Indentation size effect and wear characteristics of spark plasma sintered,hard MWCNT/Al 2 O 3 nanocomposites

Abstract

Magnesia doped multiwalled carbon nanotube (CNT)/α-alumina nanocomposites have been fabricated by spark plasma sintering at 1500°C under 50 MPa in argon. Owing to combined grain refining effect of nanotube and magnesia, nanocomposites possessed smaller matrix grains and extensively lower matrix crystallites than pure alumina. Thermal expansion mismatch between matrix and filler rendered up to four times higher compressive lattice microstrain to the nanocomposites over pure alumina. Despite very low CNT loading (e.g. 0·13 wt-%), nanocomposites offered considerably higher hardness (as high as 24·42 GPa), negligible indentation size effect (Meyer exponent = 1·906 − 1·941) and enhanced elastic response over pure alumina. Up to 0·27 wt-% nanotube loading, much higher wear resistance was observed for the nanocomposites over pure alumina. The presence of uniformly dispersed and structurally intact nanotubes coupled with lower matrix grains and crystallites having compressive lattice strain were the key factors behind achieving such improved mechanical properties of the present nanocomposites.

Keywords

Nanocomposite CNT Structural Hardness ISE Wear

Introduction

Aluminium oxide (or so called alumina having a chemical formula of Al₂O₃) is probably the most widely utilised oxide ceramic in various fields starting from chemical absorbent to biomedical implants to advanced structural components and many more.^1–5 This is due to the fact that the most stable phase, i.e. α-Al₂O₃ (commonly known as corundum), has several advantages over other oxide ceramics, like low density (3·97 g cm^− 3), high hardness (15–18 GPa), high flexural strength (210–500 MPa), low thermal expansion coefficient (5·4–8·1 × 10^− 6/K), extreme chemical inertness, high temperature dimensional stability (up to 1700°C in ambient), high thermal shock resistance, high dielectric strength and biocompatibility, and above all, apart from its wide availability, this oxide ceramic can be found at a reasonable price.^6–9 However, extreme brittleness of Al₂O₃ has led towards unpredictable catastrophic failures during use. Coupled to this, poor fracture toughness (K_IC) and high temperature mechanical property retentivity are some of the critical issues that restrict the suitability of monolithic α-Al₂O₃ as advanced high performance structural ceramic.^10–12 To surmount such issues, researchers have always tried to fabricate advanced structural composites containing a multiphase Al₂O₃ matrix. These composites are expected to exhibit engineered microstructure–property relationship and also an improved performance over monolithic Al₂O₃.^13–15 Soon after the discovery of carbon nanotube (CNT) in 1991, the outstanding capability of single walled CNT and multiwalled CNT (MWCNT) as effective filler in ceramic matrices triggered the global R&D activities on fabrication and multiaspect characterisations of CNT/Al₂O₃ nanocomposites to meet the state of the art high performance application demands.^11,16–29 As far as mechanical property evaluation of CNT/Al₂O₃ nanocomposites is concerned, both at room temperature and up to sufficiently high temperature, reports are available to illustrate the improved structural performance of this nanocomposite over monolithic Al₂O₃ at different CNT concentrations.²⁴ Conversely, deterioration in properties of CNT/Al₂O₃ nanocomposites compared to monolithic Al₂O₃ also exists. Such degradation in property values resulted from many factors, e.g. non-uniform distribution of nanotubes at higher loading, structural collapse of nanotubes at high sintering temperature and pressure, poor matrix densification at high CNT concentration, and presence of agglomerated CNTs that act as pores of similar dimension.^{18,22,24,29,30} Thus, further research is required to optimise the performance and utilisation of CNT/Al₂O₃ nanocomposites as a real life, advanced, high performance structural material. On the other hand, in spite of the extreme significance of hardness values, fracture toughness and wear characteristics of these particular nanocomposites, reports and data are quite limited for wear component applications.^18,19,27,28 In the present work, CNT reinforced Al₂O₃ matrix nanocomposites containing 0·07–1·09 wt-% MWCNT have been fabricated using simple wet mixing of as received raw materials followed by spark plasma sintering (SPS). The nanocomposites and pure Al₂O₃ have been characterised in terms of microstructure, Vickers hardness (HV) from 0·2 to 2·0 kgf indentation loads and dry sliding wear behaviour in the range of 10–20 N normal loads (F_N). In addition, to trace out the effect of nanotube addition in Al₂O₃, matrix crystallite sizes (L_c) and lattice microstrain (ε_c) values have been evaluated using Williamson–Hall (W–H) technique from the corresponding X-ray diffraction (XRD) profiles.³¹ Finally, performances of nanocomposite specimens were compared with those of pure Al₂O₃ to evaluate the extent of improvement of present SPS-ed nanocomposites.

Experimental

Raw materials

The raw materials used in this work are M/s Shenzhen Nano Tech Port, China make MWCNT having >95 wt-% purity, 60–100 nm outer diameter and 5–15 μm length and M/s Almatis Alumina Pvt. Ltd., India make A-16-SG grade reactive polycrystalline Al₂O₃ powder having 99·8 wt-% purity and 0·5 μm median particle size.

Batch preparation and sintering

The as received MWCNT was first dispersed in isopropyl alcohol (IPA) using ultrasonic bath sonication (Micro Clean 109, M/s Oscar Ultrasonics Pvt. Ltd, India) for 1 h. The dispersed CNT was mixed with Al₂O₃ powder and 0·5 wt-% MgO powder (M/s Merck Specialties Pvt. Ltd, India) as sintering additive using attrition milling for another 3 h in IPA. Al₂O₃ balls (Φ = 3 mm) were used as milling media. Finally, the mixed slurry was dried at 70°C in an air oven for volatile removal and sieved through 60 mesh BS screen and collected. The powder mixtures were then consolidated using SPS (HP-D-25, M/s FCT GmbH, Germany) in a Φ = 20 mm graphite mould at 1500°C and 50 MPa uniaxial pressure with a dwell of 3 min in static argon atmosphere.

Characterisations

Bulk densities of the consolidated specimens were measured using Archimedes water immersion technique. Composition, specimen ID, theoretical density and relative density values of all the studied batches are given in Table 1. Microstructures of the polished and thermally etched specimens were viewed through a desktop SEM (Phenom^TM proX, M/s Phenom-World BV, The Netherlands) up to 15 000 × magnification. CNT/Al₂O₃ interactions in sintered nanocomposites have been viewed through a field emission SEM (FESEM: Supra-35, VP-Carl Zeiss, Germany). Phase, L_c and ε_c analyses of the specimens were performed using XRD patterns obtained from Bruker D8 Advance DA Vinci XRD System in between 10° and 90° 2θ values with a step size of 0·05°. Hardness values of the specimens were measured using a micro-Vickers hardness tester (402 MVD, M/s Wolpert-Wilson, Germany) equipped with Minuteman-1 software and 1·3 MP USB camera, with indentation loads ranging from 0·2 to 2·0 kgf and a dwell of 10 s. At least 15 indents were made on each sample at each of the test loads for obtaining a reproducible data. For analysing and comparing the indentation size effect (ISE) of MWCNT/Al₂O₃ nanocomposites with that of pure Al₂O₃, three different models were utilised, namely, Meyer's law, proportional specimen's resistance (PSR) model and modified PSR (M-PSR) model. Room temperature dry sliding wear characteristics of the specimens were evaluated using a DUCOM (India) TR-101 Tribometer equipped with a standard diamond Vickers indenter. Linear scratches (5 mm) were made on each of the specimens at three different normal loads (F_N) from 10 to 20 N at 0·1 mm s^− 1 scratching velocity. Scar topography and dimension were viewed through a non-contact surface profilometer (Contour GT, M/s Bruker, USA).

Table 1

Physical properties of SPS1500 series specimens

Composition	Sample ID	TD*/g cm^− 3	RD^†/%
(Al₂O₃ +0·5 wt-% MgO)	A0SP15	3·968	99·99
(Al₂O₃ +0·5 wt-% MgO) +0·07 wt-% MWCNT	A1SP15	3·965	99·81
(Al₂O₃ +0·5 wt-% MgO) +0·13 wt-% MWCNT	A2SP15	3·961	99·67
(Al₂O₃ +0·5 wt-% MgO) +0·27 wt-% MWCNT	A3SP15	3·955	99·85
(Al₂O₃ +0·5 wt-% MgO) +0·54 wt-% MWCNT	A4SP15	3·942	99·99
(Al₂O₃ +0·5 wt-% MgO) +1·09 wt-% MWCNT	A5SP15	3·916	99·57

TD, theoretical density (calculated using ‘rule of mixture’ and taking ρ_Al2O3 = 3·970 g cm^− 3; ρ_MgO = 3·581 g cm^− 3; ρ_MWCNT = 1·775 g cm^− 3).

†

RD, relative density [(bulk density/theoretical density) × 100%].

Results and discussion

From Table 1, it may be visualised that all specimens including pure Al₂O₃ were almost theoretically dense at the present SPS schedule. Although CNT generally acts as sintering inhibitor for ceramics,^{11,17,19,26,29} Milsom et al.³² reported the effectiveness of this nanofiller towards obtaining highly dense ceramic matrix composites, especially through SPS processing. According to the report, addition of only 2 vol.-% CNT reduced the sintering activation energy by >62% compared to the monolithic ceramic that has an activation energy of 456 kJ mol^− 1.³² This was primarily achieved through the formation of an electrical percolating network that rendered higher grain boundary diffusion through local heating of matrix particles during SPS. Attainment of near theoretical densities of present SPS-ed nanocomposites, especially beyond 0·27 wt-% MWCNT loading (Table 1), evidently corroborates the observations made by Milsom et al.³² Microstructure analyses of sintered specimens indicated that although population of ≥ 2·0 μm grains was the highest in pure Al₂O₃ (Fig. 1a), increasing CNT content up to 1·09 wt-% resulted in considerable decrease in matrix grain size ( ≤ 1·5 μm), indicating combined grain refining effect by MgO and CNT (Fig. 1b). However, with increasing CNT content, especially at 1·09 wt-% CNT loading, highly segregated regions in the sintered mass were identified due to the presence of large CNT agglomerates (Fig. 1c). Such segregated regions in high nanotube loaded nanocomposites have also been reported by others, and this CNT agglomerates are responsible for significant deterioration of structural properties of CNT/Al₂O₃ nanocomposites.^{22–24,26,29} On the other hand, FESEM images of sintered nanocomposites revealed bridging of multiple Al₂O₃ grains by well dispersed CNTs at low as well as at high nanotube concentrations (Fig. 2a and b). However, as mentioned above, segregated regions have also been identified in high CNT loaded specimens, especially at 1·09 wt-%, where the presence of entangled CNTs at the grain boundary regions was clearly identified (Fig. 2c).

Microstructure of a pure Al₂O₃ (scale bar, 5 μm), b 0·27 wt-% MWCNT/Al₂O₃ nanocomposite (scale bar, 5 μm) and c 1·09 wt-% MWCNT/Al₂O₃ nanocomposite (scale bar, 10 μm)

FESEM images of sintered nanocomposites: a bridging of multiple Al₂O₃ grains (numbered) by individual CNTs (marked by pointed arrows) in 0·27 wt-% MWCNT/Al₂O₃ specimen (scale bar, 200 nm); b bridging of multiple Al₂O₃ grains by almost individually dispersed CNTs (marked by pointed arrows) in 1·09 wt-% MWCNT/Al₂O₃ specimen (scale bar, 400 nm); c segregated matrix regions and agglomerated CNTs (indicated by dashed region) at grain boundary of 1·09 wt-% MWCNT/Al₂O₃ specimen (scale bar, 400 nm)

XRD phase analysis revealed that the sintered specimens mainly consisted of α-Al₂O₃ phase (JCPDS file 76-0144) of high crystallinity (Fig. 3). Owing to very low CNT concentrations and strong matrix peaks, identification of sp² carbon phase was not possible. Figure 4 shows matrix L_c and ε_c values of sintered specimens evaluated using W–H technique. It may be visualised from the figure that at up to 0·13 wt-% CNT loading, crystallite size decreased steadily (∼26% lower than pure Al₂O₃; ^A0SP15L_C = 36·3 nm) and increased slightly at 0·27 wt-% CNT loading compared to A2SP15 specimen followed by a gradual decrease up to 1·09 wt-% CNT loading (∼31% lower than A0SP15). This was due to the fact that at up to 0·13 wt-% nanotube concentration, CNTs were dispersed homogeneously in Al₂O₃ that resulted in uniform matrix grain refining effect and, eventually, steady decrease in L_C. On the contrary, beyond 0·13 wt-% CNT loading, nanotubes started to form agglomerates and heterogeneous dispersion of individual and clustered nanotubes produced mixed crystallite size in the matrix grains. This ultimately produced matrix crystallites comparatively larger than A2SP15 specimen but lower than A0SP15 (Fig. 4). Finally, at 1·09 wt-% CNT loading, the presence of large number of clustered CNTs in the matrix phase rendered significant hindrance towards grain boundary diffusion and produced smallest matrix crystallites (Fig. 4). The present L_C values are significantly lower than that obtained for pressureless sintered CNT/Al₂O₃ nanocomposites due to dramatic reduction in total sintering time in SPS over pressureless sintering technique.²⁰ Similar trend was also noticed in lattice microstrain values due to the reason stated above (Fig. 4). Furthermore, the negative sign of strain values indicates formation of compressive ε_C in the matrix crystallites (Fig. 4), which can effectively contribute towards improvement of mechanical properties of the present nanocomposites. In addition, due to the large thermal expansion mismatch between Al₂O₃ (CTE = 8 × 10^− 6/K) and CNT (CTE = − 1·6 × 10^− 6/K),^33,34 the magnitude of the compressive crystallite strain increased significantly in the nanocomposites over pure Al₂O₃ with increasing CNT concentration. The 1·09 wt-% MWCNT/Al₂O₃ specimen had about four times higher ε_C than pure Al₂O₃ (Fig. 4).

XRD patterns of all SPS-ed specimens and JCPDS file 76-0144 for α-Al₂O₃

Matrix crystallite size (L_c) and lattice microstrain (ε) versus CNT concentration plots obtained from W–H analysis

Figure 5 shows the HV data of all specimens within the indentation test span, i.e. from 0·2 to 2·0 kgf. Except A5SP15 specimen, all other nanocomposites offered improved HV values at all indentation loads. The poor hardness data obtained for the 1·09 wt-% MWCNT/Al₂O₃ specimen were possibly due to the presence of clustered CNTs in the matrix phase that failed to offer any reinforcing effect and acted as pores of equivalent dimensions and poor CNT/Al₂O₃ interface performance. On the other hand, the highest HV values obtained for A2SP15 (HV_0·2 = 24·42 ± 2·33 GPa; HV_2·0 = 22·72 ± 1·75 GPa) were almost 18% higher than pure Al₂O₃ (Fig. 5). The hardness values of present nanocomposites and extent of improvement in HV data over pure Al₂O₃ are perhaps the highest among others.²⁴ This was achieved through uniform dispersion of CNT in the matrix phase at least up to 0·13 wt-% loading, proper SPS schedule, structural retention of nanotubes in sintered specimens and adequate CNT/Al₂O₃ interface performance that helped in proper load sharing between matrix and the filler. To find out the extent of indentation loads on HV data, Meyer's exponent (n) values were evaluated from the linear fit of log(d) versus log(P) data points, where d is the average length of indentation diagonal and P is the indentation load (Fig. 6). It may be seen from the data given in Table 2 that all the specimens including pure Al₂O₃ had almost similar extent of ISE (n>1·9) within indentation loading span. Furthermore, since the n values lie well above 1·9, at this loading region (i.e. 0·2 kgf < P ≤ 2·0 kgf), all the specimens showed negligible ISE, suggesting worthy use of the present nanocomposites as structural component.^35,36 Results obtained from PSR and M-PSR plots (Figs. 7 and 8), which were utilised to quantitatively describe the ISE of the present specimens, revealed the following (Tables 3 and 4): (i)

improved elastic response (increased a₁/a₂ ratio) of the nanocomposites over pure Al₂O₃ within the test span because of enhanced reinforcing effect of structurally intact highly elastic CNTs that ultimately helped in HV value retention to a much higher level compared to A0SP15,

(ii)

formation of gradually increased machining induced compressive surface residual stress in the nanocomposites over pure Al₂O₃ because of increased concentration of reinforcing CNTs that rendered higher pullout resistance to the matrix grain during machining that effectively contributed towards improvement in HV values of the nanocomposites and

(iii)

lower H_T (true hardness) values in the absence of thermal and/or machining induced residual stresses in the specimens.

Variation in HV values of studied specimens within indentation loading range

Meyer's plots of studied specimens

Table 2

Meyer exponent of studied specimens

Sample ID	Meyer's exponent (n)
A0SP15	1·939
A1SP15	1·906
A2SP15	1·940
A3SP15	1·935
A4SP15	1·941
A5SP15	1·941

PSR plots of studied specimens

M-PSR plots of studied specimens

Table 3

Results obtained from PSR model fitting of studied specimens

Sample ID	a ₁ /N mm^− 1	a ₂ /N mm^− 2	H _T /GPa
A0SP15	16·25	9995·27	18·54
A1SP15	22·37	10865·25	20·14
A2SP15	17·87	11780·79	21·85
A3SP15	19·72	10861·75	20·14
A4SP15	17·85	10533·86	19·53
A5SP15	17·12	10060·19	18·66

Table 4

Results obtained from M-PSR model fitting of studied specimens

Sample ID	P ₀ /N	a ₁ /N mm^− 1	a ₂ /N mm^− 2	( a _1/ a ₂)
A0SP15	− 0·09	24·39	9843·67	0·0025
A1SP15	− 0·15	36·31	10600·92	0·0034
A2SP15	− 0·25	41·97	11301·99	0·0037
A3SP15	− 0·47	63·40	10024·06	0·0063
A4SP15	− 0·76	86·50	9250·55	0·0094
A5SP15	− 0·39	50·67	9444·91	0·0054

Figure 9a–c shows the scar profiles of A0SP15, A2SP15 and A4SP15 scratched under unlubricated condition at F_N = 20 N. It may be seen from Fig. 9a that pure Al₂O₃ underwent comparatively higher extent lateral cracking than the nanocomposites containing at least up to 0·27 wt-% CNT (Fig. 9b). This suggested that beside higher hardness of the nanocomposites over pure Al₂O₃ (Fig. 5), the uniformly dispersed reinforcing CNTs offered higher pullout resistance to the matrix grains during scratch test up to 20 N normal load that significantly lowered the extent of martial removal through lateral cracking. However, at higher CNT loading (especially at >0·27 wt-%), agglomerated nanotubes in matrix phase failed to offer any significant pullout resistance to the matrix grains, and as a result for A4SP15 and A5SP15, both scar width and depth were found to be higher than that of A0SP15 specimen (Fig. 9c). Figure 10a–f shows the variation in coefficient of friction (COF) with scratching distance of all the specimens. It may be seen from the data given in the figures that the average values of COF at different normal loads varied over a narrow range (0·50–0·59), which are found to be almost similar or lower than those available in literature on Al₂O₃ or Al₂O₃ matrix composites, suggesting possible improvement in frictional behaviour of the present nanocomposites even in unlubricated condition.^{18,19,27,28,37} Furthermore, at 0·27 wt-% CNT loading, COF values were found to be decreased than A0SP15 specimen (Fig. 10d), indicating possible exploitation of self-lubricating properties of nanotube in the composites. However, since all other nanocomposites offered almost similar extent of COF values as that of pure Al₂O₃ (Fig. 10a–c, e and f), confirmation on self-lubricating behaviour of MWCNT in the present nanocomposites requires further study. Table 5 represents the wear rate of studied specimens at three scratching loads. From the table, it is clear that nanocomposites containing up to 0·27 wt-% MWCNT offered improved wear resistance than pure Al₂O₃ up to F_N = 20 N. Table 5 further indicates that from 0·27 wt-% CNT concentration in the matrix phase, the wear resistance of the nanocomposites started to decrease compared to A1SP15 and A2SP15 nanocomposites due to the same reason mentioned during crystallite size analyses. Among all spark plasma sintered nanocomposites, specimen A2SP15 offered the highest resistance to sliding wear up to 20 N (Table 5). Depending on F_N, W_R of 0·13 wt-% MWCNT/Al₂O₃ nanocomposite was found to be ∼28% (at F_N = 10 N) and ∼40% (at F_N = 20 N) lower than those of pure Al₂O₃ (Table 5). Analogous to the results obtained during ISE analyses of present specimen, wear results also suggested that even at higher scratching load, where pure Al₂O₃ failed to offer adequate resistance to the scratching force, specimen A2SP15 offered significant resistance to the applied normal load up to 20 N, indicating useful reinforcing effect of structurally intact CNTs in the specimen (Table 5).

Scar profiles of a pure Al₂O₃, b 0·27 wt-% MWCNT/Al₂O₃ nanocomposite and c 0·54 wt-% MWCNT/Al₂O₃ nanocomposite at F_N = 20 N

a–f COF values of all studied specimens at three different F_N‘s against scratching distance

Table 5

Rate of wear (W_R) in mm³ (N m)^− 1 of studied specimens

Load/N	A0SP15	A1SP15	A2SP15	A3SP15	A4SP15	A5SP15
10	0·0257	0·0203	0·0186	0·0228	0·0326	0·0339
15	0·0260	0·0205	0·0203	0·0275	0·0338	0·0343
20	0·0368	0·0261	0·0224	0·0314	0·0391	0·0459

Conclusions

MWCNT/α-Al₂O₃ nanocomposites (RD >99%) doped with 0·5 wt-% MgO have been fabricated by simple wet mixing of as received raw materials followed by SPS at 1500°C under 50 MPa for 3 min in static argon ambience. Combined grain refining effect of CNT and MgO resulted in much smaller matrix grains in the nanocomposites over pure Al₂O₃. W–H analyses also revealed significantly lowered matrix L_C values in the nanocomposites than that of pure Al₂O₃. Furthermore, the large CTE mismatch between CNT and Al₂O₃ resulted in the formation of increased compressive lattice microstrain in the nanocomposite specimens. Microhardness analyses indicated negligible ISE of the present nanocomposites (Meyer's exponent ≥ 1·90). HV value as high as ∼24·5 GPa was obtained only at 0·13 wt-% CNT loading, which was ∼18% higher than monolithic Al₂O₃. While such high HV value of the present nanocomposite is one of the best values available in published literature, this was achieved at only 0·13 wt-% CNT loading due to uniform dispersion of structurally intact nanotubes in the SPS-ed specimens. In addition, PSR and M-PSR model fitting results revealed much improved elastic response of the nanocomposites over pure Al₂O₃ due to effective reinforcing effect rendered by the structurally intact and highly elastic CNTs. The ratio of elastic to plastic deformation (i.e. a₁/a₂) steadily increased from 0·0025 for pure Al₂O₃ to the maximum of 0·0094 for 0·54 wt-% MWCNT/Al₂O₃ nanocomposite. Unlubricated scratch test suggested much higher resistance to wear of the nanocomposites over pure Al₂O₃ at least up to 0·27 wt-% CNT loading. W_R at F_N = 20 N decreased from 0·0368 mm³ (N m)^− 1 for pure Al₂O₃ to a minimum of 0·0224 mm³ (N m)^− 1 for 0·13 wt-% MWCNT/Al₂O₃ nanocomposite. The present results suggest enhanced mechanical properties of studied nanocomposites over pure alumina and have potential for advanced high performance structural applications.

Footnotes

Acknowledgements

The authors express their sincere gratitude to the Director, CSIR-CG&CRI, India, for his kind permission to publish this work. The authors thankfully acknowledge financial support received from the Council of Scientific and Industrial Research (CSIR), India (WP3·0 of ESC-0104).

References

Kern

and Gadow

: ‘Alumina toughened zirconia from yttria coated powders’, J. Eur. Ceram. Soc., 2012, 32, (15), 3911–3918. doi: 10.1016/j.jeurceramsoc.2012.03.014

Chi

, Serjouei

, Sridhar

and Tan

G. E. B.

: ‘Ballistic impact on bilayer alumina/aluminium armor: a semianalytical approach’, Int. J. Impact Eng., 2013, 52, 37–46. doi: 10.1016/j.ijimpeng.2012.10.001

Takahashi

, Matsuzaki

, Funai

, Wada

, Fukunaga

, Takatsuka

and Mano

: ‘Effects of membrane properties on CO₂ desorption from chemical absorbents using a membrane flash process’, Energy Procedia, 2013, 37, 1060–1066. doi: 10.1016/j.egypro.2013.05.202

Wadley

H. N. G.

, O'Masta

M. R.

, Dharmasena

K. P.

, Compton

B. G.

, Gamble

E. A.

and Zok

F. W.

: ‘Effect of core topology on projectile penetration in hybrid aluminum/alumina sandwich structures’, Int. J. Impact Eng., 2013, 62, 99–113. doi: 10.1016/j.ijimpeng.2013.05.008

Medvedovski

: ‘Alumina ceramics for ballistic protection. Part 2’, Am. Ceram. Soc. Bull., 2002, 81, (4), 45–50.

Auerkari

,; 1996;. Mechanical and physical properties of engineering alumina ceramics, Technical Research Centre of Finland, VTT Tiedotteita, Meddelanden, Espoo.

Liu

, Shao

, Hu

and Wang

: ‘Surface modification and microstructure design of alumina ceramics’, J. Am. Ceram. Soc., 2011, 94, (9), 2764–2766. doi: 10.1111/j.1551-2916.2011.04705.x

Septawendar

and Edwin

: ‘Nanocrystalline α-Al₂O₃ powder preparation with sucrose as a template through a chemical route’, J. Ceram. Process. Res., 2011, 12, (4), 365–370.

Matori

K. A.

, Wah

L. C.

, Hashim

, Ismail

and Zaid

M. H. M.

: ‘Phase transformations of α-Alumina made from waste aluminum via a precipitation technique’, Int. J. Mol. Sci., 2012, 13, 16812–16821. doi: 10.3390/ijms131216812

10.

Korinek

S. L.

and Castaing

: ‘Slip and twinning in polycrystalline alumina (a-Al₂O₃) deformed under hydrostatic pressure between 600° and 1000°C’, J. Am. Ceram. Soc., 2003, 86, (4), 566–573. doi: 10.1111/j.1151-2916.2003.tb03342.x

11.

Sarkar

and Das

P. K.

: ‘Temperature and load dependent mechanical properties of pressureless sintered carbon nanotube/alumina nanocomposites’, Mater. Sci. Eng. A, 2012, A531, 61–69. doi: 10.1016/j.msea.2011.10.024

12.

Chin

Y. L.

and Tuan

W. H.

: ‘Use of MAX particles to improve the toughness of brittle ceramics’, in ‘Advances in Science and Technology of Mn+1Axn Phases’, (ed. Low

I. M.

., 26; 2012, Cambridge, Woodhead Publishing.

13.

Zhang

X. H.

, Liu

C. X.

, Li

M. S.

, Zhang

J. H.

and Sun

J. L.

: ‘Toughening mechanism of alumina-matrix ceramic composites with the addition of AlTiC master alloys and ZrO₂’, Ceram. Int., 2009, 35, 93–97. doi: 10.1016/j.ceramint.2007.09.109

14.

Rodriguez-Suarez

, Díaz

L. A.

, Torrecillas

, Lopez-Esteban

, Tuan

W. -H.

, Nygren

and Moya

J. S.

: ‘Alumina/tungsten nanocomposites obtained by Spark Plasma Sintering’, Compos. Sci. Technol., 2009, 69, 2467–2473. doi: 10.1016/j.compscitech.2009.06.022

15.

Reveron

, Zaafrani

and Fantozzi

: ‘Microstructure development, hardness, toughness and creep behaviour of pressureless sintered alumina/SiC micro–nanocomposites obtained by slip-casting’, J. Eur. Ceram. Soc., 2010, 30, 1351–1357. doi: 10.1016/j.jeurceramsoc.2009.12.008

16.

Iijima

: ‘Helical microtubules of graphitic carbon’, Nature, 1991, 354, (7), 56–58. doi: 10.1038/354056a0

17.

Michálek

, Bodišová

, Michálková

, Sedláček

and Galusek

: ‘Alumina/MWCNTs composites by aqueous slip casting and pressureless sintering’, Ceram. Int., 2013, 39, 6543–6550. doi: 10.1016/j.ceramint.2013.01.087

18.

Lee

K.-S.

, Jang

B.-K.

and Sakka

: ‘Damage and wear resistance of Al₂O₃–CNT nanocomposites fabricated by spark plasma sintering’, J. Ceram. Soc. Jpn, 2013, 121, (10), 867–872. doi: 10.2109/jcersj2.121.867

19.

Lim

D.-S.

, You

D.-H.

, Choi

H.-J.

, Lim

S.-H.

and Jang

: ‘Effect of CNT distribution on tribological behavior of alumina–CNT composites’, Wear, 2005, 259, 539–544. doi: 10.1016/j.wear.2005.02.031

20.

Sarkar

and Das

P. K.

: ‘Effect of sintering temperature and nanotube concentration on microstructure and properties of carbon nanotube/alumina nanocomposites’, Ceram. Int., 2014, 40, 7449–7458. doi: 10.1016/j.ceramint.2013.12.092

21.

, Su

, Hou

, Liu

, Song

W.-L.

and Cao

M.-S.

: ‘Electrical conductivity and microwave absorption of shortened multi-walled carbon nanotube/alumina ceramic composites’, Ceram. Int., 2013, 39, 5979–5983. doi: 10.1016/j.ceramint.2012.12.057

22.

Michálek

, Sedláček

, Parchoviansky

, Michálková

and Galusek

: ‘Mechanical properties and electrical conductivity of alumina/MWCNT and alumina/zirconia/MWCNTcomposites’, Ceram. Int., 2014, 40, 1289–1295. doi: 10.1016/j.ceramint.2013.07.008

23.

Yamamoto

, Omori

, Yokomizo

and Hashida

: ‘Mechanical properties and structural characterization of carbon nanotube/alumina composites prepared by precursor method’, Diamond Relat. Mater., 2008, 17, 1554–1557. doi: 10.1016/j.diamond.2008.01.059

24.

Sarkar

and Das

P. K.

: ‘Processing and properties of carbon nanotube/alumina nanocomposites: a review’, Rev. Adv. Mater. Sci., 2014, 37, 53–82.

25.

Sarkar

and Das

P. K.

: ‘Role of interface on electrical conductivity of carbon nanotube/alumina nanocomposite’, Ceram. Int., 2014, 40, (2), 2723–2729. doi: 10.1016/j.ceramint.2013.10.049

26.

Bakhsh

, Khalid

F. A.

and Hakeem

A. S.

: ‘Synthesis and characterization of pressureless sintered carbon nanotube reinforced alumina nanocomposites’, Mater. Sci. Eng. A, 2013, A578, 422–429. doi: 10.1016/j.msea.2013.04.020

27.

J.-W.

, You

D.-H.

and Lim

D.-S.

: ‘Tribological properties of hot-pressed alumina–CNT composites’, Wear, 2003, 255, 677–681. doi: 10.1016/S0043-1648(03)00216-3

28.

Keshri

A. K.

and Agarwal

: ‘Wear behavior of plasma-sprayed carbon nanotube-reinforced aluminum oxide coating in marine and high-temperature environments’, J. Therm. Spray Technol., 2011, 20, (6), 1217–1230. doi: 10.1007/s11666-011-9669-2

29.

Puchy

, Hvizdos

, Dusza

, Kovac

, Inam

and Reece

M. J.

: ‘Wear resistance of Al₂O₃–CNT ceramic nanocomposites at room and high temperatures’, Ceram. Int., 2013, 39, 5821–5826. doi: 10.1016/j.ceramint.2012.12.100

30.

Thomson

K. E.

, Jiang

, Ritchie

R. O.

and Mukherjee

A. K.

: ‘A preservation study of carbon nanotubes in alumina-based nanocomposites via Raman spectroscopy and nuclear magnetic resonance’, Appl. Phys. A, 2007, 89, (3), 651–654. doi: 10.1007/s00339-007-4253-9

31.

Williamson

G. K.

and Hall

W. H.

: ‘X-ray line broadening from filed aluminium and wolfram’, Acta Metall. Mater., 1953, 1, 22–31. doi: 10.1016/0001-6160(53)90006-6

32.

Milsom

, Viola

, Gao

, Inam

, Peijs

and Reece

M. J.

: ‘The effect of carbon nanotubes on the sintering behaviour of zirconia’, J. Eur. Ceram. Soc., 2012, 32, 4149–4156. doi: 10.1016/j.jeurceramsoc.2012.07.028

33.

Kumar

A. S.

, Durai

A. R.

and Sornakumar

: ‘Development of yttria and ceria toughened alumina composite for cutting tool application’, Int. J. Refract. Met. Hard Mater., 2007, 25, 214–219. doi: 10.1016/j.ijrmhm.2006.05.002

34.

Pipes

R. B.

and Hubert

: ‘Helical carbon nanotube arrays: thermal expansion’, Compos. Sci. Technol., 2003, 63, 1571–1579. doi: 10.1016/S0266-3538(03)00075-7

35.

Csehova

, Andrejovská

, Limpichaipanit

, Dusza

and Todd

: ‘Indentation load-size effect in Al₂O₃–SiC nanocomposites’, J. Electr. Eng., 2010, 61, (5), 305–307.

36.

Sebastian

and Khadar

M. A.

: ‘Microhardness indentation size effect studies in 60B₂O₃–(40 − x)PbO–xMCl₂ and 50B₂O₃(50 − x)PbO–xMCl₂ (M = Pb, Cd) glasses’, J. Mater. Sci., 2005, 40, (7), 1655–1659. doi: 10.1007/s10853-005-0666-z

37.

Kim

S.-H.

and Lee

S. W.

: ‘Wear and friction behavior of self-lubricating alumina–zirconia–fluoride composites fabricated by the PECS technique’, Ceram. Int., 2014, 40, 779–790. doi: 10.1016/j.ceramint.2013.06.068