Contents lists available at ScienceDirect

Ceramics International

journal homepage: www.elsevier.com/locate/ceramint

Preparation of mullite-TiB2-CNTs hybrid composite through spark plasmasinteringYasin Oroojia, Ehsan Ghasalia,b,∗, Mostafa Moradic, Mohammad Reza Derakhshandehb,Masoud Alizadehb, Mehdi Shahedi Asld, Touradj Ebadzadehba College of Materials Science and Engineering, Nanjing Forestry University, No. 159, Longpan Road, Nanjing, 210037, Jiangsu, PR Chinab Ceramic Dept., Materials and Energy Research Center, Alborz, Iranc Department of Materials, Science and Engineering, Sharif University of Technology, Tehran, IrandDepartment of Mechanical Engineering, University of Mohaghegh Ardabili, Ardabil, Iran

A R T I C L E I N F O

Keywords:MulliteTitanium diborideCompositeHybridSpark plasma sintering

A B S T R A C T

A near fully dense mullite-TiB2-CNTs hybrid composite was prepared successfully trough spark plasma sintering.1 wt%CNT and 10wt%TiB2 were mixed with nano-sized mullite powders using a high energy mixer mill. Sparkplasma sintering was carried out at 1350 °C under the primary and final pressure of 10MPa and 30MPa, re-spectively. XRD results showed mullite and TiB2 as dominant crystalline phases accompanied by tiny peaks ofalumina. The microstructure of prepared composites demonstrated uniform distribution of TiB2 reinforcementsin mullite matrix without any pores and porosities as a result of near fully densified spark plasma sinteredcomposite. The fracture surface of composite revealed a proper bonding of TiB2 with mullite matrix and alsoareas with CNTs tunneling and superficies as a result of pulling-out phenomenon. The flexural strength of531 ± 28MPa, Vickers harness of 18.31 ± 0.3 GPa, and fracture toughness of 5.46 ± 0.12MPam−1/2 wereachieved for prepared composites as the measured mechanical properties.

1. Introduction

Desirable toughness, low thermal expansion coefficient, properthermal shock resistance, good creep resistance, and superior chemicaland thermal stability have made mullite-based composites high-tem-perature structural materials over the past decades [1–4]. Whilemonolithic mullite ceramics show very inadequate fracture toughness,normally, in the order of 1.5–2.5MPam1/2, using reinforcing agentssuch as zirconia and alumina can improve a limited fracture by ap-proximately up to 4.5–5MPam1/2 [5,6]. Among different mullite-basedcomposites, zirconia is a popular reinforcement that has attracted muchattention of researchers due to its enhanced fracture toughness, desir-able creep and flexural strength, and simple and early productionprocess of the reaction sintering of alumina and zircon [7,8]. Moreover,the effect of other reinforcement particles such as B4C [9], Ta2O5 [10],TiO2 [11], SiC [12], Si3N4 [13], WC [14], TiC [15], etc., on mechanicalproperties of mullite has been frequently studied by researchers. Re-cently, graphene nano plates (GNPs) [16] and carbon nano tubes(CNTs) [17], which provide proper mechanical properties and superior

applications for mullite composites including ceramic membranes [18],electrically conductive ceramics [16,19] and heavy metal ions removal[20], have been regarded as another advanced reinforcement in mullitematerials.

In an attempt to reduce the brittleness and at the same time tomaintain attractive mechanical properties, hybrid reinforced compo-sites have been considered as a promising candidate to improve themechanical properties of mullite-based materials [21,22]. Meanwhile,using particle reinforcements alongside CNTs can ensure the ad-vantages of two types of reinforcement in one product [23]. TiB2, as anonoxide ceramic, can serve as a useful part in the reinforcement phase,as shown in the research work [24]. C. T. HO [25], for example, pre-pared a mullite-TiB2 composite through combustion synthesis fromblends of TiO2, boron and carbon powders, finding that 20 vol% of TiB2in mullite matrix led to a fracture toughness and bending strength of4.3MPaml/2 and 427MPa, respectively. On the other hand, regardingthe reports, CNTs as reinforcement can admirably improve the me-chanical behavior of mullite-based composites. J. Wang et al. also in-vestigated the microstructural and mechanical characteristics of

https://doi.org/10.1016/j.ceramint.2019.05.154Received 24 April 2019; Received in revised form 11 May 2019; Accepted 14 May 2019

∗ Corresponding author. Ceramic Dept., Materials and Energy Research Center, Alborz, Iran.E-mail addresses: e.ghasali@merc.ac.ir, ehsan_ghasali@yahoo.com, ehsan.ghasali@gmail.com (E. Ghasali).

Ceramics International xxx (xxxx) xxx–xxx

0272-8842/ © 2019 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

Please cite this article as: Yasin Orooji, et al., Ceramics International, https://doi.org/10.1016/j.ceramint.2019.05.154

mullite-CNTs composites fabricated through hot press method; theyshowed the highest fracture toughness for 5 vol% CNTs addition. Inaddition, A. Weibel et al. [26] addressed the preparation of a mullite-iron-CNT composite through reduction in H2–CH4 and consolidation ina spark plasma sintering (SPS) process. They compared the mechanicalproperties of prepared composites with monolithic mullite ceramic.

Modern sintering methods such as spark plasma and microwaveheating have been broadly used for the production of structural ceramicmaterials [27–29]. Spark plasma sintering, as a pressure-assistedmethod with unique advantages, includes vacuum condition, fastheating rate, sparks between particles, Joule's heating, etc.; it is con-sidered as the proper process to consolidate a wide range of composites[30–33]. Various studies have been conducted on the influence of SPSprocess on the final properties of mullite-base composites; most of theresearches have obtained a proper densification by using this sinteringmethod [34–37].

To the best of our knowledge, no attempt has yet been made toprepare a mullite-TiB2-CNT hybrid composite through spark plasmasintering. Moreover, in the present work, a simple route was applied toobtain the almost uniform distribution of reinforcement components.Also, the microstructure of prepared composites was studied in detailby the polished and fractured surfaces after the bending test.

2. Experimental procedures

The synthesis of mullite nano-sized powders has been reported inthe authors' previous work [9] through a sol-gel method of tetraethylorthosilicate (Merck: 800658), Al(NO3)3.9H2O (Merck: 101086) and Si(C2H5O)4 as the staring mullite powders. Fig. 1 shows FESEM micro-graphs of as-synthesized nano-sized powders. First, 0.15 g of MWCNTs(OD: 5–15 nm,>95%, US Research nanomaterials Inc.) was dispersedin ethanol. After that, sodium dodecylbenzene sulfonate (SDS) was in-serted into the mixture and high-energy ultrasonication was used tointensify the dispersion of carbonaceous nano-additive in ethanol, ac-cording to the authors' previous work [38]. 1 wt% CNT and 10wt% oftitanium diboride powders (US Research nanomaterials Inc, purity:>98%, 2–12 μm mean particle size) were added to the mullite nano-sizedpowders and the mixing was carried out using a high-energy mixer millin ethanol media with five alumina balls in a polyethylene container for10min. The mixture was dehumidified on a heater-stirrer at 70 °C. In

order to fabricate SPS (SPS-20T-10, China) samples, the dried mixtureswere inserted directly into a graphite die and sintering process began ina vacuum condition (25 Pa) under an initial load of 10MPa and fina-lized at 1350 °C under the applied load of 30MPa. The densitometry(Archimede's principle), three bending strength (Santam-STm 20) andmicrohardness (MKV-h21, Microhardness Tester) analyses were doneon prepared samples. In the case of bending strength test, samples werethen cut into pieces of 5×5×25mm in dimension and the three pointbending strength measurements were done for three samples with18mm gage length and loading rate of 0.5 mm/min at room tempera-ture. The X-ray diffraction (XRD) test (Philips X'Pert: Cu kα) and fieldemission scanning electron microscopy (FESEM: MIRA 3 TESCAN,Czech Republic) were then implemented to identify the phases andmicrostructures of as-sintered mullite-based composites. Indentationfracture toughness was also estimated from the measurement of cracklengths after 10 successful indentations based on ISO 28079:2009, ac-cording to the following formula:

= Pa

K 0.0028 (H ) ,IC v1/2

1/2

where Σa is the total crack length (in mm), Hv is Vickers hardness (inMPa), and P is the load of indentation (in N).

3. Results and discussion

Fig. 2 presents FESEM micrographs of powder mixtures after highenergy milling and drying process. As can be observed, an almostuniform distribution of TiB2 was obtained at this mixing process.Moreover, the enlarged area with EDS elemental mapping at Fig. 2shows the presence of TiB2 and CNTs with mullite powders.

Fig. 3 demonstrates XRD pattern of sintered (SPSed) hybrid compo-site at 1350 °C. Regarding Fig. 3, the dominant peaks of mullite and TiB2were identified as the crystalline phases. The phase identification wascarried out using XRD data ICDD cards (01-089- 2644 (mullite), 01-075-0782 (Al2O3) and 01-089-3923 (TiB2)). Also, the tiny peaks related toalumina were realized from the corresponding patterns. As the use ofsynthetic mullite powders with almost complete mullitization has beenreported from the authors’ previous work, it seemed that the presence ofalumina could be related to the nature of SPS process. SPS process deals

Fig. 1. FESEM micrographs of the synthesized mullite nano-sized powders.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

2

with the advantages of vacuum conditions and sparks between particles.Moreover, there are some contradictions regarding the formation ofplasma during sintering [39,40]; however, it seems that complex eventstake place at the microscopic level in sintering process. In the case ofmullite sintering, one of the critical issues is assumed to be SiO2 de-composition (reducing condition) and its turning to SiO with its volati-zation at high temperatures [41]. In other words, the sparks can preparea local temperature higher than that detected by a pyrometer [42];

therefore, there is a high possibility of the evaporation phenomenon of acomponent such as mullite (3Al2O3·2SiO2) with a comparatively highvapor pressure at the solid solution [43]. By decreasing the SiO2 contentof mullite, Al2O3 could appear in the XRD pattern.

Fig. 4 depicts displacement, displacement rate and temperature vs.time during spark plasma sintering process. Regarding the curves pre-sented in Fig. 4, a tiny shrinkage at the first level of sintering could berelated to degassing air between powders as temperature was raised(red dashed circle). The second level could be followed by an increasein temperature, but the constant shrinkage up to 25min (almost relatedto 1000 °C) showed the warmed powders. Afterwards, the componentsof green body showed the first stage of sintering process with the for-mation of necks and their growth. This trend continued with an in-crease in shrinkages aligned with sintering time and temperature (reddashed oval). Finally, a noticeable change in shrinkage due to the in-crease of applied load from 10 to 30MPa at a maximum sinteringtemperature of 1350 °C could be detected in Fig. 4 (blue dashed oval).Meanwhile, this increase in pressure helped to complete a normaldensification by mass transportation in the sample. Finally, with regardto typical sintering temperature of a sample and the constant amount ofshrinkage at soaking time, the densification reached the highest levelunless sample had more complex reactions or needed a higher sinteringtemperature far from the applied temperature [42,44,45].

Fig. 5 displays backscattered (BSE) and secondary electrons (SE)Fig. 3. XRD pattern of the mullite-TiB2-CNT composite fabricated by sparkplasma sintering.

Fig. 2. FESEM images of mullite, TiB2 and CNTs mixture after high-energy milling and the corresponding EDS elemental mapping.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

3

Fig. 5. FESEM micrographs of the polished surface of mullite-TiB2-CNT composite.

Fig. 4. Displacement, displacement rate and temperature variations vs. sintering time of SPS route.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

4

FESEM micrographs of the polished surface of mullite-TiB2-CNT com-posite at different magnifications. The bright phases with a higherdensity and uniform distribution represent TiB2 phases at BSE images.On the other hand, the gray phase as mullite matrix depicted a properdensification. However, there were some dark areas which looked likeporosities, but SEM images at the same region showed that this darkarea was formed as a result of the height difference and particlespulling-out during surface polishing [46,47].

Fig. 6 reveals FESEM micrographs and corresponding EDS elementalmapping of mullite-TiB2-CNT composite. As can be observed, brightphases showed the presence of Ti and B elements as TiB2 compounds. Itis worth mentioning that the detection of B is difficult by EDS elementalmapping, but the presence of Ti can be assumed as TiB2 particles

considering this fact that there is no detected reaction product betweenmullite and TiB2 regarding to XRD patterns [48–50]. Moreover, thegray matrix demonstrated mullite phase in the presence of Al, Si and Oelements. Also, there were some dark regions with higher carbon con-tent. It seemed that CNTs were present in these regions or the polishingprocess with diamond paste, resulting in carbon remaining in theseareas.

Figs. 7 and 8 display FESEM fractographs of mullite-TiB2-CNTsample at low and high magnifications, respectively. With regard toFig. 7, the flat fracture surface of prepared composite could be attrib-uted to the ceramic nature of mullite as the matrix in composite.However, at higher magnification images, as shown in Fig. 7, particlepulling-out could be seen at the mullite matrix. Moreover, it seemed

Fig. 6. FESEM micrograph and corresponding EDS elemental mapping of mullite-TiB2-CNT composite.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

5

that TiB2 reinforcement had a proper bonding to the mullite matrixafter fracture. Fig. 8, with high magnification FESEM images from thefracture surface of prepared composite, demonstrates two different re-gions in the presence of CNTs and the affected area with CNTs pulling-out tunnels, as well as superficies patterns. According to Fig. 8, theconfiguration of CNTs and their positions could affect the fracturesurface of composite. Meanwhile, CNTs parallel to the fracture direction(fracture force) created superficies at surface. The agglomerate regionled to the formation of areas with different heights; furthermore, CNTswith an angle perpendicular to the force direction left tunnels at thefracture surface of composite. All of the above mentioned phenomenacould lead to the introduction of different mechanisms of fracture, re-sulting in an increase in fracture toughness due to the fact that each ofthe events would use the required energy of fracture.

Fig. 9 illustrates FESEM images of the indent diagonal impressions

of prepared composite after Vickers hardness test. As shown, the di-agonal at BSE image could not be detected, but SE images showed thediagonal with cracks. The straight crack path showed a brittle nature ofmullite matrix composite. However, it seemed that the crack length wasinfluenced by the presence of TiB2 as reinforcement. On the other hand,Fig. 9 demonstrates another indent diagonal impression which was animperfect shape at the right side of diagonal. It could probably happendue to the presence of CNTs at this area which looked like deformationat the sides of diagonal.

Table 1 depicts the physical and mechanical properties of sparkplasma sintered mullite-TiB2-CNT composite. As can be seen, the highVickers harness of 18.31 ± 0.3 GPa was obtained from preparedcomposite in comparison to the literature, as a result of using SPSmethod with low grain growth and nearly full densification, as well asTiB2 reinforcement as a hard ceramic [5,51]. In the case of flexural

Fig. 7. FESEM micrographs of the fracture surface of mullite-TiB2-CNT composite at low magnifications.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

6

strength and fracture toughness, the prepared hybrid composite showedproper mechanical properties which seemed to be higher than reportedvalues in other works [52–55]. So, the nearly full densification, ultra-low water absorption and the use of TiB2 and CNT as hybrid re-inforcement could lead to the preparation of mullite-based hybridcomposite and superior mechanical properties, as compared to mullitecomposite reinforced with either TiB2 or CNT separately [25,56].

Fig. 10 reveals the force-extension curve and adsorbed energyduring the bending strength test. As can been seen, the acceptablefracture load and extension were obtained for prepared composite, ascompared to the authors’ previous work. Moreover, the remarkableadsorbed energy as a result of using a hybrid reinforcement was cal-culated for SPSed composite [9,10].

4. Conclusions

Mullite-TiB2-CNT hybrid composite was prepared successfullythrough spark plasma sintering. The XRD patterns showed the presenceof composite component with tiny amounts of alumina, probably as aresult of SiO2/SiO volatization at high temperatures and the specialsintering condition of SPS process. The fracture surface of preparedcomposite demonstrated different areas such as: particles pulling-out,CNTs remaining tunneling, superficies, and some agglomeration ofCNTs, while the polished surface of composite depicted uniform dis-tribution of TiB2 reinforcement. The superior mechanical properties ofproduced composite, including the flexural strength of 531 ± 28MPa,Vickers harness of 18.31 ± 0.3 GPa, and fracture toughness of5.46 ± 0.12MPam−1/2, could suggest using both TiB2 particles andCNTs to produce the hybrid composite.

Fig. 8. FESEM micrographs of the fracture surface of mullite-TiB2-CNT composite at high magnifications.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

7

References

[1] A.H. Pakseresht, E. Ghasali, M. Nejati, K. Shirvanimoghaddam, A.H. Javadi,R. Teimouri, Development empirical-intelligent relationship between plasma sprayparameters and coating performance of Yttria-Stabilized Zirconia, Int. J. Adv.Manuf. Technol. 76 (2015) 1031–1045.

[2] A.H. Pakseresht, A.H. Javadi, M. Nejati, K. Shirvanimoghaddam, E. Ghasali,R. Teimouri, Statistical analysis and multiobjective optimization of process para-meters in plasma spraying of partially stabilized zirconia, Int. J. Adv. Manuf.Technol. 75 (2014) 739–753.

[3] J. Roy, S. Das, S. Maitra, Solgel‐processed mullite coating—a review, Int. J. Appl.Ceram. Technol. 12 (2015) E71–E77.

[4] H. Schneider, R.X. Fischer, J. Schreuer, Mullite: crystal structure and relatedproperties, J. Am. Ceram. Soc. 98 (2015) 2948–2967.

[5] C. Sadik, I.-E. El Amrani, A. Albizane, Recent advances in silica-alumina refractory:a review, J. Asian Ceram. Soc. 2 (2014) 83–96.

[6] H. Schneider, J. Schreuer, B. Hildmann, Structure and properties of mullite—a re-view, J. Eur. Ceram. Soc. 28 (2008) 329–344.

[7] S. Lathabai, D.G. Hay, F. Wagner, N. Claussen, Reaction‐bonded mullite/zirconiacomposites, J. Am. Ceram. Soc. 79 (1996) 248–256.

[8] R. Sarkar, K. Bishoyi, Reaction sintering of zirconia-mullite composites in the pre-sence of SrO, Interceram-International Ceram. Rev. 63 (2014) 304–306.

Fig. 9. FESEM images of indent diagonals impression on the polished surface of prepared composite.

Table 1Physical and mechanical properties of the prepared composite.

Water absorption (%) Relative density (%) Fracture toughness (MPa.m−1/2) Flexural strength (MPa) Hardness (GPa) Properties

0.081 ± 0.06 99.1 ± 0.3 5.46 ± 0.12 531 ± 28 18.31 ± 0.3 Mullite-TiB2-CNT

Fig. 10. Force-extension curve and absorbed energy during the flexuralstrength test.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

8

[9] E. ghasali, E.K. Saeidabadi, M. Alizadeh, A. Fazili, H. Rajaei, A. Jam,H. Kazemzadeh, T. Ebadzadeh, Preparation of mullite/B4C composites: a com-parative study on the effect of heating methods, Ceram. Int. 44 (2018)18743–18751, https://doi.org/10.1016/j.ceramint.2018.07.104.

[10] S.M.R. Derakhshandeh, M.S. Gohari, E.K. Saeidabadi, A. Jam, A. Fazili, M. Alizadeh,E. Ghasali, A.H. Pakseresht, T. Ebadzadeh, Comparison of spark plasma and mi-crowave sintering of mullite based composite: mullite/Ta2O5 reaction, Ceram. Int.44 (2018) 13176–13181.

[11] N. Montoya, F.J. Serrano, M.M. Reventós, J.M. Amigo, J. Alarcón, Effect of TiO2 onthe mullite formation and mechanical properties of alumina porcelain, J. Eur.Ceram. Soc. 30 (2010) 839–846.

[12] Y. Jing, X. Deng, J. Li, C. Bai, W. Jiang, Fabrication and properties of SiC/mullitecomposite porous ceramics, Ceram. Int. 40 (2014) 1329–1334.

[13] E.K. Saeidabadi, E. Salahi, T. Ebadzadeh, Preparation mullite/Si3N4 composites byreaction spark plasma sintering and their characterization, Ceram. Int. 45 (2019)5367–5383.

[14] H. Rajaei, M. Farvizi, I. Mobasherpour, M. Zakeri, Effect of spark plasma sinteringtemperature on microstructure and mechanical properties of mullite-WC compo-sites, Int. J. Refract. Metals Hard Mater. 70 (2018) 197–201.

[15] D. Ghahremani, T. Ebadzadeh, A. Maghsodipour, Densification, microstructure andmechanical properties of Mullite-TiC composites prepared by spark plasma sin-tering, Ceram. Int. 41 (2015) 1957–1962, https://doi.org/10.1016/j.ceramint.2014.07.146.

[16] J. Ru, Y. Fan, W. Zhou, Z. Zhou, T. Wang, R. Liu, J. Yang, X. Lu, J. Wang, C. Ji,Electrically conductive and mechanically strong graphene/mullite ceramic com-posites for high-performance electromagnetic interference shielding, ACS Appl.Mater. Interfaces 10 (2018) 39245–39256.

[17] A. Azarniya, S. Sovizi, A. Azarniya, M.R.R.T. Boyuk, T. Varol, P. Nithyadharseni,H.R.M. Hosseini, S. Ramakrishna, M.V. Reddy, Physicomechanical properties ofspark plasma sintered carbon nanotube-containing ceramic matrix nanocomposites,Nanoscale 9 (2017) 12779–12820.

[18] Y. Dong, L. Ma, C.Y. Tang, F. Yang, X. Quan, D. Jassby, M.J. Zaworotko,M.D. Guiver, Stable superhydrophobic ceramic-based carbon nanotube compositedesalination membranes, Nano Lett. 18 (2018) 5514–5521.

[19] P. Čapková, V. Matějka, J. Tokarský, P. Peikertová, L. Neuwirthová, L. Kulhánková,J. Beňo, V. Stýskala, Electrically conductive aluminosilicate/graphene nano-composite, J. Eur. Ceram. Soc. 34 (2014) 3111–3117.

[20] M.A. Tofighy, T. Mohammadi, Copper ions removal from water using functionalizedcarbon nanotubes–mullite composite as adsorbent, Mater. Res. Bull. 68 (2015)54–59.

[21] K.P. Gadkaree, K.C. Chyung, M.P. Taylor, Hybrid ceramic matrix composites, J.Mater. Sci. 23 (1988) 3711–3720, https://doi.org/10.1007/BF00540519.

[22] E. Ghasali, R. Yazdani-rad, K. Asadian, T. Ebadzadeh, Production of Al-SiC-TiChybrid composites using pure and 1056 aluminum powders prepared through mi-crowave and conventional heating methods, J. Alloy. Comp. 690 (2017) 512–518.

[23] M.N. Gururaja, A.N.H. Rao, A review on recent applications and future prospectusof hybrid composites, Int. J. Soft Comput. Eng. 1 (2012) 352–355.

[24] Z.I. Zaki, Combustion synthesis of mullite–titanium boride composite, Ceram. Int.35 (2009) 673–678.

[25] C.T. Ho, In situ reacted TiB 2-reinforced mullite, J. Mater. Sci. 30 (1995)1338–1342.

[26] A. Weibel, A. Peigney, G. Chevallier, C. Estournès, C. Laurent, Toughened carbonnanotube–iron–mullite composites prepared by spark plasma sintering, Ceram. Int.39 (2013) 5513–5519.

[27] H. Majidian, E. Ghasali, T. Ebadzadeh, M. Razavi, Effect of heating method onmicrostructure and mechanical properties of zircon reinforced aluminum compo-sites, Mater. Res. 19 (2016) 1443–1448.

[28] E. Ghasali, R. Yazdani-Rad, A. Rahbari, T. Ebadzadeh, Microwave sintering ofaluminum-ZrB2 composite: focusing on microstructure and mechanical properties,Mater. Res. 19 (2016) 765–769.

[29] E. Ghasali, M. Alizadeh, A.H. Pakseresht, T. Ebadzadeh, Preparation of siliconcarbide/carbon fiber composites through high-temperature spark plasma sintering,J. Asian Ceram. Soc. 5 (2017) 472–478, https://doi.org/10.1016/j.jascer.2017.10.004.

[30] E. Ghasali, Y. Palizdar, A. Jam, H. Rajaei, T. Ebadzadeh, Effect of Al and Mo ad-dition on phase formation, mechanical and microstructure properties of sparkplasma sintered iron alloy, Mater. Today Commun. 13 (2017) 221–231, https://doi.org/10.1016/j.mtcomm.2017.10.005.

[31] E. Ghasali, M. Alizadeh, K. Shirvanimoghaddam, R. Mirzajany, M. Niazmand,A. Faeghi-Nia, T. Ebadzadeh, Porous and non-porous alumina reinforced magne-sium matrix composite through microwave and spark plasma sintering processes,Mater. Chem. Phys. 212 (2018) 252–259.

[32] E. Ghasali, T. Ebadzadeh, M. Alizadeh, M. Razavi, Spark plasma sintering of WC-based cermets/titanium and vanadium added composites: a comparative study onthe microstructure and mechanical properties, Ceram. Int. 44 (2018) 10646–10656.

[33] E. Ghasali, T. Ebadzadeh, M. Alizadeh, M. Razavi, Mechanical and microstructuralproperties of WC-based cermets: a comparative study on the effect of Ni and Mobinder phases, Ceram. Int. 44 (2018) 2283–2291, https://doi.org/10.1016/j.ceramint.2017.10.189.

[34] W. Lerdprom, A. Bhowmik, S. Grasso, E. Zapata‐Solvas, D.D. Jayaseelan,M.J. Reece, W.E. Lee, Impact of spark plasma sintering (SPS) on mullite formationin porcelains, J. Am. Ceram. Soc. 101 (2018) 525–535.

[35] E. Nsiah-Baafi, A. Andrews, Fabrication of mullite from lithomargic clay via sparkplasma sintering, Ceram. Int. 43 (2017) 14277–14280.

[36] E.K. Saeidabadi, T. Ebadzadeh, E. Salahi, Preparation of mullite from alumina/aluminum nitrate and kaolin clay through spark plasma sintering process, Ceram.Int. 44 (17) (2018) 21053–21066.

[37] R.E. Tressler, Recent developments in fibers and interphases for high temperatureceramic matrix composites, Compos. Part A Appl. Sci. Manuf. 30 (1999) 429–437.

[38] E. Ghasali, P. Sangpour, A. Jam, H. Rajaei, K. Shirvanimoghaddam, T. Ebadzadeh,Microwave and spark plasma sintering of carbon nanotube and graphene reinforcedaluminum matrix composite, Arch. Civ. Mech. Eng. 18 (2018) 1042–1054, https://doi.org/10.1016/j.acme.2018.02.006.

[39] D.M. Hulbert, A. Anders, D. V Dudina, J. Andersson, D. Jiang, C. Unuvar,U. Anselmi-Tamburini, E.J. Lavernia, A.K. Mukherjee, The absence of plasma in“spark plasma sintering,”, J. Appl. Phys. 104 (2008) 33305.

[40] R. Marder, C. Estournès, G. Chevallier, R. Chaim, Plasma in spark plasma sinteringof ceramic particle compacts, Scripta Mater. 82 (2014) 57–60.

[41] M.M. Shul’ts, S.I. Shornikov, V.L. Stolyarova, Evaporation processes and thermo-dynamic properties of mullite, Dokl. Phys. Chem. Consultants Bureau., New York,1994, p. 95.

[42] Y.C. Wang, Z.Y. Fu, Q.J. Zhang, SPS temperature distribution of different con-ductivity materials, Key Eng. Mater. Trans Tech Publ, 2002, pp. 717–720.

[43] H. Schneider, S. Komarneni, Mullite, John Wiley & Sons, 2006.[44] U. Anselmi-Tamburini, S. Gennari, J.E. Garay, Z.A. Munir, Fundamental in-

vestigations on the spark plasma sintering/synthesis process: II. Modeling of currentand temperature distributions, Mater. Sci. Eng. A 394 (2005) 139–148.

[45] M. Barsoum, M.W. Barsoum, Fundamentals of Ceramics, CRC press, 2002.[46] R.B. Azhiri, J.F. Sola, R.M. Tekiyeh, F. Javidpour, A.S. Bideskan, Analyzing of joint

strength, impact energy, and angular distortion of the ABS friction stir welded jointsreinforced by nanosilica addition, Int. J. Adv. Manuf. Technol. 100 (2019)2269–2282, https://doi.org/10.1007/s00170-018-2761-8.

[47] R.B. Azhiri, A.S. Bideskan, F. Javidpour, R.M. Tekiyeh, Study on material removalrate, surface quality, and residual stress of AISI D2 tool steel in electrical dischargemachining in presence of ultrasonic vibration effect, Int. J. Adv. Manuf. Technol.101 (2019) 2849–2860, https://doi.org/10.1007/s00170-018-3023-5.

[48] T.U. Probst, Methods for boron analysis in boron neutron capture therapy (BNCT).A review, Fresenius J. Anal. Chem. 364 (1999) 391–403.

[49] R.M. Tekiyeh, M. Najafi, S. Shahraki, Machinability of AA7075-T6/carbon nano-tube surface composite fabricated by friction stir processing, Proc. Inst. Mech. Eng.Part E J. Process Mech. Eng. 0 (2018) 1–10, https://doi.org/10.1177/0954408918809618.

[50] R.B. Azhiri, R. Mehdizad Tekiyeh, E. Zeynali, M. Ahmadnia, F. Javidpour,Measurement and evaluation of joint properties in friction stir welding of ABSsheets reinforced by nanosilica addition, Meas. J. Int. Meas. Confed. 127 (2018)198–204, https://doi.org/10.1016/j.measurement.2018.05.005.

[51] B. Saruhan, Oxide-Based Fiber-Reinforced Ceramic-Matrix Composites: Principlesand Materials, Springer Science & Business Media, 2013.

[52] X. Chen, L. Luo, L. Liu, J. Li, H. Yu, W. Li, Y. Chen, Microstructure and mechanicalproperties of hot-pressed Al2O3–mullite–ZrO2–SiC composites, Mater. Sci. Eng. A740 (2019) 390–397.

[53] P.B. Sistani, A. Kiani-Rashid, S.M. Beidokhti, Microstructural and diametral tensilestrength evaluation of the zirconia-mullite composite, Ceram. Int. 45 (2019)7127–7136.

[54] J.-S. Ha, K.K. Chawla, Mechanical behaviour of mullite composites reinforced withmullite fibres, Mater. Sci. Eng. A 203 (1995) 171–176.

[55] A. Cascales, N. Tabares, J.F. Bartolomé, A. Cerpa, A. Smirnov, R. Moreno,M.I. Nieto, Processing and mechanical properties of mullite and mullite–aluminacomposites reinforced with carbon nanofibers, J. Eur. Ceram. Soc. 35 (2015)3613–3621 https://doi.org/10.1016/j.jeurceramsoc.2015.05.011.

[56] J. Wang, H.M. Kou, Y.B. Pan, J.K. Guo, The microstructure, mechanical and di-electric properties of CNTs/mullite ceramics composites, Key Eng. Mater. TransTech Publ, 2006, pp. 145–150.

Y. Orooji, et al. Ceramics International xxx (xxxx) xxx–xxx

9