- Email: [email protected]
Influence of Ti3SiC2 content on tribological properties of NiAl matrix self-lubricating composites
Materials and Design 45 (2013) 179–189
Contents lists available at SciVerse ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Influence of Ti3SiC2 content on tribological properties of NiAl matrix self-lubricating composites Xiaoliang Shi a,b,⇑, Mang Wang a, Wenzheng Zhai b, Zengshi Xu a, Qiaoxin Zhang a,b, Ying Chen c a
School of Mechanical and Electronic Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, China School of Materials Science and Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, China c School of Mechanical and Electronic Engineering, Beijing Institute of Technology, 5 South Zhongguancun Street, Beijing 100081, China b
a r t i c l e
i n f o
Article history: Received 2 July 2012 Accepted 22 August 2012 Available online 23 September 2012 Keywords: NiAl matrix Ti3SiC2 Spark plasma sintering In situ Tribology
a b s t r a c t NiAl matrix self-lubricating composites (NMCs) with various contents of Ti3SiC2 were fabricated by in situ technique using spark plasma sintering. The effects of Ti3SiC2 content on tribological properties of NMC were investigated. The results showed that NMC were composed of the matrix phase of NiAl alloy, enhanced phase of TiC and lubricating phases of Ti3SiC2 and C. NMC with 10 wt.% Ti3SiC2 exhibited low friction coefficient of 0.60 and wear rate of 5.45 10 5 mm3 (N m) 1 at the condition of 10 N– 0.234 m/s at room temperature. The optimum content of Ti3SiC2 was 10 wt.%. The excellent tribological performance of NMC could be attributed to the balance between strength and lubricity, as well as synergetic effect of enhanced phase and lubricating phases. The wear mechanisms changed with the increasing of the doped content of Ti3SiC2. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Thanks to high melting point (1638 °C), high thermal conductivity (70 W m 1 K 1), low density (5.95 g/cm3) and excellent oxidation resistance [1–4], NiAl intermetallic compound is regarded as potential high temperature structural material having promising applications including turbochargers, high temperature dies and molds, furnace fixtures, rollers in steel slab heating furnaces, hydroturbines, cutting tools, pistons, valves and various components within gas turbines [3,5,6]. However, like most intermetallic compounds, NiAl is rather brittle for an engineering material at low temperatures and exhibits a brittle-to-ductile transition temperature (BDTT) at 500–700 °C, depending upon composition, grain size, processing conditions and strain rate. Surprisingly, the creep resistance of NiAl is relatively poor above the BDTT despite the fact that NiAl has an ordered crystal structure with limited slip systems. Composite strengthening is one approach that can improve both toughness and creep resistance. Many studies have been carried out to synthesize reinforcing phase in NiAl matrix. The addition of TiB2 as a particulate reinforcement to NiAl increases the hardness of the composite with respect to NiAl, and reduces the wear rates (WRs) at all volume fractions on garnet and Al2O3 abra-
⇑ Corresponding author at: School of Mechanical and Electronic Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, China. Tel./fax: +86 27 87651793. E-mail address: [email protected] (X.L. Shi). 0261-3069/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.matdes.2012.08.060
sives. Abrasion on SiC resulted in a minimum of the wear rate for the composite with 20 vol.% TiB2 for most conditions [7]. Wear resistance of NiAl is the challenging factor which has to be considered since some of the proposed applications, e.g. turbine blade tips, are subject to sliding type wear [6]. Wear testing of three NiAl alloys containing 45 at.%, 48 at.% and 50 at.% aluminum indicated that all of them had low coefficient of friction (COF) and WRs. Wear rate is hardness-dependent for these NiAl alloys, and higher hardness leads to lower wear. Meanwhile, wear process of NiAl is dominated by plastic deformation for all compositions [8]. From the tribological point of view, TiC reinforced NiAl intermetallic matrix composites are expected to process excellently both abrasive and adhesive wear resistance because of high hardness of the TiC reinforcements as well as the strong atomic bonding characteristic and abnormal hardness-temperature relationship of nickel aluminides. As a result, TiC reinforced nickel aluminide matrix composite coatings are potential candidates of wear-resistant coating materials [9]. NiAl intermetallic matrix composites containing solid lubricants have attracted much attention as high-temperature selflubricating materials. NiAl–31BaF2–19CaF2 (mass%) prepared by spark plasma sintering (SPS) exhibited the low friction coefficients and wear rates at 873–1073 K [10]. Zhu et al. [11] investigated the high temperature sliding tribological behavior of NiAl matrix composites with addition of oxides (ZnO/CuO). At 1000 °C, the wear rate of NiAl matrix composites with ZnO was about 7 10 6 mm3 (N m) 1, and the superior wear resistance was due to the formation of the ZnO layer on the worn surface. Moreover, the NiAl
180
X. Shi et al. / Materials and Design 45 (2013) 179–189
matrix composites with CuO showed self-lubricating performance at 800 °C, which was attributed to the presence of the glaze layer containing CuO and MoO3. As a new solid self-lubricating material, Ti3SiC2 has attracted much attention recently. Like graphite, Ti3SiC2 has a hexagonal structure with a space group of P63/mmc. Besides, Ti3SiC2 is readily machined as graphite and possesses excellent thermostability. Thus, it was proposed that Ti3SiC2 could be a new solid self-lubricating material with low COF and WRs from low to high temperature ranges [12–15]. However, meager information is available as regards the development of NiAl matrix self-lubricating composites by use of the Ti3SiC2 and C dual lubrication phases and TiC enhanced phase. Recently, in situ technique has been developed to fabricate metal matrix composites. The composites exhibit a clean matrix/ reinforcement interface, which leads to better improvement in mechanical properties and tribological properties of the in situ composites [16,17]. Compared with other traditional sintering methods, the main advantages of SPS are the fast heating and cooling speeds, short sintering time, fine grain size of the prepared materials, controlled organizational structure, energy saving, environmental protection and so on [18–20]. In this study, NiAl matrix self-lubricating composites (NMCs) with different contents of Ti3SiC2 were prepared by in situ technique using SPS. The effects of Ti3SiC2 content on the dry sliding friction and wear behavior of NiAl matrix self-lubricating composites at the load of 10 N and sliding speed of 0.234 m/s for 20 min at room temperature were investigated through the determination of COF and WRs and the analysis of the compositions of wear debris, worn surfaces of both NMC and Si3N4 ceramic ball friction pair. 2. Experimental details
lue was given. The density of as-prepared specimens was determined by Archimedes’ principle according to the ASTM Standard B962-08 [24,25]. 2.3. Tribological test The tribological test was conducted on a HT-1000 ball-on-disk high temperature tribometer according to the ASTM Standard G99-95 [26,27]. The disks, which were the as-prepared materials, were cleaned with acetone and then dried in hot air before test. The counterpart ball was the commercial Si3N4 ceramic ball with a diameter of 5 mm (about HV 15 GPa). The test temperature was room temperature. The sliding speed and applied load were 0.234 m/s and 10 N respectively. The friction radius was 2 mm. The testing time was 20 min. The COF was automatically measured and recorded in real time by the computer system of the friction tester. The wear quantity of NMC was measured by the weighting method, which measured the mass loss of the samples for every friction process. The tests for every given conditions were repeated three times to obtain reliable data. The average value was used as the evaluating data. 2.4. Analysis The surfaces of the as-prepared specimens were examined by XRD with Cu Ka radiation at 30 kV and 40 mA at a scanning speed of 0.01 s 1 for the identification of the phase constitution. The morphologies and compositions of worn surfaces of NMC and Si3N4 ceramic ball friction pair, wear debris were analyzed by a SIRION 200 field emission scanning electron microscope (FESEM) equipped with energy dispersive spectroscopy (EDS) and a JAX8230 electron probe microanalyzer equipped with EDS.
2.1. Materials 3. Results and discussion NMC with different contents of Ti3SiC2 were prepared by in situ technique using SPS. The composite powders of NiAl matrix were composed of commercially available Ni, Al, Mo, Nb, Fe and B powders by atomic ratio of 48:50:1:1:0.5:0.02, whose average particle sizes were about 30–50 lm. The weight fractions of Ti3SiC2 in the composites were fixed at 0 wt.%, 5 wt.%, 10 wt.%, 15 wt.% and 20 wt.%, respectively. The fabrication process of Ti3SiC2 powder (5 lm in average size, 99.5% in purity) was described in detail elsewhere [21]. NiAl alloy without Ti3SiC2 was denoted as NA. While NB, NC, ND, NE represented NMC with 5 wt.%, 10 wt.%, 15 wt.% and 20 wt.% Ti3SiC2, respectively. The starting powders were mixed by high energy ball-milling in vacuum. Balls and vials were made of hard alloy. The charge ratio (ball to powder mass ratio) employed was 10:1. The milling time and speed are 10 h and 200 rpm respectively. After being mixed and dried, the mixtures were then sintered by SPS using a D.R. SinterÒ SPS3.20 (Sumitomo Coal & Mining, now SPS Syntex Inc.) apparatus at 1100 °C under a pressure of 30 MPa for 5 min in pure Ar atmosphere protection. The heating rate was 100 °C/min. The cylindrical graphite molds with an inner diameter of 20 mm were used. The as-prepared specimen surfaces were ground to remove the layer on the surface and polished mechanically with emery papers down to 1200 grad, and then with 0.05 lm wet polishing diamond pastes.
3.1. Compositions of NMC Fig. 1 was the XRD patterns of NMC with different contents of Ti3SiC2 prepared by in situ technique using SPS. It could be found that the diffraction peaks primarily belonged to the NiAl, Ni3Al and Ti3SiC2 phases. Additionally, it was clear that there were TiC and C phases existing in the NiAl matrix composites, which were attributed to the decomposition reaction of Ti3SiC2 and carburization role of graphite atmosphere caused by the graphite mold.
2.2. Vicker’s microhardness and density The Vicker’s microhardness of each as-received specimen was measured, according to the ASTM standard E92-82 [22,23], using an HVS-1000 Vicker’s hardness instrument with a load of 1 kg and a dwell time of 8 s. Five tests were conducted and the mean va-
Fig. 1. XRD patterns of NMC doped with varying weight fractions of Ti3SiC2.
X. Shi et al. / Materials and Design 45 (2013) 179–189
181
As shown in Fig. 1, there were very obvious diffraction peaks of Ti3SiC2 phase at 2h = 34.096° (1 0 1), 39.548° (1 0 4) and 60.283° (1 1 0) existing in the samples of NB, NC, ND and NE. Meanwhile, the peaks belonging to C phase at 2h = 26.381° (0 0 2) could also be found. The peak intensities of Ni3Al phase of NB at both 2h = 43.253° and 53.373° was the much stronger than that of NC, ND and NE. Moreover, as shown in Fig. 1, the intensities of TiC phase at 2h = 35.858° (1 1 1), 41.643° (2 0 0) and 60.357° (2 2 0) in NB when compared with NC, ND and NE were much weaker. 3.2. Vicker’s microhardness and density As shown in Table 1, NMC doped with Ti3SiC2 when compared with NA exhibited the higher Vicker’s microhardness. And the Vicker’s microhardness of NMC increased with the increasing of the doped content of Ti3SiC2. The addition of Ti3SiC2 could improve the strength of NiAl. Moreover, Ti3SiC2 addition further enhanced the hardness, which was attributed to the formation of TiC during the fabricated process. The density of NMC deceased with the increasing of the doped contents of Ti3SiC2. It was attributed to the lower density of Ti3SiC2 (4.52 g/cm3) when compared with NiAl [21]. NB had a higher density due to that the Ni3Al phase were formed. The measured density and Vicker’s microhardness of NC were 5.873 g/cm3 and 593.5 HV1 respectively.
Fig. 2. The relationship of friction coefficient of NMC versus sliding time at the condition of 10 N–0.234 m/s.
3.3. Friction and wear behaviors Fig. 2 shows the typical measuring curves of the dynamic COF of NMC against Si3N4 ceramic ball friction pair at the constant sliding speed of 0.234 m/s and load of 10 N at room temperature. It was obvious that all curves were relatively smooth except that of NE. And the friction coefficients of NMC against Si3N4 ceramic ball pair were in the range of 0.5–0.8. NC had the lowest COF among the samples. Fig. 3 shows the variation of wear rate and friction coefficient of NMC with content of Ti3SiC2 at the condition of 10 N–0.234 m/s. As shown in Fig. 3, NC against Si3N4 ceramic ball pair had the best tribological properties among the samples. The COF and wear rate of NC were about 0.60 and 5.45 10 5 mm3 (N m) 1 respectively. And both of them were the lowest among that of all the samples. However, NE against Si3N4 pair had the worst tribological performance among the samples. The COF and wear rate of NE were about 0.75 and 3.67 10 3 mm3 (N m) 1 respectively. And both of them were highest among that of all the samples. According to the curve of WRs, it was concluded that the WRs increased with the increasing of the doped content of Ti3SiC2 except 10 wt.% Ti3SiC2. 3.4. Microstructure analysis Fig. 4 exhibited the FESEM morphologies and EDS patterns of fractured surfaces of NA and NC. Microstructure of NA was
Table 1 Densities and Vicker’s microhardness of NMC prepared by SPS. Sample
Composition
Measured density (g/ cm3)
Vicker’s microhardness (HV1)
NA NB
NiAl NiAl–5 wt.% Ti3SiC2 NiAl–10 wt.% Ti3SiC2 NiAl–15 wt.% Ti3SiC2 NiAl–20 wt.% Ti3SiC2
5.896 6.249
553.2 580.9
5.873
593.5
5.557
619.8
5.452
678.0
NC ND NE
Fig. 3. The variation of COF and WRs of NMC with varying weight fractions of Ti3SiC2 at the condition of 10 N–0.234 m/s.
presented in Fig. 4a. As shown in Fig. 4a, NA had a dense and homogeneous microstructure. Thus, the NiAl intermetallic should have excellent mechanical properties. Fig. 4c was a result of EDS analysis for Fig. 4a, and the EDS pattern exhibited the presence of Ni, Al, Nb, Mo and a small amount of Fe. As shown in Fig. 4b, it seemed that NC had denser microstructure than NA. Fig. 4d was a result of EDS analysis for Fig. 4b, and the EDS pattern exhibited the presence of Al, Si, Nb, Mo, Ti, Cr and Ni. Meanwhile, the contents of Ti and Si elements were 8.65 at.% and 4.84 at.% respectively. And there was a clean matrix/Ti3SiC2 interface existing. A clean matrix/reinforcement interface would lead to better and improved mechanical and tribological properties of the in situ composites [16,28]. Ramesh and Ahamed [16] reported that there was homogeneity in the distribution of in situ TiB2 particles within the matrix alloy with a clear matrix-reinforcement interface. The in situ TiB2 particles significantly improved the tribological properties of the 6063/ TiB2 composites. 3.5. Morphologies, compositions of wear debris and the worn surfaces of NA, NC, NE and Si3N4 ceramic ball Fig. 5a–c exhibited the typical electron probe morphologies of worn surfaces of NA at the condition of 10 N–0.234 m/s for
182
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 4. FESEM morphologies and EDS patterns of fractured surfaces of NA and NC. (a and b) Morphologies of fractured surfaces of NA and NC; (c and d) EDS patterns exhibiting the chemical compositions of NA and NC; (c) versus (a) and (d) versus (b).
20 min at room temperature. As shown in Fig. 5a–c, it was apparent that there was the evidence of a significant amount of plastic flow on the worn surfaces. Johnson et al. [8] studied the dry sliding wear behavior of the B2-structured (ordered body-centered cubic) compound NiAl. The results showed that surface cracking was all but nonexistent, meaning that the brittle nature of the material did not seem to be affecting wear. Plasticity appeared in two forms: extrusion of material across and over the trailing edge of the wear scar, and deep grooving or microploughing of material across the wear scar from leading to trailing edge. The NiAl worn surfaces showed signs of chunking, or separation of (relatively) large pieces of material followed by the formation of a groove. Fig. 5d was a result of EDS analysis for the A area as shown in Fig. 5c. The EDS pattern of the surface of NA exhibited the presence of Al, Ni, Nb, Mo, C and a small amount of O. The content of O was 1.23 wt.%. Fig. 5e was a result of EDS analysis performed on the worn track for the B area as shown in Fig. 5c, which revealed that there were O, Ni and Al elements existing. And the content of O in B area when compared with A area increased obviously, which was 4.52 wt.%. The content of C was 1.64 wt.%, which could be produced by the carburization role of graphite atmosphere from the graphite mold. Shaji and Radhakrishnan [29] investigated the possibility of using graphite as a lubricating medium to reduce the heat generated at the grinding zone in surface grinding with a newly developed experimental setup. Cho et al. [30] researched the effect of three different solid lubricants (graphite, Sb2S3, and
MoS2) in the brake friction material on various aspects of friction characteristics was investigated. The results showed that graphite was an excellent lubricant at lower temperature. The existing graphite lubricant could be beneficial for the tribological performance of NA at the room temperature. It is conceivable that the frictional oxide film was formed on the worn surfaces for the higher oxygen content of 10.75 at.%. It was no doubt that the Si was transferred from the counterpart of Si3N4 ceramic ball. It was attributed to severe abrasive action of tribological pairs during sliding process. In this area, obvious crack was observed. Fig. 5f was a result of EDS analysis for the C area as shown in Fig. 5c. There were little particles detached from the surface of this area. The abundant amount of O element was 4.84 wt.%, which confirmed the forming of the oxide film. However, the thickness of oxide film was thin. Stott [31] reviewed some of the models developed to account for the generation of oxide during sliding and the effects of such oxides on the rates of wear. He claimed that the wear-protective layers consisted mainly of loosely-compacted particles at low temperatures. The oxidation layer occurred on worn surface decreased the wear rate. As shown in Fig. 3, the wear rate of NA was 9.02 10 5 mm3 (N m) 1. Some particles were detached from the worn surface, which remained between the surfaces of NiAl and Si3N4 ball. Because of the high surface temperature originated from the friction force, they these particles were oxidized. Meanwhile, it could be concluded that the wear process of NiAl was dominated by plastic deformation.
X. Shi et al. / Materials and Design 45 (2013) 179–189
183
Fig. 5. Electron probe morphologies and EDS patterns of worn surface of NA at the condition of 10 N–0.234 m/s. (a–c) Morphologies of worn surface of NA; (d–f) EDS patterns exhibiting the chemical compositions of NA; (d) versus A area in (c), (e) versus B area in (c) and (f) versus C area in (c).
Fig. 6a–c exhibited the typical electron probe morphologies of worn surfaces of NC at the condition of 10 N–0.234 m/s. As shown in Fig. 6a, the tribo-layers called ‘glazes’ [32] had been formed on the worn surfaces of NC. The glazes were not complete, and the delamination of the glazes happened. Stott et al. [33,31,34] have investigated the interaction between sliding wear and oxidation in superalloys. The results showed that the glazes could temporar-
ily protect the surfaces from further contact damage. If they happened to wear off, new glazes could be formed to take their place. This progression of formation, loss, and reformation could result in short-term friction or wear transients. As shown in Fig. 6b, the grooves on the worn surface of NC became coarser when compared with the relatively finer grooves of NA, and the corresponding wear mechanism transferred from a mixed
184
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 6. Electron probe morphologies and EDS patterns of worn surface of NC at the condition of 10 N–0.234 m/s. (a–c) Morphologies of worn surface of NC; (d–f) EDS patterns exhibiting the chemical compositions of NC; (d) versus A area in (c), (e) versus B area in (c) and (f) versus C area in (c).
abrasion-plastic deformation to microploughing. Moreover, the delaminated glazes were observed. Apparently, the worn surfaces were characterized by long discontinuous grooves and delaminated glazes. During the sliding friction and wear process, the Si3N4 ceramic ball ploughed across the surface of NC, and eventually removed or pushed material into ridges along sides of the grooves. Meanwhile, the tribo-layers were formed during the contact
process, which were mixture of lubricants, NiAl matrix and oxides. And delamination was attributed to the tangential shears during sliding, which caused the phenomenon of discontinuous grooves. Fig. 6d was a result of EDS analysis for the A area as shown in Fig. 6c. The EDS pattern exhibited the presence of Al, Ni, Ti, Mo, C, Si and O. The content of O was 3.10 wt.%, which was lower than 4.52 wt.% of worn surface of NA. It showed that the oxidation
X. Shi et al. / Materials and Design 45 (2013) 179–189
185
Fig. 7. Electron probe morphologies and EDS patterns of worn surfaces of NE at the condition of 10 N–0.234 m/s. (a–c) Morphologies of worn surface of NE; (d–h) EDS patterns exhibiting the chemical compositions of NE; (d) versus D area in (a), (e) versus E area in (a), (f) versus F area in (a), (g) versus G area in (b) and (h) versus H area in (b).
186
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 8. FESEM morphology and EDS patterns of worn surface of Si3N4 friction pair against NC at the condition of 10 N–0.234 m/s. (a) Morphology of worn surface of Si3N4 friction pair, (b and c) EDS patterns exhibiting the chemical compositions of Si3N4 friction pair, (b) versus B point in (a) and (c) versus C point in (a).
phenomenon of worn surface of NC when compared with NA was mild at the condition of 10 N–0.234 m/s. The friction force between NC and Si3N4 ceramic ball was smaller for the Ti3SiC2 lubricant existing, thus the friction heat was also lower. And the oxidation phenomenon was mild. Fig. 6e was the EDS analysis for the B area as shown in Fig. 6c. As shown in Fig. 6e, the EDS pattern exhibited the presence of Al, Ni, Ti, C, Si and O. The content of O was 3.41 wt.%, which was higher than 3.10 wt.% of A area on worn surface. It showed the oxidation phenomenon of B area when compared with A area on the worn surface was more serious. As the lubricant layer gradually disappeared, microploughing became more serious. Some particles were detached from the surface, and oxidation phenomenon increased. However, the glaze flaking of NC when compared with NA was not more serious. Thus, the wear rate of NC was lower than that of NA at the condition of 10 N–0.234 m/s. Fig. 6f was a result of EDS analysis for the C area (see Fig. 6c), which was formed by the ground and polished process before the tribological test. The EDS pattern exhibited the presence of Al, Ni, Ti and Si. A small amount of O (0.89 wt.%) also existed in this area, which was much lower than that of worn surface obtained by the dry sliding friction and wear process of NC against Si3N4 ceramic ball. Fig. 7a–c exhibited the typical electron probe morphologies of worn surfaces of NE at the condition of 10 N–0.234 m/s. It could be observed that the worn surface turned much coarser when the doped content of Ti3SiC2 was 20 wt.%, and there were a lot of pits
and mounds exiting on worn surfaces of NE. It indicated that the main wear mechanisms obviously were ploughing and abrasive wear. Moreover, as shown in area G of Fig. 7b, a few discontinuous glazes were also formed. And the areas of the glazes when compared with the whole worn surfaces were smaller. The worn surface was very rough, and the wear resistance should be higher. Thus, the COF of NE was 0.75, which was the highest in NMC. Fig. 7d–h were the EDS analysis for the D, E and F areas as shown in Fig. 7a, as well as G and H areas as shown in Fig. 7b, respectively. As shown in Fig. 7d–h, the Ti content was approximately three times of Si content. Moreover, the contents of O in the areas of E, G and H were much higher than that in the areas of D and F. It showed that the oxidation phenomena in the E, G and H areas of the worn surfaces were more serious. However, the content of Ti in NE was nearly six to nine times as much as that of NC, which was much higher than the content of Ni and Al. Meanwhile, the content of C was also higher. Thus, we could get a conclusion that a lot of TiC and Ti3SiC2 existed on the worn surfaces of NE. TiC enhanced the hardness of NE (678 HV1). When the doped content of Ti3SiC2 was 20 wt.%, the homogeneous distribution of Ti3SiC2 lubricant in NMC could not be realized. The lubricant enrichment happened in the local parts of NMC. And mechanical and tribological properties of NMC could be deteriorated by the inhomogeneity of lubricant. Thus, the NE had the highest wear rate of 3.67 10 3 mm3 (N m) 1. Fig. 8 shows the FESEM morphologies and EDS of worn surfaces of Si3N4 friction pair against NC at the condition of 10 N–0.234 m/s.
X. Shi et al. / Materials and Design 45 (2013) 179–189
187
Fig. 9. FESEM morphology and EDS pattern of wear debris of NA at the condition of 10 N–0.234 m/s with Si3N4 friction pair. (a) Morphology of wear debris of NA and (b) EDS pattern exhibiting the chemical compositions of wear debris of NA.
Fig. 10. FESEM morphology and EDS pattern of wear debris of NC at the condition of 10 N–0.234 m/s with Si3N4 friction pair. (a) Morphology of wear debris of NC and (b) EDS pattern exhibiting the chemical compositions of wear debris of NC.
As shown in Fig. 8a, there were a lot of mounds and some adhesive wear debris existing on the worn surface. Fig. 8b was the EDS analysis for the B point as shown in Fig. 8a. The Si content was 8.87 wt.%, and no N was detected on this point. The contents of O, Al and Ni were 8.63 wt.%, 1.87 wt.% and 6.09 wt.%, respectively, which confirmed the existing of oxides of aluminum and nickel. It also showed that matter translocation between NC and Si3N4 ball had happened. Moreover, the elements of K and Cl appeared in Fig. 8b. It could be caused by the sweat of the operator during the specimen preparation. Fig. 8c was the EDS analysis for the C point as shown in Fig. 8a. The Si content was 66.49 wt.%, and the N content was 20.26 wt.%. The contents O and Al were 9.77 wt.% and 3.50 wt.% respectively, which confirmed the existing of aluminum oxide. It was attributed to severe abrasive action and material transfer of tribological pairs during sliding friction and wear process. Fig. 9 shows the FESEM morphologies and EDS of wear debris of NA at the condition of 10 N–0.234 m/s. As shown in Fig. 9a, the sizes of wear debris with irregular shape were from 100 nm to 10 lm. It showed that most of wear debris were directly generated from NA rather than the glazes. The wear resistance was higher, thus the COF was also higher. Fig. 9b was a result of EDS analysis
for the wear debris. The EDS pattern exhibited the presence of Al, Ni, Ti, C, Si, Mo, Fe and O. The contents of O, Al and Ni were 25.43 wt.%, 17.34 wt.% and 37.99 wt.%, respectively, which confirmed that the wear debris were composed of oxides of aluminum and nickel, as well as wear debris particles of NA. The Si content was 0.55 wt.%. Fig. 10 shows the FESEM morphologies and EDS of wear debris of NC obtained at the condition of 10 N–0.234 m/s. As shown in Fig. 10a, it indicated that the wear debris was mainly a product of the glazes rather than having been directly generated from NC. The wear debris were composed of lots of finer particles, whose average size was less than 1 lm. And the particle size distribution was uniform. Fig. 10b was a result of EDS analysis for the wear debris. The EDS pattern exhibited the presence of Al, Ni, Ti, C, Si, Mo, Fe and O. The contents of O, Al, Si, Ti and Ni were 5.28 wt.%, 17.76 wt.%, 3.28 wt.%, 11.23 wt.% and 48.21 wt.%, respectively. The content of Ti was 11.23 wt.%, and the content of Si was 3.28 wt.%. The debris should be composed of oxides and wear debris particles of NC. Sun et al. [35] investigated the dry sliding friction and wear behavior of Ti2AlN/TiAl composite at room temperature. They reported that the wear debris formed at earlier friction stage move between the two surfaces, and the major wear
188
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 11. FESEM morphology and EDS pattern of wear debris of NE at the condition of 10 N–0.234 m/s with Si3N4 friction pair. (a) Morphology of wear debris of NE and (b) EDS pattern exhibiting the chemical compositions of wear debris of NE.
mechanism changes from two-body abrasion to three-body abrasion. Fine equiaxed wear debris were compressed and broken into finer particles by the normal stress. Meanwhile, the high surface temperature that originated from the friction heat during the dry sliding friction process made some fine particles be oxidized in the air. Oxides of fine equiaxed wear debris and unoxidized fine equiaxed wear debris were trapped on the worn surface of NC to form the glazes, which acted as an antifriction. The glazes actually had not only a significant antifriction effect, but also a protective action for the friction surfaces. Thus, the COF of NC was about 0.60, and the wear rate was 5.45 10 5 mm3 (N m) 1. Fig. 11 shows the FESEM morphologies and EDS of wear debris of NE obtained at the condition of 10 N–0.234 m/s. As shown in Fig. 11a, the sizes of wear debris with irregular shape were from 1 lm to 10 lm. The wear resistance of forming large size of wear debris was much higher, thus the COF of NE was about 0.75. Meanwhile, the larger size wear debris meant higher wear rate, which was 3.67 10 3 mm3 (N m) 1. Fig. 11b was a result of EDS analysis for the wear debris. The EDS pattern exhibited the presence of Al, Ni, Ti, Si, Nb, Fe, O and a small amount of C. The contents of O, Al, Si, Ti and Ni were 21.40 wt.%, 15.94 wt.%, 4.85 wt.%, 22.90 wt.% and 31.25 wt.%, respectively. The content of O was much higher. And the Ti content was almost three times of Si content. The wear debris should be composed of oxides and wear debris particles of NE. 4. Conclusions (1) NC fabricated by the in situ technique using SPS at 1100 °C for 5 min under 30 MPa had the excellent tribological properties. At the condition of 10 N–0.234 m/s at room temperature, the COF of NC was about 0.60, and the wear rate was 5.45 10 5 mm3 (N m) 1. Moreover, the tribological noise was quite low during the dry sliding friction and wear test process. (2) The wear mechanisms changed with the increasing of the doped content of Ti3SiC2. The wear mechanism of NA was a mixed abrasion-plastic deformation. Microploughing and delamination were dominant wear mechanisms in case of NC. While the wear mechanisms of NE were ploughing and abrasive wear. (3) The glazes of NC actually had not only a significant antifriction effect, but also a protective action for the friction and wear surfaces. Oxides of fine equiaxed wear debris and
unoxidized fine equiaxed wear debris were trapped on the worn surface of NC to form the glazes, which acted as an antifriction. The glazes were composed of lots of finer particles, whose average size was less than 1 lm. And the particle size distribution was uniform.
Acknowledgments This work was supported by the Fundamental Research Funds for the Central Universities (2010-II-020); the National Natural Science Foundation of China (51275370); the Project for Science and Technology Plan of Wuhan City; the Program for New Century Excellent Talents in University of Ministry of Education of China; the Nature Science Foundation of Hubei Province; and the Academic Leader Program of Wuhan City (201150530146).
References [1] Miracle DB. The physical and mechanical properties of NiAl. Acta Metall Mater 1993;41:649–84. [2] Zhou LZ, Guo JT, Li GS, Xiong LY, Wang SH, Li CG. Investigation of annealing behavior of nanocrystalline NiAl. Mater Des 1997;18:373–7. [3] Liu CT. Recent advances in ordered intermetallics. Mater Chem Phys 1996;42:77–86. [4] Stoloff NS, Liu CT, Deevi SC. Emerging applications of intermetallics. Intermetallics 2000;8:1313–20. [5] Lasalmonie A. Intermetallics: why is it so difficult to introduce them in gas turbine engines. Intermetallics 2006;14:1123–9. [6] Ozdemir O, Zeytin S, Bindal C. Tribological properties of NiAl produced by pressure-assisted combustion synthesis. Wear 2008;265:979–85. [7] Hawk JA, Alman DE. Abrasive wear behavior of NiAl and NiAl–TiB2 composites. Wear 1999;225–229:544–56. [8] Johnson BJ, Kennedy FE, Baker I. Dry sliding wear of NiAl. Wear 1996;192:241–7. [9] Chen Y, Wang HM. Microstructure of laser clad TiC/NiAl–Ni3(Al, Ti, C) wearresistant intermetallic matrix composite coatings. Mater Lett 2003;57:2029–36. [10] Murakami T, Ouyang JH, Sasaki S, Umeda K, Yoneyama Y. High-temperature tribological properties of Al2O3, Ni-20 mass% Cr and NiAl spark-plasmasintered composites containing BaF2–CaF2 phase. Wear 2005;259:626–33. [11] Zhu SY, Bi QL, Niu MY, Yang J, Liu WM. Tribological behavior of NiAl matrix composites with addition of oxides at high temperatures. Wear 2012;274– 275:423–34. [12] El-Raghy T, Blau P, Barsoum MW. Effect of grain size on friction and wear behavior of Ti3SiC2. Wear 2000;238:125–30. [13] Sun ZM, Zhou YC, Li S. Tribological behavior of Ti3SiC2-based material. J Mater Sci Technol 2000;18:142–5. [14] Zhang Y, Ding GP, Zhou YC, Cai BC. Ti3SiC2-a self-lubricating ceramic. Mater Lett 2002;55:285–9.
X. Shi et al. / Materials and Design 45 (2013) 179–189 [15] Gupta S, Filimonov D, Zaitsev V, Palanisamy T, Barsoum MW. Ambient and 550 °C tribological behavior of select MAX phases against Ni-based superalloys. Wear 2008;264:270–8. [16] Ramesh CS, Ahamed A. Friction and wear behaviour of cast Al 6063 based in situ metal matrix composites. Wear 2011;271:1928–39. [17] Tjong SC, Ma ZY. Microstructural and mechanical characteristics of in situ metal matrix composites. Mater Sci Eng R 2000;29:49–113. [18] Murakami T, Sasaki S, Ito K. Oxidation behavior and thermal stability of Crdoped Nb(Si, Al)2 and Nb3Si5Al2 matrix compacts prepared by spark plasma sintering. Intermetallics 2003;11:269–78. [19] Murakami T, Sasaki S, Ichikawa K, Kitahara A. Microstructure, mechanical properties and oxidation behavior of Nb–Si–Al and Nb–Si–N powder compacts prepared by spark plasma sintering. Intermetallics 2001;9:621–7. [20] Kim H, Kawahara M, Tokita M. Specimen temperature and sinterability of Ni powder by spark plasma sintering. J Jpn Soc Powder Metall 2000;47:887–91. [21] Peng MC, Shi XL, Zhu ZW, Wang M, Zhang QX. Facile synthesis of Ti3SiC2 powder by high energy ball-milling and vacuum pressureless heat-treating process from Ti–TiC–SiC–Al powder mixtures. Ceram Int 2012;38:2027–33. [22] Loganathan TM, Purbolaksono J, Inayat-Hussain JI, Wahab N. Effects of carburization on expected fatigue life of alloys steel shafts. Mater Des 2011;32:3544–7. [23] American society for testing and materials. Standard test method for vickers hardness of metallic materials. ASTM E92-82 e2, 2003. [24] Pellizzari M, Fedrizzi A, Zadra M. Influence of processing parameters and particle size on the properties of hot work and high speed tool steels by Spark Plasma Sintering. Mater Des 2011;32:1796–805.
189
[25] American society for testing and materials. Standard test methods for density of compacted or sintered powder metallurgy (PM) products using Archimedes’ principle, ASTM B962-08, 2008. [26] Koksal S, Ficici F, Kayikci R, Savas O. Experimental optimization of dry sliding wear behavior of in situ AlB2/Al composite based on Taguchi’s method. Mater Des 2012;42:124–30. [27] American society for testing and materials. Standard test method for wear testing with a pin-on-disk apparatus, ASTM G99-95, 1995. [28] Kumar S, Chakraborty M, Subramanya Sarma V, Murthy BS. Tensile and wear behaviour of in situ Al-7Si/TiB2 particulate composites. Wear 2008;26:134–42. [29] Shaji S, Radhakrishnan V. Analysis of process parameters in surface grinding with graphite as lubricant based on the Taguchi method. J Mater Process Technol 2003;141:51–9. [30] Cho MH, Ju J, Kim SJ, Jang H. Tribological properties of solid lubricants (graphite, Sb2S3, MoS2) for automotive brake friction materials. Wear 2006;260:855–60. [31] Stott FH. The role of oxidation in the wear of alloys. Tribol Int 1998;31:61–71. [32] Blau PJ. Elevated-temperature tribology of metallic materials. Tribol Int 2010;43:1203–8. [33] Stott FH, Lin DC, Wood GC. Structure and mechanism of formation of the ‘Glaze’ oxide layers produced on Ni-based alloys during wear at high temperatures. Corros Sci 1973;13:449–69. [34] Stott FH, Lin DC, Wood GC. ‘Glazes’ produced on nickel-base alloys during high temperature wear. Nature 1973;242:75–7. [35] Sun T, Wang Q, Sun DL, Wu GH, Na Y. Study on dry sliding friction and wear properties of Ti2AlN/TiAl composite. Wear 2010;268:693–9.
Contents lists available at SciVerse ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/matdes
Influence of Ti3SiC2 content on tribological properties of NiAl matrix self-lubricating composites Xiaoliang Shi a,b,⇑, Mang Wang a, Wenzheng Zhai b, Zengshi Xu a, Qiaoxin Zhang a,b, Ying Chen c a
School of Mechanical and Electronic Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, China School of Materials Science and Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, China c School of Mechanical and Electronic Engineering, Beijing Institute of Technology, 5 South Zhongguancun Street, Beijing 100081, China b
a r t i c l e
i n f o
Article history: Received 2 July 2012 Accepted 22 August 2012 Available online 23 September 2012 Keywords: NiAl matrix Ti3SiC2 Spark plasma sintering In situ Tribology
a b s t r a c t NiAl matrix self-lubricating composites (NMCs) with various contents of Ti3SiC2 were fabricated by in situ technique using spark plasma sintering. The effects of Ti3SiC2 content on tribological properties of NMC were investigated. The results showed that NMC were composed of the matrix phase of NiAl alloy, enhanced phase of TiC and lubricating phases of Ti3SiC2 and C. NMC with 10 wt.% Ti3SiC2 exhibited low friction coefficient of 0.60 and wear rate of 5.45 10 5 mm3 (N m) 1 at the condition of 10 N– 0.234 m/s at room temperature. The optimum content of Ti3SiC2 was 10 wt.%. The excellent tribological performance of NMC could be attributed to the balance between strength and lubricity, as well as synergetic effect of enhanced phase and lubricating phases. The wear mechanisms changed with the increasing of the doped content of Ti3SiC2. Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction Thanks to high melting point (1638 °C), high thermal conductivity (70 W m 1 K 1), low density (5.95 g/cm3) and excellent oxidation resistance [1–4], NiAl intermetallic compound is regarded as potential high temperature structural material having promising applications including turbochargers, high temperature dies and molds, furnace fixtures, rollers in steel slab heating furnaces, hydroturbines, cutting tools, pistons, valves and various components within gas turbines [3,5,6]. However, like most intermetallic compounds, NiAl is rather brittle for an engineering material at low temperatures and exhibits a brittle-to-ductile transition temperature (BDTT) at 500–700 °C, depending upon composition, grain size, processing conditions and strain rate. Surprisingly, the creep resistance of NiAl is relatively poor above the BDTT despite the fact that NiAl has an ordered crystal structure with limited slip systems. Composite strengthening is one approach that can improve both toughness and creep resistance. Many studies have been carried out to synthesize reinforcing phase in NiAl matrix. The addition of TiB2 as a particulate reinforcement to NiAl increases the hardness of the composite with respect to NiAl, and reduces the wear rates (WRs) at all volume fractions on garnet and Al2O3 abra-
⇑ Corresponding author at: School of Mechanical and Electronic Engineering, Wuhan University of Technology, 122 Luoshi Road, Wuhan 430070, China. Tel./fax: +86 27 87651793. E-mail address: [email protected] (X.L. Shi). 0261-3069/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.matdes.2012.08.060
sives. Abrasion on SiC resulted in a minimum of the wear rate for the composite with 20 vol.% TiB2 for most conditions [7]. Wear resistance of NiAl is the challenging factor which has to be considered since some of the proposed applications, e.g. turbine blade tips, are subject to sliding type wear [6]. Wear testing of three NiAl alloys containing 45 at.%, 48 at.% and 50 at.% aluminum indicated that all of them had low coefficient of friction (COF) and WRs. Wear rate is hardness-dependent for these NiAl alloys, and higher hardness leads to lower wear. Meanwhile, wear process of NiAl is dominated by plastic deformation for all compositions [8]. From the tribological point of view, TiC reinforced NiAl intermetallic matrix composites are expected to process excellently both abrasive and adhesive wear resistance because of high hardness of the TiC reinforcements as well as the strong atomic bonding characteristic and abnormal hardness-temperature relationship of nickel aluminides. As a result, TiC reinforced nickel aluminide matrix composite coatings are potential candidates of wear-resistant coating materials [9]. NiAl intermetallic matrix composites containing solid lubricants have attracted much attention as high-temperature selflubricating materials. NiAl–31BaF2–19CaF2 (mass%) prepared by spark plasma sintering (SPS) exhibited the low friction coefficients and wear rates at 873–1073 K [10]. Zhu et al. [11] investigated the high temperature sliding tribological behavior of NiAl matrix composites with addition of oxides (ZnO/CuO). At 1000 °C, the wear rate of NiAl matrix composites with ZnO was about 7 10 6 mm3 (N m) 1, and the superior wear resistance was due to the formation of the ZnO layer on the worn surface. Moreover, the NiAl
180
X. Shi et al. / Materials and Design 45 (2013) 179–189
matrix composites with CuO showed self-lubricating performance at 800 °C, which was attributed to the presence of the glaze layer containing CuO and MoO3. As a new solid self-lubricating material, Ti3SiC2 has attracted much attention recently. Like graphite, Ti3SiC2 has a hexagonal structure with a space group of P63/mmc. Besides, Ti3SiC2 is readily machined as graphite and possesses excellent thermostability. Thus, it was proposed that Ti3SiC2 could be a new solid self-lubricating material with low COF and WRs from low to high temperature ranges [12–15]. However, meager information is available as regards the development of NiAl matrix self-lubricating composites by use of the Ti3SiC2 and C dual lubrication phases and TiC enhanced phase. Recently, in situ technique has been developed to fabricate metal matrix composites. The composites exhibit a clean matrix/ reinforcement interface, which leads to better improvement in mechanical properties and tribological properties of the in situ composites [16,17]. Compared with other traditional sintering methods, the main advantages of SPS are the fast heating and cooling speeds, short sintering time, fine grain size of the prepared materials, controlled organizational structure, energy saving, environmental protection and so on [18–20]. In this study, NiAl matrix self-lubricating composites (NMCs) with different contents of Ti3SiC2 were prepared by in situ technique using SPS. The effects of Ti3SiC2 content on the dry sliding friction and wear behavior of NiAl matrix self-lubricating composites at the load of 10 N and sliding speed of 0.234 m/s for 20 min at room temperature were investigated through the determination of COF and WRs and the analysis of the compositions of wear debris, worn surfaces of both NMC and Si3N4 ceramic ball friction pair. 2. Experimental details
lue was given. The density of as-prepared specimens was determined by Archimedes’ principle according to the ASTM Standard B962-08 [24,25]. 2.3. Tribological test The tribological test was conducted on a HT-1000 ball-on-disk high temperature tribometer according to the ASTM Standard G99-95 [26,27]. The disks, which were the as-prepared materials, were cleaned with acetone and then dried in hot air before test. The counterpart ball was the commercial Si3N4 ceramic ball with a diameter of 5 mm (about HV 15 GPa). The test temperature was room temperature. The sliding speed and applied load were 0.234 m/s and 10 N respectively. The friction radius was 2 mm. The testing time was 20 min. The COF was automatically measured and recorded in real time by the computer system of the friction tester. The wear quantity of NMC was measured by the weighting method, which measured the mass loss of the samples for every friction process. The tests for every given conditions were repeated three times to obtain reliable data. The average value was used as the evaluating data. 2.4. Analysis The surfaces of the as-prepared specimens were examined by XRD with Cu Ka radiation at 30 kV and 40 mA at a scanning speed of 0.01 s 1 for the identification of the phase constitution. The morphologies and compositions of worn surfaces of NMC and Si3N4 ceramic ball friction pair, wear debris were analyzed by a SIRION 200 field emission scanning electron microscope (FESEM) equipped with energy dispersive spectroscopy (EDS) and a JAX8230 electron probe microanalyzer equipped with EDS.
2.1. Materials 3. Results and discussion NMC with different contents of Ti3SiC2 were prepared by in situ technique using SPS. The composite powders of NiAl matrix were composed of commercially available Ni, Al, Mo, Nb, Fe and B powders by atomic ratio of 48:50:1:1:0.5:0.02, whose average particle sizes were about 30–50 lm. The weight fractions of Ti3SiC2 in the composites were fixed at 0 wt.%, 5 wt.%, 10 wt.%, 15 wt.% and 20 wt.%, respectively. The fabrication process of Ti3SiC2 powder (5 lm in average size, 99.5% in purity) was described in detail elsewhere [21]. NiAl alloy without Ti3SiC2 was denoted as NA. While NB, NC, ND, NE represented NMC with 5 wt.%, 10 wt.%, 15 wt.% and 20 wt.% Ti3SiC2, respectively. The starting powders were mixed by high energy ball-milling in vacuum. Balls and vials were made of hard alloy. The charge ratio (ball to powder mass ratio) employed was 10:1. The milling time and speed are 10 h and 200 rpm respectively. After being mixed and dried, the mixtures were then sintered by SPS using a D.R. SinterÒ SPS3.20 (Sumitomo Coal & Mining, now SPS Syntex Inc.) apparatus at 1100 °C under a pressure of 30 MPa for 5 min in pure Ar atmosphere protection. The heating rate was 100 °C/min. The cylindrical graphite molds with an inner diameter of 20 mm were used. The as-prepared specimen surfaces were ground to remove the layer on the surface and polished mechanically with emery papers down to 1200 grad, and then with 0.05 lm wet polishing diamond pastes.
3.1. Compositions of NMC Fig. 1 was the XRD patterns of NMC with different contents of Ti3SiC2 prepared by in situ technique using SPS. It could be found that the diffraction peaks primarily belonged to the NiAl, Ni3Al and Ti3SiC2 phases. Additionally, it was clear that there were TiC and C phases existing in the NiAl matrix composites, which were attributed to the decomposition reaction of Ti3SiC2 and carburization role of graphite atmosphere caused by the graphite mold.
2.2. Vicker’s microhardness and density The Vicker’s microhardness of each as-received specimen was measured, according to the ASTM standard E92-82 [22,23], using an HVS-1000 Vicker’s hardness instrument with a load of 1 kg and a dwell time of 8 s. Five tests were conducted and the mean va-
Fig. 1. XRD patterns of NMC doped with varying weight fractions of Ti3SiC2.
X. Shi et al. / Materials and Design 45 (2013) 179–189
181
As shown in Fig. 1, there were very obvious diffraction peaks of Ti3SiC2 phase at 2h = 34.096° (1 0 1), 39.548° (1 0 4) and 60.283° (1 1 0) existing in the samples of NB, NC, ND and NE. Meanwhile, the peaks belonging to C phase at 2h = 26.381° (0 0 2) could also be found. The peak intensities of Ni3Al phase of NB at both 2h = 43.253° and 53.373° was the much stronger than that of NC, ND and NE. Moreover, as shown in Fig. 1, the intensities of TiC phase at 2h = 35.858° (1 1 1), 41.643° (2 0 0) and 60.357° (2 2 0) in NB when compared with NC, ND and NE were much weaker. 3.2. Vicker’s microhardness and density As shown in Table 1, NMC doped with Ti3SiC2 when compared with NA exhibited the higher Vicker’s microhardness. And the Vicker’s microhardness of NMC increased with the increasing of the doped content of Ti3SiC2. The addition of Ti3SiC2 could improve the strength of NiAl. Moreover, Ti3SiC2 addition further enhanced the hardness, which was attributed to the formation of TiC during the fabricated process. The density of NMC deceased with the increasing of the doped contents of Ti3SiC2. It was attributed to the lower density of Ti3SiC2 (4.52 g/cm3) when compared with NiAl [21]. NB had a higher density due to that the Ni3Al phase were formed. The measured density and Vicker’s microhardness of NC were 5.873 g/cm3 and 593.5 HV1 respectively.
Fig. 2. The relationship of friction coefficient of NMC versus sliding time at the condition of 10 N–0.234 m/s.
3.3. Friction and wear behaviors Fig. 2 shows the typical measuring curves of the dynamic COF of NMC against Si3N4 ceramic ball friction pair at the constant sliding speed of 0.234 m/s and load of 10 N at room temperature. It was obvious that all curves were relatively smooth except that of NE. And the friction coefficients of NMC against Si3N4 ceramic ball pair were in the range of 0.5–0.8. NC had the lowest COF among the samples. Fig. 3 shows the variation of wear rate and friction coefficient of NMC with content of Ti3SiC2 at the condition of 10 N–0.234 m/s. As shown in Fig. 3, NC against Si3N4 ceramic ball pair had the best tribological properties among the samples. The COF and wear rate of NC were about 0.60 and 5.45 10 5 mm3 (N m) 1 respectively. And both of them were the lowest among that of all the samples. However, NE against Si3N4 pair had the worst tribological performance among the samples. The COF and wear rate of NE were about 0.75 and 3.67 10 3 mm3 (N m) 1 respectively. And both of them were highest among that of all the samples. According to the curve of WRs, it was concluded that the WRs increased with the increasing of the doped content of Ti3SiC2 except 10 wt.% Ti3SiC2. 3.4. Microstructure analysis Fig. 4 exhibited the FESEM morphologies and EDS patterns of fractured surfaces of NA and NC. Microstructure of NA was
Table 1 Densities and Vicker’s microhardness of NMC prepared by SPS. Sample
Composition
Measured density (g/ cm3)
Vicker’s microhardness (HV1)
NA NB
NiAl NiAl–5 wt.% Ti3SiC2 NiAl–10 wt.% Ti3SiC2 NiAl–15 wt.% Ti3SiC2 NiAl–20 wt.% Ti3SiC2
5.896 6.249
553.2 580.9
5.873
593.5
5.557
619.8
5.452
678.0
NC ND NE
Fig. 3. The variation of COF and WRs of NMC with varying weight fractions of Ti3SiC2 at the condition of 10 N–0.234 m/s.
presented in Fig. 4a. As shown in Fig. 4a, NA had a dense and homogeneous microstructure. Thus, the NiAl intermetallic should have excellent mechanical properties. Fig. 4c was a result of EDS analysis for Fig. 4a, and the EDS pattern exhibited the presence of Ni, Al, Nb, Mo and a small amount of Fe. As shown in Fig. 4b, it seemed that NC had denser microstructure than NA. Fig. 4d was a result of EDS analysis for Fig. 4b, and the EDS pattern exhibited the presence of Al, Si, Nb, Mo, Ti, Cr and Ni. Meanwhile, the contents of Ti and Si elements were 8.65 at.% and 4.84 at.% respectively. And there was a clean matrix/Ti3SiC2 interface existing. A clean matrix/reinforcement interface would lead to better and improved mechanical and tribological properties of the in situ composites [16,28]. Ramesh and Ahamed [16] reported that there was homogeneity in the distribution of in situ TiB2 particles within the matrix alloy with a clear matrix-reinforcement interface. The in situ TiB2 particles significantly improved the tribological properties of the 6063/ TiB2 composites. 3.5. Morphologies, compositions of wear debris and the worn surfaces of NA, NC, NE and Si3N4 ceramic ball Fig. 5a–c exhibited the typical electron probe morphologies of worn surfaces of NA at the condition of 10 N–0.234 m/s for
182
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 4. FESEM morphologies and EDS patterns of fractured surfaces of NA and NC. (a and b) Morphologies of fractured surfaces of NA and NC; (c and d) EDS patterns exhibiting the chemical compositions of NA and NC; (c) versus (a) and (d) versus (b).
20 min at room temperature. As shown in Fig. 5a–c, it was apparent that there was the evidence of a significant amount of plastic flow on the worn surfaces. Johnson et al. [8] studied the dry sliding wear behavior of the B2-structured (ordered body-centered cubic) compound NiAl. The results showed that surface cracking was all but nonexistent, meaning that the brittle nature of the material did not seem to be affecting wear. Plasticity appeared in two forms: extrusion of material across and over the trailing edge of the wear scar, and deep grooving or microploughing of material across the wear scar from leading to trailing edge. The NiAl worn surfaces showed signs of chunking, or separation of (relatively) large pieces of material followed by the formation of a groove. Fig. 5d was a result of EDS analysis for the A area as shown in Fig. 5c. The EDS pattern of the surface of NA exhibited the presence of Al, Ni, Nb, Mo, C and a small amount of O. The content of O was 1.23 wt.%. Fig. 5e was a result of EDS analysis performed on the worn track for the B area as shown in Fig. 5c, which revealed that there were O, Ni and Al elements existing. And the content of O in B area when compared with A area increased obviously, which was 4.52 wt.%. The content of C was 1.64 wt.%, which could be produced by the carburization role of graphite atmosphere from the graphite mold. Shaji and Radhakrishnan [29] investigated the possibility of using graphite as a lubricating medium to reduce the heat generated at the grinding zone in surface grinding with a newly developed experimental setup. Cho et al. [30] researched the effect of three different solid lubricants (graphite, Sb2S3, and
MoS2) in the brake friction material on various aspects of friction characteristics was investigated. The results showed that graphite was an excellent lubricant at lower temperature. The existing graphite lubricant could be beneficial for the tribological performance of NA at the room temperature. It is conceivable that the frictional oxide film was formed on the worn surfaces for the higher oxygen content of 10.75 at.%. It was no doubt that the Si was transferred from the counterpart of Si3N4 ceramic ball. It was attributed to severe abrasive action of tribological pairs during sliding process. In this area, obvious crack was observed. Fig. 5f was a result of EDS analysis for the C area as shown in Fig. 5c. There were little particles detached from the surface of this area. The abundant amount of O element was 4.84 wt.%, which confirmed the forming of the oxide film. However, the thickness of oxide film was thin. Stott [31] reviewed some of the models developed to account for the generation of oxide during sliding and the effects of such oxides on the rates of wear. He claimed that the wear-protective layers consisted mainly of loosely-compacted particles at low temperatures. The oxidation layer occurred on worn surface decreased the wear rate. As shown in Fig. 3, the wear rate of NA was 9.02 10 5 mm3 (N m) 1. Some particles were detached from the worn surface, which remained between the surfaces of NiAl and Si3N4 ball. Because of the high surface temperature originated from the friction force, they these particles were oxidized. Meanwhile, it could be concluded that the wear process of NiAl was dominated by plastic deformation.
X. Shi et al. / Materials and Design 45 (2013) 179–189
183
Fig. 5. Electron probe morphologies and EDS patterns of worn surface of NA at the condition of 10 N–0.234 m/s. (a–c) Morphologies of worn surface of NA; (d–f) EDS patterns exhibiting the chemical compositions of NA; (d) versus A area in (c), (e) versus B area in (c) and (f) versus C area in (c).
Fig. 6a–c exhibited the typical electron probe morphologies of worn surfaces of NC at the condition of 10 N–0.234 m/s. As shown in Fig. 6a, the tribo-layers called ‘glazes’ [32] had been formed on the worn surfaces of NC. The glazes were not complete, and the delamination of the glazes happened. Stott et al. [33,31,34] have investigated the interaction between sliding wear and oxidation in superalloys. The results showed that the glazes could temporar-
ily protect the surfaces from further contact damage. If they happened to wear off, new glazes could be formed to take their place. This progression of formation, loss, and reformation could result in short-term friction or wear transients. As shown in Fig. 6b, the grooves on the worn surface of NC became coarser when compared with the relatively finer grooves of NA, and the corresponding wear mechanism transferred from a mixed
184
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 6. Electron probe morphologies and EDS patterns of worn surface of NC at the condition of 10 N–0.234 m/s. (a–c) Morphologies of worn surface of NC; (d–f) EDS patterns exhibiting the chemical compositions of NC; (d) versus A area in (c), (e) versus B area in (c) and (f) versus C area in (c).
abrasion-plastic deformation to microploughing. Moreover, the delaminated glazes were observed. Apparently, the worn surfaces were characterized by long discontinuous grooves and delaminated glazes. During the sliding friction and wear process, the Si3N4 ceramic ball ploughed across the surface of NC, and eventually removed or pushed material into ridges along sides of the grooves. Meanwhile, the tribo-layers were formed during the contact
process, which were mixture of lubricants, NiAl matrix and oxides. And delamination was attributed to the tangential shears during sliding, which caused the phenomenon of discontinuous grooves. Fig. 6d was a result of EDS analysis for the A area as shown in Fig. 6c. The EDS pattern exhibited the presence of Al, Ni, Ti, Mo, C, Si and O. The content of O was 3.10 wt.%, which was lower than 4.52 wt.% of worn surface of NA. It showed that the oxidation
X. Shi et al. / Materials and Design 45 (2013) 179–189
185
Fig. 7. Electron probe morphologies and EDS patterns of worn surfaces of NE at the condition of 10 N–0.234 m/s. (a–c) Morphologies of worn surface of NE; (d–h) EDS patterns exhibiting the chemical compositions of NE; (d) versus D area in (a), (e) versus E area in (a), (f) versus F area in (a), (g) versus G area in (b) and (h) versus H area in (b).
186
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 8. FESEM morphology and EDS patterns of worn surface of Si3N4 friction pair against NC at the condition of 10 N–0.234 m/s. (a) Morphology of worn surface of Si3N4 friction pair, (b and c) EDS patterns exhibiting the chemical compositions of Si3N4 friction pair, (b) versus B point in (a) and (c) versus C point in (a).
phenomenon of worn surface of NC when compared with NA was mild at the condition of 10 N–0.234 m/s. The friction force between NC and Si3N4 ceramic ball was smaller for the Ti3SiC2 lubricant existing, thus the friction heat was also lower. And the oxidation phenomenon was mild. Fig. 6e was the EDS analysis for the B area as shown in Fig. 6c. As shown in Fig. 6e, the EDS pattern exhibited the presence of Al, Ni, Ti, C, Si and O. The content of O was 3.41 wt.%, which was higher than 3.10 wt.% of A area on worn surface. It showed the oxidation phenomenon of B area when compared with A area on the worn surface was more serious. As the lubricant layer gradually disappeared, microploughing became more serious. Some particles were detached from the surface, and oxidation phenomenon increased. However, the glaze flaking of NC when compared with NA was not more serious. Thus, the wear rate of NC was lower than that of NA at the condition of 10 N–0.234 m/s. Fig. 6f was a result of EDS analysis for the C area (see Fig. 6c), which was formed by the ground and polished process before the tribological test. The EDS pattern exhibited the presence of Al, Ni, Ti and Si. A small amount of O (0.89 wt.%) also existed in this area, which was much lower than that of worn surface obtained by the dry sliding friction and wear process of NC against Si3N4 ceramic ball. Fig. 7a–c exhibited the typical electron probe morphologies of worn surfaces of NE at the condition of 10 N–0.234 m/s. It could be observed that the worn surface turned much coarser when the doped content of Ti3SiC2 was 20 wt.%, and there were a lot of pits
and mounds exiting on worn surfaces of NE. It indicated that the main wear mechanisms obviously were ploughing and abrasive wear. Moreover, as shown in area G of Fig. 7b, a few discontinuous glazes were also formed. And the areas of the glazes when compared with the whole worn surfaces were smaller. The worn surface was very rough, and the wear resistance should be higher. Thus, the COF of NE was 0.75, which was the highest in NMC. Fig. 7d–h were the EDS analysis for the D, E and F areas as shown in Fig. 7a, as well as G and H areas as shown in Fig. 7b, respectively. As shown in Fig. 7d–h, the Ti content was approximately three times of Si content. Moreover, the contents of O in the areas of E, G and H were much higher than that in the areas of D and F. It showed that the oxidation phenomena in the E, G and H areas of the worn surfaces were more serious. However, the content of Ti in NE was nearly six to nine times as much as that of NC, which was much higher than the content of Ni and Al. Meanwhile, the content of C was also higher. Thus, we could get a conclusion that a lot of TiC and Ti3SiC2 existed on the worn surfaces of NE. TiC enhanced the hardness of NE (678 HV1). When the doped content of Ti3SiC2 was 20 wt.%, the homogeneous distribution of Ti3SiC2 lubricant in NMC could not be realized. The lubricant enrichment happened in the local parts of NMC. And mechanical and tribological properties of NMC could be deteriorated by the inhomogeneity of lubricant. Thus, the NE had the highest wear rate of 3.67 10 3 mm3 (N m) 1. Fig. 8 shows the FESEM morphologies and EDS of worn surfaces of Si3N4 friction pair against NC at the condition of 10 N–0.234 m/s.
X. Shi et al. / Materials and Design 45 (2013) 179–189
187
Fig. 9. FESEM morphology and EDS pattern of wear debris of NA at the condition of 10 N–0.234 m/s with Si3N4 friction pair. (a) Morphology of wear debris of NA and (b) EDS pattern exhibiting the chemical compositions of wear debris of NA.
Fig. 10. FESEM morphology and EDS pattern of wear debris of NC at the condition of 10 N–0.234 m/s with Si3N4 friction pair. (a) Morphology of wear debris of NC and (b) EDS pattern exhibiting the chemical compositions of wear debris of NC.
As shown in Fig. 8a, there were a lot of mounds and some adhesive wear debris existing on the worn surface. Fig. 8b was the EDS analysis for the B point as shown in Fig. 8a. The Si content was 8.87 wt.%, and no N was detected on this point. The contents of O, Al and Ni were 8.63 wt.%, 1.87 wt.% and 6.09 wt.%, respectively, which confirmed the existing of oxides of aluminum and nickel. It also showed that matter translocation between NC and Si3N4 ball had happened. Moreover, the elements of K and Cl appeared in Fig. 8b. It could be caused by the sweat of the operator during the specimen preparation. Fig. 8c was the EDS analysis for the C point as shown in Fig. 8a. The Si content was 66.49 wt.%, and the N content was 20.26 wt.%. The contents O and Al were 9.77 wt.% and 3.50 wt.% respectively, which confirmed the existing of aluminum oxide. It was attributed to severe abrasive action and material transfer of tribological pairs during sliding friction and wear process. Fig. 9 shows the FESEM morphologies and EDS of wear debris of NA at the condition of 10 N–0.234 m/s. As shown in Fig. 9a, the sizes of wear debris with irregular shape were from 100 nm to 10 lm. It showed that most of wear debris were directly generated from NA rather than the glazes. The wear resistance was higher, thus the COF was also higher. Fig. 9b was a result of EDS analysis
for the wear debris. The EDS pattern exhibited the presence of Al, Ni, Ti, C, Si, Mo, Fe and O. The contents of O, Al and Ni were 25.43 wt.%, 17.34 wt.% and 37.99 wt.%, respectively, which confirmed that the wear debris were composed of oxides of aluminum and nickel, as well as wear debris particles of NA. The Si content was 0.55 wt.%. Fig. 10 shows the FESEM morphologies and EDS of wear debris of NC obtained at the condition of 10 N–0.234 m/s. As shown in Fig. 10a, it indicated that the wear debris was mainly a product of the glazes rather than having been directly generated from NC. The wear debris were composed of lots of finer particles, whose average size was less than 1 lm. And the particle size distribution was uniform. Fig. 10b was a result of EDS analysis for the wear debris. The EDS pattern exhibited the presence of Al, Ni, Ti, C, Si, Mo, Fe and O. The contents of O, Al, Si, Ti and Ni were 5.28 wt.%, 17.76 wt.%, 3.28 wt.%, 11.23 wt.% and 48.21 wt.%, respectively. The content of Ti was 11.23 wt.%, and the content of Si was 3.28 wt.%. The debris should be composed of oxides and wear debris particles of NC. Sun et al. [35] investigated the dry sliding friction and wear behavior of Ti2AlN/TiAl composite at room temperature. They reported that the wear debris formed at earlier friction stage move between the two surfaces, and the major wear
188
X. Shi et al. / Materials and Design 45 (2013) 179–189
Fig. 11. FESEM morphology and EDS pattern of wear debris of NE at the condition of 10 N–0.234 m/s with Si3N4 friction pair. (a) Morphology of wear debris of NE and (b) EDS pattern exhibiting the chemical compositions of wear debris of NE.
mechanism changes from two-body abrasion to three-body abrasion. Fine equiaxed wear debris were compressed and broken into finer particles by the normal stress. Meanwhile, the high surface temperature that originated from the friction heat during the dry sliding friction process made some fine particles be oxidized in the air. Oxides of fine equiaxed wear debris and unoxidized fine equiaxed wear debris were trapped on the worn surface of NC to form the glazes, which acted as an antifriction. The glazes actually had not only a significant antifriction effect, but also a protective action for the friction surfaces. Thus, the COF of NC was about 0.60, and the wear rate was 5.45 10 5 mm3 (N m) 1. Fig. 11 shows the FESEM morphologies and EDS of wear debris of NE obtained at the condition of 10 N–0.234 m/s. As shown in Fig. 11a, the sizes of wear debris with irregular shape were from 1 lm to 10 lm. The wear resistance of forming large size of wear debris was much higher, thus the COF of NE was about 0.75. Meanwhile, the larger size wear debris meant higher wear rate, which was 3.67 10 3 mm3 (N m) 1. Fig. 11b was a result of EDS analysis for the wear debris. The EDS pattern exhibited the presence of Al, Ni, Ti, Si, Nb, Fe, O and a small amount of C. The contents of O, Al, Si, Ti and Ni were 21.40 wt.%, 15.94 wt.%, 4.85 wt.%, 22.90 wt.% and 31.25 wt.%, respectively. The content of O was much higher. And the Ti content was almost three times of Si content. The wear debris should be composed of oxides and wear debris particles of NE. 4. Conclusions (1) NC fabricated by the in situ technique using SPS at 1100 °C for 5 min under 30 MPa had the excellent tribological properties. At the condition of 10 N–0.234 m/s at room temperature, the COF of NC was about 0.60, and the wear rate was 5.45 10 5 mm3 (N m) 1. Moreover, the tribological noise was quite low during the dry sliding friction and wear test process. (2) The wear mechanisms changed with the increasing of the doped content of Ti3SiC2. The wear mechanism of NA was a mixed abrasion-plastic deformation. Microploughing and delamination were dominant wear mechanisms in case of NC. While the wear mechanisms of NE were ploughing and abrasive wear. (3) The glazes of NC actually had not only a significant antifriction effect, but also a protective action for the friction and wear surfaces. Oxides of fine equiaxed wear debris and
unoxidized fine equiaxed wear debris were trapped on the worn surface of NC to form the glazes, which acted as an antifriction. The glazes were composed of lots of finer particles, whose average size was less than 1 lm. And the particle size distribution was uniform.
Acknowledgments This work was supported by the Fundamental Research Funds for the Central Universities (2010-II-020); the National Natural Science Foundation of China (51275370); the Project for Science and Technology Plan of Wuhan City; the Program for New Century Excellent Talents in University of Ministry of Education of China; the Nature Science Foundation of Hubei Province; and the Academic Leader Program of Wuhan City (201150530146).
References [1] Miracle DB. The physical and mechanical properties of NiAl. Acta Metall Mater 1993;41:649–84. [2] Zhou LZ, Guo JT, Li GS, Xiong LY, Wang SH, Li CG. Investigation of annealing behavior of nanocrystalline NiAl. Mater Des 1997;18:373–7. [3] Liu CT. Recent advances in ordered intermetallics. Mater Chem Phys 1996;42:77–86. [4] Stoloff NS, Liu CT, Deevi SC. Emerging applications of intermetallics. Intermetallics 2000;8:1313–20. [5] Lasalmonie A. Intermetallics: why is it so difficult to introduce them in gas turbine engines. Intermetallics 2006;14:1123–9. [6] Ozdemir O, Zeytin S, Bindal C. Tribological properties of NiAl produced by pressure-assisted combustion synthesis. Wear 2008;265:979–85. [7] Hawk JA, Alman DE. Abrasive wear behavior of NiAl and NiAl–TiB2 composites. Wear 1999;225–229:544–56. [8] Johnson BJ, Kennedy FE, Baker I. Dry sliding wear of NiAl. Wear 1996;192:241–7. [9] Chen Y, Wang HM. Microstructure of laser clad TiC/NiAl–Ni3(Al, Ti, C) wearresistant intermetallic matrix composite coatings. Mater Lett 2003;57:2029–36. [10] Murakami T, Ouyang JH, Sasaki S, Umeda K, Yoneyama Y. High-temperature tribological properties of Al2O3, Ni-20 mass% Cr and NiAl spark-plasmasintered composites containing BaF2–CaF2 phase. Wear 2005;259:626–33. [11] Zhu SY, Bi QL, Niu MY, Yang J, Liu WM. Tribological behavior of NiAl matrix composites with addition of oxides at high temperatures. Wear 2012;274– 275:423–34. [12] El-Raghy T, Blau P, Barsoum MW. Effect of grain size on friction and wear behavior of Ti3SiC2. Wear 2000;238:125–30. [13] Sun ZM, Zhou YC, Li S. Tribological behavior of Ti3SiC2-based material. J Mater Sci Technol 2000;18:142–5. [14] Zhang Y, Ding GP, Zhou YC, Cai BC. Ti3SiC2-a self-lubricating ceramic. Mater Lett 2002;55:285–9.
X. Shi et al. / Materials and Design 45 (2013) 179–189 [15] Gupta S, Filimonov D, Zaitsev V, Palanisamy T, Barsoum MW. Ambient and 550 °C tribological behavior of select MAX phases against Ni-based superalloys. Wear 2008;264:270–8. [16] Ramesh CS, Ahamed A. Friction and wear behaviour of cast Al 6063 based in situ metal matrix composites. Wear 2011;271:1928–39. [17] Tjong SC, Ma ZY. Microstructural and mechanical characteristics of in situ metal matrix composites. Mater Sci Eng R 2000;29:49–113. [18] Murakami T, Sasaki S, Ito K. Oxidation behavior and thermal stability of Crdoped Nb(Si, Al)2 and Nb3Si5Al2 matrix compacts prepared by spark plasma sintering. Intermetallics 2003;11:269–78. [19] Murakami T, Sasaki S, Ichikawa K, Kitahara A. Microstructure, mechanical properties and oxidation behavior of Nb–Si–Al and Nb–Si–N powder compacts prepared by spark plasma sintering. Intermetallics 2001;9:621–7. [20] Kim H, Kawahara M, Tokita M. Specimen temperature and sinterability of Ni powder by spark plasma sintering. J Jpn Soc Powder Metall 2000;47:887–91. [21] Peng MC, Shi XL, Zhu ZW, Wang M, Zhang QX. Facile synthesis of Ti3SiC2 powder by high energy ball-milling and vacuum pressureless heat-treating process from Ti–TiC–SiC–Al powder mixtures. Ceram Int 2012;38:2027–33. [22] Loganathan TM, Purbolaksono J, Inayat-Hussain JI, Wahab N. Effects of carburization on expected fatigue life of alloys steel shafts. Mater Des 2011;32:3544–7. [23] American society for testing and materials. Standard test method for vickers hardness of metallic materials. ASTM E92-82 e2, 2003. [24] Pellizzari M, Fedrizzi A, Zadra M. Influence of processing parameters and particle size on the properties of hot work and high speed tool steels by Spark Plasma Sintering. Mater Des 2011;32:1796–805.
189
[25] American society for testing and materials. Standard test methods for density of compacted or sintered powder metallurgy (PM) products using Archimedes’ principle, ASTM B962-08, 2008. [26] Koksal S, Ficici F, Kayikci R, Savas O. Experimental optimization of dry sliding wear behavior of in situ AlB2/Al composite based on Taguchi’s method. Mater Des 2012;42:124–30. [27] American society for testing and materials. Standard test method for wear testing with a pin-on-disk apparatus, ASTM G99-95, 1995. [28] Kumar S, Chakraborty M, Subramanya Sarma V, Murthy BS. Tensile and wear behaviour of in situ Al-7Si/TiB2 particulate composites. Wear 2008;26:134–42. [29] Shaji S, Radhakrishnan V. Analysis of process parameters in surface grinding with graphite as lubricant based on the Taguchi method. J Mater Process Technol 2003;141:51–9. [30] Cho MH, Ju J, Kim SJ, Jang H. Tribological properties of solid lubricants (graphite, Sb2S3, MoS2) for automotive brake friction materials. Wear 2006;260:855–60. [31] Stott FH. The role of oxidation in the wear of alloys. Tribol Int 1998;31:61–71. [32] Blau PJ. Elevated-temperature tribology of metallic materials. Tribol Int 2010;43:1203–8. [33] Stott FH, Lin DC, Wood GC. Structure and mechanism of formation of the ‘Glaze’ oxide layers produced on Ni-based alloys during wear at high temperatures. Corros Sci 1973;13:449–69. [34] Stott FH, Lin DC, Wood GC. ‘Glazes’ produced on nickel-base alloys during high temperature wear. Nature 1973;242:75–7. [35] Sun T, Wang Q, Sun DL, Wu GH, Na Y. Study on dry sliding friction and wear properties of Ti2AlN/TiAl composite. Wear 2010;268:693–9.