- Email: [email protected]
Highly toughened dense TiC–Ni composite by in situ decomposition of (Ti,Ni)C solid solution
Available online at www.sciencedirect.com
CERAMICS INTERNATIONAL
Ceramics International 41 (2015) 4656–4661 www.elsevier.com/locate/ceramint
Highly toughened dense TiC–Ni composite by in situ decomposition of (Ti,Ni)C solid solution Hanjung Kwona, Sun-A. Junga, Chang-Yul Suha, Ki-Min Roha, Wonbaek Kima, Jiwoong Kima,b,n a
Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources, Yuseong-gu, Daejeon 305-350, Republic of Korea b Korea University of Science and Technology, Gajeong-ro, Yuseong-gu, Daejeon, Republic of Korea Received 17 September 2014; received in revised form 29 October 2014; accepted 2 December 2014 Available online 10 December 2014
Abstract A highly-toughened TiC–Ni composite was produced by the in situ decomposition during sintering of a nonequilibrim (Ti,Ni)C solid-solution phase. The solid solution was synthesized by the mechanical milling of Ti–Ni alloy/graphite mixtures, which then decomposed to finely dispersed TiC and Ni upon heating in a vacuum. To take advantage of the fine microstructure, samples were sintered to facilitate the in situ decomposition. Densification of the (Ti,Ni)C during the sintering was achieved through the coalescence of fine particles and the concurrent decomposition of the nonequilibrium (Ti,Ni)C phase. Additional densification was obtained through liquid-phase sintering by the eutectic melting of Ni. The fracture toughness of the sintered TiC–Ni composite was notably higher than that of conventional TiC/Ti(CN)–Ni or of comparable Ti(CN)–WC–Ni composites. Its fine and dense microstructure is believed to account for the enhanced toughness. The method suggested here might represent a valuable option for the preparation of TiC–Ni composites with desirable mechanical properties. & 2014 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: B. Composites; C. Hardness; D. Carbides; E. Cutting tools
1. Introduction TiC/Ti(CN)–Ni composites are widely used in cutting tools owing to their high hardness, wear resistance, chemical stability, and low density. They are generally produced by the liquid-phase sintering of a mixture of TiC/Ti(CN) and Ni powders. Nevertheless, the formation of a dense TiC/Ti(CN)– Ni composite by liquid-phase sintering is generally difficult due to the relatively poor wettability of Ni on a TiC/Ti(CN) surface. (The wetting angles are 301 and 171 in a vacuum and in hydrogen, respectively.) The poor wettability is attributed to the formation of surface oxide; therefore, the elimination of pores by the penetration of liquid Ni into the TiC/Ti(CN)–Ni n Corresponding author at: Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources, Yuseong-gu, Daejeon 305-350, Republic of Korea. Tel.: þ 82 42 868 3927; fax: þ82 42 868 3415. E-mail addresses: [email protected] (H. Kwon), [email protected] (S.-A. Jung), [email protected] (C.-Y. Suh), [email protected] (K.-M. Roh), [email protected] (W. Kim), [email protected] (J. Kim).
http://dx.doi.org/10.1016/j.ceramint.2014.12.011 0272-8842/& 2014 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
system becomes difficult [1–3]. In some cases, Mo2C is added to improve the wettability of Ni on the TiC/Ti(CN). The full densification of TiC/Ti(CN)–Ni composites requires their sintering at above 1450 1C [4–6]. This difficulty in sintering often limits the applicability of the TiC/Ti(CN)–Ni composites in cutting tools despite Ti and Ni being less expensive than the more commonly used W and Co. Mechanical alloying is a useful method for the synthesis of various nonequilibrium phases such as supersaturated solid solutions, metastable crystalline and quasicrystalline phases, nanostructures, and amorphous alloys [7]. It has widely been used to fabricate supersaturated binary solid-solution such as Zr–Al, Fe–Cu, Al–Mn, Al–Fe, Al–Si, Al–Ge, and Ni–Ag, which exhibit very limited mutual solubility [8–15]. Such supersaturated solid solutions decompose exothermically to thermodynamically stable phases with sufficient activation energy. Beyond the binary alloys, the ternary solid-solution phase of the constituents of the TiC–Ni composite (i.e., Ti, Ni, and C) does not exist at equilibrium [16–18]. Therefore, if a nonequilibrium solid-solution phase of Ti, Ni, and C
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
(i.e., (Ti,Ni)C) could be synthesized by mechanical alloying, it would spontaneously decompose into TiC and Ni upon proper heat treatment. This work reports a new method to produce a dense TiC–Ni composite via the decomposition of a nonequilibrium (Ti,Ni)C phase. A (Ti,Ni)C phase, the nonequilibrium solid-solution phase of Ti–Ni–C, was initially synthesized by the mechanical alloying of a Ti–Ni alloy/graphite mixture. The sintering behavior of (Ti, Ni)C powders was investigated in terms of the shrinkage rate and microstructure evolution and compared with that of Ti(CN)–Ni prepared by conventional mixing. The mechanical properties of the TiC–Ni composites prepared by the sintering of (Ti,Ni)C powders were compared with previously reported properties of TiC–Ni and Ti(CN)–WC–Ni composites.
4657
(FE-SEM) (MLA650F, FEI, Oregon, USA) and field-emission transmission electron microscopy (FE-TEM) (JEM-2100 F, JEOL, Tokyo, Japan). Vickers hardness was measured with an indenter load of 30 kg, and fracture toughness was calculated using the expression derived by Shetty et al. [19]. The sintering behavior and the microstructure of the (Ti,Ni)C compact were compared with those of a Ti(CN)–Ni sample prepared by the conventional mixing of powders of Ti(CN) (molar ratio of C:N¼ 1:1; purity499%; average particle size: 1 μm) and Ni (purity499%; average particle size: 4.2 μm).
3. Results and discussion 3.1. Preparation of the nonequilibrium (Ti,Ni)C phase
2. Experimental procedure Nonequilibrium solid-solution (Ti,Ni)C powders were produced using powders of a Ti–Ni alloy (Ti:Ni weight ratio=7:3; purity 4 99%; particle size o 75 μm) and of graphite (purity 4 99%; average particle size: 7–10 μm; Alfa Aesar). The Ti–Ni alloy powders were obtained by the hydrogenation and dehydrogenation of a Ti–Ni alloy ingot, which was prepared by the arc melting of Ti and Ni (purity 4 99.5%; average size: 3–20 mm; RNDKOREA) under an Ar atmosphere. The Ti–Ni alloy was transferred into a vacuum furnace for hydrogenation at a temperature of 873 K. Hydrogen gas (99.9999%) was introduced into the furnace to maintain a pressure between 5 105 and 7 105 Pa for 2 h. The hydrogenated Ti–Ni alloy was mechanically crushed into a size of o 75 μm. To remove the hydrogen, the Ti–Ni alloy powder was annealed at 973 K for 2 h in vacuum. The amount of graphite was adjusted to match the molar ratio of Ti in the Ti–Ni alloy, thus allowing TiC to form when the (Ti,Ni)C decomposed into TiC and Ni. The powders were then subjected to high-energy milling using a planetary mill (Pulverisette 5, Fritsch, Germany). Tungsten carbide balls were mixed with the Ti–Ni and graphite powders in a ball-to-powder weight ratio of 40:1. The powders were milled in a stainless steel bowl at 250 rpm in an argon atmosphere for 20 h. The phases of the milled powders were analyzed using X-ray diffraction (XRD) (SmartLab, Rigaku, Japan) with monochromated Cu Kα radiation (λ=1.5418 Å), which employed Si as a calibration standard. Differential scanning calorimetry (DSC) (SETSYS Evolution, SETARAM, France) was carried out in an Ar flow to analyze the reactions that occurred during the heat treatment of the (Ti,Ni)C phase. Heating rate of 5 1C/min was applied to all samples for the DSC analysis. The chemical composition was determined by inductively coupled plasma atomic emission spectrometry (ICP-AES, OPTIMA 5300DV, PerkinElmer) and CNO analysis (TCH-600, CS-600, LECO, Michigan, USA). The (Ti,Ni)C powder was then compacted at 125 MPa into discs. The sintering behavior of the (Ti,Ni)C compacts was monitored using a dilatometer (DIL 420 PC, Netzsch Geraetebau GmbH, Selb, Germany). The compacts were sintered at 1250 1C and 1330 1C for 5 min and 30 min. Their microstructures were examined using field-emission scanning electron microscopy
The nonequilibrium (Ti,Ni)C phase was synthesized by highenergy milling. Fig. 1 shows X-ray diffraction patterns of the phases evolved during the high-energy milling and the subsequent heat treatment of a mixture of Ti–Ni alloy and C. The as received Ti–Ni alloy was composed of Ti2Ni and pure Ti (Fig. 1 (a)). The sample milled for 20 h revealed only a B1-structured phase (NaCl-like structure), suggesting that the reaction for carbide formation had run to completion (Fig. 1(b)). It was found that the amount of contaminants introduced from the jar and balls were not much (Table 1). Milling for 20 h led to the formation of a (Ti,Ni)C phase, as evidenced by its lattice parameter being smaller than that of TiC, thus indicating the participation of Ni in the reaction (Table 2). Fig. 1(c) shows the XRD pattern of the (Ti, Ni)C powder after heat treatment at 1200 1C. Only TiC and Ni peaks are present, suggesting that the (Ti,Ni)C phase decomposed to TiC and Ni during the heat treatment. The lattice parameter of the B1-structured phase after the heat treatment was larger than that of the milled powder. The extraction of Ni from the (Ti,Ni)C phase increased the lattice parameter, because the atomic size of Ni is smaller than that of Ti. The lattice parameter of the heattreated B1-structured phase was smaller than that of the TiC phase (i.e., 4.3274 Å). This may be explained by the formation of
Fig. 1. X-ray diffraction patterns of (a) Ti–Ni raw material, (b) synthesized (Ti,Ni)C powder, and (c) (Ti,Ni)C powder heat treated for 2 h at 1200 1C.
4658
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
Table 1 Elemental analysis of the milled powdera. Element
Ti
Ni
W
Fe
Cr
Content (wt%)
70(2)
29(2)
0.40(6)
0.42(1)
–
a
Standard deviation in parentheses.
Table 2 Lattice parameters of the B1-structured phase and the carbon/nitrogen/oxygen content of the powders and the sintered bodiesa. Powder
Sintered body
Milled Heat- 1250 1C, treated 5 min
1330 1C, 5 min
1330 1C, 30 min
Lattice parameter of B1-structured phase (Å) Ti(CN)–Ni – – 4.28568(3) 4.28920(5) 4.28943(5) (Ti,Ni)C 4.2540 4.2755 4.32682(3) 4.3325(1) 4.3458(6) (5) (2) C/N/O content (wt%) Ti(CN)–Ni Carbon Nitrogen Oxygen (Ti,Ni)C Carbon
– – – 13.4 (2) Nitrogen 0.99 (3) Oxygen 3.52 (3)
Theoretical density (g/cm3) Ti(CN)–Ni – (Ti,Ni)C – a
– – – 12.8 (2) 0.73 (1) 1.04 (1)
– –
7.74(3) 11.79(7) 0.271(4) 12.5(1)
7.84(2) 11.7(3) 0.133(7) 12.6(2)
7.84(3) 12.2(1) 0.134(7) 12.5(1)
0.2704(4)
0.27(2)
0.29(2)
0.99(2)
0.83(9)
0.95(8)
5.9599 5.8720
5.9408 5.8501
5.9783 5.8111
Standard deviation in parentheses.
a carbon-deficient carbide phase owing to the reaction of oxygen with carbon. 3.2. Thermal decomposition of the nonequilibrium (Ti,Ni)C phase Our previous thermodynamic research confirmed that the nonequilibrium (Ti,Ni)C phase decomposes to TiC and Ni upon heating [20]. DSC was used here to examine in detail the thermal decomposition of the phase during heat treatment. The DSC curve for (Ti,Ni)C shows a continuous decrease from the initial analysis temperature (i.e., 600 1C) to 1200 1C, which corresponds to the endothermic reaction of carbon shown in Fig. 2(a). This agrees with the observed decreases of carbon and oxygen contents of the powder after the heat treatment (Table 2). The rate of decrease of the amount of heat generated during the carbothermal reduction reaction reduced as the temperature reached approximately 900 1C. This is related to the onset of the decomposition of the (Ti,Ni)C phase due to the exothermic nature of the phase decomposition reaction. The amount of heat generated during the carbothermal reduction reaction reached a minimum at 1200 1C, implying that the (Ti,Ni)C phase was not carbothermally reduced at higher
Fig. 2. DSC curves for (a) (Ti,Ni)C and (b) Ti(CN)–Ni powders.
temperatures. This was further substantiated by the lowering of the oxygen content of the powders (from 3.52 wt% to 1.04 wt%) observed after heat treatment at 1200 1C (Table 2). The DSC thermogram for (Ti,Ni)C shows a minimum at 1325 1C due to the endothermic melting of Ni in the (Ti,Ni)C phase. Note that the Ni melted below the melting point of pure Ni (1455 1C). As shown in the DSC thermogram of Fig. 2(b), a lower melting temperature of Ni (1360 1C) was also found in the case of Ti(CN)–Ni prepared by the mixing of Ti(CN) and Ni (Fig. 2(b)). This observation agrees with previously reported eutectic melting between Ni and TiC/Ti (CN) [21]. 3.3. Sintering of the (Ti,Ni)C phase The sintering behavior of (Ti,Ni)C and Ti(CN)–Ni compacts was investigated by measuring their linear shrinkage rate using a dilatometer (Fig. 3). At all temperatures the (Ti,Ni)C compact showed a higher linear shrinkage rate than the Ti(CN)–Ni compact, which may have been due to the (Ti,Ni)C particles (below 1 μm) being considerably finer than the Ti(CN) particles ( 1 μm) (Fig. 4), because smaller particles lead to larger surface areas and thus a higher driving force for their coalescence [22]. Unexpectedly, however, the (Ti,Ni)C compact stopped shrinking and began to expand rapidly at 1300 1C, while the Ti(CN)–Ni compact continued shrinking. The expansion implies the melting of the Ni in the (Ti,Ni)C compact, because the pores in the compact had already been eliminated by the solid-state sintering at 1300 1C. By comparison, the shrinkage of the Ti(CN)–Ni compact had not finished even at 1400 1C, a temperature higher than the starting point of Ni melting (1330 1C), because pores in the Ti(CN)–Ni compact had not been removed completely. As a result, Ni melting is expected to fill the pores in the compact at a temperature higher than the melting point of Ni, implying that full densification requires the sintering of Ti(CN)–Ni at above the melting point of Ni. Fig. 5 shows SEM micrographs showing backscattered electron images of the (Ti,Ni)C and Ti(CN)–Ni compacts sintered at 1250 1C and 1330 1C for 5 min. The particles of the (Ti,Ni)C
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
compact coalesced to become bigger after sintering at 1250 1C, while those of the Ti(CN)–Ni compact remained almost unchanged (Ti(CN): 1 μm and Ni: 4.2 μm) after similar sintering. Given that the sintering at 1250 1C was below the melting temperature of Ni (1325 1C for (Ti,Ni)C and 1360 1C for Ti(CN)–Ni), the densification was expected to occur through solid-state sintering. This suggests that the solid-state sintering of the (Ti,Ni)C compact occurred by the coalescence of the particles before the melting of the Ni. By contrast, the unchanging particle size of the Ti(CN)–Ni compact during sintering suggests that it became denser mainly by the rearrangement of particles. Another important observation concerning the microstructure of the (Ti,Ni)C compact sintered at 1250 1C is that Ni had already incorporated into the spaces between the particles before it melted. This occurred by the decomposition reaction suggested by the DSC result. The melting of Ni after the solid-state sintering affected significantly the microstructure of both samples. Fig. 5(c) and (d) show backscattered electron images of the (Ti,Ni)C and Ti (CN)–Ni compacts after sintering at 1330 1C. Liquid Ni is depicted as bright areas in the figures filling the gaps between darker carbide particles. This observation suggests that Ni had melted at 1330 1C by a eutectic reaction, which is further substantiated by the amount of carbon in the Ni matrix of the sintered bodies measured by energy dispersive X-ray spectrometry analysis (Fig. 6). The carbon content of the Ni in the sintered (Ti,Ni)C was 4.77 wt%; that in the Ti(CN)–Ni was 3.71 wt%. Both values are higher than the eutectic composition
4659
of 2.42 wt% C in the Ni–C system [23]. These results suggest that Ni in the (Ti,Ni)C and Ti(CN)–Ni compacts melted at about the eutectic point (1326 1C). The relative densities of the sintered bodies were calculated as follows and listed in Table 2: Theoretical densityðg=cm3 Þ n o in unit cellatomic weight Sum of atomic ratio Avogadro's number cell volume ¼ Unit where the atomic ratios of Ti, Ni, C, N, and O were obtained by inductively coupled plasma atomic emission spectroscopy and C/N/O analysis, and unit cell volume was measured by XRD. Fig. 7 shows the relative densities of the (Ti,Ni)C and Ti(CN)– Ni compacts after sintering at 1250 1C and at 1330 1C. The relative densities of the (Ti,Ni)C compacts were consistently significantly higher than those of the Ti(CN)–Ni compacts at each sintering condition. Both compacts showed a much greater increase in density after the hotter sintering. This suggests that pores in the compacts were filled with Ni, which melted by a eutectic reaction at 1326 1C. In summary, it can be concluded that the densification of the TiC–Ni composite was achieved here not only by particle coalescence and the in situ decomposition of the (Ti,Ni)C phase during solid-state sintering but also by the filling of pores with liquid Ni. 3.4. Mechanical properties of TiC–Ni composites
Fig. 3. Linear shrinkage rates of (Ti,Ni)C and Ti(CN)–Ni compacts.
The dense TiC–Ni composite revealed excellent mechanical properties, particularly toughness. The fracture toughness of TiC/Ti(CN)–Ni composites (6–7 MPa m1/2) is generally lower than that of WC–Co composites (10–12 MPa m1/2) [1,24]. The shorter lifetimes of the generally weaker components of TiC/Ti (CN)–Ni composites limit their broad applicability and generally leads to the addition of WC to improve their toughness for commercial use [25–28]. The fracture toughness of the TiC–Ni composite of this study (10–11 MPa m1/2) compares favorably with previously reported values for similar composites (6.8–7.4 MPa m1/2, Table 3) [24]. Furthermore, it is comparable to the highest values reported for Ti(CN)–WC–Ni composites (10–12 MPa m1/2). Therefore, the synthesis method presented here is shown effectively to produce a highly toughened TiC–Ni composite without needing the addition
Fig. 4. SEM micrographs of (a) (Ti,Ni)C and (b) Ti(CN)–Ni powders.
4660
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
Fig. 5. FE-SEM micrographs of (a) (Ti,Ni)C and (b) Ti(CN)–Ni sintered at 1250 1C for 5 min, (c) (Ti,Ni)C and (d) Ti(CN)–Ni sintered at 1330 1C for 5 min.
Fig. 6. TEM micrographs of (a) (Ti,Ni)C compact and (b) Ti(CN)–Ni compact sintered at 1330 1C for 5 min.
of WC. The composite's dense and finely dispersed microstructure —which occurred owing to the in situ decomposition of (Ti,Ni)C —led to its enhanced toughness. While this work demonstrates the fabrication of a tough new TiC–Ni composite, future work is required to elucidate in detail the mechanism of its formation. The uniform distribution of fine Ni and TiC particles that resulted from phase separation and relatively cool sintering at 1330 1C were possibly responsible for the enhanced toughness of the composite prepared from the (Ti,Ni)C phase. In any case, the suggested method would provide a valuable option for the preparation of TiC–Ni suitable to replace the commercial Ti(CN)–WC–Ni and similar composites that are currently used.
4. Conclusions A method to produce a dense TiC–Ni composite is proposed here. Such composites can be difficult to sinter due to the poor wettability of Ni on TiC. However, sintering was facilitated here through the synthesis of a (Ti,Ni)C phase—which was unstable at equilibrium—by the high-energy milling of a Ti–Ni alloy/graphite mixture. The (Ti,Ni)C compact was almost completely sintered at 1330 1C, below the temperature of conventional sintering. Enhanced densification occurred through solid-state sintering due to particle coalescence and in situ phase decomposition before the Ni melted. The solid-state sintering of the Ti(CN)–Ni compact
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
Fig. 7. Relative densities of (Ti,Ni)C and Ti(CN)–Ni after sintering.
Table 3 Mechanical properties of the TiC–Ni composites along with those of other TiC/ Ti(CN)–Ni based composites. Composite
HV (GPa)
KIC (MPa m1/2)
TiC–Ni (sintered at 1330 1C, for 5 min) TiC–Ni (sintered at 1330 1C, for 30 min) Ti(CN)–Ni (sintered at 1330 1C, for 30 min) TiC–Ni (20 30 wt% of Ni) [24] Ti(CN) (0.7 μm)–WC (0.4 μm)–Ni (4.1 μm) [25] Ti(CN) (0.7 μm)–WC (0.4 μm)–Ni (0.01 μm) [26] Ti(CN) (0.3 μm)–WC (0.2 μm)–Ni (4.1 μm) [27] Ti(CN) (0.08 μm)–WC (0.2 μm)–Ni (0.08 μm) [28]
13.5770.15 13.0070.11 9.677 0.32 13.239.21 13.6–14.2
10.1070.05 11.2270.15 – 6.8–7.4 7.25–8.8
12.5–15.3
8.4–9.1
14–15
8–10
12–13
10–12
occurred simply by the rearrangement of particles without their size changing. Further densification was achieved by the filling of pores with liquid Ni formed by a eutectic reaction. The fracture toughness of the TiC–Ni composite (10–11 MPa m1/2) was significantly higher than values previously reported for TiC–Ni composites. The fine and dense microstructure that originated from the phase decomposition was responsible for the enhanced toughness of the composite. A dense TiC–Ni compact with decent mechanical properties was demonstrated to be fabricated by the manipulation of the in situ decomposition of a nonequilibrium (Ti,Ni)C solid-solution phase. Acknowledgment This work was supported by a Grant-in-Aid from the Basic Research Project of the Korea Institute of Geoscience and Mineral Resources (KIGAM), funded by the Ministry of Science, ICT and Future Planning (GP2012-019). References [1] P. Ettmayer, Hardmetals and cermets, Annu. Rev. Mater. Sci. 19 (1989) 145–164.
4661
[2] N.M. Parikh, M. Humenik, Cermets: II, wettability and microstructure studies in liquid-phase sintering, J. Am. Ceram. Soc. 40 (1957) 315–320. [3] D.J. Miller, J.A. Pask, Liquid-phase sintering of TiC–Ni composites, J. Am. Ceram. Soc. 66 (1983) 841–846. [4] H. Suzuki, K. Hayashi, O. Terada, Two-phase region in Tic–Mo-30% Ni alloys, J Jpn. Inst Met. 35 (1971) 146–150. [5] S.Y. Ahn, S.W. Kim, S. Kang, Microstructure of Ti (CN)–WC–NbC–Ni cermets, J. Am. Ceram. Soc. 84 (2001) 843–849. [6] S. Kim, K.-H. Min, S. Kang, Rim structure in Ti(C0.7N0.3)–WC–Ni cermets, J. Am. Ceram. Soc. 86 (2003) 1761–1766. [7] C. Suryanarayana, Mechanical alloying and milling, Prog. Mater. Sci. 46 (2001) 1–184. [8] H. Fecht, G. Han, Z. Fu, W. Johnson, Metastable phase formation in the Zr–Al binary system induced by mechanical alloying, J. Appl. Phys. 67 (1990) 1744–1748. [9] C. Suryanarayana, R. Sundaresan, Metastable phases in mechanically alloyed Al Mn powder mixtures, Mater. Sci. Eng.: A 131 (1991) 237–242. [10] E. Ma, M. Atzmon, F. Pinkerton, Thermodynamic and magnetic properties of metastable FexCu100 x solid solutions formed by mechanical alloying, J. Appl. Phys. 74 (1993) 955–962. [11] J. Xu, U. Herr, T. Klassen, R. Averback, Formation of supersaturated solid solutions in the immiscible Ni–Ag system by mechanical alloying, J. Appl. Phys. 79 (1996) 3935–3945. [12] K. Chattopadhyay, X.-M. Wang, K. Aoki, T. Masumoto, Metastable phase formation during mechanical alloying of AlGe and AlSi alloys, J. Alloy. Compd. 232 (1996) 224–231. [13] B.-L. Huang, R. Perez, E. Lavernia, M. Luton, Formation of supersaturated solid solutions by mechanical alloying, Nanostruct. Mater. 7 (1996) 67–79. [14] B. Huang, K. Ishihara, P. Shingu, Metastable phases of Al–Fe system by mechanical alloying, Mater. Sci. Eng.: A 231 (1997) 72–79. [15] C. Suryanarayana, E. Ivanov, V. Boldyrev, The science and technology of mechanical alloying, Mater. Sci. Eng.: A 304 (2001) 151–158. [16] Y. Du, J. Schuster, Experimental investigation and thermodynamic modeling of the Ni–Ti–C system, Z. Metallkd. 89 (1998) 399–410. [17] D. Bandyopadhyay, R. Sharma, N. Chakraborti, The Ti–Ni–C system (titanium–nickel–carbon), J. ph. equilib. 21 (2000) 186–191. [18] I. Isomäki, M. Hämäläinen, Thermodynamic re-evaluation of the C–Ni– Ti system, J. Alloy. Compd. 416 (2006) 120–124. [19] D. Shetty, I. Wright, P. Mincer, A. Clauer, Indentation fracture of WC–Co cermets, J. Mater. Sci. 20 (1985) 1873–1882. [20] H. Kwon, J. Kim, S.-A. Jung, C.-Y. Suh, D.-S. Kil, K.-M. Roh, W. Kim, J.-W. Lim, Premixed TiC–Ni composite powder prepared by mechanical milling of Ti–Ni alloy/graphite mixture and subsequent heat treatment, Powder Technol. 253 (2014) 681–685. [21] L. Chen, W. Lengauer, P. Ettmayer, K. Dreyer, H.W. Daub, D. Kassel, Fundamentals of liquid phase sintering for modern cermets and functionally graded cemented carbonitrides (FGCC), Int. J. Refract. Met. Hard Mater. 18 (2000) 307–322. [22] R.L. Coble, Sintering crystalline solids. I. intermediate and final state diffusion models, J. Appl. Phys. 32 (2004) 787–792. [23] B.-J. Lee, On the stability of Cr carbides, Calphad 16 (1992) 121–149. [24] X.-H. Zhang, J.-C. Han, S.-Y. Du, J. Wood, Microstructure and mechanical properties of TiC–Ni functionally graded materials by simultaneous combustion synthesis and compaction, J. Mater. Sci. 35 (2000) 1925–1930. [25] E. Jeon, J. Joardar, S. Kang, Microstructure and tribo-mechanical properties of ultrafine Ti (CN) cermets, Int. J. Refract. Met. Hard Mater. 20 (2002) 207–211. [26] J. Joardar, S. Kim, S. Kang, Effect of nanocrystalline binder on the microstructure and mechanical properties of ultrafine Ti (CN) cermets, Mater. Sci. Eng.: A 360 (2003) 385–389. [27] J. Jung, S. Kang, Effect of ultra-fine powders on the microstructure of Ti (CN)–xWC–Ni cermets, Acta Mater. 52 (2004) 1379–1386. [28] J. Jung, S. Kang, Effect of nano-size powders on the microstructure of Ti (C, N)–xWC–Ni ‐cermets, J. Am. Ceram. Soc. 90 (2007) 2178–2183.
CERAMICS INTERNATIONAL
Ceramics International 41 (2015) 4656–4661 www.elsevier.com/locate/ceramint
Highly toughened dense TiC–Ni composite by in situ decomposition of (Ti,Ni)C solid solution Hanjung Kwona, Sun-A. Junga, Chang-Yul Suha, Ki-Min Roha, Wonbaek Kima, Jiwoong Kima,b,n a
Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources, Yuseong-gu, Daejeon 305-350, Republic of Korea b Korea University of Science and Technology, Gajeong-ro, Yuseong-gu, Daejeon, Republic of Korea Received 17 September 2014; received in revised form 29 October 2014; accepted 2 December 2014 Available online 10 December 2014
Abstract A highly-toughened TiC–Ni composite was produced by the in situ decomposition during sintering of a nonequilibrim (Ti,Ni)C solid-solution phase. The solid solution was synthesized by the mechanical milling of Ti–Ni alloy/graphite mixtures, which then decomposed to finely dispersed TiC and Ni upon heating in a vacuum. To take advantage of the fine microstructure, samples were sintered to facilitate the in situ decomposition. Densification of the (Ti,Ni)C during the sintering was achieved through the coalescence of fine particles and the concurrent decomposition of the nonequilibrium (Ti,Ni)C phase. Additional densification was obtained through liquid-phase sintering by the eutectic melting of Ni. The fracture toughness of the sintered TiC–Ni composite was notably higher than that of conventional TiC/Ti(CN)–Ni or of comparable Ti(CN)–WC–Ni composites. Its fine and dense microstructure is believed to account for the enhanced toughness. The method suggested here might represent a valuable option for the preparation of TiC–Ni composites with desirable mechanical properties. & 2014 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
Keywords: B. Composites; C. Hardness; D. Carbides; E. Cutting tools
1. Introduction TiC/Ti(CN)–Ni composites are widely used in cutting tools owing to their high hardness, wear resistance, chemical stability, and low density. They are generally produced by the liquid-phase sintering of a mixture of TiC/Ti(CN) and Ni powders. Nevertheless, the formation of a dense TiC/Ti(CN)– Ni composite by liquid-phase sintering is generally difficult due to the relatively poor wettability of Ni on a TiC/Ti(CN) surface. (The wetting angles are 301 and 171 in a vacuum and in hydrogen, respectively.) The poor wettability is attributed to the formation of surface oxide; therefore, the elimination of pores by the penetration of liquid Ni into the TiC/Ti(CN)–Ni n Corresponding author at: Mineral Resources Research Division, Korea Institute of Geoscience and Mineral Resources, Yuseong-gu, Daejeon 305-350, Republic of Korea. Tel.: þ 82 42 868 3927; fax: þ82 42 868 3415. E-mail addresses: [email protected] (H. Kwon), [email protected] (S.-A. Jung), [email protected] (C.-Y. Suh), [email protected] (K.-M. Roh), [email protected] (W. Kim), [email protected] (J. Kim).
http://dx.doi.org/10.1016/j.ceramint.2014.12.011 0272-8842/& 2014 Elsevier Ltd and Techna Group S.r.l. All rights reserved.
system becomes difficult [1–3]. In some cases, Mo2C is added to improve the wettability of Ni on the TiC/Ti(CN). The full densification of TiC/Ti(CN)–Ni composites requires their sintering at above 1450 1C [4–6]. This difficulty in sintering often limits the applicability of the TiC/Ti(CN)–Ni composites in cutting tools despite Ti and Ni being less expensive than the more commonly used W and Co. Mechanical alloying is a useful method for the synthesis of various nonequilibrium phases such as supersaturated solid solutions, metastable crystalline and quasicrystalline phases, nanostructures, and amorphous alloys [7]. It has widely been used to fabricate supersaturated binary solid-solution such as Zr–Al, Fe–Cu, Al–Mn, Al–Fe, Al–Si, Al–Ge, and Ni–Ag, which exhibit very limited mutual solubility [8–15]. Such supersaturated solid solutions decompose exothermically to thermodynamically stable phases with sufficient activation energy. Beyond the binary alloys, the ternary solid-solution phase of the constituents of the TiC–Ni composite (i.e., Ti, Ni, and C) does not exist at equilibrium [16–18]. Therefore, if a nonequilibrium solid-solution phase of Ti, Ni, and C
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
(i.e., (Ti,Ni)C) could be synthesized by mechanical alloying, it would spontaneously decompose into TiC and Ni upon proper heat treatment. This work reports a new method to produce a dense TiC–Ni composite via the decomposition of a nonequilibrium (Ti,Ni)C phase. A (Ti,Ni)C phase, the nonequilibrium solid-solution phase of Ti–Ni–C, was initially synthesized by the mechanical alloying of a Ti–Ni alloy/graphite mixture. The sintering behavior of (Ti, Ni)C powders was investigated in terms of the shrinkage rate and microstructure evolution and compared with that of Ti(CN)–Ni prepared by conventional mixing. The mechanical properties of the TiC–Ni composites prepared by the sintering of (Ti,Ni)C powders were compared with previously reported properties of TiC–Ni and Ti(CN)–WC–Ni composites.
4657
(FE-SEM) (MLA650F, FEI, Oregon, USA) and field-emission transmission electron microscopy (FE-TEM) (JEM-2100 F, JEOL, Tokyo, Japan). Vickers hardness was measured with an indenter load of 30 kg, and fracture toughness was calculated using the expression derived by Shetty et al. [19]. The sintering behavior and the microstructure of the (Ti,Ni)C compact were compared with those of a Ti(CN)–Ni sample prepared by the conventional mixing of powders of Ti(CN) (molar ratio of C:N¼ 1:1; purity499%; average particle size: 1 μm) and Ni (purity499%; average particle size: 4.2 μm).
3. Results and discussion 3.1. Preparation of the nonequilibrium (Ti,Ni)C phase
2. Experimental procedure Nonequilibrium solid-solution (Ti,Ni)C powders were produced using powders of a Ti–Ni alloy (Ti:Ni weight ratio=7:3; purity 4 99%; particle size o 75 μm) and of graphite (purity 4 99%; average particle size: 7–10 μm; Alfa Aesar). The Ti–Ni alloy powders were obtained by the hydrogenation and dehydrogenation of a Ti–Ni alloy ingot, which was prepared by the arc melting of Ti and Ni (purity 4 99.5%; average size: 3–20 mm; RNDKOREA) under an Ar atmosphere. The Ti–Ni alloy was transferred into a vacuum furnace for hydrogenation at a temperature of 873 K. Hydrogen gas (99.9999%) was introduced into the furnace to maintain a pressure between 5 105 and 7 105 Pa for 2 h. The hydrogenated Ti–Ni alloy was mechanically crushed into a size of o 75 μm. To remove the hydrogen, the Ti–Ni alloy powder was annealed at 973 K for 2 h in vacuum. The amount of graphite was adjusted to match the molar ratio of Ti in the Ti–Ni alloy, thus allowing TiC to form when the (Ti,Ni)C decomposed into TiC and Ni. The powders were then subjected to high-energy milling using a planetary mill (Pulverisette 5, Fritsch, Germany). Tungsten carbide balls were mixed with the Ti–Ni and graphite powders in a ball-to-powder weight ratio of 40:1. The powders were milled in a stainless steel bowl at 250 rpm in an argon atmosphere for 20 h. The phases of the milled powders were analyzed using X-ray diffraction (XRD) (SmartLab, Rigaku, Japan) with monochromated Cu Kα radiation (λ=1.5418 Å), which employed Si as a calibration standard. Differential scanning calorimetry (DSC) (SETSYS Evolution, SETARAM, France) was carried out in an Ar flow to analyze the reactions that occurred during the heat treatment of the (Ti,Ni)C phase. Heating rate of 5 1C/min was applied to all samples for the DSC analysis. The chemical composition was determined by inductively coupled plasma atomic emission spectrometry (ICP-AES, OPTIMA 5300DV, PerkinElmer) and CNO analysis (TCH-600, CS-600, LECO, Michigan, USA). The (Ti,Ni)C powder was then compacted at 125 MPa into discs. The sintering behavior of the (Ti,Ni)C compacts was monitored using a dilatometer (DIL 420 PC, Netzsch Geraetebau GmbH, Selb, Germany). The compacts were sintered at 1250 1C and 1330 1C for 5 min and 30 min. Their microstructures were examined using field-emission scanning electron microscopy
The nonequilibrium (Ti,Ni)C phase was synthesized by highenergy milling. Fig. 1 shows X-ray diffraction patterns of the phases evolved during the high-energy milling and the subsequent heat treatment of a mixture of Ti–Ni alloy and C. The as received Ti–Ni alloy was composed of Ti2Ni and pure Ti (Fig. 1 (a)). The sample milled for 20 h revealed only a B1-structured phase (NaCl-like structure), suggesting that the reaction for carbide formation had run to completion (Fig. 1(b)). It was found that the amount of contaminants introduced from the jar and balls were not much (Table 1). Milling for 20 h led to the formation of a (Ti,Ni)C phase, as evidenced by its lattice parameter being smaller than that of TiC, thus indicating the participation of Ni in the reaction (Table 2). Fig. 1(c) shows the XRD pattern of the (Ti, Ni)C powder after heat treatment at 1200 1C. Only TiC and Ni peaks are present, suggesting that the (Ti,Ni)C phase decomposed to TiC and Ni during the heat treatment. The lattice parameter of the B1-structured phase after the heat treatment was larger than that of the milled powder. The extraction of Ni from the (Ti,Ni)C phase increased the lattice parameter, because the atomic size of Ni is smaller than that of Ti. The lattice parameter of the heattreated B1-structured phase was smaller than that of the TiC phase (i.e., 4.3274 Å). This may be explained by the formation of
Fig. 1. X-ray diffraction patterns of (a) Ti–Ni raw material, (b) synthesized (Ti,Ni)C powder, and (c) (Ti,Ni)C powder heat treated for 2 h at 1200 1C.
4658
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
Table 1 Elemental analysis of the milled powdera. Element
Ti
Ni
W
Fe
Cr
Content (wt%)
70(2)
29(2)
0.40(6)
0.42(1)
–
a
Standard deviation in parentheses.
Table 2 Lattice parameters of the B1-structured phase and the carbon/nitrogen/oxygen content of the powders and the sintered bodiesa. Powder
Sintered body
Milled Heat- 1250 1C, treated 5 min
1330 1C, 5 min
1330 1C, 30 min
Lattice parameter of B1-structured phase (Å) Ti(CN)–Ni – – 4.28568(3) 4.28920(5) 4.28943(5) (Ti,Ni)C 4.2540 4.2755 4.32682(3) 4.3325(1) 4.3458(6) (5) (2) C/N/O content (wt%) Ti(CN)–Ni Carbon Nitrogen Oxygen (Ti,Ni)C Carbon
– – – 13.4 (2) Nitrogen 0.99 (3) Oxygen 3.52 (3)
Theoretical density (g/cm3) Ti(CN)–Ni – (Ti,Ni)C – a
– – – 12.8 (2) 0.73 (1) 1.04 (1)
– –
7.74(3) 11.79(7) 0.271(4) 12.5(1)
7.84(2) 11.7(3) 0.133(7) 12.6(2)
7.84(3) 12.2(1) 0.134(7) 12.5(1)
0.2704(4)
0.27(2)
0.29(2)
0.99(2)
0.83(9)
0.95(8)
5.9599 5.8720
5.9408 5.8501
5.9783 5.8111
Standard deviation in parentheses.
a carbon-deficient carbide phase owing to the reaction of oxygen with carbon. 3.2. Thermal decomposition of the nonequilibrium (Ti,Ni)C phase Our previous thermodynamic research confirmed that the nonequilibrium (Ti,Ni)C phase decomposes to TiC and Ni upon heating [20]. DSC was used here to examine in detail the thermal decomposition of the phase during heat treatment. The DSC curve for (Ti,Ni)C shows a continuous decrease from the initial analysis temperature (i.e., 600 1C) to 1200 1C, which corresponds to the endothermic reaction of carbon shown in Fig. 2(a). This agrees with the observed decreases of carbon and oxygen contents of the powder after the heat treatment (Table 2). The rate of decrease of the amount of heat generated during the carbothermal reduction reaction reduced as the temperature reached approximately 900 1C. This is related to the onset of the decomposition of the (Ti,Ni)C phase due to the exothermic nature of the phase decomposition reaction. The amount of heat generated during the carbothermal reduction reaction reached a minimum at 1200 1C, implying that the (Ti,Ni)C phase was not carbothermally reduced at higher
Fig. 2. DSC curves for (a) (Ti,Ni)C and (b) Ti(CN)–Ni powders.
temperatures. This was further substantiated by the lowering of the oxygen content of the powders (from 3.52 wt% to 1.04 wt%) observed after heat treatment at 1200 1C (Table 2). The DSC thermogram for (Ti,Ni)C shows a minimum at 1325 1C due to the endothermic melting of Ni in the (Ti,Ni)C phase. Note that the Ni melted below the melting point of pure Ni (1455 1C). As shown in the DSC thermogram of Fig. 2(b), a lower melting temperature of Ni (1360 1C) was also found in the case of Ti(CN)–Ni prepared by the mixing of Ti(CN) and Ni (Fig. 2(b)). This observation agrees with previously reported eutectic melting between Ni and TiC/Ti (CN) [21]. 3.3. Sintering of the (Ti,Ni)C phase The sintering behavior of (Ti,Ni)C and Ti(CN)–Ni compacts was investigated by measuring their linear shrinkage rate using a dilatometer (Fig. 3). At all temperatures the (Ti,Ni)C compact showed a higher linear shrinkage rate than the Ti(CN)–Ni compact, which may have been due to the (Ti,Ni)C particles (below 1 μm) being considerably finer than the Ti(CN) particles ( 1 μm) (Fig. 4), because smaller particles lead to larger surface areas and thus a higher driving force for their coalescence [22]. Unexpectedly, however, the (Ti,Ni)C compact stopped shrinking and began to expand rapidly at 1300 1C, while the Ti(CN)–Ni compact continued shrinking. The expansion implies the melting of the Ni in the (Ti,Ni)C compact, because the pores in the compact had already been eliminated by the solid-state sintering at 1300 1C. By comparison, the shrinkage of the Ti(CN)–Ni compact had not finished even at 1400 1C, a temperature higher than the starting point of Ni melting (1330 1C), because pores in the Ti(CN)–Ni compact had not been removed completely. As a result, Ni melting is expected to fill the pores in the compact at a temperature higher than the melting point of Ni, implying that full densification requires the sintering of Ti(CN)–Ni at above the melting point of Ni. Fig. 5 shows SEM micrographs showing backscattered electron images of the (Ti,Ni)C and Ti(CN)–Ni compacts sintered at 1250 1C and 1330 1C for 5 min. The particles of the (Ti,Ni)C
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
compact coalesced to become bigger after sintering at 1250 1C, while those of the Ti(CN)–Ni compact remained almost unchanged (Ti(CN): 1 μm and Ni: 4.2 μm) after similar sintering. Given that the sintering at 1250 1C was below the melting temperature of Ni (1325 1C for (Ti,Ni)C and 1360 1C for Ti(CN)–Ni), the densification was expected to occur through solid-state sintering. This suggests that the solid-state sintering of the (Ti,Ni)C compact occurred by the coalescence of the particles before the melting of the Ni. By contrast, the unchanging particle size of the Ti(CN)–Ni compact during sintering suggests that it became denser mainly by the rearrangement of particles. Another important observation concerning the microstructure of the (Ti,Ni)C compact sintered at 1250 1C is that Ni had already incorporated into the spaces between the particles before it melted. This occurred by the decomposition reaction suggested by the DSC result. The melting of Ni after the solid-state sintering affected significantly the microstructure of both samples. Fig. 5(c) and (d) show backscattered electron images of the (Ti,Ni)C and Ti (CN)–Ni compacts after sintering at 1330 1C. Liquid Ni is depicted as bright areas in the figures filling the gaps between darker carbide particles. This observation suggests that Ni had melted at 1330 1C by a eutectic reaction, which is further substantiated by the amount of carbon in the Ni matrix of the sintered bodies measured by energy dispersive X-ray spectrometry analysis (Fig. 6). The carbon content of the Ni in the sintered (Ti,Ni)C was 4.77 wt%; that in the Ti(CN)–Ni was 3.71 wt%. Both values are higher than the eutectic composition
4659
of 2.42 wt% C in the Ni–C system [23]. These results suggest that Ni in the (Ti,Ni)C and Ti(CN)–Ni compacts melted at about the eutectic point (1326 1C). The relative densities of the sintered bodies were calculated as follows and listed in Table 2: Theoretical densityðg=cm3 Þ n o in unit cellatomic weight Sum of atomic ratio Avogadro's number cell volume ¼ Unit where the atomic ratios of Ti, Ni, C, N, and O were obtained by inductively coupled plasma atomic emission spectroscopy and C/N/O analysis, and unit cell volume was measured by XRD. Fig. 7 shows the relative densities of the (Ti,Ni)C and Ti(CN)– Ni compacts after sintering at 1250 1C and at 1330 1C. The relative densities of the (Ti,Ni)C compacts were consistently significantly higher than those of the Ti(CN)–Ni compacts at each sintering condition. Both compacts showed a much greater increase in density after the hotter sintering. This suggests that pores in the compacts were filled with Ni, which melted by a eutectic reaction at 1326 1C. In summary, it can be concluded that the densification of the TiC–Ni composite was achieved here not only by particle coalescence and the in situ decomposition of the (Ti,Ni)C phase during solid-state sintering but also by the filling of pores with liquid Ni. 3.4. Mechanical properties of TiC–Ni composites
Fig. 3. Linear shrinkage rates of (Ti,Ni)C and Ti(CN)–Ni compacts.
The dense TiC–Ni composite revealed excellent mechanical properties, particularly toughness. The fracture toughness of TiC/Ti(CN)–Ni composites (6–7 MPa m1/2) is generally lower than that of WC–Co composites (10–12 MPa m1/2) [1,24]. The shorter lifetimes of the generally weaker components of TiC/Ti (CN)–Ni composites limit their broad applicability and generally leads to the addition of WC to improve their toughness for commercial use [25–28]. The fracture toughness of the TiC–Ni composite of this study (10–11 MPa m1/2) compares favorably with previously reported values for similar composites (6.8–7.4 MPa m1/2, Table 3) [24]. Furthermore, it is comparable to the highest values reported for Ti(CN)–WC–Ni composites (10–12 MPa m1/2). Therefore, the synthesis method presented here is shown effectively to produce a highly toughened TiC–Ni composite without needing the addition
Fig. 4. SEM micrographs of (a) (Ti,Ni)C and (b) Ti(CN)–Ni powders.
4660
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
Fig. 5. FE-SEM micrographs of (a) (Ti,Ni)C and (b) Ti(CN)–Ni sintered at 1250 1C for 5 min, (c) (Ti,Ni)C and (d) Ti(CN)–Ni sintered at 1330 1C for 5 min.
Fig. 6. TEM micrographs of (a) (Ti,Ni)C compact and (b) Ti(CN)–Ni compact sintered at 1330 1C for 5 min.
of WC. The composite's dense and finely dispersed microstructure —which occurred owing to the in situ decomposition of (Ti,Ni)C —led to its enhanced toughness. While this work demonstrates the fabrication of a tough new TiC–Ni composite, future work is required to elucidate in detail the mechanism of its formation. The uniform distribution of fine Ni and TiC particles that resulted from phase separation and relatively cool sintering at 1330 1C were possibly responsible for the enhanced toughness of the composite prepared from the (Ti,Ni)C phase. In any case, the suggested method would provide a valuable option for the preparation of TiC–Ni suitable to replace the commercial Ti(CN)–WC–Ni and similar composites that are currently used.
4. Conclusions A method to produce a dense TiC–Ni composite is proposed here. Such composites can be difficult to sinter due to the poor wettability of Ni on TiC. However, sintering was facilitated here through the synthesis of a (Ti,Ni)C phase—which was unstable at equilibrium—by the high-energy milling of a Ti–Ni alloy/graphite mixture. The (Ti,Ni)C compact was almost completely sintered at 1330 1C, below the temperature of conventional sintering. Enhanced densification occurred through solid-state sintering due to particle coalescence and in situ phase decomposition before the Ni melted. The solid-state sintering of the Ti(CN)–Ni compact
H. Kwon et al. / Ceramics International 41 (2015) 4656–4661
Fig. 7. Relative densities of (Ti,Ni)C and Ti(CN)–Ni after sintering.
Table 3 Mechanical properties of the TiC–Ni composites along with those of other TiC/ Ti(CN)–Ni based composites. Composite
HV (GPa)
KIC (MPa m1/2)
TiC–Ni (sintered at 1330 1C, for 5 min) TiC–Ni (sintered at 1330 1C, for 30 min) Ti(CN)–Ni (sintered at 1330 1C, for 30 min) TiC–Ni (20 30 wt% of Ni) [24] Ti(CN) (0.7 μm)–WC (0.4 μm)–Ni (4.1 μm) [25] Ti(CN) (0.7 μm)–WC (0.4 μm)–Ni (0.01 μm) [26] Ti(CN) (0.3 μm)–WC (0.2 μm)–Ni (4.1 μm) [27] Ti(CN) (0.08 μm)–WC (0.2 μm)–Ni (0.08 μm) [28]
13.5770.15 13.0070.11 9.677 0.32 13.239.21 13.6–14.2
10.1070.05 11.2270.15 – 6.8–7.4 7.25–8.8
12.5–15.3
8.4–9.1
14–15
8–10
12–13
10–12
occurred simply by the rearrangement of particles without their size changing. Further densification was achieved by the filling of pores with liquid Ni formed by a eutectic reaction. The fracture toughness of the TiC–Ni composite (10–11 MPa m1/2) was significantly higher than values previously reported for TiC–Ni composites. The fine and dense microstructure that originated from the phase decomposition was responsible for the enhanced toughness of the composite. A dense TiC–Ni compact with decent mechanical properties was demonstrated to be fabricated by the manipulation of the in situ decomposition of a nonequilibrium (Ti,Ni)C solid-solution phase. Acknowledgment This work was supported by a Grant-in-Aid from the Basic Research Project of the Korea Institute of Geoscience and Mineral Resources (KIGAM), funded by the Ministry of Science, ICT and Future Planning (GP2012-019). References [1] P. Ettmayer, Hardmetals and cermets, Annu. Rev. Mater. Sci. 19 (1989) 145–164.
4661
[2] N.M. Parikh, M. Humenik, Cermets: II, wettability and microstructure studies in liquid-phase sintering, J. Am. Ceram. Soc. 40 (1957) 315–320. [3] D.J. Miller, J.A. Pask, Liquid-phase sintering of TiC–Ni composites, J. Am. Ceram. Soc. 66 (1983) 841–846. [4] H. Suzuki, K. Hayashi, O. Terada, Two-phase region in Tic–Mo-30% Ni alloys, J Jpn. Inst Met. 35 (1971) 146–150. [5] S.Y. Ahn, S.W. Kim, S. Kang, Microstructure of Ti (CN)–WC–NbC–Ni cermets, J. Am. Ceram. Soc. 84 (2001) 843–849. [6] S. Kim, K.-H. Min, S. Kang, Rim structure in Ti(C0.7N0.3)–WC–Ni cermets, J. Am. Ceram. Soc. 86 (2003) 1761–1766. [7] C. Suryanarayana, Mechanical alloying and milling, Prog. Mater. Sci. 46 (2001) 1–184. [8] H. Fecht, G. Han, Z. Fu, W. Johnson, Metastable phase formation in the Zr–Al binary system induced by mechanical alloying, J. Appl. Phys. 67 (1990) 1744–1748. [9] C. Suryanarayana, R. Sundaresan, Metastable phases in mechanically alloyed Al Mn powder mixtures, Mater. Sci. Eng.: A 131 (1991) 237–242. [10] E. Ma, M. Atzmon, F. Pinkerton, Thermodynamic and magnetic properties of metastable FexCu100 x solid solutions formed by mechanical alloying, J. Appl. Phys. 74 (1993) 955–962. [11] J. Xu, U. Herr, T. Klassen, R. Averback, Formation of supersaturated solid solutions in the immiscible Ni–Ag system by mechanical alloying, J. Appl. Phys. 79 (1996) 3935–3945. [12] K. Chattopadhyay, X.-M. Wang, K. Aoki, T. Masumoto, Metastable phase formation during mechanical alloying of AlGe and AlSi alloys, J. Alloy. Compd. 232 (1996) 224–231. [13] B.-L. Huang, R. Perez, E. Lavernia, M. Luton, Formation of supersaturated solid solutions by mechanical alloying, Nanostruct. Mater. 7 (1996) 67–79. [14] B. Huang, K. Ishihara, P. Shingu, Metastable phases of Al–Fe system by mechanical alloying, Mater. Sci. Eng.: A 231 (1997) 72–79. [15] C. Suryanarayana, E. Ivanov, V. Boldyrev, The science and technology of mechanical alloying, Mater. Sci. Eng.: A 304 (2001) 151–158. [16] Y. Du, J. Schuster, Experimental investigation and thermodynamic modeling of the Ni–Ti–C system, Z. Metallkd. 89 (1998) 399–410. [17] D. Bandyopadhyay, R. Sharma, N. Chakraborti, The Ti–Ni–C system (titanium–nickel–carbon), J. ph. equilib. 21 (2000) 186–191. [18] I. Isomäki, M. Hämäläinen, Thermodynamic re-evaluation of the C–Ni– Ti system, J. Alloy. Compd. 416 (2006) 120–124. [19] D. Shetty, I. Wright, P. Mincer, A. Clauer, Indentation fracture of WC–Co cermets, J. Mater. Sci. 20 (1985) 1873–1882. [20] H. Kwon, J. Kim, S.-A. Jung, C.-Y. Suh, D.-S. Kil, K.-M. Roh, W. Kim, J.-W. Lim, Premixed TiC–Ni composite powder prepared by mechanical milling of Ti–Ni alloy/graphite mixture and subsequent heat treatment, Powder Technol. 253 (2014) 681–685. [21] L. Chen, W. Lengauer, P. Ettmayer, K. Dreyer, H.W. Daub, D. Kassel, Fundamentals of liquid phase sintering for modern cermets and functionally graded cemented carbonitrides (FGCC), Int. J. Refract. Met. Hard Mater. 18 (2000) 307–322. [22] R.L. Coble, Sintering crystalline solids. I. intermediate and final state diffusion models, J. Appl. Phys. 32 (2004) 787–792. [23] B.-J. Lee, On the stability of Cr carbides, Calphad 16 (1992) 121–149. [24] X.-H. Zhang, J.-C. Han, S.-Y. Du, J. Wood, Microstructure and mechanical properties of TiC–Ni functionally graded materials by simultaneous combustion synthesis and compaction, J. Mater. Sci. 35 (2000) 1925–1930. [25] E. Jeon, J. Joardar, S. Kang, Microstructure and tribo-mechanical properties of ultrafine Ti (CN) cermets, Int. J. Refract. Met. Hard Mater. 20 (2002) 207–211. [26] J. Joardar, S. Kim, S. Kang, Effect of nanocrystalline binder on the microstructure and mechanical properties of ultrafine Ti (CN) cermets, Mater. Sci. Eng.: A 360 (2003) 385–389. [27] J. Jung, S. Kang, Effect of ultra-fine powders on the microstructure of Ti (CN)–xWC–Ni cermets, Acta Mater. 52 (2004) 1379–1386. [28] J. Jung, S. Kang, Effect of nano-size powders on the microstructure of Ti (C, N)–xWC–Ni ‐cermets, J. Am. Ceram. Soc. 90 (2007) 2178–2183.