The study of the preparation and tribological behavior of TiAl matrix composites containing 1 wt% multi-walled carbon nanotubes

Abstract

The objective of searching for an optimized applied load was to minimize friction and decrease energy dissipation in rotating mechanical components. TiAl matrix self-lubricating composites containing 1.0 wt% multi-walled carbon nanotubes (TiAl-1.0 wt% MWCNTs), were evaluated over 80 min on a ball-on-disk tribometer at 1.65, 4.15, 6.65, 9.15 and 11.65 N, for their sliding friction and wear behaviors. The testing results showed that TiAl-1.0 wt% MWCNTs obtained excellent sliding friction and wear behaviors at 9.15 N with small friction coefficients and low wear rates, compared to those at 1.65, 4.15, 6.65 and 11.65 N. It was found that the small mean wear rates of TiAl-1.0 wt% MWCNTs were attributed to the high subsurface hardness of the wear scar. The low standard deviation (STDEV) of the wear rates was mainly determined by the homogeneous thickness of the compacted layer at 9.15 N.

---

1. Introduction

In recent years, the rapidly increasing attention given to TiAl alloys was attributed to their excellent combination of mechanical behaviors and thermal properties.^1–4 However, the poor tribological behaviors of TiAl alloys led to limited development in the applied fields of the aerospace, automotive and energy industries, etc.^5–7 Hence, an effective method was proposed to improve the friction and wear behaviors of TiAl alloys by fabricating TiAl matrix self-lubricating composites containing solid lubricants, such as MoS₂,⁸ multilayer graphene,⁹ graphite,¹⁰ boron nitride,¹¹ Ti₃SiC₂ (ref. 12 and 13) and multi-walled carbon nanotubes (MWCNTs).¹⁴

It was universally acknowledged that the friction and wear behaviors of a material could be significantly improved by adding solid lubricant MWCNTs. Ahmad et al.¹⁴ studied the wear resistant properties of Al₂O₃ nano-composites containing MWCNTs, and found that the mean weight loss of the Al₂O₃ nano-composites significantly increased with the increase of the applied load from 14 to 35 N. In addition, Sun et al.¹⁵ investigated the sliding friction and wear behavior of Ti₂AlN/TiAl composites. The testing results showed that the mean wear rates of Ti₂AlN/TiAl were gradually improved with the increase of the applied load between 0–5 N. Hence, the tribological behavior of materials was considered to have a close relationship with the chosen applied loads. However, research on the lubricating mechanism of TiAl matrix self-lubricating composites containing MWCNTs (TiAl-MWCNTs) at different applied loads was rarely reported.

In this study, the method of spark plasma sintering (SPS) using a D.R. Sinter® SPS3.20 system was adopted to fabricate TiAl-MWCNTs. According to the chosen applied loads of Yang et al.,¹⁶ the 80 min sliding friction and wear tests of TiAl-MWCNTs were carried out on a HT-1000 ball-on-disk tribometer at 1.65, 4.15, 6.65, 9.15 and 11.65 N. The excellent sliding friction and wear behaviors at the optimized applied load were evaluated by studying the subsurface hardness of the wear scar, the distribution and lubrication of MWCNTs, and the thickness of the compacted layers.

2. Experimental details

2.1 Material fabrication

In accordance with the research results of Melk et al.,¹⁷ the method of SPS was adopted to fabricate TiAl-MWCNTs ((48 at% Ti , 47 at% Al, 2 at% Cr, 2 at% Nb, 1 at% B)-MWCNTs) containing various amounts of MWCNTs (0.5 wt%, 1.0 wt% and 1.5 wt%). The commercially available powders (less than 20 μm in average size, 99.9% in purity) of Ti, Al, B, Nb, Cr and MWCNTs were chosen as raw materials. Fig. 1 exhibits the typical multi-walled molecular structure and FESEM (field emission scanning electron microscopy) morphology of MWCNTs. The multilayer molecular structure of MWCNTs is clearly exhibited, as shown in Fig. 1a. The morphology of the chosen MWCNT powder was better observed with the help of FESEM, as shown in Fig. 1b. Fig. 1c shows the typical XRD pattern of MWCNT powders. Vibration milling at a frequency of 45 Hz was adopted to mix the as-prepared powders in the Teflon vials. After being dried, the mixed powders were loaded into the cylindrical graphite molds with a 25 mm inner diameter, and then sintered for 5 min using the apparatus of D.R. Sinter® SPS3.20 (Sumitomo Coal & Mining, now SPS Syntex Inc.) at a temperature of 1100 °C. The protection atmosphere was pure Ar gas, the sintered pressure was 30 MPa, and the heating rate was 100 °C min⁻¹.

Fig. 1 Typical multi-walled molecular structure (a), FESEM morphology (b), and XRD pattern (c) of MWCNTs.

2.2 Vickers microhardness and density tests

According to the ASTM standard E92-82,¹⁸ TiAl-1.0 wt% MWCNTs obtained a mean Vickers hardness of 6.02 GPa using HVS-1000 apparatus (Changzhou detu Precision Instrument Co Ltd., China). Compared to the mean Vickers hardness (5.57 GPa) of TiAl alloy,¹⁹ the mean Vickers hardness of TiAl-1.0 wt% MWCNTs was significantly improved, after adding MWCNTs. According to the ASTM standard B962-08,²⁰ a mean density of 3.54 g cm⁻³ was obtained using Archimedes’ principle.

2.3 Friction and wear tests

The sliding friction and wear tests were carried out on a HT-1000 ball-on-disk tribometer (Zhong Ke Kai Hua Corporation, China) according to the ASTM standard G99-95.²¹ The as-prepared samples served as the disks rotating against the GCr15 ball of 6 mm. The 80 min sliding wear test was carried out at a sliding speed of 0.5 m s⁻¹ and at a relative humidity of 55–75%. The sliding friction coefficients were automatically recorded using the computer-controlled system of the HT-1000 at the applied loads of 1.65, 4.15, 6.65, 9.15 and 11.65 N. The wear rate W of TiAl-MWCNTs could be defined as formula (1):^22,23where V is the wear volume in mm³, F is the applied load in N, and L is the total sliding distance in mm.

2.4 Microstructure analysis

The surfaces of as-prepared samples were examined using XRD at a scanning speed of 0.01° s⁻¹. The morphologies of the wear scars were analyzed by electron probe microanalysis (EPMA) using a JAX-8230 analyser (JEOL Corporation, Japan). The morphologies of the wear scar cross-sections were characterized with the help of a FEI-SIRION 200 (JEOL Corporation, Japan) field emission scanning electron microscope (FESEM).

3. Results and discussion

3.1 Compositions of TiAl-MWCNTs

Fig. 2a exhibits the typical XRD pattern of the TiAl based alloy prepared by SPS. Fig. 2b shows the typical FESEM image of the TiAl based alloy. As shown in Fig. 2, the TiAl based alloy fabricated by SPS contained a small quantity of TiC.

Fig. 2 Typical XRD pattern (a) and FESEM image (b) of the TiAl based alloy prepared by SPS.

Fig. 3a shows the typical XRD pattern of TiAl-1.0 wt% MWCNTs prepared by SPS. As is clear in Fig. 3a, the TiAl-1.0 wt% MWCNTs were mainly composed of TiAl, MWCNTs and a small amount of TiC according to the intensities of the diffraction peaks of different phases. Fig. 3b exhibits the representative FESEM image of MWCNTs in the TiAl-1.0 wt% MWCNT cross-section. As shown in Fig. 3b, MWCNTs were tightly combined with the TiAl based alloy.

Fig. 3 Typical XRD pattern of TiAl-1.0 wt% MWCNTs prepared by SPS (a) and the FESEM image of MWCNTs in the TiAl-1.0 wt% MWCNT cross-section (b).

According to the second law of thermodynamics,^24–26 the synthesis reaction of Ti, Al and C powders during the SPS process could be written as formulae (2)–(8):

where ΔG⁰₁, ΔG⁰₂, ΔG⁰₃, ΔG⁰₄, ΔG⁰₅ and ΔG⁰₆ are known as the Gibbs free energy. As shown in formulae (2)–(8), Al₄C₃, Ti₃Al and TiAl₃ acted as the intermediate reaction compounds during the synthesis reaction process and resulted in the formation of a small amount of TiC.

3.2 Friction coefficient and wear rate

Fig. 4 shows the representative friction coefficients and mean wear rates of TiAl-MWCNTs. As shown in Fig. 4, TiAl-1.0 wt% MWCNTs had small sliding friction coefficients and low mean wear rates, as compared to TiAl-0.5 wt% MWCNTs and TiAl-1.5 wt% MWCNTs.

Fig. 4 Representative friction coefficients (a) and mean wear rates (b) of TiAl-MWCNTs.

An optimized applied load was beneficial to the decrease of friction and mechanical energy dissipation in rotating mechanical assemblies.

Fig. 5 shows the typical friction coefficients, mean wear rates, and standard deviation (STDEV) of the wear rates of TiAl-1.0 wt% MWCNTs at different applied loads. As shown in Fig. 5, the small friction coefficient and low mean wear rate of TiAl-1.0 wt% MWCNTs were obtained at 9.15 N. Fig. 5c exhibits the typical STDEV of five of the measured wear rates of TiAl-1.0 wt% MWCNTs after 80 min of sliding wear at different applied loads. As can be seen in Fig. 5c, TiAl-1.0 wt% MWCNTs obtained the lowest STDEV (0.013) of wear rates at 9.15 N. As shown in Fig. 5, TiAl-1.0 wt% MWCNTs obtained smaller sliding friction coefficients, lower mean wear rates and lower STDEV of the wear rates at 9.15 N, compared to those at 1.65, 4.15, 6.65 and 11.65 N.

Fig. 5 Typical friction coefficients (a), mean wear rates (b), and STDEV of the wear rates (c) of TiAl-1.0 wt% MWCNTs at different applied loads.

3.3 Influence of the subsurface hardness of the wear scars

Fig. 6 shows the representative eight testing locations of one wear scar marked by eight yellow spheres. Fig. 7 exhibits the Vickers hardness of one wear scar distributed in eight testing locations at different applied loads. Table 1 indicates the typical distribution of the subsurface Vickers hardness of the eight testing locations at 1.65, 4.15, 6.65, 9.15 and 11.65 N. As shown in Fig. 6 and 7 and Table 1, the higher mean Vickers hardness of 6.31 was more uniformly distributed in the wear scar subsurface with a low STDEV of 0.16 at 9.15 N, compared to those at 1.65, 4.15, 6.65 and 11.65 N.

Fig. 6 Eight representative testing locations of one wear scar marked by eight yellow spheres.

Fig. 7 Vickers hardness of one wear scar distributed in eight testing locations at different applied loads.

Table 1 Typical distribution of the subsurface Vickers hardness of the eight testing locations at 1.65, 4.15, 6.65, 9.15 and 11.65 N

Applied load (N)	Hardness of eight testing locations (GPa)								Mean hardness (GPa)	STDEV
1.65	5.53	5.85	5.87	5.96	6.04	6.11	6.44	6.48	6.04	0.31
4.15	5.56	5.88	6.03	6.14	6.22	6.31	6.34	6.42	6.11	0.28
6.65	5.89	5.94	5.97	6.23	6.28	6.41	6.46	6.51	6.21	0.25
9.15	6.01	6.18	6.24	6.31	6.36	6.38	6.47	6.49	6.31	0.16
11.65	5.52	5.70	5.88	6.34	6.42	6.54	6.90	6.96	6.28	0.54

---

According to the discussions of Ahmad et al.,²⁷ the wear volume V of TiAl-1.0 wt% MWCNTs could be written as formula (9):

where F is the applied load in N, σ_s is the contact stress in MPa, A_c is the contact area in mm², K_IC is the fracture toughness in MPa, H is the Vickers hardness of the wear scar subsurface in GPa, and a is the constant independent of the material type.

According to the descriptions about formulae (9) and (1), the wear rate W of TiAl-1.0 wt% MWCNTs could be redefined as formula (10) by substituting formula (9) into (1):

As shown in formula (10), the wear rate of TiAl-1.0 wt% MWCNTs was inversely proportional to the Vickers hardness of the wear scar subsurface. Hence, it can be concluded that the applied load of 9.15 N was beneficial to the acquisition of high subsurface hardness of the wear scar (6.31 GPa). The high subsurface hardness led to the low mean wear rate (1.57 × 10⁻⁴ mm³ N⁻¹ m⁻¹).

3.4 Wear mechanisms and structural features

Fig. 8 exhibits the typical electron probe morphologies of the wear scars after 80 min of sliding wear at 1.65, 4.15, 6.65, 9.15 and 11.65 N. As noted in Fig. 8a, massive abrasive debris existed on the wear scar at 1.65 N. It was apparent that the primary wear mechanism was abrasive abrasion. As can be seen from Fig. 8b, massive isolated island-like (patchy) layers and discontinuous wear grooves appeared on the wear scar at 4.15 N. It was evident that the dominant wear mechanism was plastic deformation at 4.15 N. As the applied load increased up to 6.65 N from 4.15 N, the main wear mechanisms were plastic deformation and severe peeling as indicated by the appearance of big peeling pits at 6.65 N (see Fig. 8c). As shown in Fig. 8d, when the chosen applied load increased from 6.65 N up to 9.15 N, small peeling pits appeared on the wear scar. This indicated a primary wear mechanism of slight peeling at 9.15 N. As can be seen in Fig. 8e, the high applied load of 11.65 N led to the formation of big peeling pits. It was noticeable that the main wear mechanism was severe peeling at 11.65 N.

Fig. 8 Typical electron probe morphologies of the wear scars after 80 min sliding wear at 1.65 N (a), 4.15 N (b), 6.65 N (c), 9.15 N (d) and 11.65 N (e).

Fig. 9a shows the typical surface textures of the wear scars of TiAl-1.0 wt% MWCNTs after 80 min of sliding at 9.15 N. As shown in Fig. 9a, TiAl-1.0 wt% MWCNTs obtained an excellent surface texture of the wear scars at 9.15 N with small height parameters (Sa: 0.29 μm, Sq: 0.25 μm and Sku: 5.81). According to the aforementioned discussions, it could be concluded that the smooth wear scar (see Fig. 8d) was beneficial to acquiring low surface roughness at 9.15 N. The low surface roughness of the wear scar led to an increase in the effective contact area between GCr15 balls and TiAl-1.0 wt% MWCNTs, resulting in the lowering of the effective contact stress. Ahmad et al.²⁷ investigated the wear resistant properties of MWCNT reinforced Al₂O₃ nanocomposites, and found that wear volume V continually increased with the increase of contact stress (see formula (9)). Formula (10) shows that the wear volume V is in proportional relation with the wear rate W. Hence, it can be concluded that the decreasing contact stress was advantageous to the acquirement of the lower wear rate at 9.15 N. Simultaneously, the homogeneous distribution of the Vickers hardness of the subsurface was also helpful for acquiring the small STDEV of wear rates (see Fig. 5c) and the smooth morphology of the wear scar (see Fig. 8d) at 9.15 N. Fig. 9b exhibits the FESEM morphologies of the wear scars after 80 min of sliding wear at 9.15 N. As can be seen from Fig. 9b, massive MWCNTs were uniformly exposed in the wear scar at 9.15 N, and formed a lubricating film with friction-reducing and anti-wear properties, leading to the small friction coefficient and low wear rate. Meanwhile, the exposed MWCNTs on the wear scar also improved the high Vickers hardness of the wear scar subsurface at 9.15 N. Fig. 9c indicates the microstructure of the lubricating film at 9.15 N. As noted in Fig. 9b and c, the MWCNT contribution to the lubricating film could be summarized into two main functions: a reinforcing function and a connective function. Reinforcing function: MWCNTs, possessing a high yield strength, high Young modulus and high Vickers hardness etc., existed in the lubricating film. This resulted in the improvement of the mechanical properties of the lubricating film at 9.15 N. Connective function: the MWCNTs of 200 μm in length were uniformly exposed in the wear scar, tightly combined with wear debris under the action of the sliding applied loads, and formed the lubricating film with a stable structure. Hence, the lubricating film with a stable structure and excellent mechanical properties enhanced the sliding friction and wear behavior of TiAl-1.0 wt% MWCNTs at 9.15 N.

Fig. 9 Typical surface textures of wear scars (a), FESEM morphologies of wear scars (b), and microstructure of the lubricating film (c).

Fig. 10 shows the representative electron probe morphologies of the wear scars at 6.65, 9.15 and 11.65 N. As shown in Fig. 10, the smooth morphology of wear scars (see Fig. 10b) was beneficial to the homogeneous distribution of the subsurface hardness of the wear scar at 9.15 N. The homogeneous subsurface hardness inhibited the formation and propagation of fatigue cracks, resulting in the acquisition of the smoother edges of the wear scar, compared to those at 6.65 (see Fig. 10a) and 11.65 N (see Fig. 10c). The smooth edges of the wear scars were beneficial to calculating accurate wear rates, resulting in the small STDEV of the wear rates at 9.15 N.

Fig. 10 Representative electron probe morphologies of the wear scars at 6.65 N (a), 9.15 N (b) and 11.65 N (c).

3.5 Analysis of wear scar cross-sections

Fig. 11 exhibits the typical FESEM micrographs of wear scar cross-sections obtained after tests at 6.65, 9.15 and 11.65 N. The positions of the wear scar cross-sections were marked by straight lines in Fig. 8c–e. As noted in Fig. 11, the significant stratification morphologies were marked as layers A–C at 6.65 N, layers D–F at 9.15 N and layers G–I at 11.65 N. Layers A, D and G were known as the lubricating films. Layers B, E and H were named as the compacted layers for the formation of submicron grains. Layers C, F and I were identified as the substrate material of TiAl-1.0 wt% MWCNTs. As is clear in Fig. 11, the homogeneous thickness of the compacted layer was beneficial to the acquisition of a more homogeneous subsurface hardness of the wear scar at 9.15 N, compared to those at 6.65 and 11.65 N, leading to the small STDEV of the wear rate at 9.15 N.

Fig. 11 Typical FESEM micrographs of wear scar cross-sections obtained after tests at 6.65 N (a), 9.15 N (b) and 11.65 N (c).

In this study, based on the HT-1000 ball-on-disk tribometer, the 80 min sliding friction and wear tests of TiAl-1.0 wt% MWCNTs were carried out against GCr15 balls at 1.65, 4.15, 6.65, 9.15 and 11.65 N. TiAl-1.0 wt% MWCNTs obtained small sliding friction coefficients, lower mean wear rates, and lower STDEV of wear rates at 9.15 N, compared to those at 1.65, 4.15, 6.65 and 11.65 N. The small mean wear rate was attributed to the high subsurface hardness of the wear scar. The low STDEV of the wear rates was mainly dependent on the homogeneous distribution of the subsurface hardness at 9.15 N. The MWCNTs on the wear scar caused the high Vickers hardness of the subsurface. The homogeneous thickness of the compacted layer led to the homogeneous distribution of the subsurface hardness and the low STDEV of the wear rate at 9.15 N.

4. Conclusions

TiAl-1.0 wt% MWCNTs (TiAl matrix self-lubricating composites containing 1.0 wt% MWCNTs) were sintered for 5 min at a temperature of 1100 °C and at a pressure of 30 MPa using SPS. The sliding friction and wear tests of TiAl-1.0 wt% MWCNTs were carried out at 1.65, 4.15, 6.65, 9.15 and 11.65 N, respectively. The conclusions obtained were as follows:

(1) TiAl-1.0 wt% MWCNTs obtained small friction coefficients, lower mean wear rates, and lower STDEV of the wear rates at 9.15 N, compared to those at 1.65, 4.15, 6.65 and 11.65 N.

(2) The small mean wear rate of TiAl-1.0 wt% MWCNTs obtained after tests at 9.15 N was attributed to the high Vickers hardness of the wear scar subsurface due to the existence of MWCNTs on the wear scar.

(3) The homogeneous compacted layer thickness was helpful for obtaining the homogeneous distribution of the Vickers hardness of the subsurface at 9.15 N.

(4) The low STDEV of the wear rates at 9.15 N was mainly dependent on the homogeneous distribution of the Vickers hardness of the subsurface.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (51275370) and the Self-determined and Innovative Research Funds of WUT (135204008); the authors also wish to gratefully thank the Material Research and Testing Center of Wuhan University of Technology for their assistance; the authors are grateful to M. J. Yang, S. L. Zhao and W. T. Zhu in Material Research and Test Center of WUT for their kind help with EPMA and FESEM.

References

1. C. M. Austin, Curr. Opin. Solid State Mater. Sci., 1999, 4, 239.
2. X. H. Wu, Intermetallics, 2006, 14, 1114.
3. S. Y. Sung and Y. J. Kim, Intermetallics, 2007, 15, 468.
4. T. Kawabata, H. Fukai and O. Izumi, Acta Mater., 1998, 46, 2185.
5. A. Rastkar, A. Bloyce and T. Bell, Wear, 2000, 240, 19.
6. C. Li, J. Xia and H. Dong, Wear, 2006, 261, 693.
7. K. Miyoshi, B. Lerch and S. Draper, Tribol. Int., 2003, 36, 145.
8. W. Z. Zhai, X. L. Shi, M. Wang, Z. S. Xu and J. Yao, Tribol. Trans., 2014, 57(3), 416.
9. Z. S. Xu, L. Chen, X. L. Shi and Q. X. Zhang, Tribol. Trans., 2015, 58(4), 668.
10. D. W. Dareing and S. Atluri, Tribol. Trans., 1997, 40(3), 413.
11. Y. Ming, X. Chen, V. P. Lin, Y. W. Dravid, M. S. Chung and W. D. Wong, Tribol. Trans., 1993, 36(3), 491.
12. Z. S. Xu, X. L. Shi, Q. X. Zhang and W. Z. Zhai, Tribol. Trans., 2014, 57(6), 1017.
13. Z. S. Xu, B. Xue, X. L. Shi and Q. X. Zhang, Tribol. Trans., 2015, 58, 87.
14. I. Ahmad, A. Kennedy and Y. Q. Zhu, Wear, 2010, 269, 71.
15. T. Sun, Q. D. Wang, L. Sun and G. H. Wu, Wear, 2010, 268, 693.
16. K. Yang, X. L. Shi, W. Z. Zhai and A. M. M. Ibrahim, RSC Adv., 2015, 5, 67102.
17. L. Melk, J. J. R. Rovira, M. L. Antti and M. Anglada, Ceram. Int., 2015, 41(1), 459.
18. ASTM E92-82, Standard test method for Vickers hardness of metallic materials, ASTM International, 2003.
19. X. L. Shi, J. Yao, Z. S. Xu and W. Z. Zhai, Mater. Des., 2014, 53, 620.
20. ASTM B962-08, Standard test methods for density of compacted or sintered powder metallurgy (PM) products using Archimedes’ principle, ASTM International, 2008.
21. ASTM G99-95, Standard test method for wear testing with a pin-on-disk apparatus, ASTM International, 2195.
22. R. D. Tyagi, S. Xiong and J. L. Li, Wear, 2011, 270, 423.
23. C. X. Li, J. Xia and H. Dong, Wear, 2006, 261, 693.
24. Y. L. Yue, H. T. Wu, B. Wu, Z. J. Wang and W. H. Tao, Journal of University of Jinan (Science and Technology), 2005, 19(2), 106.
25. Y. Y. Li, P. F. Xiang, Y. Long and Q. Y. Qu, Journal of south china university of technology, 1999, 27(12), 7.
26. R. A. Rapp and X. J. Zheng, Metall. Trans. A, 1991, 22(12), 3071.
27. I. Ahmad, A. Kennedy and Y. Q. Zhu, Wear, 2010, 269, 71.

---