Thin Solid Films 545 (2013) 601–607

Contents lists available at ScienceDirect

Thin Solid Films

j ourna l homepage: www.e lsev ie r .com/ locate / ts f

Superconducting niobium nitride thin films by reactive pulsedlaser deposition

Y. Ufuktepe a,⁎, A.H. Farha b,h, S.I. Kimura c,d, T. Hajiri c,e, K. Imura c,f, M.A. Mamun g,F. Karadag a, A.A. Elmustafa g, H.E. Elsayed-Ali b

a Department of Physics, Cukurova University, Adana, 01330, Turkeyb Department of Electrical and Computer Engineering and the Applied Research Center, Old Dominion University, Norfolk, VA 23529, USAc UVSOR Facility, Institute for Molecular Science, Okazaki, 444-8585, Japand School of Physical Sciences, the Graduate University for Advanced Studies (SOKENDAI), Okazaki, 444-8585, Japane Graduate School of Engineering, Nagoya University, Nagoya, 464-8601, Japanf Department of Physics, Nagoya University, Nagoya, 464-8601, Japang Department of Mechanical and Aerospace Engineering and the Applied Research Center, Old Dominion University, Norfolk, VA 23529, USAh Department of Physics, Faculty of Science, Ain Shams University, Cairo 11566, Egypt

⁎ Corresponding author. Tel.: +90 3223386084; fax: +E-mail address: ufuk@cu.edu.tr (Y. Ufuktepe).

0040-6090/$ – see front matter © 2013 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.tsf.2013.08.051

a b s t r a c t

a r t i c l e i n f o

Article history:Received 20 June 2012Received in revised form 18 April 2013Accepted 9 August 2013Available online 17 August 2013

Keywords:NbNPulsed laser depositionThin filmsX-ray spectroscopy

The structural, electronic, and nanomechanical properties of cubic niobium nitride thin films were investigated.The films were deposited on Si(100) under different background nitrogen gas pressures (26.7-66.7 Pa) at con-stant substrate temperature of 800 °C by reactive pulsed laser deposition. Our results reveal that the NbNx

films exhibit a cubic δ-NbNwith strong (111) orientation and highly-oriented textured structures.We find nitro-gen background pressure to be an important factor in determining the structure of the NbNx films. The depen-dence of the electronic structure as well as that of the superconducting transition temperature (Tc) on thenitrogen gas background pressure is studied. A correlation between surface morphology, electronic andsuperconducting properties is found for the deposited NbNx thin films. The highly-textured δ-NbN films havea Tc up to 15.07 K. Nanoindentation with continuous stiffness method is used to evaluate the hardness andmodulus of the NbNx thin films as a function of depth. The film deposited at nitrogen background pressure of66.7 Pa exhibits improved superconducting properties and shows higher hardness values as compared to filmsdeposited at lower nitrogen pressures.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

The superconducting nature of niobium nitride (NbNx) thin filmsmakes thempotential candidates for use in superconducting electronics,such as tunnel junctions [1,2], coatings for superconductive cables [3],and single photon detectors [4–6]. Moreover, good mechanical proper-ties, such as their high hardness [7,8] and corrosion resistance [9,10],makes NbNx suitable as a protective layer in active environments suchas in turbine engines and spacecrafts [11,12]. The superconducting andmechanical properties of NbNx are strongly affected by the Nb-N com-position and the crystal structure. The Nb-N system presents severalnitrides and the corresponding phase diagram is complex. As the nitro-gen content in the system is changed, various and interrelated types ofcrystallographic structures are formed. In fact, NbNx is often observedto be in mixed phases. Most studies related to NbNx deal with itssuperconducting properties. The highest transition temperature mea-sured is ~17.3 K for the delta cubic phase [13]. To achieve high Tc, it is

90 3223386070.

ghts reserved.

necessary to find the conditions for growth of highly-textured NbNx

films with the structure mainly in the cubic phase.Pulsed laser deposition (PLD) is a well-established and highly flexi-

ble thin-film growth technique used to grow various kinds of materials,including nitrides and other semiconductors. One of the importantadvantages of the PLD method is the stoichiometric transfer from thetarget to substrate and the ability to introduce gases in the chamberfor reactive deposition. In this paper, NbNx films were deposited onSi(100) substrate at different nitrogen gas pressures by using reactivePLD. Depending on the nitrogen pressure, NbNx with different phasecompositions and different Tc were grown at 800 °C substrate tempera-ture. In addition to measuring Tc of the NbNx films grown at differentbackground nitrogen pressures, the morphology, electronic, and me-chanical properties of the NbNx films were studied.

2. Experimental details

NbNx films were grown on Si(100) single-crystal substrate by ab-lating a 1-inch diameter Nb target (99.995% pure). The chamber wasequipped with a turbo-molecular and ion pump, operated at a basepressure of ~1.3 × 10−7 Pa. A pulsed Nd:YAG laser beam (wavelength

Fig. 1.XRD patterns of NbNx thinfilms deposited on Si(100) substrate at different nitrogenbackground pressures.

Fig. 2.Variation of lattice parameter and crystallite size of (111) planewith nitrogen back-ground pressure. The solid lines are drawn as a guide for the eye.

602 Y. Ufuktepe et al. / Thin Solid Films 545 (2013) 601–607

λ = 1064 nm, repetition rate 10 Hz, pulse duration 40 ns) was fo-cused with a 50 cm focal length lens at 45° onto a rotating (25 rpm)Nb target. The distance between the target and the substrate wasfixed at 6 cm. The nitrogen gas operating pressure was achieved byfilling the vacuum chamber with the right amount of gas and keepingit in static condition.

The Si(100) substrate was 3.0 × 12 × 0.5 mm3, p-type boron dopedwith resistivity ranging between 0.060 and 0.075 Ω cm and had amiscut angle 0.38°. The Si substrate was chemically etched using modi-fied Shiraki method [14] before being loaded into the chamber. In orderto deposit cubic structure NbNx films, we optimized the PLD parameterssuch as the laser energy density, pulse repetition rate, and target-to-substrate distance. The highly-oriented crystalline structure of ourNbNx films was obtained with 15 J/cm2 laser fluence and 800 °C sub-strate deposition temperature. The thickness of the deposited films is25 ± 5 nm, as measured by performing cross-sectional transmissionelectron microscopy measurements.

X-ray diffraction (XRD) was made using graphite-monochromatedCuKα radiation on a Bruker-AXS three-circle diffractometer, equippedwith a SMART Apex II CCD detector. X-ray photoelectron spectrosco-py (XPS) was performed on 100-mm radius hemispherical photo-electron analyzer (VG Scienta SES-100) with Mg Kα X-ray radiation(hυ = 1253.6 eV). The base pressure of the photoemission chamberwas b2.0 × 10−8 Pa. Surface cleaning of the samples was performedat room temperature by using Ar+ sputtering. The sputtering wascarried out at 3 kV and 10 mA cm−2 beam current density, with anargon partial pressure of 1.5 × 10−5 Pa. The energy scale of the X-rayphotoemission spectrum is calibrated by the binding energy of Au 4flevel. The morphology of the films was examined using a Digital Instru-ments Dimension 3100 atomic force microscope (AFM). AFM imageswere taken in air using tapping mode. In addition, conventional four-point probe method measurements of thin films with a temperatureaccuracy of 0.01 K were used to obtain the superconducting transitiontemperature. The nanoindentation experiments probe the hardnessand elasticmodulus of thefilmswhich is correlated to the structural, elec-tronic, and superconducting properties of the thinfilms. Nanoindentationis an effective technique to investigate the mechanical properties ofthin films and nano materials. This technique is capable of measuringthe moduli and hardness of thin films as a function of depth [15–17].Nanoindentation experiments were conducted using a Nano Indenter®

XP in conjunction with a continuous stiffness measurement (CSM)equipped with a three-sided diamond Berkovich indenter tip. Duringthe CSM indentation testing, a small and well-controlled oscillation is in-troduced into the normal loading sequence of the nanoindenter, whichenables the CSM to monitor the contact depth and the contact stiffnessthroughout the indentation loading. Hence, the CSMmaintains the mea-surement of material properties as a continuous function of depth fora single indentation, as detailed elsewhere [18,19]. A G-Series CSMStandard Hardness, Modulus, and Tip Cal method was adopted forthe indentation experiments in this study. The indenter penetratedthe sample surface with a harmonic displacement oscillation target of1 nm at a frequency of 45 Hz to a depth of 500 nm with a strain rateof 0.05 per second. Surface detection criterion was set at 160 N/m forthese tests. A total of 10 indents were performed on each sample atdifferent locations.

3. Results and discussion

3.1. Structure and surface morphology

The effect of nitrogen background pressure on the crystallinity ofNbNx thin films was examined by XRD measurements. Fig. 1 showsthe XRD patterns of NbNx films prepared at different nitrogen back-ground pressures. The substrate temperature and laser energy densityremained constant and only the nitrogen pressure was varied from26.7 to 66.7 Pa. The influence of nitrogen pressure can be seen in

Fig. 1. Two different crystalline structures are observed; the films werehighly textured and were indexed with mainly the cubic phase ofthe δ-NbN [20] and the tetragonal γ-Nb4N3 [21] phase. Variations ofthe N/Nb ratio in the films may have contributed to the changes in thecrystal structure for different nitrogen pressures.

Fig. 2 shows the variation of the lattice parameter and the crystallitesize of the cubic phase calculated on the basis of the (111) peak as afunction of nitrogen background pressure. The mean crystallite size ofδ-NbN was determined by the Debye–Scherrer formula [22]. The calcu-lated values are in the range of 8.46 to 9.31 nm. The cubic δ-NbN phasecan formover a specific stoichiometric range of 0.85 b x b 1.06with thechange of lattice parameter a from 4.37 to 4.39 Å. The calculated valuesare in the range of 4.365 ≤ a ≤ 4.395 and are consistentwith the previ-ous result [23]. Different lattice parameters were observed becausecrystal imperfections affects the lattice parameter; thehigher the vacan-cies in NbNx films are, the lower the a values [23,24].

The size of the (111) crystallite size in the film decreases with in-creasing background nitrogen pressure. The nitrogen background pres-sure reduces the kinetic energy and flux of the ablated materials.Therefore, for the lower deposition pressure, the surface mobility ofadatoms is enhanced due to higher kinetic energy of atoms in the PLDflux. This enhance surface mobility results in larger crystallite size. InFig. 1, the integrated area of the (111) peak for the sample grown at26.7 Pa is higher than that for the other samples. Deposition at thelower pressures gives more dense and highly (111) oriented NbNx

Fig. 4. Dependence of surface roughness RMS values on nitrogen background pressure.

603Y. Ufuktepe et al. / Thin Solid Films 545 (2013) 601–607

film consisting of larger crystallites. Higher background gas pressuregives rise to not only lowering the kinetic energy of the ablated Nb butalso the confinement of the plasma plume. This results in an increasein the probability of multiple collisions between ablated materialsthat yields formation of small crystallite size [25]. As a consequence,higher molecular nitrogen pressure is causing lower deposition rateand resulting in less nitrogen ratio in the NbNx film.

AFM images of films prepared at different nitrogen backgroundpressures are shown in Fig. 3. The scanning area was 2 × 2 μm2. TheAFM image of the film grown at 26.7 Pa consists of triangular islandsof 100–200 nm sizes and heights of 15 nm, as shown in Fig. 3(a).As the nitrogen pressure was increased to 53.3 Pa, the island size andmorphologies became more irregular in shape and with average heightof ~20 nm.

For growth at a nitrogen pressure of 66.7 Pa, the size of islands in-creases to submicrometer as shown in Fig. 3(c). However, the averageheight of grains is ~21 nmwhich is about the same as for film depositedat 53.3 Pa. When we used 66.7 Pa nitrogen pressure, the island size in-creased significantly compared to lower pressures. At lower nitrogenpressures, the niobium andmolecular nitrogen flux incident on the sub-strate surface is increased. The higher nucleation rate can cause smallerislands to grow. The reduced nitrogen content in the thin film withincreased molecular nitrogen background could be due to reducedatomic nitrogen flux reaching the surfacewith pressure. As the nitrogenpressure is increased to 66.7 Pa, the diffusion length of the atomic nitro-gen, generated from electron collision in the plasma plume with thebackground molecular nitrogen, is reduced and recombination rate inincreased.

To check surface roughness, diagonal line scans on the AFM imageswere obtained as shown in Fig. 3(d). The root mean square surfaceroughness (RMS) was obtained from line scans of AFM images. The sur-face roughness as a function of nitrogen background pressure is shownin Fig. 4. The surface roughness increases with increasing the nitrogenbackground pressure. The film grown at 26.7 Pa showed an RMS valueof 4.4 nm. Roughness increased to 9.8 and 11.2 nm as the nitrogen gaspressure increased to 53.3 and 66.7 Pa, respectively. The increase in sur-face roughness is expected with increasing the background pressure.In PLD, the background pressure plays a crucial role in controlling thekinetic energy of adatoms and their diffusion into the surface. Low am-bient gas pressure contributes to higher adatoms kinetic energy and this

0

0 2.001.000

2.00

1.00

0 2.001.000

2.00

1.00

0 nm

50 nm

100 nm

0 nm

50 nm

100 nm

µm

µm

Fig. 3. The 3D AFM images of films grown at (a) 26.7 Pa, (b) 53.3 Pa, and (c

provides low surface roughness for the NbNx films. The surface rough-ness results are consistent with that reported for β-Nb2N films grownby PLD on Nb substrates for similar nitrogen gas pressure range [26].

3.2. Electronic structure by X-ray photoemission spectroscopy

X-ray photoemission spectroscopy was used for electronic structureanalysis. Fig. 5 shows XPS spectra of Nb 3d core levels for NbNx films de-posited at various nitrogen background pressures. The spectra are nor-malized so that maximum peak intensities are equal in each spectrum.Binding energies are givenwith respect to the Fermi level. TheXPS spec-tra of the three NbNx films show a strong pair of peaks due to Nb 3d3/2and 3d5/2 doublets. The corresponding peak positions are summarizedin Table 1.

Comparing NbNx film spectra with that from pure Nb (205.5 and202.3 eV), it appears that Nb 3d3/2 and 3d5/2 peaks are shifted to higherbinding energies as a result of Nb–N bonding, indicating the transfer ofelectrons fromniobium tonitrogen. The peak at ~210 eV corresponds to

2.001.00

0 nm

50 nm

100 nm

µµm0

2.00

1.00

) 66.7 Pa; (d) Line scans of AFM images at different nitrogen pressures.

Fig. 5. Nb 3d XPS spectra of NbNx films deposited at various N2 background pressures. Fig. 6. XPS spectra of Nb 3p core level for NbNx films deposited at different backgroundpressures.

604 Y. Ufuktepe et al. / Thin Solid Films 545 (2013) 601–607

niobium oxide which suggests that the samples were contaminated byoxygen after deposition as theywere exposed to air. The binding energyvalues of Nb 3d3/2 and 3d5/2 doublets were previously determinedfor NbNx, NbN(1 − y)Oy and NbO [27,28]. Our results indicate that the3d level of all the deposited films is not supporting the structure of aniobium oxide or oxynitride and the spectra are consistent with thatof NbNx [29–32].

Nb 3p core level spectra of NbNx films are shown in Fig. 6. The spec-tra consist of two 3p3/2 and 3p1/2 doublets and strong N 1s peak. Thecorresponding binding energy values are given in Table 2, which areconsistent with the previous NbNx studies [30,31]. Note that the 3pbinding energy doublets are shifted to the lower energy side as com-pared to the Nb 3p peak of pure Nb. This shift is in agreementwith elec-tron donation to the nitrogen atoms and a clear indication of the NbNx

structure.The intensity of photoelectron peaks is directly related to the con-

tent of Nb and N in the NbNx film and can be used for determinationof the chemical composition of the surface. We have calculated relativeconcentration of nitrogen in NbNx as a function of background pressureby using the ratio of the background corrected integrated areasunder the N 1s and Nb 3d peaks in the XPS spectra and normalizingwith atomic cross-section of each level at Mg Kα radiation [33]. Calcu-lated x (x = N/Nb) values are 0.90, 0.88 and 0.80 for 26.7, 53.3 and66.7 Pa samples, respectively, which are in agreement with previouslyreported cubic and tetragonal phases of the Nb–N systems [34,35].The content of N in NbNx films is slightly decreasing as the tetragonalphase becomes visible with the increase of nitrogen backgroundpressure during deposition. It is well known that Nb4N3 has a distortedδ-NbN structure and the δ–γ phase transition could be observed whenx in NbNx has changed [11,36]. Shirley background corrected N 1s spec-trum showed that the peak position slightly shifted to lower bindingenergy (0.2 eV) when the nitrogen background pressure changedfrom 26.7 to 66.7 Pa. Broadening of the N 1s lines was observed due toincreasing of the N/Nb ratio (x) and is related to the greater number of

Table 1XPS binding energies of Nb 3d in NbNx films deposited at different N2 backgroundpressures.

Background N2

pressure (Pa)Nb 3d5/2(±0.05) (eV)

Nb 3d3/2(±0.05) (eV)

Nb―N bond(±0.05) (eV)

26.7 204.00 206.81 209. 8353.3 204.09 206.92 209.6466.7 204.08 206.90 209.72

sites occupied by N atoms in NbNx films [37]. In Table 2, a similar effectwas observed for the energy separation between N 1s and Nb 3d5/2peaks (Δ). We find a decrease in Δ due to an increase in charge transferfrom Nb to N when higher nitrogen background pressure is used. It isreported in the literature that the binding energy value Δ associatedwith N 1s and Nb 3d indicates the covalency level of the NbNx film [38].

3.3. Ultraviolet photoelectron spectroscopy (UPS) of NbN valence band

Fig. 7(a) shows the UPS valence band (VB) spectra of the three dif-ferent NbNx films. The valence band spectra were recorded using He IIradiation (40.8 eV) from a discharge lamp. The VB spectra are mainlydominated by two peaks below the Fermi level. Strong and weak hy-bridization of Nb 4d and N 2p levels are about 1 and 6 eV, respectively.In the case of the sample grown at 26.7 Pa, the peak in the region of 6 eVis narrowed and shifted to lower binding energy in comparisonwith theother two films, while the emission just below the Fermi edge is charac-terized by broad emission.

In order to compare two different phases of NbNx films, we carriedout the VB spectral subtraction of the samples grown at 26.7 and66.7 Pa nitrogen as shown in Fig. 7(b). The difference between the spec-tra of films grown at 26.7 and 66.7 Pa exhibits positive and negativepeaks at 3.5 and 8 eV, respectively, due to the difference in nitrogencontent and NbNx phases of these two films.

In order to better understand the valence band electronic structureof theNbNx system, we have calculated the cubic and tetragonal densityof states (DOS) of NbNx as shown in Fig. 8. Our density functional theory(DFT) calculations were carried out using the Abinit code [39], onthe basis of DFT and Many-Body Perturbation Theory. It is clear fromthe figure that the electronic structures of these two phases are quitesimilar and two distinctive regions are visible by the peak centered at−6 eV and the second region between 0 and 4 eV. The first region is be-tween−4 to−8 eV andmainly arises from the degeneracy between N2p and Nb 4d states. The overlap of the nitrogen p and niobium d levels

Table 2Nb 3p doublet binding energies of NbN films deposited at different N2 backgroundpressures.

BackgroundN2 pressure(Pa)

Nb 3p3/2(±0.05) (eV)

Nb 3p1/2(±0.05) (eV)

N 1s(±0.05) (eV)

Δ = N 1s − Nb 3d5/2(±0.05) eV

26.7 362.47 378.00 397.50 193.2953.3 362.63 378.10 397.29 193.2166.7 362.85 378.20 397.30 193.17

Fig. 7. (a) Valence band spectra of NbNx thin films. (b) In each spectrum, the UPS signal ofthe Fermi edgewas set to the same point. Difference VB spectrum obtained by subtracting26.7 and 66.7 Pa spectra shown at the bottom.

Fig. 8. Calculated partial and total density of state (DOS) for (a) cubic δ-NbN and(b) tetragonal γ-Nb4N3. Fermi levels are set to 0 eV.

Fig. 9. Resistance versus temperature of NbNx films deposited at different backgroundpressures.

605Y. Ufuktepe et al. / Thin Solid Films 545 (2013) 601–607

is associated with the covalent bonding between metal and non-metalelements. The second region, just below and above the Fermi level,originates mainly from the Nb 4d state. In Fig. 8(a) the Fermi level is lo-cated just near a pseudo gap of the DOS, but in the case of the tetragonalγ-Nb4N3 phase, the pseudo gap is split in bonding andantibonding statesnear the Fermi energy (Fig. 8b) and can lead to amore pronounced cova-lency and higher stability [40]. Larger covalent contribution to the Nb–Nbonding enhances the mechanical properties of γ-Nb4N3.

3.4. Superconductor phase transition temperature

In order to determine the superconductor behavior of the NbNx

films, resistance measurements were performed. The temperature de-pendence of the electrical resistivity for films deposited at differentnitrogen background pressures is shown in Fig. 9. All three sampleshad the transition to the superconducting state in the temperaturerange of 7.66–15.07 K. Different transition temperatures are related tothe crystalline phases of the NbNx films.

Typical residual resistance ratio values for the films are less thanunity. The highest superconducting transition temperature (Tc) of15.07 K is observed for NbNx film deposited at 66.7 Pa nitrogen back-ground pressure. Our XPS analysis showed that all films are nearlystoichiometric in composition with very small change of x in NbNx

observed with the increase of nitrogen background pressure. Lowestsuperconducting transition temperature observed for the film depos-ited at 26.7 Pa can be understood on the basis of the rich cubic phaseof the film, which has a large amount of vacancies. Cubic δ-NbN has

Fig. 11. Hardness versus depth of indentation.

606 Y. Ufuktepe et al. / Thin Solid Films 545 (2013) 601–607

the highest superconducting transition temperature. It should benoted that we did not observe pure cubic phase at 26.7 Pa nitrogenpressure. The presence of mixed phases affects the critical tempera-ture. As the nitrogen background pressure is increased, the numberof vacancies in the film decreases, the lattice constant becomes larger,and the superconductor properties of the NbNx film improves.

The superconducting transition temperature (Tc) for each film in-creases from 7.66 to 15.07 K by varying the nitrogen background pres-sure from 26.7 to 66.7 Pa as shown in Fig. 10, while the resistivitymeasured at 20 K increases from 60 × 10−3 to 120 × 10−3 Ω cm. Ourexperimental results can be understood on the basis of the crystal andelectronic structure of the NbNx films. The microstructure of the NbNx

films has great effect on the superconducting Tc even when the stoichi-ometry is close. Changes in texture and granular structure of the NbNx

films influence the Tc of the films [32]. The lower Tc observed for thesample deposited at 26.7 Pa is a consequence of mainly the film struc-ture being δ-NbN phase with a large number of vacancies. For deposi-tion at 66.7 Pa nitrogen background pressure, the film had a mixedphase structure of δ-NbN and γ-Nb4N3 with reduced vacancies. Thisfilm exhibited a higher Tc value. Moreover, for deposition at 66.7 Pa ni-trogen, the lattice parameter becomes very close to the bulk (4.393 Å)of fcc δ-NbN which favors higher Tc. It was reported that metal andnonmetal vacancies are effective in reducing Tc in NbNx [40]. When asufficient number of vacancies are removed from NbNx, electron trans-fers to the conduction band and increases the lattice size. As a result ofthe effect on the electron concentration, Tc increases.

From Fig. 9, for all three films the resistivity curves show a negativetemperature coefficient of resistivity measured above ~16 K. The trendshows that the slope is most negative for films deposited at 66.7 Paand this slope becomes less negative as the background nitrogen pres-sure was reduced. These slope measurements correlate well with theobserved variation of Tc between the three films and confirms previousobservations that filmswith negative slope of resistivity versus temper-ature showhigh Tc [32]. This slope of the resistivity curve is an indicationof film transport quality; a more negative slope indicates reduced elec-tron scattering by grain boundaries and point defects such as vacancies.

3.5. Evaluation of hardness and modulus using nanoindentation

The hardness is defined as

H ¼ LA

ð1Þ

where L is the indentation load and A is the projected area of the hard-ness impression. The nanohardness of NbNx thin films on Si substrate

Fig. 10. Superconductor transition temperature (Tc) and the resistivity measured at 20 K(ρ20) as a function of nitrogen background pressure.

deposited at different ambient nitrogen pressures was determinedand the results are presented in Fig. 11. The Meyer hardness is definedas the maximum load (from the load-depth curve) over the projectedarea. Themeasured hardness has an average of 12 GPa for deep indents,which represents the hardness of the Si substrate. At shallow depth ofindentation, i.e., a depth of 20 nm, the hardness increases to an averagevalue of 14, 16, and 18 GPa for the samples deposited with nitrogenpressures of 26.7, 53.3, and 66.7 Pa, respectively. The average hardnessof the Si substrate agrees well with literature-reported values [41].

The reduced modulus, Er can be defined as

1Er

¼ 1−ν2s

Esþ 1−ν2

d

Ed: ð2Þ

where Es and Ed are Young's moduli, and vs and vd are Poisson's ratios ofthe specimen and the indenter, respectively. Poisson's ratio and modu-lus of the diamond tip are vd = 0.07 and Ed = 1137 GPa, respectively.The modulus versus depth of indentation is shown in Fig. 12. It is alsoobserved that the moduli remain flat at an average of 180–200 GPafor the three samples from deep to shallow indents to an average

Fig. 12.Modulus versus depth of indentation.

607Y. Ufuktepe et al. / Thin Solid Films 545 (2013) 601–607

shallow depth of indentation of 20 nm. This can clearly be verified fromthe sub-plots of Figs. 11 and 12 of hardness and modulus for data be-tween 0 and 50 nm depth of indentation. The error bars representthree times the standard error. The average modulus of the Si substrateis 180 GPa, which is considered higher than the average modulus forcubic Si reported in the literature due to pile-up [42].

The superconducting and mechanical properties of NbNx films arestrongly affected by the Nb-N composition and the crystal structure.The hardness and elastic modulus correlate with the variation of theconcentration of the hexagonal δ′-NbN phase [2,10]. The films withhigh concentration of the hexagonal δ′-NbN phase exhibit high hard-ness and high elastic modulus [43]. Since FCC δ-NbN possesses superiorsuperconducting properties, it is therefore essential to control thebiaxial compressive stress in the film that might produce phase trans-formation [44]. The correlation between the nanomechanical propertiesand Tc of theNbNxfilmswas previusly attributed to number of nonmetalvacancies found in the film. As number of vacancies increases, the num-ber of chemical bonds is reduced and this decreases the strength ofthe material [32]. Work done on group IVb nitrides (TiNx, ZrNx andHfNx) related the deceases in hardness and elastic module to the in-crease in the numbers of nonmetal vacancies in these films [45,46].The film deposited at 66.7 Pa nitrogen pressure, which had highestTc, also showed highest hardness values compared to the other twosamples. This increased hardness is related to the reduced vacancydensity and the increase in the γ-Nb4N3 phase mixed with δ-NbNfor deposition at 66.7 Pa nitrogen.

4. Conclusions

We have studied niobium nitride thin films grown on Si(100) at dif-ferent nitrogen background pressures by PLD. The reactive PLD processyields stoichiometric NbNfilms, with high Tc 15.07 K at 66.7 Pa ambientgas pressure, onto a silicon substrate held at 800 °C. Changing of thebackground gas pressure during deposition shows significant effectson the phase orientations, morphologies, and superconducting transi-tion temperature. The lowest Tc values of 7.66 K can be attributed tothe existence of vacancies in cubic NbNx. A systematic increase in latticeparameters with nitrogen background pressure along with decrease inthe size of the crystallite was observed and attributed to the secondtetragonal NbNx phase. The binding energy shift in Nb core level isdirectly related to the occupied number of electrons. Nevertheless,highly textured NbNx layers with good superconducting propertiescan be prepared on silicon using PLD. It is important to optimize alldeposition parameters of PLD. The result reveals that controlling thenitrogen pressure in the deposition chamber in a suitable range couldimprove the crystalline quality to show better superconductivity.

Acknowledgments

One of the authors (Y.U.) thanks the Cukurova University, theCouncil of Higher Education of Turkey, and the Japan Society for thePromotion of Science (JSPS) for their support. H.E.A. acknowledgessupport by the National Science Foundation Grant Nos. DMR-9988669and MRI-0821180. A.H.F. was supported by a Jefferson Lab fellowship.

References

[1] L.E. Toth, Transition Metal Carbides and Nitrides, Academic Press, New York, 1971.[2] M. Benkahoul, E. Martinez, A. Karimi, R. Sanjines, F. Levy, Surf. Coat. Technol.

180–181 (2004) 178.[3] V.E. Shaternik, S.Y. Larkin, T.A. Khachaturova, Physica C 435 (2006) 96.[4] K.L. Rutherford, P.W. Hatto, C. Davies, I.M. Hutchings, Surf. Coat. Technol. 86–87

(1996) 472.[5] S.P. Chockalingam, M. Chand, J. Jesudasan, V. Tripathi, P. Raychaudhuri, Phys. Rev. B

77 (2008) 214503.[6] G.N. Gol'tsman, A.D. Semenov, Y.P. Gousev,M.A. Zorin, I.G. Godidze, E.M. Gershenzon,

P.T. Lang, W.J. Knott, K.F. Renk, Supercond. Sci. Technol. 4 (1991) 453.[7] S.A. Barnett, M. Shinn, Annu. Rev. Mater. Sci. 24 (1994) 481.[8] M.A. Mamun, A.H. Farha, A.O. Er, Y. Ufuktepe, D. Gu, H.E. Elsayed-Ali, A.A. Elmustafa,

Appl. Surf. Sci. 258 (2012) 4308.[9] G. Oya, Y. Onodera, J. Appl. Phys. 47 (1976) 2833.

[10] N. Cansever, M. Danisman, K. Kazmanli, Surf. Coat. Technol. 202 (2008) 5919.[11] W. Lengauer, M. Bohn, B. Wollein, K. Lisak, Acta Mater. 48 (2000) 2633.[12] R. Sanjines, M. Benkahoul, C.S. Sandu, P.E. Schmid, F. Levy, Thin Solid Films 494

(2006) 190.[13] K.S. Keskar, T. Yamashita, Y. Onodera, Jpn. J. Appl. Phys. 10 (1971) 370.[14] A. Ishizaka, Y. Shiraki, J. Electrochem. Soc. 133 (1986) 666.[15] L. Xiaodong, B. Bhushan, Thin Solid Films 340 (1999) 210.[16] W.C. Oliver, C.J. McHargue, Thin Solid Films 161 (1988) 117.[17] M.F. Tambwe, D.S. Stone, A.J. Griffin, H. Kung, Y. Cheng, M. Nastasi, J. Mater. Res. 14

(1999) 407.[18] W.C. Oliver, G.M. Pharr, J. Mater. Res. 7 (1992) 1564.[19] W.C. Oliver, G.M. Pharr, J. Mater. Res. 19 (2004) 3.[20] Powder Diffraction File 00-038-1155, International Center for Diffraction Data,

PDF-2/Release 2004, 2004.[21] Powder Diffraction File 01-071-0096, International Center for Diffraction Data,

PDF-2/Release 2004, 2004.[22] B.D. Cullity, Elements of X-ray Diffraction, Addison-Wesley, New York, 1956.[23] G. Brauer, H. Kirner, Z. Anorg. Allg. Chem. 328 (1964) 34.[24] W. Lengauer, Surf. Interface Anal. 15 (1990) 377.[25] D. Riabinina, E. Irissou, B. Le Drogoff, M. Chaker, D. Guay, J. Appl. Phys. 108 (2010)

034322.[26] A.H. Farha, A.O. Er, Y. Ufuktepe, G. Myneni, H.E. Elsayed-Ali, Surf. Coat. Technol. 206

(2011) 1168.[27] N. Ozer, D.G. Chen, C.M. Lampert, Thin Solid Films 277 (1996) 162.[28] K.S. Harvey, J.S. Zabinski, S.D. Walck, Thin Solid Films 303 (1997) 238.[29] G. Jouve, C. Séverac, S. Cantacuzène, Thin Solid Films 287 (1996) 146.[30] M. Lucci, S. Sanna, G. Contini, N. Zema, V. Merlo, M. Salvato, H.N. Thanh, I. Davoli,

Surf. Sci. 601 (2007) 2647.[31] R. Krishnan, C. David, P.K. Ajikumar, S. Dash, A.K. Tyagi, V. Jayaram, B. Raj, Surf. Coat.

Technol. 206 (2011) 1196.[32] J.J. Olaya, L. Huerta, S.E. Rodil, R. Escamilla, Thin Solid Films 516 (2008) 8768.[33] J.J. Yeh, I. Lindau, At. Data Nucl. Data Tables 32 (1985) 1.[34] R. Musenich, P. Fabbricatore, G. Gemme, R. Parodi, M. Viviani, B. Zhang, V. Buscaglia,

C. Bottino, J. Alloys Compd. 209 (1994) 319.[35] G. Brauer, J. Less-Common Met. 2 (1960) 131.[36] V.I. Ivashchenko, P.E.A. Turchi, E.I. Olifan, Phys. Rev. B 82 (2010) 054109.[37] G. Oya, Y. Onodera, J. Appl. Phys. 45 (1974) 1389.[38] R. Sanjinés, M. Benkahoul, M. Papagno, F. Lévy, D. Music, J. Appl. Phys. 99 (2006)

044911.[39] X. Gonze, B. Amadon, P.M. Anglade, J.-M. Beuken, F. Bottin, P. Boulanger, F. Bruneval,

D. Caliste, R. Caracas, M. Cote, T. Deutsch, L. Genovese, Ph. Ghosez, M. Giantomassi, S.Goedecker, D. Hamann, P. Hermet, F. Jollet, G. Jomard, S. Leroux, M. Mancini, S.Mazevet, M.J.T. Oliveira, G. Onida, Y. Pouillon, T. Rangel, G.-M. Rignanese, D. Sangalli,R. Shaltaf, M. Torrent, M.J. Verstraete, G. Zérah, J.W. Zwanziger, Comput. Phys.Commun. 180 (2009) 2582.

[40] C.D. Gelatt Jr., A.R. Williams, V.L. Moruzzi, Phys. Rev. B 27 (1983) 2005.[41] P. Mishra, S.R. Bhattacharyya, D. Ghose, Nucl. Instrum. Methods B 266 (2008) 1629.[42] S. Ruffell, J.E. Bradby, J.S. Williams, D. Munoz-Paniagua, S. Tadayyon, L.L. Coatsworth,

P.R. Norton, Nanotechnology 20 (2009) 135603.[43] M.A. Mamun, A.H. Farha, Y. Ufuktepe, H.E. Elsayed-Ali, A.A. Elmustafa, J. Mater. Res.

27 (2012) 1725.[44] M. Wen, C.Q. Hu, C. Wang, T. An, Y.D. Su, Q.N. Meng, W.T. Zheng, J. Appl. Phys. 104

(2008) 023527.[45] X. Jiang, M.Wang, K. Schmidt, E. Dunlop, J. Haupt, W. Gissler, J. Appl. Phys. 69 (1991)

3053.[46] H. Holleck, J. Vac. Sci. Technol. A 4 (1986) 2661.