- Email: [email protected]
Nb5Si3 composites alloyed with W, Mo and W–Mo fabricated by spark plasma sintering
Materials Science & Engineering A 606 (2014) 68–73
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Microstructures and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo and W–Mo fabricated by spark plasma sintering Bowen Xiong n, Changchun Cai, Zhenjun Wang National Defence Key Discipline Laboratory of Light Alloy Processing Science and Technology, Nanchang Hangkong University, Nanchang 330063, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 1 March 2014 Received in revised form 22 March 2014 Accepted 24 March 2014 Available online 28 March 2014
Microstructures and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo and W–Mo fabricated by spark plasma sintering were investigated. The microstructures were examined using scanning electron microscope (SEM). X-ray diffraction (XRD) was performed on the bulk specimens for identification of phases. The chemical species were analyzed using electron-probe micro-analysis (EPMA). Results indicated that the microstructures of Nb/Nb5Si3 composites alloyed with W or Mo is unaltered, including primary Nb and eutectic mixtures of Nb and Nb5Si3, and the coarse and fine eutectic mixtures. The W and Mo elements solid solution in Nb and Nb5Si3 phase are detected. But that alloyed with W and Mo together. The microstructures are changed obviously, including Nb phase, the solid solubility phases of W and Mo atoms in Nb, and the solid solubility phases of Nb atoms in W are also found, but the solid solubility phenomenon of Nb5Si3 phases is not detected. The microhardness of Nb and Nb5Si3 phases increases obviously because of solid solution strengthening. The excellent high temperature tensile strength of Nb 20Si 10W 10Mo sample was obtained. It is attributable to granular Nb surrounding by Nb5Si3 phase in eutectic, which different with interlamellar eutectic. & 2014 Elsevier B.V. All rights reserved.
Keywords: Spark plasma sintering Microstructure Nb/Nb5Si3 composites Tensile strength
1. Introduction Niobium silicide based alloys have been extensively studied as high-temperature structural materials [1–3]. Nb5Si3 has high melting point (2515 1C), low density (7.16 g/cm3) and excellent specific mechanical properties at high temperature [4–6], but Nb5Si3 has low fracture toughness at room temperature and poor deformability at high temperature, these drawbacks hinder the application of Nb5Si3. The Nb has high melting point (2472 1C), excellent fracture toughness and good deformability. According to Nb Si binary phase diagram, Nb and Nb5Si3 coexist in wide range of temperature and silicon content. These features of the phase diagram allow us to fabricate Nb/Nb5Si3 in situ composites with desirable mechanical properties and excellent thermo-chemical stability [5,7]. Most studies succeeded in improving the fracture toughness at room temperature with the sacrifice of high temperature strength of Nb/Nb5Si3 in situ composites by increasing the volume fraction of the ductile Nb phase. Recent researches [8–12] have shown that the significant improvement in the strength of Nb/Nb5Si3 composites has been achieved by solid solution of Nb and Nb5Si3 on alloying with specific elements, such as Ti, Cr, Hf, W and Mo [7,13]. Currently, the Nb/Nb5Si3 composite was prepared mainly by arc
n
Corresponding author. Tel./fax: þ 86 791 86453167. E-mail address: [email protected] (B. Xiong).
http://dx.doi.org/10.1016/j.msea.2014.03.085 0921-5093/& 2014 Elsevier B.V. All rights reserved.
melting [7,14], self-propagating high-temperature synthesis [4,15], directional solidification [1,14] and powder metallurgy [1,16]. Previous literatures have detected some difference in microstructures of Nb/Nb5Si3 composite fabricated by different preparation method [1], for example, at near eutectic composition, arc melting alloys exhibit a dispersed microstructure consisting of fine Nb phase in a Nb5Si3 matrix, while the microstructures of powder metallurgy alloys are characterized by aggregate structure and equiaxed structure [1]. Recently, the spark plasma sintering (hereafter abbreviated as SPS) has been drawing attention as a new process for fabricating advanced materials [17]. The most important feature of SPS is that the pressed powders are heated by the spark plasma; as a result, the samples can be sintered uniformly and rapidly, and the dense materials with fine grains can be obtained in a very short holding time [5,17]. Therefore, SPS is considered to be a potential method for fabricating Nb/Nb5Si3 composites. However, from the available literatures, the investigations on microstructures and mechanical properties of Nb/Nb5Si3 composite alloyed with specific elements fabricated by SPS are still rare. Therefore, in the present paper, using Nb–20Si, Nb 20Si 10W, Nb 20Si 10Mo and Nb 20Si10W 10Mo as raw materials, Nb/Nb5Si3 in situ composites were fabricated by SPS. The objective of the present study was to experimentally investigate microstructure and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo, and W Mo. The significant information was expected to guide the fabrication of Nb/Nb5Si3 composites by SPS.
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
2. Experimental procedure
3. Results and discussion
Nb 20Si, Nb 20Si 10W, Nb 20Si 10Mo and Nb 20Si 10W 10Mo elemental powder mixtures (at%) were used to produce the Nb/Nb5Si3 composites by SPS respectively. The raw materials used in the present study were Nb (325 mesh), Si (200 mesh), W (300 mesh) and Mo (130 mesh) powders with purities of 99.9%, 99.6%, 99.5% and 99.5% (wt%) respectively. These powders were dry mixed in a mechanical mixer for 24 h, and the mixed powders were packed into a graphite die with an inside diameter of 15 mm. The sintering was carried out on a SPS machine manufactured by Sumitomo Coal Mining Co. Ltd., Japan. A uniaxial load of 6 kN acted on the powders to realize sintering and densification simultaneously. A heating speed of 150 1C/min and the sintering temperature of 1500 1C were adopted under fixed sintering holding time of 10 min in the present study. The specimens were metallographically polished and etched using a reagent comprising 5 ml HNO3, 10 ml HF, 15 ml H2SO4 and 50 ml distilled water. The microstructures and fracture surfaces have been examined using a JSMT-200 scanning electron microscope. D/MAX-IIIB X-ray diffraction was performed on the bulk specimens for identification of phases. The distribution of phase and chemical species has been analyzed using JXA-800R electron-probe micro-analysis. The tensile samples whose gage has the dimensions of 1.5 mm 2 mm 10.5 mm were prepared from the center of ingot. The tensile test was performed in an Instron-type material-testing machine at room temperature, 1573 K and 1773 K, and the tensile test was carried out at a strain rate of 1 10 4 s 1 in an argon atomosphere. Microhardness of the primary phases and the eutectic microstructures was measured with the help of a Leica microhardness tester operated using Vickers diamond indenter at loads of 10 gf. At least 10 measurements were taken for each alloy and condition.
3.1. Microstructure
69
The microstructures of Nb–20Si, Nb–20Si–10W, Nb–20Si–10Mo and Nb–20Si–10W–10Mo samples by SPS are presented in Fig. 1. Compared with the conventional sintering process, the SPS technology has unique advantages. The pulse current was directly yielded by special power acting on the reactant powders, which induces rarefied gas ionization and discharge at the gaps of powders. The local high temperature created by plasma and Joule heating at the gaps of powders leads to impurities gasification and surfaces activation taking place on the particle surfaces [18]. At the beginning stage of SPS, the bed of reactant powders was very loose, which induces the low degree of rarefied gas ionization and discharge. The powder was rearranged by the uniaxial load of 6 kN. With the powders getting tight, the degree of rarefied gas ionization and discharge improves; simultaneously, a large amount of discharge heating as well as Joule heating is got. When the sintering temperature increases to the melting temperature of silicon (1414 1C), resulting in the increase of molten silicon, the liquid phase sintering takes place under the uniaxial load of 6 kN [19]. With increasing the sintering temperature, the density of samples improves and pores disappear, resulting in the function of rarefied gas ionization and discharge debases even disappear finally. The samples with nearly full density are achieved by Joule heating and the uniaxial load. The pulse current uniformly passes through the powder and the reaction of Nb and Si simultaneously takes place everywhere. Furthermore, the reactant powders satisfactorily vibrated by the pulse current wallop and intermission discharge pressure leads the particle phase to take on uniform distribution [15]. Therefore, the dispersed microstructures of Nb/ Nb5Si3 composites can be obtained.
Fig. 1. Microstructures of Nb/Nb5Si3 composites (a) Nb–20Si, (b) Nb–20Si–10W, (c) Nb–20Si–10Mo, and (d) Nb–20Si–10W–10Mo.
70
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
The microstructures of Nb–20Si are indicated in Fig. 1(a). It is evident that the microstructures include small amount of primary Nb and eutectic mixtures of Nb and Nb5Si3. By comparing Fig. 1 (b) and (c) with Fig. 1(a), it is evident that the effects of W and Mo on the constituent phases in microstructure of Nb/Nb5Si3 composites are not obvious and the constituent phases is not changed, while by alloying with W and Mo together, the microstructure of Nb/Nb5Si3 composites changed significantly in Fig. 1(d). EPMA results of point 1 and point 2 in Nb–20Si–10W sample are given in Table 1, which indicate that the particle phase in Fig. 1(b) (i.e. point 1) contains Nb and W, and the matrix (point 2) contains both Nb and Si. The XRD pattern of Nb–20Si–10W sample is shown in Fig. 2, individual phase was identified by matching the characteristic XRD peaks against JCPDS data. It is obvious from Fig. 2 that the peaks of Nb and Nb5Si3 are detected but not peaks of W, Si and Nb3Si. Thus, combining the EPMA results with the analyzed results of XRD (Fig. 2), it is shown that the particle phases (point 1) are the solid solution of W in primary Nb and the matrix is the solid solution of W in Nb5Si3 phase. Mo and W are strong solid solution strengtheners of Nb and improve its high temperature strength. Atomic radii of Nb and W in pure substances are 0.143 nm and 0.136 nm, respectively. The atomic radii are very close, showing the Nb can be substituted by W. Many investigations have indicated that W element as the solid solution strengthening can been dissolved into Nb/Nb5Si3 composites [20,21]. The content of W in Nb5Si3 phase is small, indicating W tends to concentrate in Nb phase, as also obtained by Sha [13]. In this study, the W phase was not found, which indicates W element was dissolved completely, so the microstructure of Nb/Nb5Si3 composites alloyed with W is not changed. The microstructures of Nb–20Si–10Mo sample are indicated in Fig. 1(c), and the XRD pattern of Nb–20Si–10Mo sample is shown in Fig. 3. Only the peaks of Nb and Nb5Si3 are detected in Fig. 3, combining the EPMA results of points 3 and 4 (in Table 1), the
Table 1 EPMA results of phases. Location
Point Point Point Point Point Point Point Point
1 2 3 4 5 6 7 8
Element (at%) Nb
Si
W
Mo
81.34 73.42 60.57 72.33 81.03 72.58 60.45 7 3.52 100 30.81 7 2.02 62.277 1.87 68.747 1.06
– 37.687 2.42 – 37.767 2.08 – – 37.737 1.09 –
18.66 71.08 1.75 7 0.09 – – – 69.197 2.31 – 6.29 7 0.45
– – 18.977 1.26 1.79 7 0.11 – – – 24.977 1.41
Fig. 2. XRD pattern of Nb–20Si–10W sample.
Fig. 3. XRD pattern of Nb–20Si–10Mo sample.
Fig. 4. XRD pattern of Nb–20Si–10W–10Mo sample.
particle phases (point 3) are the solid solution of Mo in primary Nb and points 4 is the solid solution of Mo in Nb5Si3 phase. Atomic radius of Mo in pure substances is 0.140 nm. The atomic radius of Mo is closer to that of Nb than W, indicating Mo (10 at%) is easier to dissolve into Nb/Nb5Si3 composites. Recently, Kim et al. have investigated in detail a partial isothermal section of Nb–Si–Mo system at 1973 K [1,14]. According to their result, Mo is preferentially dissolved in Nb solid solution than Nb5Si3 phase in the twophase Nb/Nb5Si3 composites. Therefore, the concentration of Mo is higher in Nb solid solution than Nb5Si3 phase. The microstructure of Nb–20Si–10W–10Mo sample by SPS is given in Fig. 1(d). Because of the solid solution of W and Mo together, the microstructure of Nb–20Si–10W–10Mo sample is different from other samples. In order to further analyze chemical species, the four points in Fig. 1(d) were analyzed using electronprobe micro-analysis (EPMA), as shown in Table 1. It is indicated that the phase (point 5) is single Nb phase, point 6 contains Nb (30.81 at%) and W (69.19 at%), point 7 comprises Nb (62.27 at%) and Si (37.73 at%), and point 8 includes Nb (68.74 at%), Mo (24.97 at%) and W (6.29 at%). The XRD pattern of Nb–20Si–10W–10Mo sample is presented in Fig. 4. It is evident from Fig. 4 that the peaks of Nb5Si3, Nb and W are found but no peaks of Mo and Si. Thus, combining the EPMA results with the analyzed results of XRD (Fig. 4), the microstructures of Nb–20Si–10W–10Mo sample consist of the single Nb phase, Nb5Si3 phase, the solid solution of Nb atoms in W and the solid solution of Mo and W in Nb. It is obvious that no W or Mo element solid solution in Nb5Si3 phases (point 7) was detected in Nb–20Si–10W–10Mo sample, although the solid solution of W and Mo in Nb5Si3 phases was detected in Nb–20Si–10W and Nb–20Si–10Mo sample, respectively. By alloying with W and Mo together, why W and Mo cannot form solid solution in Nb5Si3 phases will be considered in detail in our future study.
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
Furthermore, the almost identical size of primary Nb is presented in Fig. 1, but the eutectic mixtures may be categorized as coarse and fine, based on the size of Nb phase present in eutectic mixtures. The eutectic solidification occurs in a ternary alloy system over a range of temperatures. The coarse eutectic is likely to form at higher temperature, while the fine eutectic forms towards the end of the solidification process at relatively lower temperatures [13]. The fine eutectic was found in Fig. 1(a)–(c), but only coarse eutectic was detected obviously in Fig. 1(a). In addition, the fine eutectic in Fig. 1(c) is finer than that in Fig. 1(b). From the obtained results above, it is indicated that the temperature of eutectic reaction in Nb/Nb5Si3 composites can be decreased by alloying with Mo or W, and it seems Mo element has more effect in decreasing the temperature of eutectic reaction than W element. The refinement of eutectic Nb phase by adding appropriate Mo and W element was also obtained by Kim [14] and Ma [20], respectively. The lamellae eutectic was observed in Fig. 1(a)–(c), but not presented in Fig. 1(d). The eutectic takes on the small granular Nb with about 5–10 μm in size (indicated by arrows) surrounded by Nb5Si3 phase in Fig. 1(d), and the large observation scope in Nb–20Si–10W–10Mo sample for surveying the eutectic was given in Fig. 5. The reason is uncertain in the present study, but one possible explanation is the effects of Mo and W together on thermodynamics or kinetics of eutectic reaction during solidification, and alloy element affects the thermodynamics or kinetics of solidification reaction in Nb/Nb5Si3 composites to change the microstructures which is also reported in previous literatures [7,22]. As hypereutectic Nb–20Si composition, the small amount of primary Nb phase was detected in Fig. 1, which is different from the microstructure obtained in previous literatures [7,14]. This phenomenon may be due to the Si content being lower than the eutectic composition at local. The size of Si particle is bigger than Nb particle, which will induce the inevitable segregation of Nb particle among Si particles in the mixed raw materials. The molten Si is beneficial to the homogenization of compositions during SPS, but the sintering time is very short about 10 min, resulting in Si content being lower than the eutectic composition at local. Therefore, the small amount of primary Nb phase was detected in Fig. 1. 3.2. Mechanical properties 3.2.1. Microhardness Table 2 shows the variation of microhardness of Nb phase and Nb5Si3 phases. The microhardness of Nb and Nb5Si3 phases changes obviously by alloying. The microhardness of Nb phases solid solution of W, Mo and W–Mo increase obviously by about 39.35%, 36.66% and 53.17%, respectively, which can be explained on the basis of solid solution strengthening. The atomic radii are very close, indicating the Nb can be substituted by Mo and W.
Fig. 5. The large observation scope in Nb–20Si–10W–10Mo sample.
71
Table 2 Microhardness (HV) of Nb and Nb5Si3 phases. Materials
Nb
Nbss
Nb5Si3
Nb–20Si Nb–20Si–10W Nb–20Si–10Mo Nb–20Si–10W–10Mo
521 76 – – 532 7 5
– 7267 6 7127 8 798 710
14377 11 1783 7 13 17607 16 14337 8
Mo and W substitutes Nb in the crystal, resulting in the decrease of lattice parameter of Nb solid solution and deformation of lattice [1]. Therefore, the microhardness of Nb phases with the solid solution of W, Mo and W–Mo increase obviously. The microhardness of W solid solution in Nb phases (point 1) is slight higher than that of Mo (point 3), although the concentration of W is slightly lower than that of Mo. This phenomenon can be explained based on the fact that atomic radius of W is smaller than that of Mo leading to greater deformation of lattice. By alloying W–Mo, the microhardness of Nb phase solid solution of W and Mo together increases greatly. Because of the solid solution of W and Mo together, the concentration of solute (in point 8) increases significantly, therefore, the microhardness increases greatly. The microhardness of Nb5Si3 phases in Nb–20Si–10W and Nb–20Si– 10Mo samples obvious increase respectively, but that in Nb–20Si– 10W–10Mo sample is almost unaltered. This phenomenon is because solid solution hardening for Nb5Si3 phase does not appear in Nb–20Si–10W–10Mo sample.
3.2.2. Tensile property The tensile stress–strain curves of the Nb/Nb5Si3 composite prepared in this study are presented in Fig. 6. From Fig. 6, the highest tensile strength of Nb/Nb5Si3 composite with same composition was obtained at room temperature, but these composites fracture before yielding, which are different from the fracture at 1573 K and 1773 K. The tensile stress–strain curves show a peak value followed by a gradual decrease, and the composite fractures at a stress of about 300 MPa and 100 MPa at 1573 K and 1773 K, respectively. The tensile stress peak value decreases continuously with the increase of test temperature, meanwhile the elongation increases markedly. The Nb–20Si–10W–10Mo sample takes on the highest tensile strength and best plastic deformability, either at room temperature or at high temperature. The tensile strength and plastic deformability of Nb–20Si–10Mo sample are only less than that of Nb–20Si–10W–10Mo sample, and the tensile strength and plastic deformability of Nb–20Si sample are the worst. The plastic deformability of Nb–20Si–10W–10Mo sample reaches about 16.25% and 21.31% at 1573 K and 1773 K, respectively. As we know, the mechanical property is relevant mainly with phase composition and size of microstructure. The contents of constituent phases including primary Nb, eutectic Nb and Nb5Si3 phase and interlamellar spacing of eutectic were measured by statistical quantitative metallographic analysis [23], as given in Table 3. The contents of Nb phase (including primary and eutectic Nb) and Nb5Si3 phase are almost same in Nb–20Si, Nb–20Si–10W and Nb–20Si–10Mo samples, but the interlamellar spacing of eutectic is obviously different. Previous literatures have indicated that interlamellar spacing of eutectic plays an important role in mechanical properties [7,22]. Although the constituent phases and its content in Nb–20Si, Nb–20Si–10W and Nb–20Si–10Mo samples are similar, the tensile strength is obviously different. This phenomenon may be explained on the basis of interlamellar spacing of eutectic and solid solution strengthening. The Nb–Nb5Si3 interfaces in the eutectic act as obstacles to dislocation motion, indicating that interlamellar spacing plays an important role in
72
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
that of Mo, which will lead to the greater strengthening effect mentioned above. However, the tensile strength of Nb–20Si–10W is lower than that of Nb–20Si–10Mo sample. This may be because interlamellar spacing has greater impact on tensile strength than solid solution strengthening of W in this study. Therefore, the tensile strength of Nb–20Si–10Mo is higher than that of Nb–20Si–10W sample. Nb–20Si–10W–10Mo sample has the highest tensile strength. This may be mainly attributed to its microstructure. The eutectic takes on the small granular Nb surrounded by Nb5Si3 phase in Nb– 20Si–10W–10Mo sample (Fig. 1(d)), which is obviously different with interlamellar eutectic, and the distribution of eutectic Nb is uniform from Fig. 5. This type of microstructure is more conducive to increase interface in eutectic than interlamellar eutectic. Therefore, the tensile strength of Nb–20Si–10W–10Mo sample is the highest. 3.2.3. Status of tensile property of Nb/Nb5Si3 composites The highest tensile strength of Nb–20Si–10W–10Mo sample is 587.687 12.5 MPa, 413.65 78.7 MPa and 24675.2 MPa at room temperature, 1573 K and 1773 K, respectively. The investigation on the high temperature tensile strength of Nb/Nb5Si3 composites is rare compared with compression strength. From available literature data, these tensile strengths are slightly higher than that of Nb–18Si–5Mo–5Hf alloy [24] and Nb–16Si–2Fe alloy [25] at same test temperature, but the deformability is far worse than that of Nb–18Si–5Mo–5Hf alloy and Nb–16Si–2Fe alloy. The Nb/Nb5Si3 composite fabricated by SPS takes on excellent high temperature tensile strength. However, its deformability at room temperature is very low. This will hinder its extensive application. In order to develop Nb–Si based structural materials, the refinement of microstructure and controlling of interface will be an important concern.
4. Conclusions In this work, using Nb–20Si, Nb–20Si–10W, Nb–20Si–10Mo and Nb–20Si–10W–10Mo elemental powder mixtures as raw materials, the microstructures and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo and W–Mo fabricated by spark plasma sintering were investigated. The conclusions of this work are as follows:
Fig. 6. Tensile strength of Nb/Nb5Si3 composites (a) at room temperature, (b) 1573 K, and (c) 1773 K.
Table 3 Volume fraction of constituent phases and interlamellar spacing. Materials
Nb–20Si Nb–20Si–10W Nb–20Si–10Mo Nb–20Si–10W–10Mo
Volume fraction (%) Primary Nb
Nb5Si3
Eutectic Nb
W
Interlamellar spacing of eutectic (μm)
10.08 10.86 11.07 7.87
50.24 49.58 49.08 48.68
39.68 40.06 39.85 37.82
– – – 5.63
1.357 0.75 0.95 7 0.15 0.83 7 0.16 –
(1) The microstructures Nb/Nb5Si3 composites alloyed with W and Mo, respectively, include primary Nb and eutectic mixtures of Nb and Nb5Si3, and the coarse and fine eutectic mixtures are detected. The W or Mo elements solid solution in Nb and Nb5Si3 phase are detected. (2) The microstructures of Nb/Nb5Si3 composites alloyed with W and Mo together obviously change, including Nb phase, the solid solubility phases of W and Mo atoms in Nb, and the solid solubility phases of Nb atoms in W are also found, but the solid solubility phenomena of Nb5Si3 phases is not detected. (3) The excellent high temperature tensile strength of Nb–20Si– 10W–10Mo sample was obtained. It is attributable to granular Nb surrounded by Nb5Si3 phase in eutectic, which is different with interlamellar eutectic.
Acknowledgements its deformation behavior. Therefore, with the decrease of interlamellar spacing, the tensile strength increases. Otherwise, the Nb phase and Nb5Si3 phase solid solution of Mo and W will bring strengthening effect, as a result of elastic interaction with solute atoms of different sizes. The atomic radius of W is smaller than
This research was financially supported by Nanchang Hangkong University Foundation of China under Grant no. EC200701021, and Jiangxi Provincial Department of Science and Technology Foundation of China under Grant no. DB200701049.
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
References [1] W.Y. Kim, H. Tanaka, S. Hannda, Intermetallics 10 (2002) 625–634. [2] H.Z. Fu, J. Aeronaut. Mater 18 (1998) 52–60. [3] J.L. Yu, Z.K. Li, K.F. Zhang, X. Zheng, H. Liu, H. Wang, W.S. Wang, Fail. Anal. Prev. 4 (2010) 231–234. [4] C.L. Yeh, W.H. Chen, J. Alloys Compd. 425 (2006) 216–222. [5] Z. Chen, Y.W. Yan, J. Alloys Compd. 413 (2006) 73–76. [6] J. Geng, P. Tsakiropoulos, G. Shao, Intermetallics 15 (2007) 69–76. [7] K. Chattopadhyay, G. Balachandran, R. Mitra, K.K. Ray, Intermetallics 14 (2006) 1452–1460. [8] P.R. Subramanian, M.G. Mendiratta, D.M. Dimiduk, M.A. Stuck, Mater. Sci. Eng. A 239 (1997) 1–3. [9] K.S. Chan, D.L. Davidson, Mater. Sci. Eng. A 30 (1999) 925–939. [10] K.S. Chan, Mater. Sci. Eng. A (2002) 513–522. [11] K.S. Chan, D.L. Davidson, Mater. Sci. Eng. A 34 (2003) 1833–1849. [12] J. Geng, P. Tsakiropoulos, G. Sha, Intermetallics 14 (2006) 227–235. [13] J. Sha, H. Hirai, T. Tabaru, A. Kitahara, H. Ueno, S. Hanada, Metall. Mater. Trans. A 34 (2003) 2861–2871.
73
[14] W.Y. Kim, H. Tanaka, A. Kasama, S.J. Hanada, Intermetallics 9 (2001) 827–834. [15] M.E. Reiss, C.M. Esber, D. Van Heerden, A.J. Gavens, M.E. Williams, T.P. Weihs, Mater. Sci. Eng. A 261 (1999) 217–222. [16] B.B. Fernandes, G. Rodrigues, G. Silva, E.C.T. Ramos, C.A. Nunes, H.R.Z. Sandim, A.S. Ramos, J. Alloys Compd. 434–435 (2007) 530–534. [17] O. Mamoru, Mater. Sci. Eng. A 287 (2000) 183–188. [18] P.B. Wang, G.J. Yang, X.N. Mao, Rare Metal. Mater. Eng. 36 (2007) 484–488. [19] Z. Chen, Y.W. Yan, Rare Metal. Mater. Eng. 35 (2006) 484–487. [20] C.L. Ma, J.G. Li, Y. Tan, R. Tanaka, S. Hanada, Mater. Sci. Eng. A 386 (2004) 375–383. [21] I. Grammenos, P. Tsakiropoulos, Intermetallics 19 (2011) 1612–1621. [22] S. Kashyap, C.S. Tiwary, K. Chattopadhyay, Intermetallics 19 (2011) 1943–1952. [23] R.T. DeHoff, F.N. Rhines, Quantitative Microscopy, first ed., McGraw-Hill, Inc., New York, 1968. [24] J.H. Kim, T. Tabaru, M. Sakamoto, S. Hanada, Mater. Sci. Eng. A 372 (2004) 137–144. [25] J.L. Yu, Z.K. Li, K.F. Zhang, X. Zheng, J.J. Zhang, R. Bai, W.S. Wang, Mater. Sci. Eng. A 527 (2010) 5230–5233.
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Microstructures and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo and W–Mo fabricated by spark plasma sintering Bowen Xiong n, Changchun Cai, Zhenjun Wang National Defence Key Discipline Laboratory of Light Alloy Processing Science and Technology, Nanchang Hangkong University, Nanchang 330063, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 1 March 2014 Received in revised form 22 March 2014 Accepted 24 March 2014 Available online 28 March 2014
Microstructures and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo and W–Mo fabricated by spark plasma sintering were investigated. The microstructures were examined using scanning electron microscope (SEM). X-ray diffraction (XRD) was performed on the bulk specimens for identification of phases. The chemical species were analyzed using electron-probe micro-analysis (EPMA). Results indicated that the microstructures of Nb/Nb5Si3 composites alloyed with W or Mo is unaltered, including primary Nb and eutectic mixtures of Nb and Nb5Si3, and the coarse and fine eutectic mixtures. The W and Mo elements solid solution in Nb and Nb5Si3 phase are detected. But that alloyed with W and Mo together. The microstructures are changed obviously, including Nb phase, the solid solubility phases of W and Mo atoms in Nb, and the solid solubility phases of Nb atoms in W are also found, but the solid solubility phenomenon of Nb5Si3 phases is not detected. The microhardness of Nb and Nb5Si3 phases increases obviously because of solid solution strengthening. The excellent high temperature tensile strength of Nb 20Si 10W 10Mo sample was obtained. It is attributable to granular Nb surrounding by Nb5Si3 phase in eutectic, which different with interlamellar eutectic. & 2014 Elsevier B.V. All rights reserved.
Keywords: Spark plasma sintering Microstructure Nb/Nb5Si3 composites Tensile strength
1. Introduction Niobium silicide based alloys have been extensively studied as high-temperature structural materials [1–3]. Nb5Si3 has high melting point (2515 1C), low density (7.16 g/cm3) and excellent specific mechanical properties at high temperature [4–6], but Nb5Si3 has low fracture toughness at room temperature and poor deformability at high temperature, these drawbacks hinder the application of Nb5Si3. The Nb has high melting point (2472 1C), excellent fracture toughness and good deformability. According to Nb Si binary phase diagram, Nb and Nb5Si3 coexist in wide range of temperature and silicon content. These features of the phase diagram allow us to fabricate Nb/Nb5Si3 in situ composites with desirable mechanical properties and excellent thermo-chemical stability [5,7]. Most studies succeeded in improving the fracture toughness at room temperature with the sacrifice of high temperature strength of Nb/Nb5Si3 in situ composites by increasing the volume fraction of the ductile Nb phase. Recent researches [8–12] have shown that the significant improvement in the strength of Nb/Nb5Si3 composites has been achieved by solid solution of Nb and Nb5Si3 on alloying with specific elements, such as Ti, Cr, Hf, W and Mo [7,13]. Currently, the Nb/Nb5Si3 composite was prepared mainly by arc
n
Corresponding author. Tel./fax: þ 86 791 86453167. E-mail address: [email protected] (B. Xiong).
http://dx.doi.org/10.1016/j.msea.2014.03.085 0921-5093/& 2014 Elsevier B.V. All rights reserved.
melting [7,14], self-propagating high-temperature synthesis [4,15], directional solidification [1,14] and powder metallurgy [1,16]. Previous literatures have detected some difference in microstructures of Nb/Nb5Si3 composite fabricated by different preparation method [1], for example, at near eutectic composition, arc melting alloys exhibit a dispersed microstructure consisting of fine Nb phase in a Nb5Si3 matrix, while the microstructures of powder metallurgy alloys are characterized by aggregate structure and equiaxed structure [1]. Recently, the spark plasma sintering (hereafter abbreviated as SPS) has been drawing attention as a new process for fabricating advanced materials [17]. The most important feature of SPS is that the pressed powders are heated by the spark plasma; as a result, the samples can be sintered uniformly and rapidly, and the dense materials with fine grains can be obtained in a very short holding time [5,17]. Therefore, SPS is considered to be a potential method for fabricating Nb/Nb5Si3 composites. However, from the available literatures, the investigations on microstructures and mechanical properties of Nb/Nb5Si3 composite alloyed with specific elements fabricated by SPS are still rare. Therefore, in the present paper, using Nb–20Si, Nb 20Si 10W, Nb 20Si 10Mo and Nb 20Si10W 10Mo as raw materials, Nb/Nb5Si3 in situ composites were fabricated by SPS. The objective of the present study was to experimentally investigate microstructure and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo, and W Mo. The significant information was expected to guide the fabrication of Nb/Nb5Si3 composites by SPS.
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
2. Experimental procedure
3. Results and discussion
Nb 20Si, Nb 20Si 10W, Nb 20Si 10Mo and Nb 20Si 10W 10Mo elemental powder mixtures (at%) were used to produce the Nb/Nb5Si3 composites by SPS respectively. The raw materials used in the present study were Nb (325 mesh), Si (200 mesh), W (300 mesh) and Mo (130 mesh) powders with purities of 99.9%, 99.6%, 99.5% and 99.5% (wt%) respectively. These powders were dry mixed in a mechanical mixer for 24 h, and the mixed powders were packed into a graphite die with an inside diameter of 15 mm. The sintering was carried out on a SPS machine manufactured by Sumitomo Coal Mining Co. Ltd., Japan. A uniaxial load of 6 kN acted on the powders to realize sintering and densification simultaneously. A heating speed of 150 1C/min and the sintering temperature of 1500 1C were adopted under fixed sintering holding time of 10 min in the present study. The specimens were metallographically polished and etched using a reagent comprising 5 ml HNO3, 10 ml HF, 15 ml H2SO4 and 50 ml distilled water. The microstructures and fracture surfaces have been examined using a JSMT-200 scanning electron microscope. D/MAX-IIIB X-ray diffraction was performed on the bulk specimens for identification of phases. The distribution of phase and chemical species has been analyzed using JXA-800R electron-probe micro-analysis. The tensile samples whose gage has the dimensions of 1.5 mm 2 mm 10.5 mm were prepared from the center of ingot. The tensile test was performed in an Instron-type material-testing machine at room temperature, 1573 K and 1773 K, and the tensile test was carried out at a strain rate of 1 10 4 s 1 in an argon atomosphere. Microhardness of the primary phases and the eutectic microstructures was measured with the help of a Leica microhardness tester operated using Vickers diamond indenter at loads of 10 gf. At least 10 measurements were taken for each alloy and condition.
3.1. Microstructure
69
The microstructures of Nb–20Si, Nb–20Si–10W, Nb–20Si–10Mo and Nb–20Si–10W–10Mo samples by SPS are presented in Fig. 1. Compared with the conventional sintering process, the SPS technology has unique advantages. The pulse current was directly yielded by special power acting on the reactant powders, which induces rarefied gas ionization and discharge at the gaps of powders. The local high temperature created by plasma and Joule heating at the gaps of powders leads to impurities gasification and surfaces activation taking place on the particle surfaces [18]. At the beginning stage of SPS, the bed of reactant powders was very loose, which induces the low degree of rarefied gas ionization and discharge. The powder was rearranged by the uniaxial load of 6 kN. With the powders getting tight, the degree of rarefied gas ionization and discharge improves; simultaneously, a large amount of discharge heating as well as Joule heating is got. When the sintering temperature increases to the melting temperature of silicon (1414 1C), resulting in the increase of molten silicon, the liquid phase sintering takes place under the uniaxial load of 6 kN [19]. With increasing the sintering temperature, the density of samples improves and pores disappear, resulting in the function of rarefied gas ionization and discharge debases even disappear finally. The samples with nearly full density are achieved by Joule heating and the uniaxial load. The pulse current uniformly passes through the powder and the reaction of Nb and Si simultaneously takes place everywhere. Furthermore, the reactant powders satisfactorily vibrated by the pulse current wallop and intermission discharge pressure leads the particle phase to take on uniform distribution [15]. Therefore, the dispersed microstructures of Nb/ Nb5Si3 composites can be obtained.
Fig. 1. Microstructures of Nb/Nb5Si3 composites (a) Nb–20Si, (b) Nb–20Si–10W, (c) Nb–20Si–10Mo, and (d) Nb–20Si–10W–10Mo.
70
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
The microstructures of Nb–20Si are indicated in Fig. 1(a). It is evident that the microstructures include small amount of primary Nb and eutectic mixtures of Nb and Nb5Si3. By comparing Fig. 1 (b) and (c) with Fig. 1(a), it is evident that the effects of W and Mo on the constituent phases in microstructure of Nb/Nb5Si3 composites are not obvious and the constituent phases is not changed, while by alloying with W and Mo together, the microstructure of Nb/Nb5Si3 composites changed significantly in Fig. 1(d). EPMA results of point 1 and point 2 in Nb–20Si–10W sample are given in Table 1, which indicate that the particle phase in Fig. 1(b) (i.e. point 1) contains Nb and W, and the matrix (point 2) contains both Nb and Si. The XRD pattern of Nb–20Si–10W sample is shown in Fig. 2, individual phase was identified by matching the characteristic XRD peaks against JCPDS data. It is obvious from Fig. 2 that the peaks of Nb and Nb5Si3 are detected but not peaks of W, Si and Nb3Si. Thus, combining the EPMA results with the analyzed results of XRD (Fig. 2), it is shown that the particle phases (point 1) are the solid solution of W in primary Nb and the matrix is the solid solution of W in Nb5Si3 phase. Mo and W are strong solid solution strengtheners of Nb and improve its high temperature strength. Atomic radii of Nb and W in pure substances are 0.143 nm and 0.136 nm, respectively. The atomic radii are very close, showing the Nb can be substituted by W. Many investigations have indicated that W element as the solid solution strengthening can been dissolved into Nb/Nb5Si3 composites [20,21]. The content of W in Nb5Si3 phase is small, indicating W tends to concentrate in Nb phase, as also obtained by Sha [13]. In this study, the W phase was not found, which indicates W element was dissolved completely, so the microstructure of Nb/Nb5Si3 composites alloyed with W is not changed. The microstructures of Nb–20Si–10Mo sample are indicated in Fig. 1(c), and the XRD pattern of Nb–20Si–10Mo sample is shown in Fig. 3. Only the peaks of Nb and Nb5Si3 are detected in Fig. 3, combining the EPMA results of points 3 and 4 (in Table 1), the
Table 1 EPMA results of phases. Location
Point Point Point Point Point Point Point Point
1 2 3 4 5 6 7 8
Element (at%) Nb
Si
W
Mo
81.34 73.42 60.57 72.33 81.03 72.58 60.45 7 3.52 100 30.81 7 2.02 62.277 1.87 68.747 1.06
– 37.687 2.42 – 37.767 2.08 – – 37.737 1.09 –
18.66 71.08 1.75 7 0.09 – – – 69.197 2.31 – 6.29 7 0.45
– – 18.977 1.26 1.79 7 0.11 – – – 24.977 1.41
Fig. 2. XRD pattern of Nb–20Si–10W sample.
Fig. 3. XRD pattern of Nb–20Si–10Mo sample.
Fig. 4. XRD pattern of Nb–20Si–10W–10Mo sample.
particle phases (point 3) are the solid solution of Mo in primary Nb and points 4 is the solid solution of Mo in Nb5Si3 phase. Atomic radius of Mo in pure substances is 0.140 nm. The atomic radius of Mo is closer to that of Nb than W, indicating Mo (10 at%) is easier to dissolve into Nb/Nb5Si3 composites. Recently, Kim et al. have investigated in detail a partial isothermal section of Nb–Si–Mo system at 1973 K [1,14]. According to their result, Mo is preferentially dissolved in Nb solid solution than Nb5Si3 phase in the twophase Nb/Nb5Si3 composites. Therefore, the concentration of Mo is higher in Nb solid solution than Nb5Si3 phase. The microstructure of Nb–20Si–10W–10Mo sample by SPS is given in Fig. 1(d). Because of the solid solution of W and Mo together, the microstructure of Nb–20Si–10W–10Mo sample is different from other samples. In order to further analyze chemical species, the four points in Fig. 1(d) were analyzed using electronprobe micro-analysis (EPMA), as shown in Table 1. It is indicated that the phase (point 5) is single Nb phase, point 6 contains Nb (30.81 at%) and W (69.19 at%), point 7 comprises Nb (62.27 at%) and Si (37.73 at%), and point 8 includes Nb (68.74 at%), Mo (24.97 at%) and W (6.29 at%). The XRD pattern of Nb–20Si–10W–10Mo sample is presented in Fig. 4. It is evident from Fig. 4 that the peaks of Nb5Si3, Nb and W are found but no peaks of Mo and Si. Thus, combining the EPMA results with the analyzed results of XRD (Fig. 4), the microstructures of Nb–20Si–10W–10Mo sample consist of the single Nb phase, Nb5Si3 phase, the solid solution of Nb atoms in W and the solid solution of Mo and W in Nb. It is obvious that no W or Mo element solid solution in Nb5Si3 phases (point 7) was detected in Nb–20Si–10W–10Mo sample, although the solid solution of W and Mo in Nb5Si3 phases was detected in Nb–20Si–10W and Nb–20Si–10Mo sample, respectively. By alloying with W and Mo together, why W and Mo cannot form solid solution in Nb5Si3 phases will be considered in detail in our future study.
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
Furthermore, the almost identical size of primary Nb is presented in Fig. 1, but the eutectic mixtures may be categorized as coarse and fine, based on the size of Nb phase present in eutectic mixtures. The eutectic solidification occurs in a ternary alloy system over a range of temperatures. The coarse eutectic is likely to form at higher temperature, while the fine eutectic forms towards the end of the solidification process at relatively lower temperatures [13]. The fine eutectic was found in Fig. 1(a)–(c), but only coarse eutectic was detected obviously in Fig. 1(a). In addition, the fine eutectic in Fig. 1(c) is finer than that in Fig. 1(b). From the obtained results above, it is indicated that the temperature of eutectic reaction in Nb/Nb5Si3 composites can be decreased by alloying with Mo or W, and it seems Mo element has more effect in decreasing the temperature of eutectic reaction than W element. The refinement of eutectic Nb phase by adding appropriate Mo and W element was also obtained by Kim [14] and Ma [20], respectively. The lamellae eutectic was observed in Fig. 1(a)–(c), but not presented in Fig. 1(d). The eutectic takes on the small granular Nb with about 5–10 μm in size (indicated by arrows) surrounded by Nb5Si3 phase in Fig. 1(d), and the large observation scope in Nb–20Si–10W–10Mo sample for surveying the eutectic was given in Fig. 5. The reason is uncertain in the present study, but one possible explanation is the effects of Mo and W together on thermodynamics or kinetics of eutectic reaction during solidification, and alloy element affects the thermodynamics or kinetics of solidification reaction in Nb/Nb5Si3 composites to change the microstructures which is also reported in previous literatures [7,22]. As hypereutectic Nb–20Si composition, the small amount of primary Nb phase was detected in Fig. 1, which is different from the microstructure obtained in previous literatures [7,14]. This phenomenon may be due to the Si content being lower than the eutectic composition at local. The size of Si particle is bigger than Nb particle, which will induce the inevitable segregation of Nb particle among Si particles in the mixed raw materials. The molten Si is beneficial to the homogenization of compositions during SPS, but the sintering time is very short about 10 min, resulting in Si content being lower than the eutectic composition at local. Therefore, the small amount of primary Nb phase was detected in Fig. 1. 3.2. Mechanical properties 3.2.1. Microhardness Table 2 shows the variation of microhardness of Nb phase and Nb5Si3 phases. The microhardness of Nb and Nb5Si3 phases changes obviously by alloying. The microhardness of Nb phases solid solution of W, Mo and W–Mo increase obviously by about 39.35%, 36.66% and 53.17%, respectively, which can be explained on the basis of solid solution strengthening. The atomic radii are very close, indicating the Nb can be substituted by Mo and W.
Fig. 5. The large observation scope in Nb–20Si–10W–10Mo sample.
71
Table 2 Microhardness (HV) of Nb and Nb5Si3 phases. Materials
Nb
Nbss
Nb5Si3
Nb–20Si Nb–20Si–10W Nb–20Si–10Mo Nb–20Si–10W–10Mo
521 76 – – 532 7 5
– 7267 6 7127 8 798 710
14377 11 1783 7 13 17607 16 14337 8
Mo and W substitutes Nb in the crystal, resulting in the decrease of lattice parameter of Nb solid solution and deformation of lattice [1]. Therefore, the microhardness of Nb phases with the solid solution of W, Mo and W–Mo increase obviously. The microhardness of W solid solution in Nb phases (point 1) is slight higher than that of Mo (point 3), although the concentration of W is slightly lower than that of Mo. This phenomenon can be explained based on the fact that atomic radius of W is smaller than that of Mo leading to greater deformation of lattice. By alloying W–Mo, the microhardness of Nb phase solid solution of W and Mo together increases greatly. Because of the solid solution of W and Mo together, the concentration of solute (in point 8) increases significantly, therefore, the microhardness increases greatly. The microhardness of Nb5Si3 phases in Nb–20Si–10W and Nb–20Si– 10Mo samples obvious increase respectively, but that in Nb–20Si– 10W–10Mo sample is almost unaltered. This phenomenon is because solid solution hardening for Nb5Si3 phase does not appear in Nb–20Si–10W–10Mo sample.
3.2.2. Tensile property The tensile stress–strain curves of the Nb/Nb5Si3 composite prepared in this study are presented in Fig. 6. From Fig. 6, the highest tensile strength of Nb/Nb5Si3 composite with same composition was obtained at room temperature, but these composites fracture before yielding, which are different from the fracture at 1573 K and 1773 K. The tensile stress–strain curves show a peak value followed by a gradual decrease, and the composite fractures at a stress of about 300 MPa and 100 MPa at 1573 K and 1773 K, respectively. The tensile stress peak value decreases continuously with the increase of test temperature, meanwhile the elongation increases markedly. The Nb–20Si–10W–10Mo sample takes on the highest tensile strength and best plastic deformability, either at room temperature or at high temperature. The tensile strength and plastic deformability of Nb–20Si–10Mo sample are only less than that of Nb–20Si–10W–10Mo sample, and the tensile strength and plastic deformability of Nb–20Si sample are the worst. The plastic deformability of Nb–20Si–10W–10Mo sample reaches about 16.25% and 21.31% at 1573 K and 1773 K, respectively. As we know, the mechanical property is relevant mainly with phase composition and size of microstructure. The contents of constituent phases including primary Nb, eutectic Nb and Nb5Si3 phase and interlamellar spacing of eutectic were measured by statistical quantitative metallographic analysis [23], as given in Table 3. The contents of Nb phase (including primary and eutectic Nb) and Nb5Si3 phase are almost same in Nb–20Si, Nb–20Si–10W and Nb–20Si–10Mo samples, but the interlamellar spacing of eutectic is obviously different. Previous literatures have indicated that interlamellar spacing of eutectic plays an important role in mechanical properties [7,22]. Although the constituent phases and its content in Nb–20Si, Nb–20Si–10W and Nb–20Si–10Mo samples are similar, the tensile strength is obviously different. This phenomenon may be explained on the basis of interlamellar spacing of eutectic and solid solution strengthening. The Nb–Nb5Si3 interfaces in the eutectic act as obstacles to dislocation motion, indicating that interlamellar spacing plays an important role in
72
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
that of Mo, which will lead to the greater strengthening effect mentioned above. However, the tensile strength of Nb–20Si–10W is lower than that of Nb–20Si–10Mo sample. This may be because interlamellar spacing has greater impact on tensile strength than solid solution strengthening of W in this study. Therefore, the tensile strength of Nb–20Si–10Mo is higher than that of Nb–20Si–10W sample. Nb–20Si–10W–10Mo sample has the highest tensile strength. This may be mainly attributed to its microstructure. The eutectic takes on the small granular Nb surrounded by Nb5Si3 phase in Nb– 20Si–10W–10Mo sample (Fig. 1(d)), which is obviously different with interlamellar eutectic, and the distribution of eutectic Nb is uniform from Fig. 5. This type of microstructure is more conducive to increase interface in eutectic than interlamellar eutectic. Therefore, the tensile strength of Nb–20Si–10W–10Mo sample is the highest. 3.2.3. Status of tensile property of Nb/Nb5Si3 composites The highest tensile strength of Nb–20Si–10W–10Mo sample is 587.687 12.5 MPa, 413.65 78.7 MPa and 24675.2 MPa at room temperature, 1573 K and 1773 K, respectively. The investigation on the high temperature tensile strength of Nb/Nb5Si3 composites is rare compared with compression strength. From available literature data, these tensile strengths are slightly higher than that of Nb–18Si–5Mo–5Hf alloy [24] and Nb–16Si–2Fe alloy [25] at same test temperature, but the deformability is far worse than that of Nb–18Si–5Mo–5Hf alloy and Nb–16Si–2Fe alloy. The Nb/Nb5Si3 composite fabricated by SPS takes on excellent high temperature tensile strength. However, its deformability at room temperature is very low. This will hinder its extensive application. In order to develop Nb–Si based structural materials, the refinement of microstructure and controlling of interface will be an important concern.
4. Conclusions In this work, using Nb–20Si, Nb–20Si–10W, Nb–20Si–10Mo and Nb–20Si–10W–10Mo elemental powder mixtures as raw materials, the microstructures and mechanical properties of Nb/Nb5Si3 composites alloyed with W, Mo and W–Mo fabricated by spark plasma sintering were investigated. The conclusions of this work are as follows:
Fig. 6. Tensile strength of Nb/Nb5Si3 composites (a) at room temperature, (b) 1573 K, and (c) 1773 K.
Table 3 Volume fraction of constituent phases and interlamellar spacing. Materials
Nb–20Si Nb–20Si–10W Nb–20Si–10Mo Nb–20Si–10W–10Mo
Volume fraction (%) Primary Nb
Nb5Si3
Eutectic Nb
W
Interlamellar spacing of eutectic (μm)
10.08 10.86 11.07 7.87
50.24 49.58 49.08 48.68
39.68 40.06 39.85 37.82
– – – 5.63
1.357 0.75 0.95 7 0.15 0.83 7 0.16 –
(1) The microstructures Nb/Nb5Si3 composites alloyed with W and Mo, respectively, include primary Nb and eutectic mixtures of Nb and Nb5Si3, and the coarse and fine eutectic mixtures are detected. The W or Mo elements solid solution in Nb and Nb5Si3 phase are detected. (2) The microstructures of Nb/Nb5Si3 composites alloyed with W and Mo together obviously change, including Nb phase, the solid solubility phases of W and Mo atoms in Nb, and the solid solubility phases of Nb atoms in W are also found, but the solid solubility phenomena of Nb5Si3 phases is not detected. (3) The excellent high temperature tensile strength of Nb–20Si– 10W–10Mo sample was obtained. It is attributable to granular Nb surrounded by Nb5Si3 phase in eutectic, which is different with interlamellar eutectic.
Acknowledgements its deformation behavior. Therefore, with the decrease of interlamellar spacing, the tensile strength increases. Otherwise, the Nb phase and Nb5Si3 phase solid solution of Mo and W will bring strengthening effect, as a result of elastic interaction with solute atoms of different sizes. The atomic radius of W is smaller than
This research was financially supported by Nanchang Hangkong University Foundation of China under Grant no. EC200701021, and Jiangxi Provincial Department of Science and Technology Foundation of China under Grant no. DB200701049.
B. Xiong et al. / Materials Science & Engineering A 606 (2014) 68–73
References [1] W.Y. Kim, H. Tanaka, S. Hannda, Intermetallics 10 (2002) 625–634. [2] H.Z. Fu, J. Aeronaut. Mater 18 (1998) 52–60. [3] J.L. Yu, Z.K. Li, K.F. Zhang, X. Zheng, H. Liu, H. Wang, W.S. Wang, Fail. Anal. Prev. 4 (2010) 231–234. [4] C.L. Yeh, W.H. Chen, J. Alloys Compd. 425 (2006) 216–222. [5] Z. Chen, Y.W. Yan, J. Alloys Compd. 413 (2006) 73–76. [6] J. Geng, P. Tsakiropoulos, G. Shao, Intermetallics 15 (2007) 69–76. [7] K. Chattopadhyay, G. Balachandran, R. Mitra, K.K. Ray, Intermetallics 14 (2006) 1452–1460. [8] P.R. Subramanian, M.G. Mendiratta, D.M. Dimiduk, M.A. Stuck, Mater. Sci. Eng. A 239 (1997) 1–3. [9] K.S. Chan, D.L. Davidson, Mater. Sci. Eng. A 30 (1999) 925–939. [10] K.S. Chan, Mater. Sci. Eng. A (2002) 513–522. [11] K.S. Chan, D.L. Davidson, Mater. Sci. Eng. A 34 (2003) 1833–1849. [12] J. Geng, P. Tsakiropoulos, G. Sha, Intermetallics 14 (2006) 227–235. [13] J. Sha, H. Hirai, T. Tabaru, A. Kitahara, H. Ueno, S. Hanada, Metall. Mater. Trans. A 34 (2003) 2861–2871.
73
[14] W.Y. Kim, H. Tanaka, A. Kasama, S.J. Hanada, Intermetallics 9 (2001) 827–834. [15] M.E. Reiss, C.M. Esber, D. Van Heerden, A.J. Gavens, M.E. Williams, T.P. Weihs, Mater. Sci. Eng. A 261 (1999) 217–222. [16] B.B. Fernandes, G. Rodrigues, G. Silva, E.C.T. Ramos, C.A. Nunes, H.R.Z. Sandim, A.S. Ramos, J. Alloys Compd. 434–435 (2007) 530–534. [17] O. Mamoru, Mater. Sci. Eng. A 287 (2000) 183–188. [18] P.B. Wang, G.J. Yang, X.N. Mao, Rare Metal. Mater. Eng. 36 (2007) 484–488. [19] Z. Chen, Y.W. Yan, Rare Metal. Mater. Eng. 35 (2006) 484–487. [20] C.L. Ma, J.G. Li, Y. Tan, R. Tanaka, S. Hanada, Mater. Sci. Eng. A 386 (2004) 375–383. [21] I. Grammenos, P. Tsakiropoulos, Intermetallics 19 (2011) 1612–1621. [22] S. Kashyap, C.S. Tiwary, K. Chattopadhyay, Intermetallics 19 (2011) 1943–1952. [23] R.T. DeHoff, F.N. Rhines, Quantitative Microscopy, first ed., McGraw-Hill, Inc., New York, 1968. [24] J.H. Kim, T. Tabaru, M. Sakamoto, S. Hanada, Mater. Sci. Eng. A 372 (2004) 137–144. [25] J.L. Yu, Z.K. Li, K.F. Zhang, X. Zheng, J.J. Zhang, R. Bai, W.S. Wang, Mater. Sci. Eng. A 527 (2010) 5230–5233.