- Email: [email protected]
Fabrication and mechanical properties of powder metallurgy tantalum prepared by hot isostatic pressing
Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
Contents lists available at ScienceDirect
Int. Journal of Refractory Metals and Hard Materials journal homepage: www.elsevier.com/locate/IJRMHM
Fabrication and mechanical properties of powder metallurgy tantalum prepared by hot isostatic pressing Youngmoo Kim a,⁎, Eun-Pyo Kim a, Joon-Woong Noh a, Sung Ho Lee a, Young-Sam Kwon b, In Seok Oh c a b c
Agency for Defense Development, P.O. Box 35, Yuseong-gu, Daejeon 305–600, Republic of Korea Cetatech, GTIC 490, Seonjingongwon-gil, Yonghyeon-myeon, Sacheon 664-953, Republic of Korea STM Co. Ltd., KEPCO Venture Center, 105 Munji-ro, Yuseong-gu, Daejeon 305–760, Republic of Korea
a r t i c l e
i n f o
Article history: Received 16 July 2014 Received in revised form 6 September 2014 Accepted 6 September 2014 Available online 16 September 2014 Keywords: Tantalum Hot isostatic pressing Mechanical property
a b s t r a c t The fabrication process of a powder metallurgy (P/M) tantalum product with full density and fine microstructure was developed by using cold and hot isostatic pressing techniques. In order to increase the compact density and make the uniform density distribution, cold isostatic pressing (CIPing) of tantalum powders was conducted. Prior to hot isostatic pressing (HIPing), the CIPed billet was encapsulated and degassed to remove the contaminants in the container. After degassing, HIPing was performed twice and full densification of the tantalum powders was accomplished, regardless of powder size. The effect of processing conditions on the microstructure and mechanical properties of P/M tantalum billets was investigated. As the number of processing steps and temperature increased, the grain size of HIPed tantalum billets increased. Moreover, contrary to the Hall–Petch relation, the mechanical strength was increased in spite of increasing the grain size. This is because the oxygen content of the billets increased with rising in temperature and the number of processing steps. Therefore, in case of tantalum, it is found that the mechanical properties of tantalum may be highly influenced by the amount of interstitial elements, especially oxygen, rather than microstructural properties. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Tantalum has been used in high strain rate applications such as shaped charges in warheads because of its high density (16.65 g/cm3) and good dynamic ductility [1]. However, there are two constraints on applying in shaped charge armaments: the high-prices of raw materials and inconsistent microstructural variation in mill products [2]. Such restrictions may limit the wide application of tantalum and cause anisotropic properties in the final product where they are used [3]. Therefore, numerous recent studies have focused on reducing fabrication costs and eliminating inhomogeneous microstructure [4–6]. Equal channel angular pressing (ECAP) was applied to remove inhomogeneous microstructure of tantalum mill products, such as texture banding, leading to controlled microstructure of tantalum liners [4]. However, the process demands great effort and cost despite eliminating the inhomogeneity successfully. Powder metallurgy (P/M) techniques, instead of thermo-mechanical processing, may enable near-net shape (NNS) processing of tantalum to final products [7]. This NNS process has the potential to reduce the manufacturing and labor costs to produce a part. Furthermore, P/M processing promotes randomly textured products, thereby minimizing unexpected variations in mechanical properties. ⁎ Corresponding author. E-mail address: [email protected] (Y. Kim).
http://dx.doi.org/10.1016/j.ijrmhm.2014.09.012 0263-4368/© 2014 Elsevier Ltd. All rights reserved.
The sintering of tantalum powders with full densification requires high temperatures and long processing times due to its high melting temperature and low thermal conductivity [8]. Moreover, unlike tungsten and molybdenum, tantalum has high affinity for interstitial atoms (O, N, C, and H); therefore, it should be consolidated under high vacuum atmosphere in a furnace with heating elements made of a refractory metal not a graphite. However, it is difficult for such conditions to be satisfied simultaneously, because under high vacuum and temperature conditions, metal heaters in a furnace would be damaged. Thus, several studies have focused on improving the sinterability of tantalum powders [8–14]. The addition of nickel enhanced the densification of tantalum powders, but the presence of the additional element may degrade the high strain rate properties [8]. Nearly full density, i.e., 95 wt% of the theoretical value, was achieved by spark plasma sintering at 1700 °C; however, the unwanted phases, like tantalum carbides, were formed by contamination from molds [9]. Nanocrystalline tantalum powders were used to enhance sinterability [10]; however, a high amount of oxygen due to their large surface area would lead to reduction in the dynamic ductility of tantalum. Hot isostatic pressing has been widely used to consolidate tantalum powders to prevent contamination and inhomogeneous microstructures. Lavernia et al. reported that the density of a HIPed tantalum sample was approximately 95.6% with an average grain size of 110 μm [11]. H.C. Starck developed a P/M tantalum billet by sintering, HIPing and additional plastic deformation [12]. Moreover, Bingert et al. and Boncoeur et al. also fabricated a
212
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
HIPed billet with homogeneous microstructures and sound mechanical properties [13,14]. However, few studies have been investigated into the effect of HIPing conditions on microstructure and mechanical properties of tantalum. In this study, the influence of the HIP process on the sintering behavior and microstructure of tantalum powders was investigated. The mechanical properties at room temperature were evaluated, and the relationship among processing conditions, microstructures, and interstitial elements was studied. 2. Experimental
Table 1 Characteristics of raw tantalum powders.
Average particle size (μm) Particle size distribution (μm)
B.E.T. surface area (m2/g) Tap density (g/cm3) Manufacturer
D10 D25 D50 D75 D90
TaA
TaB
25.57 12.16 17.53 24.45 34.63 44.92 0.0486 7.767 H.C. Starck (GER)
16.89 3.29 5.51 11.41 21.65 38.94 0.1501 5.231 Ningxia Orient Tantalum Industries Co. Ltd (CHN)
2.1. Characterization of raw materials The raw powders used in this study were obtained from two different sources: H.C. Starck (designated as TaA) and Ningxia Orient Tantalum Industry Co. Ltd. (referred to as TaB). The scanning electron micrographs and characteristics of the powders are shown in Fig. 1 and Table 1, respectively. The particle size distributions were measured by a laser diffraction method (Beckman Coulter LS 230 model). The particle morphologies of both powders were irregular, and the average particle size of TaA was larger than that of TaB. The chemical compositions
supplied by manufacturers are shown in Table 2. The oxygen content of TaA and TaB powders, one of the important factors in determining the properties, was 257 and 550 ppm, respectively. The difference reflects the larger specific surface area of TaB compared with TaA. 2.2. Consolidation The powders were initially cold isostatically pressed (CIPed) at 2000 bar. The compacts were machined to cylinders; the green densities of the TaA and TaB powder compacts were 12.82 and 13.27 g/cm3, respectively. They were placed into a titanium (Grade 2) container of 25 mm internal diameter, 100 mm length, and 1 mm wall thickness. Following compact loading, degassing was performed at 250 °C to remove vapor and contaminants in the powder. When the vacuum level in the capsule reached 10− 2 torr, one end of the stem was crimped and welded. The encapsulated specimen was heated to 1500 °C, pressurized to 1000 bar, and maintained for 2 h under Ar atmosphere. Detailed conditions of hot isostatic pressing are shown in Table 3. The pressure and temperature were increased linearly up to the target values, where the pressure increased due to an increase in the temperature of the gas under constant volume. After HIPing, the container was removed by machining. The de-canned component was HIPed again without encapsulation to reach full density, as the pre-sintered product is already coherent, and the very small percentage of residual porosity ensures that there is little surface-connected porosity. 2.3. Evaluation of mechanical properties The sintered densities of consolidated specimens were evaluated using Archimedes' principle, and their microstructures were also observed with an optical microscope. The tensile properties, such as yield, tensile strengths, and ductility, were characterized by the ASTM E 8 method. The oxygen content of the sintered components was measured by an elemental analyzer LECO® 836 Series. The effects of powder
Table 2 Chemical analysis of raw tantalum powders. (Unit: ppm)
Fig. 1. Morphologies of (a) TaA and (b) TaB powders.
Element
TaA
TaB
C H N O Fe Ni Si Nb Ti Mo W Ta
19 86 39 257 4 b2 b7 b5 – b4 6 Bal.
16 33 88 550 32 8 40 b30 b1 17 10 Bal.
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
213
Table 3 Consolidation conditions of P/M tantalum components. Specimens
CIP
1st HIP
2nd HIP
TaA-1 TaA-2 TaA-3 TaB-1 TaB-2 TaB-3
2000 bar
1500 °C/1000 bar/1 h
– 1600 1700 – 1600 1700
°C/500 bar/2 h °C/500 bar/2 h °C/500 bar/2 h °C/500 bar/2 h
size, processing conditions and oxygen content on the sintering behaviors and microstructure were investigated.
3. Results and discussion 3.1. Fabrication of powder metallurgy tantalum Fig. 2 shows the shape change of titanium cans containing CIPed tantalum billets, revealing that they were well contracted without defects after HIPing. The nearly fully dense components were successfully produced by HIPing regardless of particle size, as shown in Fig. 3(a). At the temperature of 1500 °C, the relative densities of HIPed TaA and TaB components were 0.97 and 0.98, respectively. Full densification, 99.9 wt% of theoretical density, of both products was achieved by additional HIPing at 1600 or 1700 °C without encapsulation. As illustrated in Fig. 3(b), with increasing HIPing temperature, the average grain sizes of HIPed TaA and TaB components were coarsened from 18.9 to 26.7 μm and from 7.89 to 17.8 μm. In all processing conditions, the grain size of the TaB was smaller than that of the TaA because of lower particle size of the TaB powder. These were supported by microstructural observations, as illustrated in Fig. 4. At HIPed at 1500 °C, the microstructures exhibited the spheroidized pore located inside grains and not at grain boundaries as shown in Fig. 4(a) and (d). These grain and pore structures were typically found in the final stage of solid state sintering even though the consolidation temperature was as low as 1500 °C [15]. When HIPed at 1600 and 1700 °C of both sintered components, few pores were observed in the microstructures, as shown in Fig. 4(b), (c), (e), and (f), which agreed with the results of sintered densities. It is generally accepted that the final stage of solid state sintering is introduced above 0.8 of the melting temperature; therefore, for tantalum, the onset temperature to final densification would be 2300 °C and higher [15]. Such drastic reduction of temperature to achieve full densification, about 600 or 700 °C, would resulted from cold and hot isostatic pressing of tantalum powders. Moreover, a low consolidation temperature may prevent the grain coarsening during sintering. Therefore, this technique would produce a full dense P/M tantalum with fine and homogeneous microstructure at even low temperature conditions. These results seemed to be outstanding compared to previous studies [9–14] as shown in Fig. 5. Lavernia failed to produce full dense components by HIPing [11], and in Hunkeler [12] and Boncoeures' studies [14], the tantalum powder was fully densified, but grains were rather larger (N 50 μm). The characteristics of components reported by Bingert
Fig. 2. Change in container shape after HIPing.
Fig. 3. Variation in (a) HIPed densities and (b) average grain sizes of TaA and TaB products with HIPing steps and temperatures.
[13] were quite similar to our results, but sintering prior to HIP may cause not only densification but also grain growth. Based on these results, cold isostatic pressing of tantalum powder, instead of powder encapsulated HIPing, may contribute to enhancing consolidation and also preventing grain coarsening. 3.2. Mechanical properties of a powder metallurgy tantalum Fig. 6 shows the stress–strain behaviors of P/M tantalum specimens. The yield-drop phenomenon and work hardening were observed regardless of particle size. These result from retarding the dislocation motion of interstitial atoms. Fig. 7 shows the variation of tensile strength and elongation with the powders and processing conditions. The ultimate tensile strengths of HIPed TaA and TaB components increased from 415 to 527 MPa and from 635 to 747 MPa, respectively, with increased pressing temperature, as shown in Fig. 7(a). The yield stresses of TaA and TaB also increased from 327 to 447 MPa and 581 to 667 MPa, respectively. Contrary to the strength, as the temperature increased, the elongations were reduced from 43 to 36% (for TaA) and from 34 to 27% (for TaB), respectively, as seen in Fig. 7(b). The Vickers hardness also increased with increasing processing temperature. The mechanical strengths of the HIPed TaA component were lower than
214
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
Fig. 4. Microstructures of HIPed at (a) 1500 °C, (b) 1600 °C, and (c) 1700 °C from TaA powders; HIPed at (d) 1500 °C, (e) 1600 °C, and (f) 1700 °C from TaB powders.
Fig. 5. Average grain size as a function of the HIPed density for P/M tantalum products.
Fig. 6. Tensile stress–strain behaviors of HIPed TaA and TaB specimens.
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
Fig. 7. Variations in (a) tensile and yield strengths and (b) elongation and Vickers hardness of HIPed TaA and TaB products as a function of HIPing steps and temperatures.
those of the TaB one, whereas its ductility showed the opposite trend, irrespective of processing conditions. These seemed to be originated from the smaller grain size of the HIPed TaB specimen compared to TaA one. It is generally accepted that as grain size decreases, the tensile strength is increased and the ductility is reduced. Whereas, in case of the identical powder, the strength increased even though the grain size was coarsened, as illustrated in Fig. 3. This seems to be associated with the content of interstitial elements, such as oxygen and carbon. It has been well known that such elements have a large influence on the mechanical properties of tantalum [16]. As illustrated in Fig. 8, for both powders, the amount of oxygen was increased with increasing the processing steps. It is natural that the specimens were gradually contaminated from external surroundings, such as moisture and pressurized argon during additional HIPing. As well, for all processing conditions, the oxygen content of HIPed TaB specimens was higher than that of the TaA ones because of the lower particle size of TaB parts, leading to a larger specific surface area compared to TaA ones. Thus, it would be reasonable to assume that impurity contents are a higher impact on the mechanical strength of tantalum rather than the grain structures. Since such interstitial atoms may act as obstacles to dislocation motions, the strength would be improved with increasing the content of impurities. Among the mechanical properties, yield strength and elongation have been considered as the main properties of tantalum for high strain
215
Fig. 8. Variations in impurity content for (a) HIPed TaA and (b) TaB specimens as a function of HIPing steps and temperatures.
rate applications [1]. Fig. 9 illustrates the comparison of these properties in our study with previous results [11–15]. According to R. M. German, the typical yield strength and elongation of a P/M tantalum are
Fig. 9. The elongation as a function of the yield strength for P/M tantalum products.
216
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
396 MPa and 30%, respectively [15]. In case of commercial products, TaM(melted grade) and TaS(sintered grade) manufactured by PLANSEE GmbH, the yield strength and elongation were 390 MPa and 31% for TaM, and 420 MPa and 18% for TaS. The TaA product exhibited excellent ductility while retaining suitable strength; otherwise, for the TaB one, the strength was superior, but ductility was relatively low compared to other results. Consequently, the HIPed TaA product seems to be more suitable than TaB one for high strain rate applications, because TaA specimens showed higher ductility and lower oxygen content with proper mechanical strengths.
to increased mechanical strength with increasing the amount of oxygen despite grain coarsening. Consequently, unlike other refractory metals, the mechanical properties of tantalum may be highly influenced by the content of interstitial elements, especially oxygen, rather than by microstructural properties. The TaA component developed in this study exhibited excellent ductility, finer grain size, and lower oxygen content compared to previous results while retaining proper strengths. Therefore, it seems to be applicable in high strain applications.
4. Conclusions
References
The characteristics of HIPed specimens in this study, such as grain size and mechanical properties, were found to be outstanding compared to those of previous studies. We investigated the effect of particle size and HIPing conditions on the relative density and microstructures of P/M tantalum products. The relationship between mechanical properties and the amount of interstitial elements were also studied. The major results can be summarized as follows.
[1] Cardonne SM, Kumar P, Muchaluk CA, Schwartz HD. Tantalum and its alloys. Int J Refract Met Hard Mater 1995;13:187–94. [2] Burns RH, Fang JC, Kumar P. Evolution of applications of tantalum. Tantalum. In: Chen A, Crowson A, Lavernia E, Ebihara W, Kumar P, editors. Proc 125th TMS Annual Meeting and Exhibition. Anaheim: TMS; February 1996. p. 273–85. [3] Chou PC, Crudza ME. The effects of liner anisotropy on warhead performance. In: Asfahani R, Chen E, Crowson A, editors. High Strain Rate Behavior of Refractory Metals and Alloys. Warrendale PA: TMS; 1992. p. 97–109. [4] House J, Flater P, Harris R, Angelis RD, Nixon M, Bingert J, O'Brien J. Annealing and mechanical properties of ECAP tantalum. AFRL-RW-EG-TR-2011-045 , Air Force Research Laboratory Technical Report; March, 2011. [5] House J, Flater P, Harris R, Angelis RD. Texure evolution during dynamic loading of ECAP tantalum. AFRL-RW-EG-TR-2011–032 , Air Force Research Laboratory Technical Report; March, 2011. [6] Flater P, House J, O'Brien J, Hosford WF. High strain rate properties of tantalum processed by equal channel angular pressing. AFRL-RW-EG-TR-2007-7412 , Air Force Research Laboratory Technical Report; August, 2007. [7] Muller JF, Dinh PM. Evaluation of powder metallurgy tantalum liners for explosively formed penetrator. Tungsten and refractory metals. In: Bose A, Dowding RJ, editors. Proc 2nd Int Conf Tungsten Refract Met. McLean: MPIF; October 1994. p. 573–86. [8] German RM, Munir ZA. The sintering of tantalum with transition metal additions. Powder Metall 1977;3:145–50. [9] Angerer P, Neubauer E, Yu LG, Khor KA. Texture and structure evolution of tantalum powder samples during spark-plasma-sintering and conventional hot-pressing. Int J Refract Met Hard Mater 2007;25:280–5. [10] Yoo SH, Sudarshan TS, Sethuram K, Subhash G, Dowding RJ. Consolidation and high strain rate mechanical behavior of nanocrystalline tantalum powder. Nanostruct Mater 1999;12:23–8. [11] Xu Q, Hayes RW, Lavernia EJ. Creep properties of ball-milled and HIPed pure tantalum. Scr Mater 2001;45:447–54. [12] Funkeler FJ. Fabricated tantalum from unsintered HIP powder. In: Bose A, Dowding RJ, editors. Tungsten and refractory metals, Proc 2nd Int Conf Tungsten Refract Met, MPIF: McLean; October 1994. p. 531–46. [13] Bingert SR, Vargas VD, Sheinberg H. Tantalum powder consolidation, modeling and properties. Tantalum. In: Chen A, Crowson A, Lavernia E, Ebihara W, Kumar P, editors. Proc 125th TMS Annual Meeting and Exhibition. Anaheim: TMS; February 1996. p. 95–105. [14] Boncoeur M, Valin F, Raisson G, Michaud H. Properties of hot isostatically pressed tantalum. Hot isostatic pressing-theory and applications. In: Koizumi M, editor. Proc 3rd Int Conf Hot Isostatic Pressing. Osaka: Springer; June 1991. p. 247–52. [15] German RM. Sintering Theory and Practice. John Wiley and Sons; 1996. [16] Efe M, Kim HJ, Chandrasekar S, Trumble KP. The chemical state and control of oxygen in powder metallurgy tantalum. Mater Sci Eng A 2012;544:1–9.
(1) The relative densities of P/M tantalum were increased with increasing number of HIPing steps and temperatures, achieving 99.9 wt% of the theoretical density for both powders. The average grain size also increased with increasing temperature and with the number of processing steps. Such tendencies were supported by microstructural observations. In the case of specimens with 95 wt% of the theoretical HIPed at 1500 °C, most spheroidized pores were located inside the grains; these pore structures were typically observed at the final stage of solid state sintering. After secondary HIPing at 1600 and 1700 °C, the full densification occurred, and few pores were observed. The microstructural characteristics of P/M tantalum including a grain size below 30 μm with full densification, appear to be superior when compared with previous studies. (2) The mechanical strengths, such as yield, tensile strength, and hardness, of TaB specimens were higher than those of TaA ones. This would result from lower grain size of TaB components compared to TaA ones. Whereas, for specimens fabricated from an identical tantalum powder, the strengths were increased with increasing the processing steps and temperature, leading to grain coarsening. The elongation revealed the opposite trend and such characteristics are not generally consistent in the Hall–Petch relation. This inconsistency may result from the oxygen content of tantalum specimens. The oxygen in tantalum would act as an inhibitor to a grain boundary motion, leading
Contents lists available at ScienceDirect
Int. Journal of Refractory Metals and Hard Materials journal homepage: www.elsevier.com/locate/IJRMHM
Fabrication and mechanical properties of powder metallurgy tantalum prepared by hot isostatic pressing Youngmoo Kim a,⁎, Eun-Pyo Kim a, Joon-Woong Noh a, Sung Ho Lee a, Young-Sam Kwon b, In Seok Oh c a b c
Agency for Defense Development, P.O. Box 35, Yuseong-gu, Daejeon 305–600, Republic of Korea Cetatech, GTIC 490, Seonjingongwon-gil, Yonghyeon-myeon, Sacheon 664-953, Republic of Korea STM Co. Ltd., KEPCO Venture Center, 105 Munji-ro, Yuseong-gu, Daejeon 305–760, Republic of Korea
a r t i c l e
i n f o
Article history: Received 16 July 2014 Received in revised form 6 September 2014 Accepted 6 September 2014 Available online 16 September 2014 Keywords: Tantalum Hot isostatic pressing Mechanical property
a b s t r a c t The fabrication process of a powder metallurgy (P/M) tantalum product with full density and fine microstructure was developed by using cold and hot isostatic pressing techniques. In order to increase the compact density and make the uniform density distribution, cold isostatic pressing (CIPing) of tantalum powders was conducted. Prior to hot isostatic pressing (HIPing), the CIPed billet was encapsulated and degassed to remove the contaminants in the container. After degassing, HIPing was performed twice and full densification of the tantalum powders was accomplished, regardless of powder size. The effect of processing conditions on the microstructure and mechanical properties of P/M tantalum billets was investigated. As the number of processing steps and temperature increased, the grain size of HIPed tantalum billets increased. Moreover, contrary to the Hall–Petch relation, the mechanical strength was increased in spite of increasing the grain size. This is because the oxygen content of the billets increased with rising in temperature and the number of processing steps. Therefore, in case of tantalum, it is found that the mechanical properties of tantalum may be highly influenced by the amount of interstitial elements, especially oxygen, rather than microstructural properties. © 2014 Elsevier Ltd. All rights reserved.
1. Introduction Tantalum has been used in high strain rate applications such as shaped charges in warheads because of its high density (16.65 g/cm3) and good dynamic ductility [1]. However, there are two constraints on applying in shaped charge armaments: the high-prices of raw materials and inconsistent microstructural variation in mill products [2]. Such restrictions may limit the wide application of tantalum and cause anisotropic properties in the final product where they are used [3]. Therefore, numerous recent studies have focused on reducing fabrication costs and eliminating inhomogeneous microstructure [4–6]. Equal channel angular pressing (ECAP) was applied to remove inhomogeneous microstructure of tantalum mill products, such as texture banding, leading to controlled microstructure of tantalum liners [4]. However, the process demands great effort and cost despite eliminating the inhomogeneity successfully. Powder metallurgy (P/M) techniques, instead of thermo-mechanical processing, may enable near-net shape (NNS) processing of tantalum to final products [7]. This NNS process has the potential to reduce the manufacturing and labor costs to produce a part. Furthermore, P/M processing promotes randomly textured products, thereby minimizing unexpected variations in mechanical properties. ⁎ Corresponding author. E-mail address: [email protected] (Y. Kim).
http://dx.doi.org/10.1016/j.ijrmhm.2014.09.012 0263-4368/© 2014 Elsevier Ltd. All rights reserved.
The sintering of tantalum powders with full densification requires high temperatures and long processing times due to its high melting temperature and low thermal conductivity [8]. Moreover, unlike tungsten and molybdenum, tantalum has high affinity for interstitial atoms (O, N, C, and H); therefore, it should be consolidated under high vacuum atmosphere in a furnace with heating elements made of a refractory metal not a graphite. However, it is difficult for such conditions to be satisfied simultaneously, because under high vacuum and temperature conditions, metal heaters in a furnace would be damaged. Thus, several studies have focused on improving the sinterability of tantalum powders [8–14]. The addition of nickel enhanced the densification of tantalum powders, but the presence of the additional element may degrade the high strain rate properties [8]. Nearly full density, i.e., 95 wt% of the theoretical value, was achieved by spark plasma sintering at 1700 °C; however, the unwanted phases, like tantalum carbides, were formed by contamination from molds [9]. Nanocrystalline tantalum powders were used to enhance sinterability [10]; however, a high amount of oxygen due to their large surface area would lead to reduction in the dynamic ductility of tantalum. Hot isostatic pressing has been widely used to consolidate tantalum powders to prevent contamination and inhomogeneous microstructures. Lavernia et al. reported that the density of a HIPed tantalum sample was approximately 95.6% with an average grain size of 110 μm [11]. H.C. Starck developed a P/M tantalum billet by sintering, HIPing and additional plastic deformation [12]. Moreover, Bingert et al. and Boncoeur et al. also fabricated a
212
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
HIPed billet with homogeneous microstructures and sound mechanical properties [13,14]. However, few studies have been investigated into the effect of HIPing conditions on microstructure and mechanical properties of tantalum. In this study, the influence of the HIP process on the sintering behavior and microstructure of tantalum powders was investigated. The mechanical properties at room temperature were evaluated, and the relationship among processing conditions, microstructures, and interstitial elements was studied. 2. Experimental
Table 1 Characteristics of raw tantalum powders.
Average particle size (μm) Particle size distribution (μm)
B.E.T. surface area (m2/g) Tap density (g/cm3) Manufacturer
D10 D25 D50 D75 D90
TaA
TaB
25.57 12.16 17.53 24.45 34.63 44.92 0.0486 7.767 H.C. Starck (GER)
16.89 3.29 5.51 11.41 21.65 38.94 0.1501 5.231 Ningxia Orient Tantalum Industries Co. Ltd (CHN)
2.1. Characterization of raw materials The raw powders used in this study were obtained from two different sources: H.C. Starck (designated as TaA) and Ningxia Orient Tantalum Industry Co. Ltd. (referred to as TaB). The scanning electron micrographs and characteristics of the powders are shown in Fig. 1 and Table 1, respectively. The particle size distributions were measured by a laser diffraction method (Beckman Coulter LS 230 model). The particle morphologies of both powders were irregular, and the average particle size of TaA was larger than that of TaB. The chemical compositions
supplied by manufacturers are shown in Table 2. The oxygen content of TaA and TaB powders, one of the important factors in determining the properties, was 257 and 550 ppm, respectively. The difference reflects the larger specific surface area of TaB compared with TaA. 2.2. Consolidation The powders were initially cold isostatically pressed (CIPed) at 2000 bar. The compacts were machined to cylinders; the green densities of the TaA and TaB powder compacts were 12.82 and 13.27 g/cm3, respectively. They were placed into a titanium (Grade 2) container of 25 mm internal diameter, 100 mm length, and 1 mm wall thickness. Following compact loading, degassing was performed at 250 °C to remove vapor and contaminants in the powder. When the vacuum level in the capsule reached 10− 2 torr, one end of the stem was crimped and welded. The encapsulated specimen was heated to 1500 °C, pressurized to 1000 bar, and maintained for 2 h under Ar atmosphere. Detailed conditions of hot isostatic pressing are shown in Table 3. The pressure and temperature were increased linearly up to the target values, where the pressure increased due to an increase in the temperature of the gas under constant volume. After HIPing, the container was removed by machining. The de-canned component was HIPed again without encapsulation to reach full density, as the pre-sintered product is already coherent, and the very small percentage of residual porosity ensures that there is little surface-connected porosity. 2.3. Evaluation of mechanical properties The sintered densities of consolidated specimens were evaluated using Archimedes' principle, and their microstructures were also observed with an optical microscope. The tensile properties, such as yield, tensile strengths, and ductility, were characterized by the ASTM E 8 method. The oxygen content of the sintered components was measured by an elemental analyzer LECO® 836 Series. The effects of powder
Table 2 Chemical analysis of raw tantalum powders. (Unit: ppm)
Fig. 1. Morphologies of (a) TaA and (b) TaB powders.
Element
TaA
TaB
C H N O Fe Ni Si Nb Ti Mo W Ta
19 86 39 257 4 b2 b7 b5 – b4 6 Bal.
16 33 88 550 32 8 40 b30 b1 17 10 Bal.
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
213
Table 3 Consolidation conditions of P/M tantalum components. Specimens
CIP
1st HIP
2nd HIP
TaA-1 TaA-2 TaA-3 TaB-1 TaB-2 TaB-3
2000 bar
1500 °C/1000 bar/1 h
– 1600 1700 – 1600 1700
°C/500 bar/2 h °C/500 bar/2 h °C/500 bar/2 h °C/500 bar/2 h
size, processing conditions and oxygen content on the sintering behaviors and microstructure were investigated.
3. Results and discussion 3.1. Fabrication of powder metallurgy tantalum Fig. 2 shows the shape change of titanium cans containing CIPed tantalum billets, revealing that they were well contracted without defects after HIPing. The nearly fully dense components were successfully produced by HIPing regardless of particle size, as shown in Fig. 3(a). At the temperature of 1500 °C, the relative densities of HIPed TaA and TaB components were 0.97 and 0.98, respectively. Full densification, 99.9 wt% of theoretical density, of both products was achieved by additional HIPing at 1600 or 1700 °C without encapsulation. As illustrated in Fig. 3(b), with increasing HIPing temperature, the average grain sizes of HIPed TaA and TaB components were coarsened from 18.9 to 26.7 μm and from 7.89 to 17.8 μm. In all processing conditions, the grain size of the TaB was smaller than that of the TaA because of lower particle size of the TaB powder. These were supported by microstructural observations, as illustrated in Fig. 4. At HIPed at 1500 °C, the microstructures exhibited the spheroidized pore located inside grains and not at grain boundaries as shown in Fig. 4(a) and (d). These grain and pore structures were typically found in the final stage of solid state sintering even though the consolidation temperature was as low as 1500 °C [15]. When HIPed at 1600 and 1700 °C of both sintered components, few pores were observed in the microstructures, as shown in Fig. 4(b), (c), (e), and (f), which agreed with the results of sintered densities. It is generally accepted that the final stage of solid state sintering is introduced above 0.8 of the melting temperature; therefore, for tantalum, the onset temperature to final densification would be 2300 °C and higher [15]. Such drastic reduction of temperature to achieve full densification, about 600 or 700 °C, would resulted from cold and hot isostatic pressing of tantalum powders. Moreover, a low consolidation temperature may prevent the grain coarsening during sintering. Therefore, this technique would produce a full dense P/M tantalum with fine and homogeneous microstructure at even low temperature conditions. These results seemed to be outstanding compared to previous studies [9–14] as shown in Fig. 5. Lavernia failed to produce full dense components by HIPing [11], and in Hunkeler [12] and Boncoeures' studies [14], the tantalum powder was fully densified, but grains were rather larger (N 50 μm). The characteristics of components reported by Bingert
Fig. 2. Change in container shape after HIPing.
Fig. 3. Variation in (a) HIPed densities and (b) average grain sizes of TaA and TaB products with HIPing steps and temperatures.
[13] were quite similar to our results, but sintering prior to HIP may cause not only densification but also grain growth. Based on these results, cold isostatic pressing of tantalum powder, instead of powder encapsulated HIPing, may contribute to enhancing consolidation and also preventing grain coarsening. 3.2. Mechanical properties of a powder metallurgy tantalum Fig. 6 shows the stress–strain behaviors of P/M tantalum specimens. The yield-drop phenomenon and work hardening were observed regardless of particle size. These result from retarding the dislocation motion of interstitial atoms. Fig. 7 shows the variation of tensile strength and elongation with the powders and processing conditions. The ultimate tensile strengths of HIPed TaA and TaB components increased from 415 to 527 MPa and from 635 to 747 MPa, respectively, with increased pressing temperature, as shown in Fig. 7(a). The yield stresses of TaA and TaB also increased from 327 to 447 MPa and 581 to 667 MPa, respectively. Contrary to the strength, as the temperature increased, the elongations were reduced from 43 to 36% (for TaA) and from 34 to 27% (for TaB), respectively, as seen in Fig. 7(b). The Vickers hardness also increased with increasing processing temperature. The mechanical strengths of the HIPed TaA component were lower than
214
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
Fig. 4. Microstructures of HIPed at (a) 1500 °C, (b) 1600 °C, and (c) 1700 °C from TaA powders; HIPed at (d) 1500 °C, (e) 1600 °C, and (f) 1700 °C from TaB powders.
Fig. 5. Average grain size as a function of the HIPed density for P/M tantalum products.
Fig. 6. Tensile stress–strain behaviors of HIPed TaA and TaB specimens.
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
Fig. 7. Variations in (a) tensile and yield strengths and (b) elongation and Vickers hardness of HIPed TaA and TaB products as a function of HIPing steps and temperatures.
those of the TaB one, whereas its ductility showed the opposite trend, irrespective of processing conditions. These seemed to be originated from the smaller grain size of the HIPed TaB specimen compared to TaA one. It is generally accepted that as grain size decreases, the tensile strength is increased and the ductility is reduced. Whereas, in case of the identical powder, the strength increased even though the grain size was coarsened, as illustrated in Fig. 3. This seems to be associated with the content of interstitial elements, such as oxygen and carbon. It has been well known that such elements have a large influence on the mechanical properties of tantalum [16]. As illustrated in Fig. 8, for both powders, the amount of oxygen was increased with increasing the processing steps. It is natural that the specimens were gradually contaminated from external surroundings, such as moisture and pressurized argon during additional HIPing. As well, for all processing conditions, the oxygen content of HIPed TaB specimens was higher than that of the TaA ones because of the lower particle size of TaB parts, leading to a larger specific surface area compared to TaA ones. Thus, it would be reasonable to assume that impurity contents are a higher impact on the mechanical strength of tantalum rather than the grain structures. Since such interstitial atoms may act as obstacles to dislocation motions, the strength would be improved with increasing the content of impurities. Among the mechanical properties, yield strength and elongation have been considered as the main properties of tantalum for high strain
215
Fig. 8. Variations in impurity content for (a) HIPed TaA and (b) TaB specimens as a function of HIPing steps and temperatures.
rate applications [1]. Fig. 9 illustrates the comparison of these properties in our study with previous results [11–15]. According to R. M. German, the typical yield strength and elongation of a P/M tantalum are
Fig. 9. The elongation as a function of the yield strength for P/M tantalum products.
216
Y. Kim et al. / Int. Journal of Refractory Metals and Hard Materials 48 (2015) 211–216
396 MPa and 30%, respectively [15]. In case of commercial products, TaM(melted grade) and TaS(sintered grade) manufactured by PLANSEE GmbH, the yield strength and elongation were 390 MPa and 31% for TaM, and 420 MPa and 18% for TaS. The TaA product exhibited excellent ductility while retaining suitable strength; otherwise, for the TaB one, the strength was superior, but ductility was relatively low compared to other results. Consequently, the HIPed TaA product seems to be more suitable than TaB one for high strain rate applications, because TaA specimens showed higher ductility and lower oxygen content with proper mechanical strengths.
to increased mechanical strength with increasing the amount of oxygen despite grain coarsening. Consequently, unlike other refractory metals, the mechanical properties of tantalum may be highly influenced by the content of interstitial elements, especially oxygen, rather than by microstructural properties. The TaA component developed in this study exhibited excellent ductility, finer grain size, and lower oxygen content compared to previous results while retaining proper strengths. Therefore, it seems to be applicable in high strain applications.
4. Conclusions
References
The characteristics of HIPed specimens in this study, such as grain size and mechanical properties, were found to be outstanding compared to those of previous studies. We investigated the effect of particle size and HIPing conditions on the relative density and microstructures of P/M tantalum products. The relationship between mechanical properties and the amount of interstitial elements were also studied. The major results can be summarized as follows.
[1] Cardonne SM, Kumar P, Muchaluk CA, Schwartz HD. Tantalum and its alloys. Int J Refract Met Hard Mater 1995;13:187–94. [2] Burns RH, Fang JC, Kumar P. Evolution of applications of tantalum. Tantalum. In: Chen A, Crowson A, Lavernia E, Ebihara W, Kumar P, editors. Proc 125th TMS Annual Meeting and Exhibition. Anaheim: TMS; February 1996. p. 273–85. [3] Chou PC, Crudza ME. The effects of liner anisotropy on warhead performance. In: Asfahani R, Chen E, Crowson A, editors. High Strain Rate Behavior of Refractory Metals and Alloys. Warrendale PA: TMS; 1992. p. 97–109. [4] House J, Flater P, Harris R, Angelis RD, Nixon M, Bingert J, O'Brien J. Annealing and mechanical properties of ECAP tantalum. AFRL-RW-EG-TR-2011-045 , Air Force Research Laboratory Technical Report; March, 2011. [5] House J, Flater P, Harris R, Angelis RD. Texure evolution during dynamic loading of ECAP tantalum. AFRL-RW-EG-TR-2011–032 , Air Force Research Laboratory Technical Report; March, 2011. [6] Flater P, House J, O'Brien J, Hosford WF. High strain rate properties of tantalum processed by equal channel angular pressing. AFRL-RW-EG-TR-2007-7412 , Air Force Research Laboratory Technical Report; August, 2007. [7] Muller JF, Dinh PM. Evaluation of powder metallurgy tantalum liners for explosively formed penetrator. Tungsten and refractory metals. In: Bose A, Dowding RJ, editors. Proc 2nd Int Conf Tungsten Refract Met. McLean: MPIF; October 1994. p. 573–86. [8] German RM, Munir ZA. The sintering of tantalum with transition metal additions. Powder Metall 1977;3:145–50. [9] Angerer P, Neubauer E, Yu LG, Khor KA. Texture and structure evolution of tantalum powder samples during spark-plasma-sintering and conventional hot-pressing. Int J Refract Met Hard Mater 2007;25:280–5. [10] Yoo SH, Sudarshan TS, Sethuram K, Subhash G, Dowding RJ. Consolidation and high strain rate mechanical behavior of nanocrystalline tantalum powder. Nanostruct Mater 1999;12:23–8. [11] Xu Q, Hayes RW, Lavernia EJ. Creep properties of ball-milled and HIPed pure tantalum. Scr Mater 2001;45:447–54. [12] Funkeler FJ. Fabricated tantalum from unsintered HIP powder. In: Bose A, Dowding RJ, editors. Tungsten and refractory metals, Proc 2nd Int Conf Tungsten Refract Met, MPIF: McLean; October 1994. p. 531–46. [13] Bingert SR, Vargas VD, Sheinberg H. Tantalum powder consolidation, modeling and properties. Tantalum. In: Chen A, Crowson A, Lavernia E, Ebihara W, Kumar P, editors. Proc 125th TMS Annual Meeting and Exhibition. Anaheim: TMS; February 1996. p. 95–105. [14] Boncoeur M, Valin F, Raisson G, Michaud H. Properties of hot isostatically pressed tantalum. Hot isostatic pressing-theory and applications. In: Koizumi M, editor. Proc 3rd Int Conf Hot Isostatic Pressing. Osaka: Springer; June 1991. p. 247–52. [15] German RM. Sintering Theory and Practice. John Wiley and Sons; 1996. [16] Efe M, Kim HJ, Chandrasekar S, Trumble KP. The chemical state and control of oxygen in powder metallurgy tantalum. Mater Sci Eng A 2012;544:1–9.
(1) The relative densities of P/M tantalum were increased with increasing number of HIPing steps and temperatures, achieving 99.9 wt% of the theoretical density for both powders. The average grain size also increased with increasing temperature and with the number of processing steps. Such tendencies were supported by microstructural observations. In the case of specimens with 95 wt% of the theoretical HIPed at 1500 °C, most spheroidized pores were located inside the grains; these pore structures were typically observed at the final stage of solid state sintering. After secondary HIPing at 1600 and 1700 °C, the full densification occurred, and few pores were observed. The microstructural characteristics of P/M tantalum including a grain size below 30 μm with full densification, appear to be superior when compared with previous studies. (2) The mechanical strengths, such as yield, tensile strength, and hardness, of TaB specimens were higher than those of TaA ones. This would result from lower grain size of TaB components compared to TaA ones. Whereas, for specimens fabricated from an identical tantalum powder, the strengths were increased with increasing the processing steps and temperature, leading to grain coarsening. The elongation revealed the opposite trend and such characteristics are not generally consistent in the Hall–Petch relation. This inconsistency may result from the oxygen content of tantalum specimens. The oxygen in tantalum would act as an inhibitor to a grain boundary motion, leading