- Email: [email protected]
Comparative study of sinterability and thermal stability in plasma-sintered niobium and vanadium beryllides
Journal of Alloys and Compounds 638 (2015) 277–281
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Comparative study of sinterability and thermal stability in plasma-sintered niobium and vanadium beryllides Jae-Hwan Kim ⇑, Masaru Nakamichi Breeding Functional Materials Development Group, Department of Blanket Systems Research, Rokkasho Fusion Institute, Fusion Research and Development Directorate, Japan Atomic Energy Agency, 2 – 166 Oaza-Obuchi-Aza-Omotedate, Rokkasho-mura, Kamikita-gun, Aomori 039-3212, Japan
a r t i c l e
i n f o
Article history: Received 3 February 2015 Received in revised form 2 March 2015 Accepted 13 March 2015 Available online 18 March 2015 Keywords: Beryllide Thermal stability Plasma sintering Be12Nb Be12V
a b s t r a c t Niobium and vanadium–beryllium intermetallic compounds (beryllides) were synthesized by plasma sintering under different sintering times at 1273 K. The beryllide with 7.7 at.% Nb mainly consisted of various phases of Be, Be12Nb, Be17Nb2, and Be2Nb, whereas that with 7.7 at.% V consisted of Be12V, Be2V, and V. As the sintering time increased, area fractions of the target compositions Be12Nb and Be12V increased while that of Be decreased. A comparative analysis demonstrated that the beryllide with 7.7 at.% Nb a showed higher density as well as a greater hardness than that with 7.7 at.% V, due to there being less difference between the sintering temperature and the melting point. In terms of thermal phase stability, the beryllide with 7.7 at.% Nb showed good thermal phase stability with fewer pores and a smaller unhomogenized area, because the beryllide contained a smaller area fraction of the Be phase, which may cause evaporation resulting in pore formation. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Beryllium intermetallic compounds (beryllides) have attracted a great deal of attention as refractory materials and advanced neutron multipliers for use in the demonstration (DEMO) fusion reactor, owing to their high stability at high temperature. However, there is no doubt that the beryllides are too brittle to be fabricated in small size. As a part of the Broader Approach activities from 2007 to 2016 on DEMO R&D for the International Fusion Energy Research Centre (IFERC) project, our group has reported that the binary beryllium and titanium intermetallic compounds were easily and successfully synthesized by a plasma-sintering method [1–5]. In previous research on preliminary synthesis of V and Nb beryllides [6], X-ray diffraction analysis clearly depicted the syntheses of Be12V and Be12Nb compounds. Furthermore, beryllides have been reported for use as neutron multipliers in both form of a disk [7] and form of pebbles [8] with a diameter of 1 mm, in light of their mechanical properties and microstructural variation. In addition, new functional materials for use in the fusion field may be considered in addition to the already-existing ones, owing to high temperature stability. Bruemmer et al. [9] reported that hot isostatically-pressed Be12Nb and Be17Nb2 indicated reasonable strength at both high and low temperature as refractory material
⇑ Corresponding author. Tel.: +81 175 71 6537; fax: +81 175 71 6502. E-mail address: [email protected] (J.-H. Kim). http://dx.doi.org/10.1016/j.jallcom.2015.03.105 0925-8388/Ó 2015 Elsevier B.V. All rights reserved.
application. Furthermore, study on the superconducting property of BeNb3 has been reported with a structural evaluation by Tuleushev et al. [10]. With regards to Be–V intermetallic compounds, Kurinskiy et al. [11] reported that Be12V intermetallic compounds indicated similar brittle behavior to that which caused the failure of Be12Ti as a neutron-multiplying material. In the present study, Be12Nb and Be12V were synthesized by the plasma-sintering method with different sintering times to investigate their sinterability and thermal stability. In addition, to evaluate the thermal stability, an annealing test at 1473 K was carried out with Be12Nb and Be12V, and the result was explained with structural evolution.
2. Materials and method The binary beryllides were synthesized by the plasma-sintering method [5]. Beryllium, niobium, and vanadium powders with high purities of 99.5%, 99.9%, and 99.0%, and particle sizes of less than 45, 45, and 75 lm, respectively, were used. As shown in Fig. 1, the powders were mixed at concentrations of 92.3 at.% Be, 7.7 at.% Nb or V to produce Be12Nb and Be12V using a mortar (RM200, Retsch, Germany) for 1 h and then loading into a graphite punch and die for cold compact. Prior to sintering, an alternating current of 500 A was applied for 30 s to activate the powder surface. The sintering was conducted at 1273 K for holding times of 5 min, 20 min, and 60 min with heating and cooling rates of 100 K/min and 200 K/min, respectively. For characterization, the beryllides were cut into 3 3 5 mm samples and polished up to 15 lm with SiC sand paper. To investigate the sintering density of the beryllide including open porosity, the Archimedes immersion method and a He gas pycno-meter (AccupycII 1340-1CC, Shimadzu, Japan) were used. Sintering
278
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
Fig. 1. Phase diagrams of Be–Nb (left) and Be–V (right).
Fig. 2. X-ray diffraction profiles of the beryllides sintered for 5, 20, and 60 min.
Fig. 3. SEM images of the niobium beryllides plasma-sintered at 1273 K for (a) 5 min, (b) 20 min, and (c) 60 min.
density was calculated by theoretical density based on the area fraction of each phase, and the apparent density measured by the immersion method with open porosity was evaluated by densities measured by immersion and He pycno-meter methods. The intermetallic phases of the beryllides were observed using an electron probe micro-analyzer (EPMA, JXA-8530F, JEOL, Japan) with electron back-scatter for the point analysis of each phase. To confirm the composition of the each phase, X-ray diffraction analysis (UltimaIV, Rigaku, Japan) and point analysis were carried out by the EPMA. The area fractions of each phase in the beryllides, were calculated by an image-analyzing program (ImageJ 1.44p, National Institutes of Health, USA). To evaluate the mechanical properties, the Vickers micro-hardness (HM-221, Mitsutoyo, Japan) was measured at 16 points with a pressure of 1 kgf/mm2. In addition, the samples were heated to 1473 K for 5 h (KEF-1600, Kaken, Japan) and observed by the EPMA to investigate the thermal stability of the beryllides.
3. Results and discussion To identify the phases in each beryllide sample, X-ray diffraction measurements were performed. Fig. 2 shows X-ray diffraction results for Be–7.7 at.% Nb (left) and Be–7.7 at.% V (right). The qualitative X-ray diffraction results implied that the Be–Nb beryllide samples consisted of four phases, Be12Nb, Be17Nb2, Be2Nb, and
Be, in a sample sintered for 5 min, and that as the sintering time increased, Be17Nb2, Be2Nb, and Be decreased whereas Be12Nb increased. Conversely, with regards to Be–V beryllide samples, Xray diffraction profiles confirmed the identification of four phases, Be12V, Be2V, Be, and V, implying that with increasing sintering time, Be2V, Be, and V peaks have a tendency to decrease while Be12V peaks relatively increase. This identification is in good agreement with phase diagrams (see Fig. 1). For quantitative evaluation of each phase, surface observation and image analysis were conducted. SEM images of the beryllides synthesized from Be–7.7 at.% Nb by plasma sintering are shown in Fig. 3, which shows that the beryllide samples consisted of four different contrasting hues: black, gray, light gray, and white, which corresponded to Be, Be12Nb, Be17Nb2, and Be2Nb, respectively (see (b)). In the case of beryllide sintered for 5 min, larger fractions of the Be and Be2Nb phases were detected, while the Be12Nb phase, which is target composition, was successfully synthesized. With regards to phase evolution according to sintering time, it was obvious that, with increased sintering time, the area fraction of the Be12Nb phase increased. The area fractions calculated by image
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
279
Fig. 4. Variation of the area fraction of each phase in niobium beryllides plasma-sintered over different amounts of time.
Fig. 5. SEM images of the vanadium beryllides plasma-sintered at 1273 K for (a) 5 min, (b) 20 min and (c) 60 min.
Fig. 6. Variation of area fraction of vanadium beryllide samples plasma-sintered for durations of 5 min, 20 min, and 60 min.
analysis program are given in Fig. 4, indicating that the beryllides Be12Nb, Be17Nb2, and Be2Nb were synthesized with a combined fraction of 90% even for a sintering time of 5 min, and that the fraction of those increased to 99% as the sintering time increased. Fig. 5 illustrates SEM micrographs of vanadium beryllides plasma-sintered at 1273 K for 5, 20, and 60 min. It indicates that the Be–7.7 at.% V beryllide samples similarly consisted of four different hues: black, gray, light gray, and white, which were identified with Be, Be12V, Be2V, and V, respectively based on the point analyses by the EPMA (see (b)). The area fractions of each phase were shown in Fig. 6, demonstrating that there seemed to be similar tendency to that of Be–7.7 at.% Nb, but the factions of beryllides Be12V and Be2V, were relatively small: 80% for the sample sintered for 60 min. In the previous study [12], it was reported that the area fraction of titanium beryllides sintered at 1273 K for 60 min was 92% in plasma-sintered Be–7.7 at.% Ti. This difference may have been caused by the fact that the lower melting point of Be12Nb (=1945 K, lower than 1973 K in case of Be12V) enabled easier diffusion and consolidation of the beryllides, indicating that the melting point of intermetallic compounds is the most important factor for diffusion in sintering processes. Density measurements of the plasma-sintered beryllides, Be–7.7 at.% Nb and Be–7.7 at.% V, were performed to confirm the sintering density and porosity as functions of sintering time using a He gas pycno-meter and the immersion method. Fig. 7 shows the sintering densities of the plasma-sintered beryllide samples and demonstrates that they were sintered with
relatively high densities with no specific tendency as a function of sintering time. As a comparative result, the sintering densities of Nb beryllide were higher than those of vanadium beryllide. This may be due to the different melting points of Be12Nb and Be12V. Although they were sintered at the same temperature, the sample with the higher melting point could be consolidated with lower density. In this study, it is thought that the densities of Nb beryllides were higher than those of V beryllides because Be12Nb had a lower melting point. This is in good agreement with the
Fig. 7. Sintering density of beryllide samples.
280
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
Fig. 8. Vickers micro-hardnesses of beryllides.
previous result [12], which demonstrated that the sintering densities of Be–7.7 at.% Ti beryllides were relatively high regardless of different sintering times. To evaluate the mechanical properties, a hardness test was conducted with a random 9.8 N loading of 16 points and average for the beryllides, Be–7.7 at.% V and Be–7.7 at.% Nb. The Hv’s of Be17Nb2 and Be12Nb were 1050 and 500 [9], respectively, while that of Be12V was 1162 [13]. Fig. 8 reveals the Vickers micro-hardnesses of Be–Nb and Be–V as functions of sintering time. Moreover, the hardness had a tendency to increase along with the time. This was due to the fact that the fraction of the Be phase, which had a lower hardness than the beryllide phase, decreased. To compare the hardnesses of each phase, micrographs with indent and hardness were investigated and are given in Fig. 9, showing that the hardnesses of Be, Be12Nb, Be17Nb2, and Be2Nb were 484, 950, 887, and 720, respectively, while those of Be, Be12V, and Be2V were 420, 1110, and 750, respectively. In a previous study [4], the
Fig. 9. Hardnesses of each phase in (a) Be–7.7 at.% Nb and (b) Be–7.7 at.% V.
Fig. 10. SEM micrographs of niobium beryllides sintered for (a) 20 min and (b) 60 min, and vanadium beryllides sintered for (c) 20 min and (d) 60 min after heat treatment at 1473 K for 5 h.
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
hardness of each phase in Ti beryllides was evaluated, showing that the hardnesses of Be12Ti, Be17Ti2, Be2Ti, and Be were 1260, 1380, 900, and 550, respectively. With regards to thermal stability, homogenization treatments at 1473 K were carried out for 5 h using the Be–7.7 at.% V and Be–7.7 at.% Nb samples plasma-sintered for 20 and 60 min. SEM surface observations after the treatment are given in Fig. 10; (a) Be–Nb beryllide sintered for 20 min, (b) Be–Nb beryllide sintered for 60 min, (c) Be–V beryllide sintered for 20 min, and (d) Be–V beryllide sintered for 60 min. The treatment led to homogenization of Be12V and Be12Nb with small portions of unhomogenized Be17Nb2 and Be2V. However, there existed many cracks and pores in Be12Nb and Be12V, respectively. It is thought that cracks in Be12Nb (see Fig. 10(a) and (b)) were formed due to different thermal expansions for each phase, because the cracks were mainly detected at phase boundaries Be12Nb and Be17Nb2. According to the previous studies [14] on homogenization of Be– Ti beryllide, several cracks and pores were identified by the treatment. Conversely, heat-treated Be–V beryllides indicated pore formation with large fractions while unhomogenized phases corresponding to Be2V remained. The formation of these pores may have been due to evaporation of the Be phase, which primarily existed with a large fraction. 4. Conclusion Plasma-sintered beryllides with niobium and vanadium were fabricated. SEM observation and X-ray diffraction analysis revealed that Be–7.7 at.% Nb consisted of four phases, Be, Be12Nb, Be17Nb2, and Be2Nb, while Be–7.7 at.% V consisted of Be, Be12V, Be2V, and V, varying with different sintering times at 1273 K. This is in good agreement with phase diagrams. As the sintering time increased, the area fractions of the target compositions of either Be12Nb or
281
Be12V increased. Because this phase variation was closely associated with the melting point, consolidation of Be12Nb, which has lower melting point than Be12V proceeded even though the sintering of samples was conducted at the same temperature. This melting point difference influenced a variation of density, with the result that a sample with a higher melting point could be consolidated with a lower relative density. Hardness increased as a function of the sintering time; this was because the increased sintering time led to consolidation of the beryllide with higher hardness. SEM observation showed that the beryllide samples annealed at 1473 K for 5 h were homogenized, but there were cracks and pores. Accordingly, variation of phase composition seemed to be considerably dependent on differing sintering temperatures and melting points. References [1] M. Nakamichi, K. Yonehara, J. Nucl. Mater. 417 (2011) 765–768. [2] M. Nakamichi, J.-H. Kim, D. Wakai, K. Yonehara, Fus. Eng. Des. 87 (2012) 896–899. [3] J.-H. Kim, M. Nakamichi, J. Nucl. Mater. 438 (2013) 218–223. [4] J.-H. Kim, M. Nakamichi, J. Alloys Comp. 546 (2013) 171–175. [5] J.-H. Kim, M. Nakamichi, J. Alloys Comp. 577 (2013) 90–96. [6] M. Nakamichi, K. Yonehara, D. Wakai, Fus. Eng. Des. 86 (9–11) (2011) 2262–2264. [7] J.-H. Kim, M. Nakamichi, J. Alloys Comp. 556 (2013) 292–295. [8] M. Nakamichi, J.-H. Kim, K. Yonehara, Fus. Eng. Des. 88 (2013) 611–615. [9] S.M. Bruemmer, J.L. Brimhall, C.H. Henager Jr., J.P. Hirth, Mater. Res. Soc. 288 (1992) 799. [10] A.Z. Tuleushev, V.N. Volodin, Y.Z. Tuleushev, J. Exp. Theor. Phys. Lett. 78 (7) (2003) 440–442. [11] P. Kurinskiy, V. Chakin, A. Moeslang, R. Rolli, E. Alves, L.C. Alves, N. Franco, Ch. Dorn, A.A. Goraieb, Fus. Eng. Des. 86 (9–11) (2011) 2454–2457. [12] J.-H. Kim, M. Nakamichi, J. Nucl. Mater. 442 (2013) S461–S464. [13] T.G. Nieh, J. Wadsworth, F.C. Grensing, J.-M. Yang, J. Nucl. Mater. 27 (1992) 2660–2664. [14] J.-H. Kim, M. Nakamichi, Fus. Eng. Des. 88 (2013) 2215–2218.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jalcom
Comparative study of sinterability and thermal stability in plasma-sintered niobium and vanadium beryllides Jae-Hwan Kim ⇑, Masaru Nakamichi Breeding Functional Materials Development Group, Department of Blanket Systems Research, Rokkasho Fusion Institute, Fusion Research and Development Directorate, Japan Atomic Energy Agency, 2 – 166 Oaza-Obuchi-Aza-Omotedate, Rokkasho-mura, Kamikita-gun, Aomori 039-3212, Japan
a r t i c l e
i n f o
Article history: Received 3 February 2015 Received in revised form 2 March 2015 Accepted 13 March 2015 Available online 18 March 2015 Keywords: Beryllide Thermal stability Plasma sintering Be12Nb Be12V
a b s t r a c t Niobium and vanadium–beryllium intermetallic compounds (beryllides) were synthesized by plasma sintering under different sintering times at 1273 K. The beryllide with 7.7 at.% Nb mainly consisted of various phases of Be, Be12Nb, Be17Nb2, and Be2Nb, whereas that with 7.7 at.% V consisted of Be12V, Be2V, and V. As the sintering time increased, area fractions of the target compositions Be12Nb and Be12V increased while that of Be decreased. A comparative analysis demonstrated that the beryllide with 7.7 at.% Nb a showed higher density as well as a greater hardness than that with 7.7 at.% V, due to there being less difference between the sintering temperature and the melting point. In terms of thermal phase stability, the beryllide with 7.7 at.% Nb showed good thermal phase stability with fewer pores and a smaller unhomogenized area, because the beryllide contained a smaller area fraction of the Be phase, which may cause evaporation resulting in pore formation. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Beryllium intermetallic compounds (beryllides) have attracted a great deal of attention as refractory materials and advanced neutron multipliers for use in the demonstration (DEMO) fusion reactor, owing to their high stability at high temperature. However, there is no doubt that the beryllides are too brittle to be fabricated in small size. As a part of the Broader Approach activities from 2007 to 2016 on DEMO R&D for the International Fusion Energy Research Centre (IFERC) project, our group has reported that the binary beryllium and titanium intermetallic compounds were easily and successfully synthesized by a plasma-sintering method [1–5]. In previous research on preliminary synthesis of V and Nb beryllides [6], X-ray diffraction analysis clearly depicted the syntheses of Be12V and Be12Nb compounds. Furthermore, beryllides have been reported for use as neutron multipliers in both form of a disk [7] and form of pebbles [8] with a diameter of 1 mm, in light of their mechanical properties and microstructural variation. In addition, new functional materials for use in the fusion field may be considered in addition to the already-existing ones, owing to high temperature stability. Bruemmer et al. [9] reported that hot isostatically-pressed Be12Nb and Be17Nb2 indicated reasonable strength at both high and low temperature as refractory material
⇑ Corresponding author. Tel.: +81 175 71 6537; fax: +81 175 71 6502. E-mail address: [email protected] (J.-H. Kim). http://dx.doi.org/10.1016/j.jallcom.2015.03.105 0925-8388/Ó 2015 Elsevier B.V. All rights reserved.
application. Furthermore, study on the superconducting property of BeNb3 has been reported with a structural evaluation by Tuleushev et al. [10]. With regards to Be–V intermetallic compounds, Kurinskiy et al. [11] reported that Be12V intermetallic compounds indicated similar brittle behavior to that which caused the failure of Be12Ti as a neutron-multiplying material. In the present study, Be12Nb and Be12V were synthesized by the plasma-sintering method with different sintering times to investigate their sinterability and thermal stability. In addition, to evaluate the thermal stability, an annealing test at 1473 K was carried out with Be12Nb and Be12V, and the result was explained with structural evolution.
2. Materials and method The binary beryllides were synthesized by the plasma-sintering method [5]. Beryllium, niobium, and vanadium powders with high purities of 99.5%, 99.9%, and 99.0%, and particle sizes of less than 45, 45, and 75 lm, respectively, were used. As shown in Fig. 1, the powders were mixed at concentrations of 92.3 at.% Be, 7.7 at.% Nb or V to produce Be12Nb and Be12V using a mortar (RM200, Retsch, Germany) for 1 h and then loading into a graphite punch and die for cold compact. Prior to sintering, an alternating current of 500 A was applied for 30 s to activate the powder surface. The sintering was conducted at 1273 K for holding times of 5 min, 20 min, and 60 min with heating and cooling rates of 100 K/min and 200 K/min, respectively. For characterization, the beryllides were cut into 3 3 5 mm samples and polished up to 15 lm with SiC sand paper. To investigate the sintering density of the beryllide including open porosity, the Archimedes immersion method and a He gas pycno-meter (AccupycII 1340-1CC, Shimadzu, Japan) were used. Sintering
278
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
Fig. 1. Phase diagrams of Be–Nb (left) and Be–V (right).
Fig. 2. X-ray diffraction profiles of the beryllides sintered for 5, 20, and 60 min.
Fig. 3. SEM images of the niobium beryllides plasma-sintered at 1273 K for (a) 5 min, (b) 20 min, and (c) 60 min.
density was calculated by theoretical density based on the area fraction of each phase, and the apparent density measured by the immersion method with open porosity was evaluated by densities measured by immersion and He pycno-meter methods. The intermetallic phases of the beryllides were observed using an electron probe micro-analyzer (EPMA, JXA-8530F, JEOL, Japan) with electron back-scatter for the point analysis of each phase. To confirm the composition of the each phase, X-ray diffraction analysis (UltimaIV, Rigaku, Japan) and point analysis were carried out by the EPMA. The area fractions of each phase in the beryllides, were calculated by an image-analyzing program (ImageJ 1.44p, National Institutes of Health, USA). To evaluate the mechanical properties, the Vickers micro-hardness (HM-221, Mitsutoyo, Japan) was measured at 16 points with a pressure of 1 kgf/mm2. In addition, the samples were heated to 1473 K for 5 h (KEF-1600, Kaken, Japan) and observed by the EPMA to investigate the thermal stability of the beryllides.
3. Results and discussion To identify the phases in each beryllide sample, X-ray diffraction measurements were performed. Fig. 2 shows X-ray diffraction results for Be–7.7 at.% Nb (left) and Be–7.7 at.% V (right). The qualitative X-ray diffraction results implied that the Be–Nb beryllide samples consisted of four phases, Be12Nb, Be17Nb2, Be2Nb, and
Be, in a sample sintered for 5 min, and that as the sintering time increased, Be17Nb2, Be2Nb, and Be decreased whereas Be12Nb increased. Conversely, with regards to Be–V beryllide samples, Xray diffraction profiles confirmed the identification of four phases, Be12V, Be2V, Be, and V, implying that with increasing sintering time, Be2V, Be, and V peaks have a tendency to decrease while Be12V peaks relatively increase. This identification is in good agreement with phase diagrams (see Fig. 1). For quantitative evaluation of each phase, surface observation and image analysis were conducted. SEM images of the beryllides synthesized from Be–7.7 at.% Nb by plasma sintering are shown in Fig. 3, which shows that the beryllide samples consisted of four different contrasting hues: black, gray, light gray, and white, which corresponded to Be, Be12Nb, Be17Nb2, and Be2Nb, respectively (see (b)). In the case of beryllide sintered for 5 min, larger fractions of the Be and Be2Nb phases were detected, while the Be12Nb phase, which is target composition, was successfully synthesized. With regards to phase evolution according to sintering time, it was obvious that, with increased sintering time, the area fraction of the Be12Nb phase increased. The area fractions calculated by image
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
279
Fig. 4. Variation of the area fraction of each phase in niobium beryllides plasma-sintered over different amounts of time.
Fig. 5. SEM images of the vanadium beryllides plasma-sintered at 1273 K for (a) 5 min, (b) 20 min and (c) 60 min.
Fig. 6. Variation of area fraction of vanadium beryllide samples plasma-sintered for durations of 5 min, 20 min, and 60 min.
analysis program are given in Fig. 4, indicating that the beryllides Be12Nb, Be17Nb2, and Be2Nb were synthesized with a combined fraction of 90% even for a sintering time of 5 min, and that the fraction of those increased to 99% as the sintering time increased. Fig. 5 illustrates SEM micrographs of vanadium beryllides plasma-sintered at 1273 K for 5, 20, and 60 min. It indicates that the Be–7.7 at.% V beryllide samples similarly consisted of four different hues: black, gray, light gray, and white, which were identified with Be, Be12V, Be2V, and V, respectively based on the point analyses by the EPMA (see (b)). The area fractions of each phase were shown in Fig. 6, demonstrating that there seemed to be similar tendency to that of Be–7.7 at.% Nb, but the factions of beryllides Be12V and Be2V, were relatively small: 80% for the sample sintered for 60 min. In the previous study [12], it was reported that the area fraction of titanium beryllides sintered at 1273 K for 60 min was 92% in plasma-sintered Be–7.7 at.% Ti. This difference may have been caused by the fact that the lower melting point of Be12Nb (=1945 K, lower than 1973 K in case of Be12V) enabled easier diffusion and consolidation of the beryllides, indicating that the melting point of intermetallic compounds is the most important factor for diffusion in sintering processes. Density measurements of the plasma-sintered beryllides, Be–7.7 at.% Nb and Be–7.7 at.% V, were performed to confirm the sintering density and porosity as functions of sintering time using a He gas pycno-meter and the immersion method. Fig. 7 shows the sintering densities of the plasma-sintered beryllide samples and demonstrates that they were sintered with
relatively high densities with no specific tendency as a function of sintering time. As a comparative result, the sintering densities of Nb beryllide were higher than those of vanadium beryllide. This may be due to the different melting points of Be12Nb and Be12V. Although they were sintered at the same temperature, the sample with the higher melting point could be consolidated with lower density. In this study, it is thought that the densities of Nb beryllides were higher than those of V beryllides because Be12Nb had a lower melting point. This is in good agreement with the
Fig. 7. Sintering density of beryllide samples.
280
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
Fig. 8. Vickers micro-hardnesses of beryllides.
previous result [12], which demonstrated that the sintering densities of Be–7.7 at.% Ti beryllides were relatively high regardless of different sintering times. To evaluate the mechanical properties, a hardness test was conducted with a random 9.8 N loading of 16 points and average for the beryllides, Be–7.7 at.% V and Be–7.7 at.% Nb. The Hv’s of Be17Nb2 and Be12Nb were 1050 and 500 [9], respectively, while that of Be12V was 1162 [13]. Fig. 8 reveals the Vickers micro-hardnesses of Be–Nb and Be–V as functions of sintering time. Moreover, the hardness had a tendency to increase along with the time. This was due to the fact that the fraction of the Be phase, which had a lower hardness than the beryllide phase, decreased. To compare the hardnesses of each phase, micrographs with indent and hardness were investigated and are given in Fig. 9, showing that the hardnesses of Be, Be12Nb, Be17Nb2, and Be2Nb were 484, 950, 887, and 720, respectively, while those of Be, Be12V, and Be2V were 420, 1110, and 750, respectively. In a previous study [4], the
Fig. 9. Hardnesses of each phase in (a) Be–7.7 at.% Nb and (b) Be–7.7 at.% V.
Fig. 10. SEM micrographs of niobium beryllides sintered for (a) 20 min and (b) 60 min, and vanadium beryllides sintered for (c) 20 min and (d) 60 min after heat treatment at 1473 K for 5 h.
J.-H. Kim, M. Nakamichi / Journal of Alloys and Compounds 638 (2015) 277–281
hardness of each phase in Ti beryllides was evaluated, showing that the hardnesses of Be12Ti, Be17Ti2, Be2Ti, and Be were 1260, 1380, 900, and 550, respectively. With regards to thermal stability, homogenization treatments at 1473 K were carried out for 5 h using the Be–7.7 at.% V and Be–7.7 at.% Nb samples plasma-sintered for 20 and 60 min. SEM surface observations after the treatment are given in Fig. 10; (a) Be–Nb beryllide sintered for 20 min, (b) Be–Nb beryllide sintered for 60 min, (c) Be–V beryllide sintered for 20 min, and (d) Be–V beryllide sintered for 60 min. The treatment led to homogenization of Be12V and Be12Nb with small portions of unhomogenized Be17Nb2 and Be2V. However, there existed many cracks and pores in Be12Nb and Be12V, respectively. It is thought that cracks in Be12Nb (see Fig. 10(a) and (b)) were formed due to different thermal expansions for each phase, because the cracks were mainly detected at phase boundaries Be12Nb and Be17Nb2. According to the previous studies [14] on homogenization of Be– Ti beryllide, several cracks and pores were identified by the treatment. Conversely, heat-treated Be–V beryllides indicated pore formation with large fractions while unhomogenized phases corresponding to Be2V remained. The formation of these pores may have been due to evaporation of the Be phase, which primarily existed with a large fraction. 4. Conclusion Plasma-sintered beryllides with niobium and vanadium were fabricated. SEM observation and X-ray diffraction analysis revealed that Be–7.7 at.% Nb consisted of four phases, Be, Be12Nb, Be17Nb2, and Be2Nb, while Be–7.7 at.% V consisted of Be, Be12V, Be2V, and V, varying with different sintering times at 1273 K. This is in good agreement with phase diagrams. As the sintering time increased, the area fractions of the target compositions of either Be12Nb or
281
Be12V increased. Because this phase variation was closely associated with the melting point, consolidation of Be12Nb, which has lower melting point than Be12V proceeded even though the sintering of samples was conducted at the same temperature. This melting point difference influenced a variation of density, with the result that a sample with a higher melting point could be consolidated with a lower relative density. Hardness increased as a function of the sintering time; this was because the increased sintering time led to consolidation of the beryllide with higher hardness. SEM observation showed that the beryllide samples annealed at 1473 K for 5 h were homogenized, but there were cracks and pores. Accordingly, variation of phase composition seemed to be considerably dependent on differing sintering temperatures and melting points. References [1] M. Nakamichi, K. Yonehara, J. Nucl. Mater. 417 (2011) 765–768. [2] M. Nakamichi, J.-H. Kim, D. Wakai, K. Yonehara, Fus. Eng. Des. 87 (2012) 896–899. [3] J.-H. Kim, M. Nakamichi, J. Nucl. Mater. 438 (2013) 218–223. [4] J.-H. Kim, M. Nakamichi, J. Alloys Comp. 546 (2013) 171–175. [5] J.-H. Kim, M. Nakamichi, J. Alloys Comp. 577 (2013) 90–96. [6] M. Nakamichi, K. Yonehara, D. Wakai, Fus. Eng. Des. 86 (9–11) (2011) 2262–2264. [7] J.-H. Kim, M. Nakamichi, J. Alloys Comp. 556 (2013) 292–295. [8] M. Nakamichi, J.-H. Kim, K. Yonehara, Fus. Eng. Des. 88 (2013) 611–615. [9] S.M. Bruemmer, J.L. Brimhall, C.H. Henager Jr., J.P. Hirth, Mater. Res. Soc. 288 (1992) 799. [10] A.Z. Tuleushev, V.N. Volodin, Y.Z. Tuleushev, J. Exp. Theor. Phys. Lett. 78 (7) (2003) 440–442. [11] P. Kurinskiy, V. Chakin, A. Moeslang, R. Rolli, E. Alves, L.C. Alves, N. Franco, Ch. Dorn, A.A. Goraieb, Fus. Eng. Des. 86 (9–11) (2011) 2454–2457. [12] J.-H. Kim, M. Nakamichi, J. Nucl. Mater. 442 (2013) S461–S464. [13] T.G. Nieh, J. Wadsworth, F.C. Grensing, J.-M. Yang, J. Nucl. Mater. 27 (1992) 2660–2664. [14] J.-H. Kim, M. Nakamichi, Fus. Eng. Des. 88 (2013) 2215–2218.