Friction and wear properties of reactive hot-pressed TiB2–TiN composites in sliding against Al2O3 ball at elevated temperatures

Wear 271 (2011) 1966–1973

Contents lists available at ScienceDirect

Wear journal homepage: www.elsevier.com/locate/wear

Friction and wear properties of reactive hot-pressed TiB2 –TiN composites in sliding against Al2 O3 ball at elevated temperatures J.H. Ouyang ∗ , Z.L. Yang, Z.G. Liu, X.S. Liang Institute for Advanced Ceramics, Department of Materials Science, Harbin Institute of Technology, Harbin 150001, China

a r t i c l e

i n f o

Article history: Received 1 September 2010 Received in revised form 30 November 2010 Accepted 30 November 2010

Keywords: TiB2 –TiN composites Reactive hot pressing Friction Wear Wear resistance

a b s t r a c t TiB2 –TiN composites with different TiB2 :TiN molar ratios of 2:1 and 1:1 were synthesized by the reactive hot-pressing process using Ti, B and BN powders as raw materials. Friction and wear properties of reactive hot-pressed TiB2 –TiN composites were evaluated in sliding against Al2 O3 ball from room temperature to 700 ◦ C in air. For TiB2 –TiN composites, typical break-in stage in evolution of friction coefficients is observed at room temperature. At 700 ◦ C, the friction coefficient of TiB2 –TiN composites is relatively low and steady, as contrasted with the data obtained at room temperature and 400 ◦ C. Under all testing conditions, TiB2 –TiN composites exhibit small wear rates, which range in the order of 10−6 –10−7 mm3 /Nm. The specific wear rates of TiB2 –TiN composites and the coupled ball decrease with testing temperature. Wear mechanisms of TiB2 –TiN composites depend mainly upon testing temperature. At room temperature, mild abrasion is the dominated wear mechanism of TiB2 –TiN composites. However, tribo-oxidation as well as mild abrasion plays an important role during wear test at 400 ◦ C. At 700 ◦ C, continuous oxidized films comprised of rutile form on the surface by thermal oxidation, and provide excellent lubrication and protection. © 2011 Elsevier B.V. All rights reserved.

1. Introduction In recent years, there is a need for wear resistant components used in mechanical systems that involve high temperatures, heavy loads and high velocities. Refractory materials such as borides, nitrides, carbides and their combinations thereof are natural candidates for these tribological applications due to their exceptional hardness, wear resistance and thermal stability at high temperatures. Titanium diboride (TiB2 ) is one kind of refractory compounds with high melting point, high elastic modulus, and high hardness. A problem in winning industrial applications, however, is that TiB2 exhibit poor sinterability [1] and relatively low fracture toughness [2]. Compositing is a possible way of avoiding some of these problems. TiB2 -containing composites as wear resistant materials have been widely studied [3–5]. Titanium nitride (TiN) is a potential choice as a toughening second phase as it has good structural and thermodynamic compatibility with TiB2 . Moreover, TiN has some attractive properties, such as high hardness, good electrical conductivity and excellent wear resistance. TiB2 –TiN composites have attracted great attention on industrial applications as jet engine parts, armor plates, cutting tools, dies, and other high-temperature

∗ Corresponding author. Tel.: +86 451 86414291; fax: +86 451 86414291. E-mail address: [email protected] (J.H. Ouyang). 0043-1648/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.wear.2010.11.057

ceramic components. TiB2 –TiN ceramics were fabricated with selfpropagating high temperature synthesis (SHS) method using Ti (or TiH2 ), B and BN powders as raw materials [6,7]. However, only a very limited amount of experiments on tribological behavior of such promising composites are available in open literatures, and high temperature wear mechanisms of TiB2 –TiN ceramics still remain unclear. The present investigation is carried out to understand the tribological performance of the promising TiB2 –TiN composites in dry sliding against the widely used hard engineering materials such as alumina. In the present study, an attempt has been made to investigate the effects of applied load and surrounding temperature on wear mechanisms of reactive hot pressed TiB2 –TiN composites. 2. Experimental procedure 2.1. Materials preparation In the present work, two TiB2 –TiN composites with TiB2 :TiN molar ratios of 2:1 (denoted as BN21) and 1:1(denoted as BN11) were prepared by a reactive hot pressing process, using commercial titanium (∼30 ␮m, 99.9%), hexagonal boron nitride (∼150 nm, 99%) and boron powders (∼1.5 ␮m, 99.9%) as the starting reactants. The starting stoichiometry of powder mixtures was in accordance with the following reaction: xTi + BN + (2x − 3)B → TiN + (x − 1)TiB2

(x = 2, 3)

(1)

J.H. Ouyang et al. / Wear 271 (2011) 1966–1973

30 mm. Elastic modulus was calculated from the stress–strain curve obtained in the strength test. Fracture toughness was measured by the single-edge-notch beam (SENB) method. All the abovementioned tests were conducted on a universal testing machine (mod. 5569, Instron Ltd., USA). Hardness of the samples was determined using Vickers indentation using a load of 98.1 N and a dwell time of 10 s. At least six specimens were tested for each experiment. Results of these experiments are listed in Table 1. Specimens with dimensions of 4 mm × 8 mm × 16 mm were cut out from the sintered bulks by electrical discharge machining. These flats were ground with 1500-grit emery paper, and were finally polished with diamond paste down to 1 ␮m finish. The surface that is perpendicular to the hot pressing direction was used as the contact surface. Prior to friction and wear test, all of the polished specimens were ultrasonically cleaned first in acetone and ethanol successively.

TiB 2

Intensity

TiN

BN11

BN21 20

30

40

50

60

70

1967

80

2 Theta (deg.) Fig. 1. XRD patterns of the reactive hot pressed TiB2 –TiN composites.

2.2. Friction and wear tests

The weighed powders were mixed by wet ball milling for 24 h in polyethylene bottle with ZrO2 balls and ethanol as media. After mixing, the slurries were dried in a rotary vacuum evaporator, and then screened through a 120-mesh screen. The powder mixtures were heated to 1300 ◦ C and held for 30 min, and then followed by uniaxial hot pressing at 1800 ◦ C for 1 h in vacuum with an applied pressure of 30 MPa. Phase composition and microstructure of the synthesized materials were investigated by X-ray diffraction (XRD) and scanning electron microscopy (SEM). For SEM observations, the polished surfaces were etched using a mixed acid of HF and HNO3 in a volume ratio of HF:HNO3 :H2 O = 1:1:1 for 20 s. As presented in Fig. 1, only TiB2 and TiN are identified as the crystalline phases in the as-synthesized composites. Typical microstructures of TiB2 –TiN composites are presented in Fig. 2. The microstructural micrographs reveal the homogeneous distribution of TiB2 grains (gray) and TiN grains (bright). Clearly, both BN21 and BN11 composites exhibit high relative density. The bulk density was measured using the Archimedes method (distilled water as the medium), while the theoretical density was estimated by applying the rule of mixture. Mechanical properties of the as-synthesized composites were measured at room temperature. The flexural strength was tested on chamfered bars with dimensions of 36 mm × 4 mm × 3 mm (length × width × thickness, respectively) in three-point bending configuration with a span of

Friction and wear tests were performed on a reciprocating ball-on-block high temperature tribometer. The test specimen was placed on a SUS310 stainless steel holder heated by a highfrequency induction heating coil. A stationary Al2 O3 ball (a ball radius of 5 mm and a Vickers hardness of 16 GPa) was fixed on the upper holder to slide against the flat specimen. The friction coefficient was calculated by measuring the friction force at 1-ms intervals, obtaining the average friction force (absolute value) of one cycle of the reciprocation (1 s) and dividing the average friction force by the load. Wear volumes of the tribo-couples are defined in Fig. 3. The wear volumes of wear track on the flat (Vf ) and the coupled ball (Vb ) can be estimated by the following formulae:



Vf = r 2 × Arccos



Vb = h2 R −

h 3

r − d



r

−W ×

with h = R −



r−d ×L 2

(2)

R2 − b2

(3)



where for the wear track on the flat, W, d and L are the width, depth and length, respectively, and r is the curvature radius of the cross section. For the coupled ball, R and b are the radius of the ball and wear scar on the ball, respectively, and h is the height of the worn segment. The different parameters were determined as follows: W and d were determined by profilometry; b was measured by optical microscopy; L and R were scaled by vernier callipers; r and h were

Fig. 2. Microstructure of the hot pressed TiB2 –TiN composites: (a) BN21 and (b) BN11. Table 1 Relevant properties of the reactive hot pressed TiB2 –TiN composites. Sample

Relative density (%)

Flexural strength (MPa)

Fracture toughness (MPa m1/2 )

Vickers hardness (GPa)

Elastic modulus (GPa)

BN11 BN21

97.4 97.0

518 ± 27 457 ± 31

5.6 ± 0.2 5.7 ± 0.3

18.2 ± 0.8 20.9 ± 0.2

460.2 446.5

1968

J.H. Ouyang et al. / Wear 271 (2011) 1966–1973

V PL

(4)

where W (mm3 /Nm), P (N) and L (m) are the specific wear rate, applied normal load and the sliding distance, respectively. The displacement stroke, normal load, frequency and testing time are the external variables associated with the tribometer. The wear tests were performed at loads of 10 and 20 N, test temperatures of room temperature, 400 and 700 ◦ C, a reciprocating frequency of 1 Hz, a linear stroke of 10 mm, and test duration of 1200 s. In addition, each specimen and its coupled ball were heated from room temperature to testing temperature by high frequency induction heating coil at a heating rate of 50 ◦ C/min, and held at testing temperatures for 5 min before starting the wear test. During wear runs, the temperature of the specimens was monitored using a thermocouple inserted at the near-surface position through a small hole in the SUS310 stainless steel holder. After wear tests, worn surfaces of tested samples were investigated by means of X-ray diffraction (XRD), scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS) in order to obtain information on wear mechanisms. XRD was performed to identify possible structural changes of worn surfaces using a computer-controlled diffractometer with Cu K␣ radiation (Rigaku D/max 2200VPC, Japan). Morphologies of worn surfaces were observed using SEM (FEI Quanta 200F, USA). Simultaneously, the compositional analysis of worn surfaces and wear debris was carried out using the EDS attachment. For SEM observations, the samples were coated with a thin Pt coating in order to obtain sufficient conductivity on the surface.

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

(a)

0

BN21-10N BN21-20N BN11-10N BN11-20N

200

400

600

800

1000

1200

Sliding time (s)

Friction coefficient

W=

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

(b)

BN21-10N BN21-20N BN11-10N BN11-20N 0

200

400

600

800

1000

1200

Sliding time (s) 0.5

Friction coefficient

obtained by calculation. The sectional profile of each track on the tested flat was measured at 5 points with a 2 mm interval in the direction parallel to the sliding direction, using a contact type of surface roughness meter (JB-4C, Shanghai Taiming Optical Instrument Co., China). Specific wear rate was calculated by the following formula:

Friction coefficient

Fig. 3. Schematic diagram showing wear volumes of the tribo-couples: (a) wear volume of tracks on the flat, (b) wear volume of the coupled ball.

(c)

BN21-10N BN21-20N BN11-10N BN11-20N

0.4 0.3 0.2 0.1

3. Results and discussion 3.1. Friction characteristics

0.0 0

200

400

600

800

1000

1200

Sliding time (s) Fig. 4 depicts the variations in friction coefficients of BN21 and BN11 composites with sliding time at different temperatures. In each graph, four frictional curves of two composites are reported at 10 N and 20 N loads. Clearly, two-stage friction curves are observed for both BN21 and BN11 composites at room temperature. The friction coefficient starts at a relatively low value and suddenly rises to a high value in less than 30 s sliding time. After fluctuating for 100–400 s, the friction coefficient gradually decreases to a steady value. Such two-stage friction curves were also observed by Hu et al. [8] in dry sliding of Ti3 SiC2 and Ti3 SiC2 –Al2 O3 composites. The

Fig. 4. Dependency of friction coefficient with sliding time for BN21 and BN11 composites against Al2 O3 ball at different temperatures and 10 N and 20 N loads: (a) room temperature, (b) 400 ◦ C; (c) 700 ◦ C.

fluctuations of friction coefficient during sliding are related to the surface status of tribo-contact regions [9]. In the initial stage, the flat and the coupled ball contact each other only at the asperities. The asperities are gradually removed by friction and fracture during sliding. For BN21 and BN11 composites, the break-in period elon-

J.H. Ouyang et al. / Wear 271 (2011) 1966–1973

TiB2 (s) + 5/2O2 (g) → TiO2 (s) + B2 O3 (l)

(5)

TiN(s) + O2 (g) → TiO2 (s) + 1/2 N2 (g)

(6)

In our previous study [12], TiB2 –TiN composites start to oxidize above 500 ◦ C, and the oxidation rate increases with temperature. The considerable formation of TiO2 phase at the tribo-contact surfaces is believed to be responsible for the sharp decrease in friction coefficient at 700 ◦ C. The newly formed rutile phase possesses a much smaller critical shear stress, which provides it excellent lubricity, as contrasted with TiB2 and TiN constituents. Similar lubricity caused by thermal oxidation was also reported in TiN or TiB2 reinforced composites [13,14] and other non-oxide ceramics [4,16,17].

TiB 2

(a)

TiN

Intensity

BN11

BN21 20

30

40

50

60

70

80

2 Theta (deg.) TiB2

(b)

Intensity

TiN

BN11

BN21 20

30

40

50

60

70

80

2 Theta (deg.) TiB2

(c)

TiN TiO2(R) Intensity

gates with increasing applied load, as shown in Fig. 4(a). Friction coefficients of BN11 are lower than those of BN21 at room temperature. Besides, the friction coefficients increase slightly with the applied loads. As the temperature increases to 400 ◦ C, the friction coefficient becomes much higher than those obtained at room temperature. The friction coefficient (0.55–0.75) of BN21 is slightly lower than that (0.65–0.85) of BN11. The effect of applied load on friction coefficient is slight for both BN21 and BN11 composites. As presented in Fig. 4(c), both BN21 and BN11 exhibit steady and low friction coefficients at 700 ◦ C. The friction coefficients decrease with increasing applied load. The average values of BN21 and BN11 are 0.27 and 0.27 at 10 N load, and 0.18 and 0.17 at 20 N load, respectively. In order to understand tribological behavior of TiB2 –TiN composites in sliding against Al2 O3 ball, structural changes of worn surface layers after wear tests at different temperatures are investigated. No new crystalline phase is identified on worn surfaces tested at room temperature and 400 ◦ C. However, TiO2 (rutile) phase is newly formed on worn surfaces of TiB2 –TiN composites subjected to 700 ◦ C wear test, as shown in Fig. 5. Besides, trace of Al2 O3 is also identified on worn surfaces, which is speculated to result from the transferred Al2 O3 debris. As for TiB2 and TiN constituents, the following isothermal oxidation reactions will take place on the surfaces of TiB2 –TiN composites at 700 ◦ C [10,11]:

1969

Al2O3 BN11

BN21 3.2. Wear characteristics

20

30

40

50

60

70

80

2 Theta (deg.) In order to estimate the specific wear rates of TiB2 –TiN composites, the cross sectional profiles of wear tracks were measured by contact-type surface roughness meter, the wear scars on the coupled Al2 O3 balls were observed by SEM. Fig. 6 shows the representative profiles of the wear tracks after wear tests against alumina at 20 N load and different temperatures. After wear tests at room temperature, gouges are detected in the wear tracks. Besides the wide and flat gouge in the wear track, narrow and sharp gouges are also observed. The top wide gouge is related to the removal of the composites in sliding against the coupled ball, however, the low narrow sharp gouges are wear plough tracks in sliding against the debris. As temperature increases to 400 ◦ C, extrusive tribofilms, besides the gouges, are also detected. As depicted in Fig. 6(e) and (f), the surface profiles are noisy at 700 ◦ C, in which the wear tracks are too shallow to make out from the near unworn surfaces. There are no significant variations in surface roughness level in Fig. 6(e) and (f) across the wear tracks. The noisy profiles of the unworn surfaces are also correlated to the formation of the oxidized products. Therefore, a semi-quantitative estimation was carried out on BN21 and BN11 after tested at room temperature and 400 ◦ C. Only the gouges were taken into account for calculating the volume loss of the wear tracks. Fig. 7 summarizes the specific wear rates of BN21, BN11 and the coupled Al2 O3 balls. Obviously, all the tribo-couples exhibit

Fig. 5. XRD patterns of worn surfaces of TiB2 –TiN composites after wear tests at different temperatures: (a) room temperature, (b) 400 ◦ C and (c) 700 ◦ C.

lower wear rates at 400 ◦ C as contrasted with those obtained at room temperature. Al2 O3 ball has higher wear rates than the composites due to its relatively low hardness. As listed in Table 1, BN21 possesses higher hardness, comparable fracture toughness, but lower flexure strength as contrasted with BN11. In consideration of the higher hardness of BN21, more Al2 O3 ball is removed in sliding against BN21. Simultaneously, wear debris of BN21 generates more easily due to its lower flexure strength. Therefore, serious wear occurs between BN21/Al2 O3 tribopairs, as depicted in Fig. 7. At room temperature, wear rates for both composites increase with the applied load; however, they are insensitive to applied load at 400 ◦ C. In contrast with the data of other TiB2 -containing composites, the TiB2 –TiN composites have comparable or lower wear rates, which range in the order of 10−6 –10−7 mm3 /Nm. The wear rates of Al2 O3 –TiB2 –SiCw composites in dry sliding at room temperature are in the order of 10−7 mm3 /Nm [17]. O.O. Ajayi et al. [15] reported the wear rates of ␣SiC–TiB2 composites in the order of 10−6 mm3 /Nm, however, higher values were also reported in the order of 10−5 mm3 /Nm by Blanc et al. [4]. In light of the shallow wear profiles in Fig. 6(e)

1970

J.H. Ouyang et al. / Wear 271 (2011) 1966–1973

Fig. 6. Two-dimensional profiles of the wear tracks of BN21 and BN11 after wear tests against alumina at 20 N load and different temperatures: (a, c, e) BN21; (b, d, f) BN11.

and (f), the wear rates of TiB2 –TiN composites are obviously low at 700 ◦ C. 3.3. Microscopic observations of worn surfaces In order to understand the wear mechanisms, observations of worn surfaces on the flats and the coupled balls were carried out by SEM. Under all test conditions, BN21 and BN11 show similar wear character. For brevity, only worn surfaces of BN21/Al2 O3 tribopairs are discussed in details. Fig. 8 shows the worn surfaces of BN21 flat and the coupled ball after wear tests at room temperature. Similar worn surfaces characterized by microfracture and spalling of the fractured grains were observed after tested at 10 N and 20 N loads. At 10 N load,

10

BN21-10N

-6

3

Wear rate (x 10 mm /Nm)

9

BN11-10N

8

BN21-20N

7

BN11-20N

6

Ball(21)-20N

5

Ball(11)-20N

4 3 2 1 0

room temperature

400 ºC

Fig. 7. Average specific wear rates of TiB2 –TiN composites and the coupled Al2 O3 balls after wear tests at different loads and testing temperatures.

dispersive debris are observed to reattach on the surface near the microfracture region. Increasing the applied load to 20 N, microfracture of BN21 flat becomes severe on tribo-contact surface. Besides, continuous transfer layer instead of dispersive debris forms near the microfracture region, as marked in Fig. 8(b). A wear scar with a radius of 700 ␮m is observed on the coupled ball, as shown in Fig. 8(d). Fracture and removal of Al2 O3 grains as well as the formation of some transfer layers are observed in the wear scar. Compositional analysis on such transfer layers in both flat and coupled ball were carried out by EDS. According to the EDS spectra depicted in Fig. 8(c) and (f), Ti, B, Al, O elements are detected in the flat, while Al, O and trace of Ti are detected in the coupled ball. Therefore, it is reasonable to assume such transfer layers to originate from agglomerated TiB2 and Al2 O3 wear debris. Due to the relatively low hardness of Al2 O3 , more alumina debris are removed from the coupled ball, and transferred to the composite flat. During reciprocating sliding, the generated debris is crushed into small grains, and is adhered to the tribocontact region tightly, which protect the tribo-pairs from serious abrasion. Typical SEM micrographs of worn surfaces on the BN21 flat and the coupled ball at 20 N load and 400 ◦ C are presented in Fig. 9. A characteristic strip of extrusive tribofilm parallel to the sliding direction is observed at the edge of the wear track, which is consistent with the profile plotted in Fig. 6(c). As presented in Fig. 9(b), the magnified image of the adherent tribofilm shows a relatively dense appearance. Some shallow craters as well as holes are located on the tribofilm. According to the EDS results, as shown in Fig. 9(d), Ti, Al and O elements are detected in a compositional proportion of 25:18:57, however, no B and N elements are detected. It suggests that the tribofilm is composed of alumina worn debris and oxidized products of BN21. Besides, presence of shallow craters is common in the smooth wear track, as shown in Fig. 9(c). As con-

J.H. Ouyang et al. / Wear 271 (2011) 1966–1973

1971

Fig. 8. Worn surfaces of the flat and the coupled ball at room temperature: (a) BN21 flat at 10 N load; (b) BN21 flat at 20 N load; (c) the corresponding EDS spectrum of the marked position in (b); (d) the coupled ball at 20 N load; (e) magnified image of the marked position A in (d); (f) the corresponding EDS spectrum of the marked position in (e).

trasted with the coupled ball at room temperature, a small wear scar with a radius of 500 ␮m is observed on the coupled ball after wear test at 400 ◦ C. The wear scar is characterised by removal of fractured alumina grains, as well as the formation of small dispersive debris and dense transfer layers. EDS analysis confirms that the transfer layers formed at 400 ◦ C are composed of crushed alumina debris. The formation of tribofilm can effectively reduce the

wear loss and enhance the wear resistances of TiB2 –TiN composites and the coupled balls. In combination of the evolution of friction with microstructural observations of worn surfaces, a mild abrasive wear is the main wear mechanism of TiB2 –TiN composites at 400 ◦ C. As thermal oxidation occurs in the surface layer of TiN–TiB2 composites at 700 ◦ C, the formation of lubricious oxidized products

Fig. 9. Worn surfaces of the flat and the coupled ball at 400 ◦ C and 20 N load: (a) BN21flat; (b) magnified image of the transfer layer marked as A in (a); (c) magnified image of the position marked as B in (a); (d) the corresponding EDS spectrum of the marked position in (b); (e) the coupled ball; (f) magnified image of the wear scar in (e) (double-headed arrow marks indicate the sliding direction).

1972

J.H. Ouyang et al. / Wear 271 (2011) 1966–1973

Fig. 10. Worn surfaces of BN21 and the coupled ball at 700 ◦ C and 20 N load: (a) worn surface of BN21; (b) magnified image of worn surface marked in (a); (c) the corresponding EDS spectrum of the marked region in (b); (d) worn surface of the coupled ball (the double-headed arrows indicate the sliding direction, while the dotted line separates the worn (right) and unworn (left) surfaces).

at the tribo-contact regions will influence the friction and wear of the tribo-couples. Fig. 10 shows SEM micrographs and EDS analysis of worn surfaces tested at 700 ◦ C and 20 N load. The worn surface of BN21 flat, as shown at the right side to the dotted line, exhibits a relatively smooth appearance, however, dispersive oxidized products are found on the surface. As shown in Fig. 10(b), the worn region is covered by continuous nanosized oxides. These oxidized products form rapidly enough to resist wiping and provide lubricity, which is supported by the neglectable wear of the coupled ball, as presented in Fig. 10(d). Shallow scratchings are also observed on the surface of oxidized layer. Mild abrasion as well as oxidative wear is responsible for the wear of TiN–TiB2 composites at 700 ◦ C. 4. Conclusions (1) The reactive hot-pressed TiB2 –TiN composites have distinctly different friction characteristic in sliding against alumina ball at room temperature, 400 ◦ C and 700 ◦ C. Break-in period in friction is only observed at room temperature. The TiN–TiB2 ceramic exhibits a distinct decrease in friction coefficient at 700 ◦ C as contrasted with the friction data obtained at room temperature and 400 ◦ C. (2) The reactive hot-pressed TiB2 –TiN composites exhibit excellent wear resistance. Their wear rates, ranging in the order of 10−6 –10−7 mm3 /Nm, decrease with increasing temperature. (3) Wear mechanisms of TiN–TiB2 composites depend mainly upon testing temperature at identical applied loads. Lubricious oxidized products formed by thermal oxidation provide excellent lubrication at 700 ◦ C. However, mild abrasive wear and tribooxidation are the dominated wear mechanisms of TiB2 –TiN composites at 400 ◦ C. Spalling of micro-fractured grains and mild abrasive scratches play an important role during roomtemperature wear test.

Acknowledgements The authors would like to thank financial supports from the National Natural Science Foundation of China (NSFC-Nos. 50572020 and 51021002), the National 863 High-Tech Project (2006AA03Z537), and the Program for New Century Excellent Talents in University (NCET-06-0339). References [1] J.R. Ramberg, C.F. Wolfe, W.S. Williams, Resistance of titanium diboride to hightemperature plastic yielding, J. Am. Ceram. Soc. 68 (1985) 78–79. [2] R.G. Munro, Material properties of titanium diboride, J. Res. Natl. Inst. Stand. Technol. 105 (2000) 709–720. [3] G.B. Raju, B. Basu, Wear mechanisms of TiB2 and TiB2 –TiSi2 at fretting contacts with steel and WC–6 wt% Co, Int. J. Appl. Ceram. Technol. 7 (2010) 89–103. [4] C. Blanc, F. Thebenot, D. Treheux, Wear resistance of ␣SiC–TiB2 composites prepared by reactive sintering, J. Eur. Ceram. Soc. 19 (1999) 571–579. [5] K. Shobu, T. Watanabe, Frictional properties of sintered TiN–TiB2 and Ti(CN)–TiB2 ceramics at high temperature, J. Am. Ceram. Soc. 70 (1987) C103–C-104. [6] R. Tomoshige, A. Murayama, T. Matsushita, Production of TiB2 –TiN composites by combustion synthesis and their properties, J. Am. Ceram. Soc. 80 (1997) 761–764. [7] J.W. Lee, Z.A. Munir, Synthesis of dense TiB2 –TiN nanocrystalline composites through mechanical and field activation, J. Am. Ceram. Soc. 84 (2001) 1209–1216. [8] C.F. Hu, Y.C. Zhou, Y.W. Bao, D.T. Wan, Tribological properties of polycrystalline Ti3 SiC2 and Al2 O3 -reinforced Ti3 SiC2 composites, J. Am. Ceram. Soc. 89 (2006) 3456–3461. [9] K. Kato, Tribology of ceramics, Wear 136 (1990) 117–133. [10] A. Tampieri, A. Bellosi, Oxidation of monolithic TiB2 and of Al2 O3 –TiB2 composite, J. Mater. Sci. 28 (1993) 649–653. [11] H.-Y. Chen, Oxidation behavior of titanium nitride films, J. Vac. Sci. Technol. 23 (2005) 1006–1009. [12] Z.L. Yang, J.H. Ouyang, Z.G. Liu, Isothermal oxidation behaviour of reactive hotpressed TiN–TiB2 ceramics at elevated temperatures, Mater. Des. 32 (2011) 29–35. [13] N. Fateh, G.A. Fontalvo, G. Gassner, C. Mitterer, Influence of high-temperature oxide formation on the tribological behavior of TiN and VN coatings, Wear 262 (2007) 1152–1158.

J.H. Ouyang et al. / Wear 271 (2011) 1966–1973 [14] T. Senda, Y. Yamamoto, Y. Ochi, Friction and wear test of titanium boride ceramics at elevated temperatures, J. Ceram. Soc. Jpn. 101 (1993) 461–465. [15] O.O. Ajayi, A. Erdemir, R.H. Lee, F.A. Nichols, Sliding wear of silicon carbide–titanium diboride ceramic–matrix composite, J. Am. Ceram. Soc. 76 (1993) 511–517.

1973

[16] S. Gupta, D. Filimonov, V. Zaitsev, T. Palanisamy, M.W. Barsoum, Ambient and 550 ◦ C tribological behavior of select MAX phases against Ni-based superalloys, Wear 264 (2008) 270–278. [17] J.X. Deng, Friction and wear behaviour of Al2 O3 /TiB2 /SiCw ceramic composites at temperatures up to 800 ◦ C, Ceram. Int. 27 (2001) 135–141.