory of thermoelasticity, J. Mech. Phys. Solids, vol. 15, pp. 299 - 309, 1967.

3. A. E. Green and K. A. Lindsay, Thermoelasticity, J. Elasticity, vol. 2, pp. 1 – 7, 1972.

4. R. S. Dhaliwal and H. Sherief, Generalized thermoelasticity for anisotropic media, Quart. J. Appl. Math., vol. 33, pp. 1 – 8, 1980.

5. D. V. Strunin, On characteristic times in generalized thermoelasticity, ASME J. Appl. Mech., vol. 68, pp. 816 – 817, 2001.

6. D. S. Chandrsekharaiah, Thermoelasticity with second sound-A review, Appl. Mech. Rev., vol. 39, pp. 355 - 376, 1986.

7. C. C. Ackerman, B. Bertman, H. A. Fairbank and R. A. Guyer, Second sound in solid Helium, Phys. Rev. Lett., vol. 6, pp. 789 - 791, 1966.

8. R. A. Guyer and J. A. Kurmhanusal, Thermal conductivity, second sound and phenon by hydrodynamic phenomena in non-metallic crystals, Phys. Rev., vol. 148, pp. 778 - 788, 1966.

9. C. C. Ackerman and W. C. Overtone, Second sound in Helium-3, Jr., Phys. Rev., lett., vol. 22, pp. 764 - 766, 1969.

10. W. Flugge, Viscoelasticity, Blasdell, London, 1967.

11. A. M. Freudenthal, Effect of Rheological behaviour on thermal stress, J. Appl. Physics, Vol. 25, 1-10, 1954.

12. A. S. EI-Karamany, Boundary integral equation formulation for the generalized micro-polar thermoviscoelasticity, Int. J. Engng. Sci., vol 42, pp 157-168, 2004.

13. S. K. Roychoudhuri and S. Mukhopdhyay, Effect of rotation and relaxation on plane waves in generalized thermo-viscoelasticity, Int. J. Math. Math. Sci., vol. 23, pp 497-505, 2002.

14. S. Mukhopadhyay, Effect of thermal relaxation on thermo viscoelastic interactions in an unbounded body with spherical cavity subjected to periodic loading on the boundary, J. Thermal Streeses, vol. 23, pp 675-684, 2000.

15. A. S. EI-Karamany, Boundary integral equation formulation in generalized thermo-viscoelasticity with rheological volume, ASME J. Applied Mech., vol. 70, pp 661-667, 2003.

16. J. N. Sharma and P. K. Sharma, Response of thermoelastic thick plate under lateral loads, J. Thermal Stresses, 2003

17. G. Honig and U. Hirdes, A method for numerical inversion of Laplace transform, J. Comp. Appl. Math, vol 10, pp 113 – 132, 1984.

18. W. H. Press, S. K. Teukolsky, W. T. Velleraing, and B. P. Flannery, Numerical recipes in FORTRAN, 2nd Ed. Cambridge University Press, Cambridge, 1992.

Thermoelastic Wave Propagation in Circumferential Direction of Transversely Isotropic Spherical Curved Plates

Nivedita Sharma and J.N. Sharma

Department of Applied Sciences, National Institute of Technology, Hamirpur – 177005 India. Email: [email protected], [email protected]

Abstract

Spherical plate-like structures are used in pressure vessels, spherical domes of power plants, and in

many other industrial applications. For non-destructive evaluation of such spherical structures, the mechanics of elastic wave propagation in spherical curved plates must be understood. The current literature shows some valuable studies on Rayleigh surface wave propagation in isotropic solids with spherical boundaries. However, the guided wave propagation problem in an anisotropic thermoelastic spherical curved plate, which has not been studied before, is solved for the first time in this paper. The wave propagation, in both isotropic and anisotropic spherical curved plates, is investigated. The differential equations of motion, heat conduction and the stress-free boundary conditions on the inner and outer surfaces of a hollow sphere are approximately solved by a unique technique i. e. the Frobenious method. This solution technique was successfully utilized by the authors for solving the wave propagation problem in cylindrical plates, in their earlier works. Dispersion curves for spherical plates made of isotropic Zinc, steel, and anisotropic composite material are presented as well. 1. Introduction

Plates with spherical curvature have numerous applications in industry. While many researchers have studied the wave propagation problem in cylindrical curved plates , the spherical plates in thermo elasticity have not received due to attention in the non-destructive evaluation (NDE) literature and is studied here. This study will enhance the knowledge base of the NDE community as well as the seismologists and geophysicists where anisotropy effects are significant. Because of the importance of pipie and cylindrical pressure vessel inspections the elastic wave propagation in cylindrical structures has received a great deal of attention [1-5]. However, similar attention has not been given to spherical plates in thermoelasticity, also it has been studied by [7] band also in [9] but in thermoelasticity it has been done in this paper. However, for wave propagation in the circumferential direction, which is essential for nondestructive testing (NDT) of large-diameter pipes, only fewer investigations exists in the literature. in case of thermoelasticity circumferential direction has been discussed in [10].but in spherical coordinates has been discussed in this paper.

2. Formulation and solution of problem We consider a homogeneous spherically

isotropic, thermally conducting sphere of radius R at uniform temperature 0T in the undisturbed state initially. In the spherical polar coordinates (r,φ , t ), the linear governing equations of motion, and heat conduction and constitutive relations in the absence of body forces and heat sources, are given as:

ujji&&ρρσ =,

, φ,, rji = (1)

( )[ ]rrerrr eeeTTCTr

KTr

TK &&&&310,21,3 (12 ββρ φφθθφφ ++=−⎟

⎠⎞

⎜⎝⎛+⎟

⎠⎞

⎜⎝⎛ +

where Tececec rr 1131211 βσ φφθθθθ −++=

Tececec rr 1131112 βσ φφθθφφ −++= (2) Tececec rrrr 3131113 βσ φφθθ −++=

θθσ rr ec442= , φφσ rr ec442= , θφθφσ ec662=

( )2

121166

ccc

−= , ( ) 3112111 ααβ ccc ++= ,

3331133 2 ααβ cc += .

Here ijc , ije ,T, 31 ,αα , eC and ρ are respectively the elastic parameters, strain tensor, temperature change ,linear thermal expansions, thermal conductivities, specific heat and density of the medium. The comma notation is used for spatial derivatives and the superposed dot


denotes time differentiation. and assume the solution of ( ) ( ) ( ) ( ) ( )( ) (3)r ,r w,r v,ru ,,,,

tiimbr eTuuu ωφ

φθ−Θ=

where ( ) ( )rwru , and ( )rΘ represent the amplitude of vibration in the radial and two tangential directions, m is the magnitude of wave vector along wave propagation direction, and ω is the angular frequency. Upon using the solution (3) in equation (1) and (2), we obtain

( )0

242

2

22212 =⎟⎟

⎠

⎞⎜⎜⎝

⎛ −+−+′+′′ v

rbmcc

vr

v ω (4))

( )

0

11222

32

22224

4

=Θ′

⎟⎟⎠

⎞⎜⎜⎝

⎛ −+′

−+⎟

⎟⎠

⎞⎜⎜⎝

⎛ −+++′+′′ w

rc

wr

cimbu

rbmcc

urc

uc ω (5.1)

( )0

121222

2132

2212 =⎟⎟

⎠

⎞⎜⎜⎝

⎛Θ−

+++′

++⎟⎟

⎠

⎞⎜⎜⎝

⎛ +−+

′+′′

ru

rcc

urc

mbwr

bmcrw

w ω

(5.2)

0222

22

=⎟⎠⎞

⎜⎝⎛ ++′+Θ⎟⎟

⎠

⎞⎜⎜⎝

⎛−+Θ+Θ′′

rwimbu

rui

rbmi

rωεω (5.3)

Where 213 cccc −−= where

ii uT

vu01

1

βρω •

=′ , rv

r1

•

=′ω ,

01Tij

ij βσ

σ =′

144 / KcC e=•ω 4402

1 / cCT eρβε =

1/ vaa •=′ ω 44111 / ccc = ,44

111 c

cc =

,

44

122 c

cc =

,44

133 c

cc = ,

44

334 c

cc = ,

21211

66ccc −

= (6)

1

3

ββ

β = ,1

3

KK

K = ,b

vb

1

•

=′ ω , tt •=′ ω ,

0TTT =′

, ρ

4421

cv =

, 1v

cc =′ , •=′

ωωω

.

3. Boundary conditions

The following types of boundary conditions are taken on the surfaces ar = and br = . The lower and upper surfaces ar = and br = of the plate are assumed to be stress free, thermally insulated / isothermal or rigidly fixed, thermally insulated isothermal. Thus (i) Stress free, which implies that

,0,0,0 === φθ σσσ rrrr (7a) (ii) Rigidly fixed, which implies that ,0,0,0 === φθ uuur (7b)

Thermal conditions 0, =+ hTT r (7c)

where h is Biot’s heat transfer coefficient. Here 0→h refers to thermally insulated boundaries and corresponds ∞→h to isothermal surfaces. 4. Extended power series solution Equation (4) which is spherical Bessel equation has the solution

( ) ( ) ( )tiimbrYrBrJrBv ωωω ηη −Φ⎟⎟⎠

⎞⎜⎜⎝

⎛+=

−−exp2

1

1221

11

(8)

Where ( )( ) 4/29 2221

2 bmcc −−=η >0, ηJ and

ηY are the Bessel function of first and second kind

and 11B and 12B are arbitrary constants to be determined by boundary conditions. The rest of the three coupled differential equations and six boundary conditions must be satisfied simultaneously. It is also observed that the the relation (5) is a coupled system of equations in ( ) ( )rwru , and ( )rΘ .This coupled system of

equations has been solved by frobenius method.

( ) ( ) ( )[ ] ( ) ( ) ( )[ ]( )∑∞

=

+=Θ0

,,k

kskkk rsDsBsArrwru ω

(9)

where s is the eigenvalue and coefficients

kk BA , and kD are to be determined by boundary conditions. Substitution solutions (9) in three equations of (5) yields.

( ) ( ) ( ) ( ) 211111 −− +=+ kk

kk

kk

kk BsQDsPBsMAsL

( ) ( ) ( ) ( ) 221222 −− +=+ kk

kk

kk

kk AsQDsPBsMAsL (10)

( ) ( ) ( ) 2313133 −−− ++= kk

kk

kk

kk DsPBMAsLDsR

where the coefficients kiL , k

iM , kiR and k

iP ,

3,2,1=i being functions of material Properties and k they are given by

( ) ( )( ) ( ) 22221341 21 ωbmccckskscsLk −−−++++=

( ) ( )( ) 22132 21 ωcckscimbsL k +++++=

( ) ( )123 ++−= ksisL k εω (11.1)

( ) ( )( ) 221331 11 ω−−−+−+= ccccksimbsM k

( ) ( )( ) 22212 21 ω−−+++= bmckskssM k

( ) 23 ωε bmsM k = (11.2)

( ) ( ),11 −+= kssR k ω ( ) εωisR k −=3 (11.3)

( ) ,2 ωimbsR k =

Thermoelastic Wave Propagation in Circumferential Direction of Transversely Isotropic 303

( ) ( )sPsP kk2

21 =−= ω ,

( ) ( )( )[ ] 2223 1 ωbmkskssP k −+++= (11.4)

0222 =−+ bmss (12.1) 0234 =++++ DCsBsAss (12.2)

Coefficients A, B, C and D are of the form

4

2c

A =

( ) ( ) 4213142

322

3

221 cccccccbmc

B −−−+−−= (13)

( )( ) ( ) 421322

142

32134

2221 ccccbmccccccc

C −−−+−+−−−=

( )( )( ) ( )

⎭⎬⎫

⎩⎨⎧

−−−⎭⎬⎫

⎩⎨⎧

−−−+−−−++

+−= 2132131

2132122441

4

422

121ccc

ccccccccc

bmbmcc

D

The roots of equation (12.2) are denoted by js ,

( )4,3,2,1=j .The roots of equation (12.1) are

denoted by ( )6,5, =js j . Moreover, it is also found that any two of the six real roots do not differ by an integer. Therefore, general solution of equations (5) has the form

( ) ( ) ( ) ( ) ( )∑∑∞

= =

+=Θ0

6

1))(),(,(,,

k j

ksjkjkjkjk

jrsDsBsACrrwru ω (14)

where jkC are arbitrary constants which can be

evaluated by boundary conditions and ( ) ( ) ( )jkjkjk sDsBsA ,, are eigenvectors

corresponding to eigenvalues 6,5,4,3,2,1, =js j and integer k?

When k = 0, one can arrive at the eigenvectors for the first four roots by substituting the eigenvalue

js and corresponding eigenvector in equations (14).

( ) ( ) ( ) ( ) 0,,1 000 === jDjQjBjA B (15)

Where ( )jB sQ are given by 00

10

1 =+ BQML , 002

02 =+ BQML (16)

The eigenvectors for the last two eigenvalues are given by

( ) ( ) ( ) ,0,0,1 000 === jBjAjD (17) Various quantities in equations (15) to (17) are functions of 6,5,4,3,2,1, =js j . When 1≥k , the eigenvectors can be obtained from the recurrence relations (10) in terms of the eigenvectors shown in equations (15) to (17). Thus all the eigenvectors are determined. The six unknowns can be evaluated by six boundary conditions at the outer and inner surfaces.

5. Derivation of secular equation 5. 1. Stress free Invoking the traction free and thermal boundary conditions at the lower and upper surfaces

bar ,= of the spherical plate we obtain the following secular equations

8,7,0 == jiE ij (18)

6.5.4,3,2,1,0 ==′ jiEij (19)

( ) ( )⎟⎟

⎠

⎞⎜⎜⎝

⎛−′= ω

ηωω ηη aJaJE

23

77

,

( )( ) ( ) ( ) ( )( )( ) kskkk asDasBimbcsAckscE +−++++=′ 1

11212311 1 ωβ

( ) ( ) ( )( ) ( ) kskk a

asBkssimbAE +−++=′ 111 11131 ω

( )( ) ksaahksE +++=′ 1151 ω ,

Here 0→h corresponds to thermally insulated boundaries and ∞→h refers to that of the isothermal one. The elements ( )8=jElj of the determinantal equation (18) can be obtained by just replacing the spherical Bessel function of first kind in ( )7=jElj with those of the second kind,

while ( )8=lElj are obtained by replacing a in

( )7=lE lj with b ,and also in equation (19) the elements ( )6,5,4,3,2=′ jE lj

of determinantal equation (19) can be obtained by just replacing

1, =js j in ( )5,3,1=′ lE lj with js

6,5,4,3,2=j while ( )6,4,2' =′ lE slj are obtained by replacing a in ( )5,3,1=′ lE lj

with

b .Equation (19) governs the motion corresponding to the case of toroidal shear where only the θu displacement occurs. These modes of vibrations are not affected by temperature change. 5. 1. Rigidly fixed plate

Invoking the rigidly fixed and thermal boundary conditions at the lower and upper surfaces bar ,= of the plate, we obtain the secular equations as

( ) ( ) ( ) ( )( ) 0=− aYbJbYaJ ωωωω ηηηη (20)

,6,5,4,3,2,1,,0 == jlFij (21)

( )( ) ksk asAF += 1

111 ω , ( )( ) ksk asBF += 1

131 ω ,

( )( ) ksaahksF +++= 1151 ω ,

Here 0→h corresponds to thermally insulated boundaries and ∞→h refers to that of the


isothermal one. The elements ( )6,5,4,3,2=jF lj

of determinantal equation (21) can be obtained by just replacing 1, =js j in ( )5,3,1=lF lj

with

6,5,4,3,2, =js j ,while ( )6,4,2' =lF slj are

obtained by replacing a in ( )5,3,1=lF lj with

b . Equation (20) again governs the motion corresponding to the case of Toroidal shear in the rigidly fixed plate where only θu displacement occurs.These modes of vibrations are not affected by temperature change. 6. Numerical results and discussions

In order to illustrate the analytical development we propose to carry out numerical calculations to compute lowest frequency of stress free with radius of the sphere for a isotropic thermoelastic material (zinc crystal). The physical data is given as:

89.0,20.4 21 == cc , 59.1,31.1 43 == cc ,

,1=K ,89.0=β 111001.5 ×=∗ω . In computer simulation results have been presented graphically in Fig.1 and 2. for thermally insulated and isothermal conditions.

00.20.40.60.8

11.21.41.61.8

2

0 0.5 1 1.5 2Radial distance

Low

est

frequ

ency

.

Insulated(n=0)

Isothermal(n=0)

Fig.1. Variation of lowest frequency of stress free for n=0 with radius of the sphere

00.20.40.60.8

11.21.41.61.8

2

0 0.5 1 1.5 2Radial distance

Low

est

frequ

ency

.

Insulated(n=1)Isothermal(n=1)

Fig.2. Variation of lowest frequency of stress free for n=1 with radius of the sphere

From fig. (1) and (2) we see that with the increaseradial distance the frequency increases.Itcan be concluded that, for a spherical curved plate ,the influence of the boundary condition i.e., for the isothermal case the increase in frequency is more than the insulated one.

Acknowledgement

The authors are thankful to the University Grants Commission, New Delhi for providing financial support via project grant no. F. 30-236/2004(SR) to complete this work.

References [1] D.C. Gazis,.Journal of the Acoustical

America 31(1959a)5. [2] D.C. Gazis,.Journal of the Acoustical

Society of America 31(1959b) 5. [3] O.D.Grace, R.R.Goodman, Journal of the

Acoustical Society of America 39, (1966)173.

[4] A.E.Armenkas, E.S.Reitz, ASME Journal of Applied Mechanics, 168 (1973).

[5] L.M.Brekhovskikh, Soviet Physics––Acoustics 13, 462 (1968).

[6] S.G.Kargl, P.L.Marston, ,. Journal of the Acoustical Society of America 88, (1990)1103 .

[7] S.Towfighi, T. Kundu, Solids and structures, 40(2003) 5495.

[8] H.Singh, J.N.Sharma, Journal of the Acoustical Society of America 77(1985) 1046.

[9] J.N.Sharma, P. K. Sharma,.J. Thermal Stresses 25(2002).

[10] J.N.Sharma, V. Pathania . Solids and structures, 281(2005) 1117.

Author Index

Abbas, Jasim M. 47, 117 Aggarwal, Sanjeev 51, 82 Angadi, Basavwraj 100 Awasthi, Pamita 170 Baghmar, Deoshree 256 Bahniwal, Suman 51 Bains, H.S. 92 Banipal, Parampaul K. 159 Banipal, Tarlok S. 159 Barman, P.B. 57 Basu, S.N. 92 Bedi, R.K. 44, 104 Bharti, Chandrahas 25 Bhattacharya, S. 70 Bijwe, Jayashree 187 Bisht, P.S. 230, 240 Chae, K.H. 100 Chand, Rattan 292 Chand, Subhash 64, 67, 278 Chauhan, Anjana 182 Choi, W.K. 100 Choudhary, R.N.P. 110 Deshpande, S.K. 51, 82 Dharamvir, Keya 260 Dhir, Vaneet 159 Dobal, Anju 38 Dobal, Pramod Singh 38 Dogra, Shilpa 170 Dutta, Alo 25, Dwivedy, Maheshwar 268 Ganeshan, V. 1 Gaur, Bharti 162 Gaur, N.K. 61, 70, 223, 256 Gill, S.S. 246 Gowda, K.V.A. 122 Goyal, N. 28, 53, 107, 141 Goyal, P.S. 51 Gupta, Ankur 243 Gupta, D.C. 256, 265 Gupta, R. 33 Gupta, Shikha 114, 141 Gupta, Vinay 126 Jaiswal, Shivendra Kumar 11 Joshi, Virshali 230 Kaith, B.S. 167, 204 Kalia, Susheel 167

Kanjilal, D. 82 Kapil, Atul 64 Katiyar, R.S. 38 Katyal, S.C. 134 Kaur, D. 283 Kaur, Davinder 104 Kaur, Inderjeet 64, 176 Kaur, Harkiran 260 Kaur, Nupinderjeet 70, 223 Khanna, Ashwarya Jyoti 200 Khanna, N. Deepika 176 Kulshrestha, Subhra 265 Kumbhakar, P. 145 Kumar, A. 297 Kumar, Akshay 114, 141 Kumar, Jitendra 11 Kumar, Mukesh 187 Kumar, R. 1 Kumar, Rajeev 44 Kumar, Rajesh 77 Kumar, Raman 85 Kumar, S. 82 Kumbhakar, P. 129, 145 Lamba, V.K. 243, 246 Mahajan, R.K. 170 Mainika 253 Maiti, Santanu K. 235 Mamgain, Ravindra 92 Manna, A. 92 Mathur, Preeti 6, 15, 19 Mehra, Sonia 265 Mehta, Charita 47, 117 Mitra, A. K. 129, 145 Mohan, Rajneesh 61, 70, 223 Mohan, S. 122 Mollah, S. 1 Murty, B.S. 110 Mustafa, Falah Ibrahim 107, 114, 141 Nagarkar, M.P. 219 Narula, Avnish 273 Navthar, R.R. 219 Negi, N.S. 126 Nidhi 82 Parashar, Kajal 110 Parashar, S.K.S. 110 Pathak, Dinesh 104

306 Author Index

Patnaik, Amar 268 Pradhan, S.K. 268 Rai, K.N. 33 Ram, P. Raja 240 Ranjan, K. 260 Rangra, V.S. 85 Ranjta, Shabnam 204 Rao, K. Narasimha 122 Roy, R.N. 129 Saini, G.S.S. 28, 47, 53, 117,

141 Sarkar, R. 145 Sen, Vikas 96 Shama, Anjali 194 Sharma, Ajay 6, 15, 19 Sharma, Annu 51, 82 Sharma, Avneesh 246 Sharma, Indu 278 Sharma, Ishu 57, 134 Sharma, J.N. 278, 283, 288, 292,

301 Sharma, Jeewan 28, Sharma, K.K. 297 Sharma, N. 15, 19 Sharma, Nisha 292 Sharma, Nitu 246 Sharma, Nivedita 301 Sharma, P.K. 297 Sharma, Pankaj 57, 134 Sharma, Raman 246 Sharma, Rishi 96 Sharma, Sudhir Kumar 122 Sharma, Tanu 82 Sharma, Tanuj 19 Sharma, Vimal 73 Sharma, Vineet 138 Shukla, D.K. 1 Singh, A.P. 100 Singh, B. 89

Singh, D.D.N. 153 Singh, F. 1 Singh, Gurinder 53, Singh, Iqbal 44, 149 Singh, K. 126 Singh, K.C. 265, Singh, Kiran 61, 70 Singh, M. 6, 15, 19 Singh, R.K. 61, 70, 223 Singh, Sadhna 256 Singha, A.S. 167, 194, 200, 208,

212 Sinha, M.M. 216 Sinha, T.P. 25, Srivastava, Sunita 227 Tankeshwar, K. 227 Tarshika, P.S. 89 Taunk, Manish 67 Thakur, Anup 138 Thakur, Atul 6, 15, 19 Thakur, Nagesh 73,77, 253 Thakur, P. 100 Thakur, Vijay Kumar 208, 212 Thakur, Vikas 240 Tiwari, D.C. 96 Tiwari, R.K. 182 Tiwari, Rachana 182 Tiwary, C.S. 129 Tripathi, S.K. 28, 47, 53, 57, 107,

114, 117, 141 Verma, K.C. 126 Verma, Munish 243, 249 Verma, U.P. 230, 240 Vishwas, M. 122 Walia, V. 288, 297 Yadav, S.K. 33 Yadav, Shyamjeet 153 Zaware, R.N. 219

raim-08 prepages to index

Documents

national conference

innovative materials

various aspects of materials

recent advances

hp india

area of materials

authorities of nit hamirpur

new delhi