- Email: [email protected]
Enhanced liquid-phase sintering of W–Cu composites by liquid infiltration
Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
Contents lists available at ScienceDirect
Int. Journal of Refractory Metals and Hard Materials journal homepage: www.elsevier.com/locate/IJRMHM
Enhanced liquid-phase sintering of W–Cu composites by liquid infiltration Hafed Ibrahim a,b, Azizan Aziz a,⁎, Azmi Rahmat c a b c
School of Materials and Mineral Resources Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal, Pinang, Malaysia Faculty of Engineering, Omar Almukhtar University, Al-Bayda, Libya School of Materials Engineering, Universiti Malaysia Perlis, Taman Muhibah-Jejawi, Arau 02600, Perlis, Malaysia
a r t i c l e
i n f o
Article history: Received 20 September 2013 Accepted 8 December 2013 Available online 14 December 2013 Keywords: W–Cu composites Liquid-phase sintering Liquid infiltration Microstructure
a b s t r a c t Full-density of consolidated W–Cu composites produced via conventional sintering method is difficult to achieve. In this work, fully-dense W–Cu composites were developed via the combination of the liquid phase sintering (LPS) and the liquid infiltration (LI) methods, which hereinafter is named as Cu-MI technique. It operates at the low sintering temperature of 1150 °C, and maximum densification was possible without requiring a sintering activator such as Ni, Co or Fe. A comparison was also made between the sintering response of W–(13–27 wt.%) Cu composites consolidated using LPS and Cu-MI techniques. The samples were characterized using SEM, EDX and XRD. It was observed that the samples prepared via the Cu-MI method demonstrated a high relative density (N 99% theoretical density). Contrary to the composite sample prepared by the LPS method, the Cu-MI technique accounted for a homogeneous microstructure almost without any pores. The significance of this finding has major industrial implications and has potential to reduce the production costs of composite materials with improved mechanical and electrical properties. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Tungsten–copper (W–Cu) composites are typical pseudo-alloys that combine the natural properties of W, such as the low thermal expansion, high density and hardness, with those of Cu possessing unique thermal properties. Such properties make W–Cu composites very useful in a variety of industrial applications, such as thermal packages (heat sinks), welding electrodes and high-voltage electric contacts [1–4]. W–Cu composites with varying tungsten contents have wide applications [5], e.g. W–Cu composites containing 80–95 wt.%W are suitable for electronic packages, whereas composites with W content of 50–75 wt.% are preferred in electrical contact materials and military applications, such as shaped charge liner and ammunitions [5,6]. In general, there are two common methods applied in fabricating W–Cu alloys, which are the liquid phase sintering (LPS) and the liquid infiltration (LI) methods. LPS involves combining, compacting, and sintering of W and Cu elements under specific conditions. Conversely the LI technique encompasses two steps: firstly, preparation of the W skeleton with desirable pores at a temperature of 1150 °C, followed by filling the W skeleton with melted copper at a temperature of 1250 °C [7–9]. The low solubility of both W and Cu causes difficulty in attaining full or near density in W–Cu composites by LPS [10,11]. Johnson and German [12] and Boonyonagmaneerat [13] showed the effect of adding ⁎ Corresponding author. Tel.: +60 4 5996174; fax: +60 4 5941011. E-mail address: [email protected] (A. Aziz). 0263-4368/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijrmhm.2013.12.004
low concentrations of transition elements (group VIII) as sintering activators to obtain a high density for W–Cu composites by LPS. However, adding transition elements led to unsatisfactory thermal and electrical properties of the W–Cu composites [14] and enhanced mechanical properties [15]. Mechanical alloying was used to enhance the sinterability of W–Cu composites by decreasing the size of the W particles and increasing its distribution [16,17]. However, this process led to the contamination of iron and cobalt by the use of stainless steel and tungsten carbide balls or jars. Such contaminations enhanced densifications, but significantly affect the thermal properties of the W–Cu alloys [14]. Many researchers [18–20] have tried to obtain the full density of W– Cu composites without using additives as a sintering activator. In 2012, Abbaszadeh et al. [21] described a method for fabricating W–15 wt.%Cu composites by conventional mixing–compacting–sintering and mechanochemical reaction–compaction sintering techniques. They obtained a relative density equal to 94% theoretical density (TD) at 1200 °C by the latter technique. Hashempour et al. [22] adopted the method of preparing W–25 wt.%Cu composite powder using thermochemical coprecipitation. Sintering was performed at 1200 °C under H2 gas, which yielded a relative density of 98% TD. In this work, fully dense W–Cu composites (13–27 wt.%Cu) were prepared by Cu-melt infiltration method (Cu-MI; combining LPS and LI) at a low sintering temperature of 1150 °C under protective hydrogen gas. This novel method combined LPS with LI techniques and was very effective in obtaining excellent densification of the W–Cu composites, without using additives as a sintering activator. The insert method
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
(Cu-MI) was compared with LPS under the same condition. Microstructure, relative density, hardness, and thermal resistivity of specimens prepared by both methods were examined.
223
measurements were taken and arranged to represent the sintered compact average electrical resistivity. Recorded resistivity values were then converted to electrical conductivity in percentages of International Annealing Copper Standard (%IACS) based on Eq. (1):
2. Experimental work Tungsten powder with an average particle size of 16 μm (99.9% purity) and copper powder with a mean particle size of 21 μm (99.9% purity) were used in this study. An oxygen-free Cu plate (99.9% purity) was used as the source of liquid infiltration. Mixing of homogeneous W powder with Cu powder to produce green compacts was conducted manually in an alumina mortar for 20 min to prevent particle segregation due to the difference in the densities of W and Cu elements. Typically, the elemental mixture was prepared at the preferred proportions to give the composite powder the desired properties. The mixed powder was die-pressed at 400 MPa to obtain a cylindrical-shaped green compact, 13.2 mm in diameter and 2.0 mm to 2.4 mm in height. The size of the green compact depended mainly on the Cu weight fraction. W– Cu green compacts with desirable weight fractions of Cu were placed on the oxygen-free copper plate (99.9% purity) and then consolidated by Cu-melt infiltration at 1150 °C. Green compacts were likewise prepared by LPS at 1150 °C and used for comparison. LPS and Cu-MI were conducted at 1150 °C using hydrogen as a protective gas. Diagrams of both processes are presented in Fig. 1. Compacted, sintered, and infiltrated densities were measured using the water displacement method (Archimedes principle) based on ASTM B328 standard. Scanning electron microscopy (SEM) was used to investigate the composites' microstructure. X-ray diffraction (XRD) was used to study the crystalline phases of the sintered compacts. Data obtained from the XRD pattern of this composite had been analyzed by X'PertHigh Source Plus software to determine the phases present and lattice parameter of Cu and W phases. Specimens were exposed to a Cu-Kα1 radiation source (λ = 1.5406 Å) and data were collected at 2θ from 10° to 90°. Hardness of specimens was measured by Vickers microhardness under 1 kg load and 15 s duration, and the microhardness value was derived from the average of 10 readings. Electrical conductivity of the sintered composites was determined using the four-point probe resistivity method [7]. It was performed in a fully automated Changmin Tech CMT-Sr2000N apparatus. The sintered compacts were placed perpendicular to the four probes and repeatedly measured across both sides of the surface until the values were constant. Five
%IACS ¼
172:41 ρ
ð1Þ
where ρ is the electrical resistivity in unit of μΩ·cm. 3. Results and discussion 3.1. Densification of W–Cu composites The relation between relative density and sintering method of W–Cu composites is presented in Fig. 2. These were significant differences in the relative density of the W–Cu system for the different sintering methods. The highest density (99.5% TD) of W–27 wt.%Cu composites was obtained by Cu-MI, compared to a low density (89.12% TD) obtained by LPS under the same conditions. In LPS, inter-particle capillary force on grain contacts comes from a liquid that wets the particles and leads to enhanced densification [23]. As a result, the capillary force causes grain rearrangement, densification, and contact flattening. However, full densification of W–Cu composites by LPS was difficult to achieve because of insolubility of W and Cu [10,11]. Several studies also demonstrated the difficulty in attaining full density of W–Cu composites by LPS due to no mutual solubility between W and Cu [10,11,24]. During the Cu-MI process, the combined LPS with LI technique caused densification to take place concurrently in two stages. First, the rearrangement of particles took place because of the formation of liquid Cu in compact bulk. The Cu liquid phase that covers W particles produces a liquid bridge among W particles that generates capillary force. This force decreases the diameter of residual porosity among particles. Therefore, very small pores with sufficient capillarity aid infiltration efficiency [25]. Moreover, both volumetric displacement force and weight of W–Cu green compact on a Cu plate as the source of liquid infiltration combined to push liquid Cu into the W–Cu compact [9,26]. The capillary force acts on the hard particles in metal-matrix composites and the viscosity of copper decreases with temperature and the copper melt surrounds the W particles. The capillary force plays a crucial role to pull the liquid copper between W particles and fill the
Fig. 1. Cycles of both LPS and Cu-MI processes.
224
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
Fig. 2. Sintering and relative densities of W–13Cu and W–27Cu composites prepared by Cu-MI and LPS under the same condition are compared.
pores, and the process depends completely on the wettability between liquid copper and solid tungsten [23,26]. A smaller contact angle means better wetting between the solid and liquid. Ho et al. found the contact angle between solid W and liquid Cu to be less than 20° [26]. The Cu plate (source of melt copper) starts melting and the action of weight of the W–Cu green compact on the Cu plate leads to enhancement of the infiltration. The forces present during the sintering process are capillary, gravity, drag and buoyancy force, according to Mohammad et al. [23].
with an arrow. Specimens prepared by Cu-MI were pores free and highly homogeneous (Fig. 3a and b), whereas the microstructures of W–Cu composites prepared by LPS had high porosity and heterogeneous microstructure (Fig. 3c and d). The microstructures were in agreement with Ahangarkani et al. [9], who found that the microstructure of W– 40 wt.%Cu composites changed with the type of sintering mechanism used.
3.2. Microstructure characteristics
XRD analysis results of W–Cu composites obtained from using Cu-MI method are presented in Fig. 4. As presented in Fig. 4, and based on the reference standard ICSD No.: 98-009-1529 and 98-008-7417 for W and Cu, respectively. Only W and Cu peaks were obtained and no impurities were detected. Lattice parameter values calculated for Cu and W were
Fig. 3 compares the microstructures obtained from sintering W– 13Cu and W–27Cu alloys via LPS and Cu-MI. Light gray and dark gray indicate W and Cu phases, whereas the residual porosity is referred to
3.3. XRD results
Fig. 3. SEM (Secondary Electron Microscopy) micrograph of W–Cu sintered compact: (a) and (b) W–13Cu and W–27Cu prepared by Cu-MI, respectively, and (c) and (d) W–13Cu and W–27Cu prepared by LPS technique, respectively, show up W (light gray) and Cu (dark gray).
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
225
Fig. 4. XRD pattern of W–Cu sintered compacts produced by Cu-MI.
3.6156 Å and 3.1672 Å, respectively. These values that closely matched the standard values of Cu and W are 3.6130 Å and 3.1650 Å, respectively. Due to insolubility between W and Cu elements, only physical bonding occurs [27]. This shows that very negligible solubility exists between W and Cu elements in the composites produced in this work. Although, intensity of Cu phase is totally dependent on its content and fabrication method [4], the high densification results obtained can be attributed to the capillary force. On the other hand, this capillary force is completely dependent on the wettability of copper onto tungsten. Besides, it is well established that the smaller the contact angle, the better the liquid wets the surface [26]. The results obtained in this study are in a good agreement with Mohammed et al. [23] and Das et al. [25], who concluded that capillary force plays a crucial role to enhance the densification of W–Cu alloys. 3.4. Electrical conductivity Electrical conductivity for a thermal package (e.g. a heat sink) and high-voltage electric contact applications are important and easy to measure. Results obtained from using different consolidation methods for W–13Cu and W–27Cu are compared in Table 1. An increase in the Cu concentration resulted in an increased electrical conductivity, which is expected from the high thermal resistivity of pure Cu. The relative density of the consolidated compact effects resistivity because a higher relative density means lower porosity. A high amount of pores restricts the movement of electrons, leading to lower electrical conductivity [7]. Dry air is nonconductive and during sintering, pores entrap carbon and oxide impurities [28], which are usually poor conductors. The relative density of the W–Cu composites had a higher effect on
thermal resistivity compared to the weight fraction of Cu. This observation is very obvious when comparing Cu-MI W–13Cu sintering compact techniques to the LPS W-27Cu composite under the same conditions. These results were in good agreement with Wang and Hwang work [7], who reported that the compacts infiltrated with hydrogen had lower resistivity than those subjected to vacuum. This is because hydrogen had lowered the oxygen content.
3.5. Hardness Vickers hardness results of W–13Cu and W–27Cu composites prepared by LPS and/or Cu-MI techniques are shown in Fig. 5. The hardness of W–13Cu and W–27Cu infiltrated compacts using Cu-MI was higher than for the compacts prepared via LPS under the same conditions. By using Cu-MI, the Vickers hardness of W–13Cu sintered compact was 300 ± 6 HV, while the same composition prepared by LPS under the same conditions gave 166 ± 11 HV. The coincidence between hardness and improved density was reported by Johnson and German [12]. The hardness of the W–Cu composites is also dependent on the relative density of the W–Cu sintered compact; the higher relative density the great hardness [21].
Table 1 Electrical resistivity, electrical conductivity and relative density of W–Cu sintered compacts resulting from LPS and Cu-M techniques at 1150 °C under H2 gas. Composites
Fabrication methods
Relative density (% of theoretical density)
Resistivity (μΩ·cm)
Electrical conductivity (%IACS)
W–13Cu
LPS Cu-MI LPS Cu-MI
80.95 98.98 89.12 99.51
4.71 4.01 4.61 3.41
36.73 42.91 37.08 50.54
W–27Cu
± ± ± ±
3.2 0.5 2.1 0.4
± ± ± ±
1.8 1.1 1.3 0.8
Fig. 5. Comparison of hardness of W–13 wt.%Cu and W–27 wt.%Cu sintered compact prepared by Cu-MI and LPS techniques.
226
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
4. Conclusion Full densities of W–13Cu and W–27Cu composites were obtained by Cu-MI at 1150 °C. The process of fabricating W–Cu by Cu-MI (combined LPS and LI) contributed better results in terms of density. Moreover, the microstructure of W–Cu composites using the Cu-MI method showed lower porosity. Sintering density, hardness, and electrical conductivity of W–Cu prepared using the insert method provided higher values than LPS for the same composition. Solubility between Cu and W was negligible in the W–Cu composites when the Cu-MI (insert method) was used.
Acknowledgments This research work is supported by the Universiti Sains Malaysia (Incentive Research grant number, 1001/227/PBAHAN/8044001) and the University of Omar Al-Mukhtar.
References [1] Kim YD, Oh NL, Oh S-T, Moon I-H. Thermal conductivity of W–Cu composites at various temperatures. Mater Lett 2001;51:420–4. [2] Ibrahim A, Abdallah M, Mostafa SF, Hegazy AA. An experimental investigation on the W–Cu composites. Mater Des 2009;30:1398–403. [3] Qu X-h, Zhang L, Wu M, Ren S-b. Review of metal matrix composites with high thermal conductivity for thermal management applications. Prog Nat Sci Mater Int 2011;21:189–97. [4] Ardestani M, Rezaie HR, Arabi H, Razavizadeh H. The effect of sintering temperature on densification of nanoscale dispersed W–20–40%wt Cu composite powders. Int J Refract Met Hard Mater 2009;27:862–7. [5] Fan J, Liu T, Zhu S, Han Y. Synthesis of ultrafine/nanocrystalline W–(30–50)Cu composite powders and microstructure characteristics of the sintered alloys. Int J Refract Met Hard Mater 2012;30:33–7. [6] Mohammad KS, Rahmat A, Ismail AB. The effects of Fe additions on the liquid phase sintering of W–bronze composites. J Alloys Compd 2009;482:447–54. [7] Wang W, Hwang K. The effect of tungsten particle size on the processing and properties of infiltrated W–Cu compacts. Metall Mater Trans A 1998;29:1509–16. [8] Hamidi AG, Arabi H, Rastegari S. Tungsten–copper composite production by activated sintering and infiltration. Int J Refract Met Hard Mater 2011;29:538–41.
[9] Ahangarkani M, Borgi S, Abbaszadeh H, Rahmani AA, Zangeneh-Madar K. The effect of additive and sintering mechanism on the microstructural characteristics of W– 40Cu composites. Int J Refract Met Hard Mater 2012;32:39–44. [10] German RM, Suri P, Park SJ. Review: liquid phase sintering. J Mater Sci 2009;44:1–39. [11] Lu P, German RM, Iacocca R. Presintering effects on ground-based and microgravity liquid phase sintering. Metall Mater Trans A 2001;32:2097–107. [12] Johnson J, German R. Phase equilibria effects on the enhanced liquid phase sintering of tungsten–copper. Metall Mater Trans A 1993;24:2369–77. [13] Boonyongmaneerat Y. Effects of low-content activators on low-temperature sintering of tungsten. J Mater Process Technol 2009;209:4084–7. [14] Hong S-H, Kim B-K. Fabrication of W–20 wt.% Cu composite nanopowder and sintered alloy with high thermal conductivity. Mater Lett 2003;57:2761–7. [15] Chen P, Shen Q, Luo G, Li M, Zhang L. The mechanical properties of W–Cu composite by activated sintering. Int J Refract Met Hard Mater 2013;36:220–4. [16] Kim J-C, Moon I-H. Sintering of nanostructured W–Cu alloys prepared by mechanical alloying. Nanostruct Mater 1998;10:283–90. [17] Raghu T, Sundaresan R, Ramakrishnan P, Rama Mohan TR. Synthesis of nanocrystalline copper–tungsten alloys by mechanical alloying. Mater Sci Eng A 2001;304–306:438–41. [18] Pintsuk G, Smid I, Döring J-E, Hohenauer W, Linke J. Fabrication and characterization of vacuum plasma sprayed W/Cu-composites for extreme thermal conditions. J Mater Sci 2007;42:30–9. [19] Hamidi AG, Arabi H, Rastegari S. A feasibility study of W–Cu composites production by high pressure compression of tungsten powder. Int J Refract Met Hard Mater 2011;29:123–7. [20] Wang CP, Lin LC, Xu LS, Xu WW, Song JP, Liu XJ, et al. Effect of blue tungsten oxide on skeleton sintering and infiltration of W–Cu composites. Int J Refract Met Hard Mater 2013;41:236–40. [21] Abbaszadeh H, Masoudi A, Safabinesh H, Takestani M. Investigation on the characteristics of micro- and nano-structured W–15wt.%Cu composites prepared by powder metallurgy route. Int J Refract Met Hard Mater 2012;30:145–51. [22] Hashempour M, Razavizadeh H, Rezaie H. Investigation on wear mechanism of thermochemically fabricated W–Cu composites. Wear 2010;269:405–15. [23] Mohammed KS, Rahmat A, Ahmad KR. Sintering behavior and microstructure evolution of mechanically alloyed W–bronze composite powders by two-step ball milling process. J Mater Sci Technol 2013;29:59–69. [24] Johnson JL, Brezovsky JJ, German RM. Effects of tungsten particle size and copper content on densification of liquid-phase-sintered W–Cu. Metall Mater Trans A 2005;36:2807–14. [25] Das J, Chakraborty A, Bagchi TP, Sarma B. Improvement of machinability of tungsten by copper infiltration technique. Int J Refract Met Hard Mater 2008;26:530–9. [26] Ho PW, Li QF, Fuh JYH. Evaluation of W–Cu metal matrix composites produced by powder injection molding and liquid infiltration. Mater Sci Eng A 2008;485:657–63. [27] Yang N, Wang Z, Chen L, Wang Y, Zhu YB. A new process for fabricating W–15wt.%Cu sheet by sintering, cold rolling and re-sintering. Int J Refract Met Hard Mater 2010;28:198–200. [28] Chen P, Luo G, Shen Q, Li M, Zhang L. Thermal and electrical properties of W–Cu composite produced by activated sintering. Mater Des 2013;46:101–5.
Contents lists available at ScienceDirect
Int. Journal of Refractory Metals and Hard Materials journal homepage: www.elsevier.com/locate/IJRMHM
Enhanced liquid-phase sintering of W–Cu composites by liquid infiltration Hafed Ibrahim a,b, Azizan Aziz a,⁎, Azmi Rahmat c a b c
School of Materials and Mineral Resources Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal, Pinang, Malaysia Faculty of Engineering, Omar Almukhtar University, Al-Bayda, Libya School of Materials Engineering, Universiti Malaysia Perlis, Taman Muhibah-Jejawi, Arau 02600, Perlis, Malaysia
a r t i c l e
i n f o
Article history: Received 20 September 2013 Accepted 8 December 2013 Available online 14 December 2013 Keywords: W–Cu composites Liquid-phase sintering Liquid infiltration Microstructure
a b s t r a c t Full-density of consolidated W–Cu composites produced via conventional sintering method is difficult to achieve. In this work, fully-dense W–Cu composites were developed via the combination of the liquid phase sintering (LPS) and the liquid infiltration (LI) methods, which hereinafter is named as Cu-MI technique. It operates at the low sintering temperature of 1150 °C, and maximum densification was possible without requiring a sintering activator such as Ni, Co or Fe. A comparison was also made between the sintering response of W–(13–27 wt.%) Cu composites consolidated using LPS and Cu-MI techniques. The samples were characterized using SEM, EDX and XRD. It was observed that the samples prepared via the Cu-MI method demonstrated a high relative density (N 99% theoretical density). Contrary to the composite sample prepared by the LPS method, the Cu-MI technique accounted for a homogeneous microstructure almost without any pores. The significance of this finding has major industrial implications and has potential to reduce the production costs of composite materials with improved mechanical and electrical properties. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Tungsten–copper (W–Cu) composites are typical pseudo-alloys that combine the natural properties of W, such as the low thermal expansion, high density and hardness, with those of Cu possessing unique thermal properties. Such properties make W–Cu composites very useful in a variety of industrial applications, such as thermal packages (heat sinks), welding electrodes and high-voltage electric contacts [1–4]. W–Cu composites with varying tungsten contents have wide applications [5], e.g. W–Cu composites containing 80–95 wt.%W are suitable for electronic packages, whereas composites with W content of 50–75 wt.% are preferred in electrical contact materials and military applications, such as shaped charge liner and ammunitions [5,6]. In general, there are two common methods applied in fabricating W–Cu alloys, which are the liquid phase sintering (LPS) and the liquid infiltration (LI) methods. LPS involves combining, compacting, and sintering of W and Cu elements under specific conditions. Conversely the LI technique encompasses two steps: firstly, preparation of the W skeleton with desirable pores at a temperature of 1150 °C, followed by filling the W skeleton with melted copper at a temperature of 1250 °C [7–9]. The low solubility of both W and Cu causes difficulty in attaining full or near density in W–Cu composites by LPS [10,11]. Johnson and German [12] and Boonyonagmaneerat [13] showed the effect of adding ⁎ Corresponding author. Tel.: +60 4 5996174; fax: +60 4 5941011. E-mail address: [email protected] (A. Aziz). 0263-4368/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijrmhm.2013.12.004
low concentrations of transition elements (group VIII) as sintering activators to obtain a high density for W–Cu composites by LPS. However, adding transition elements led to unsatisfactory thermal and electrical properties of the W–Cu composites [14] and enhanced mechanical properties [15]. Mechanical alloying was used to enhance the sinterability of W–Cu composites by decreasing the size of the W particles and increasing its distribution [16,17]. However, this process led to the contamination of iron and cobalt by the use of stainless steel and tungsten carbide balls or jars. Such contaminations enhanced densifications, but significantly affect the thermal properties of the W–Cu alloys [14]. Many researchers [18–20] have tried to obtain the full density of W– Cu composites without using additives as a sintering activator. In 2012, Abbaszadeh et al. [21] described a method for fabricating W–15 wt.%Cu composites by conventional mixing–compacting–sintering and mechanochemical reaction–compaction sintering techniques. They obtained a relative density equal to 94% theoretical density (TD) at 1200 °C by the latter technique. Hashempour et al. [22] adopted the method of preparing W–25 wt.%Cu composite powder using thermochemical coprecipitation. Sintering was performed at 1200 °C under H2 gas, which yielded a relative density of 98% TD. In this work, fully dense W–Cu composites (13–27 wt.%Cu) were prepared by Cu-melt infiltration method (Cu-MI; combining LPS and LI) at a low sintering temperature of 1150 °C under protective hydrogen gas. This novel method combined LPS with LI techniques and was very effective in obtaining excellent densification of the W–Cu composites, without using additives as a sintering activator. The insert method
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
(Cu-MI) was compared with LPS under the same condition. Microstructure, relative density, hardness, and thermal resistivity of specimens prepared by both methods were examined.
223
measurements were taken and arranged to represent the sintered compact average electrical resistivity. Recorded resistivity values were then converted to electrical conductivity in percentages of International Annealing Copper Standard (%IACS) based on Eq. (1):
2. Experimental work Tungsten powder with an average particle size of 16 μm (99.9% purity) and copper powder with a mean particle size of 21 μm (99.9% purity) were used in this study. An oxygen-free Cu plate (99.9% purity) was used as the source of liquid infiltration. Mixing of homogeneous W powder with Cu powder to produce green compacts was conducted manually in an alumina mortar for 20 min to prevent particle segregation due to the difference in the densities of W and Cu elements. Typically, the elemental mixture was prepared at the preferred proportions to give the composite powder the desired properties. The mixed powder was die-pressed at 400 MPa to obtain a cylindrical-shaped green compact, 13.2 mm in diameter and 2.0 mm to 2.4 mm in height. The size of the green compact depended mainly on the Cu weight fraction. W– Cu green compacts with desirable weight fractions of Cu were placed on the oxygen-free copper plate (99.9% purity) and then consolidated by Cu-melt infiltration at 1150 °C. Green compacts were likewise prepared by LPS at 1150 °C and used for comparison. LPS and Cu-MI were conducted at 1150 °C using hydrogen as a protective gas. Diagrams of both processes are presented in Fig. 1. Compacted, sintered, and infiltrated densities were measured using the water displacement method (Archimedes principle) based on ASTM B328 standard. Scanning electron microscopy (SEM) was used to investigate the composites' microstructure. X-ray diffraction (XRD) was used to study the crystalline phases of the sintered compacts. Data obtained from the XRD pattern of this composite had been analyzed by X'PertHigh Source Plus software to determine the phases present and lattice parameter of Cu and W phases. Specimens were exposed to a Cu-Kα1 radiation source (λ = 1.5406 Å) and data were collected at 2θ from 10° to 90°. Hardness of specimens was measured by Vickers microhardness under 1 kg load and 15 s duration, and the microhardness value was derived from the average of 10 readings. Electrical conductivity of the sintered composites was determined using the four-point probe resistivity method [7]. It was performed in a fully automated Changmin Tech CMT-Sr2000N apparatus. The sintered compacts were placed perpendicular to the four probes and repeatedly measured across both sides of the surface until the values were constant. Five
%IACS ¼
172:41 ρ
ð1Þ
where ρ is the electrical resistivity in unit of μΩ·cm. 3. Results and discussion 3.1. Densification of W–Cu composites The relation between relative density and sintering method of W–Cu composites is presented in Fig. 2. These were significant differences in the relative density of the W–Cu system for the different sintering methods. The highest density (99.5% TD) of W–27 wt.%Cu composites was obtained by Cu-MI, compared to a low density (89.12% TD) obtained by LPS under the same conditions. In LPS, inter-particle capillary force on grain contacts comes from a liquid that wets the particles and leads to enhanced densification [23]. As a result, the capillary force causes grain rearrangement, densification, and contact flattening. However, full densification of W–Cu composites by LPS was difficult to achieve because of insolubility of W and Cu [10,11]. Several studies also demonstrated the difficulty in attaining full density of W–Cu composites by LPS due to no mutual solubility between W and Cu [10,11,24]. During the Cu-MI process, the combined LPS with LI technique caused densification to take place concurrently in two stages. First, the rearrangement of particles took place because of the formation of liquid Cu in compact bulk. The Cu liquid phase that covers W particles produces a liquid bridge among W particles that generates capillary force. This force decreases the diameter of residual porosity among particles. Therefore, very small pores with sufficient capillarity aid infiltration efficiency [25]. Moreover, both volumetric displacement force and weight of W–Cu green compact on a Cu plate as the source of liquid infiltration combined to push liquid Cu into the W–Cu compact [9,26]. The capillary force acts on the hard particles in metal-matrix composites and the viscosity of copper decreases with temperature and the copper melt surrounds the W particles. The capillary force plays a crucial role to pull the liquid copper between W particles and fill the
Fig. 1. Cycles of both LPS and Cu-MI processes.
224
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
Fig. 2. Sintering and relative densities of W–13Cu and W–27Cu composites prepared by Cu-MI and LPS under the same condition are compared.
pores, and the process depends completely on the wettability between liquid copper and solid tungsten [23,26]. A smaller contact angle means better wetting between the solid and liquid. Ho et al. found the contact angle between solid W and liquid Cu to be less than 20° [26]. The Cu plate (source of melt copper) starts melting and the action of weight of the W–Cu green compact on the Cu plate leads to enhancement of the infiltration. The forces present during the sintering process are capillary, gravity, drag and buoyancy force, according to Mohammad et al. [23].
with an arrow. Specimens prepared by Cu-MI were pores free and highly homogeneous (Fig. 3a and b), whereas the microstructures of W–Cu composites prepared by LPS had high porosity and heterogeneous microstructure (Fig. 3c and d). The microstructures were in agreement with Ahangarkani et al. [9], who found that the microstructure of W– 40 wt.%Cu composites changed with the type of sintering mechanism used.
3.2. Microstructure characteristics
XRD analysis results of W–Cu composites obtained from using Cu-MI method are presented in Fig. 4. As presented in Fig. 4, and based on the reference standard ICSD No.: 98-009-1529 and 98-008-7417 for W and Cu, respectively. Only W and Cu peaks were obtained and no impurities were detected. Lattice parameter values calculated for Cu and W were
Fig. 3 compares the microstructures obtained from sintering W– 13Cu and W–27Cu alloys via LPS and Cu-MI. Light gray and dark gray indicate W and Cu phases, whereas the residual porosity is referred to
3.3. XRD results
Fig. 3. SEM (Secondary Electron Microscopy) micrograph of W–Cu sintered compact: (a) and (b) W–13Cu and W–27Cu prepared by Cu-MI, respectively, and (c) and (d) W–13Cu and W–27Cu prepared by LPS technique, respectively, show up W (light gray) and Cu (dark gray).
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
225
Fig. 4. XRD pattern of W–Cu sintered compacts produced by Cu-MI.
3.6156 Å and 3.1672 Å, respectively. These values that closely matched the standard values of Cu and W are 3.6130 Å and 3.1650 Å, respectively. Due to insolubility between W and Cu elements, only physical bonding occurs [27]. This shows that very negligible solubility exists between W and Cu elements in the composites produced in this work. Although, intensity of Cu phase is totally dependent on its content and fabrication method [4], the high densification results obtained can be attributed to the capillary force. On the other hand, this capillary force is completely dependent on the wettability of copper onto tungsten. Besides, it is well established that the smaller the contact angle, the better the liquid wets the surface [26]. The results obtained in this study are in a good agreement with Mohammed et al. [23] and Das et al. [25], who concluded that capillary force plays a crucial role to enhance the densification of W–Cu alloys. 3.4. Electrical conductivity Electrical conductivity for a thermal package (e.g. a heat sink) and high-voltage electric contact applications are important and easy to measure. Results obtained from using different consolidation methods for W–13Cu and W–27Cu are compared in Table 1. An increase in the Cu concentration resulted in an increased electrical conductivity, which is expected from the high thermal resistivity of pure Cu. The relative density of the consolidated compact effects resistivity because a higher relative density means lower porosity. A high amount of pores restricts the movement of electrons, leading to lower electrical conductivity [7]. Dry air is nonconductive and during sintering, pores entrap carbon and oxide impurities [28], which are usually poor conductors. The relative density of the W–Cu composites had a higher effect on
thermal resistivity compared to the weight fraction of Cu. This observation is very obvious when comparing Cu-MI W–13Cu sintering compact techniques to the LPS W-27Cu composite under the same conditions. These results were in good agreement with Wang and Hwang work [7], who reported that the compacts infiltrated with hydrogen had lower resistivity than those subjected to vacuum. This is because hydrogen had lowered the oxygen content.
3.5. Hardness Vickers hardness results of W–13Cu and W–27Cu composites prepared by LPS and/or Cu-MI techniques are shown in Fig. 5. The hardness of W–13Cu and W–27Cu infiltrated compacts using Cu-MI was higher than for the compacts prepared via LPS under the same conditions. By using Cu-MI, the Vickers hardness of W–13Cu sintered compact was 300 ± 6 HV, while the same composition prepared by LPS under the same conditions gave 166 ± 11 HV. The coincidence between hardness and improved density was reported by Johnson and German [12]. The hardness of the W–Cu composites is also dependent on the relative density of the W–Cu sintered compact; the higher relative density the great hardness [21].
Table 1 Electrical resistivity, electrical conductivity and relative density of W–Cu sintered compacts resulting from LPS and Cu-M techniques at 1150 °C under H2 gas. Composites
Fabrication methods
Relative density (% of theoretical density)
Resistivity (μΩ·cm)
Electrical conductivity (%IACS)
W–13Cu
LPS Cu-MI LPS Cu-MI
80.95 98.98 89.12 99.51
4.71 4.01 4.61 3.41
36.73 42.91 37.08 50.54
W–27Cu
± ± ± ±
3.2 0.5 2.1 0.4
± ± ± ±
1.8 1.1 1.3 0.8
Fig. 5. Comparison of hardness of W–13 wt.%Cu and W–27 wt.%Cu sintered compact prepared by Cu-MI and LPS techniques.
226
H. Ibrahim et al. / Int. Journal of Refractory Metals and Hard Materials 43 (2014) 222–226
4. Conclusion Full densities of W–13Cu and W–27Cu composites were obtained by Cu-MI at 1150 °C. The process of fabricating W–Cu by Cu-MI (combined LPS and LI) contributed better results in terms of density. Moreover, the microstructure of W–Cu composites using the Cu-MI method showed lower porosity. Sintering density, hardness, and electrical conductivity of W–Cu prepared using the insert method provided higher values than LPS for the same composition. Solubility between Cu and W was negligible in the W–Cu composites when the Cu-MI (insert method) was used.
Acknowledgments This research work is supported by the Universiti Sains Malaysia (Incentive Research grant number, 1001/227/PBAHAN/8044001) and the University of Omar Al-Mukhtar.
References [1] Kim YD, Oh NL, Oh S-T, Moon I-H. Thermal conductivity of W–Cu composites at various temperatures. Mater Lett 2001;51:420–4. [2] Ibrahim A, Abdallah M, Mostafa SF, Hegazy AA. An experimental investigation on the W–Cu composites. Mater Des 2009;30:1398–403. [3] Qu X-h, Zhang L, Wu M, Ren S-b. Review of metal matrix composites with high thermal conductivity for thermal management applications. Prog Nat Sci Mater Int 2011;21:189–97. [4] Ardestani M, Rezaie HR, Arabi H, Razavizadeh H. The effect of sintering temperature on densification of nanoscale dispersed W–20–40%wt Cu composite powders. Int J Refract Met Hard Mater 2009;27:862–7. [5] Fan J, Liu T, Zhu S, Han Y. Synthesis of ultrafine/nanocrystalline W–(30–50)Cu composite powders and microstructure characteristics of the sintered alloys. Int J Refract Met Hard Mater 2012;30:33–7. [6] Mohammad KS, Rahmat A, Ismail AB. The effects of Fe additions on the liquid phase sintering of W–bronze composites. J Alloys Compd 2009;482:447–54. [7] Wang W, Hwang K. The effect of tungsten particle size on the processing and properties of infiltrated W–Cu compacts. Metall Mater Trans A 1998;29:1509–16. [8] Hamidi AG, Arabi H, Rastegari S. Tungsten–copper composite production by activated sintering and infiltration. Int J Refract Met Hard Mater 2011;29:538–41.
[9] Ahangarkani M, Borgi S, Abbaszadeh H, Rahmani AA, Zangeneh-Madar K. The effect of additive and sintering mechanism on the microstructural characteristics of W– 40Cu composites. Int J Refract Met Hard Mater 2012;32:39–44. [10] German RM, Suri P, Park SJ. Review: liquid phase sintering. J Mater Sci 2009;44:1–39. [11] Lu P, German RM, Iacocca R. Presintering effects on ground-based and microgravity liquid phase sintering. Metall Mater Trans A 2001;32:2097–107. [12] Johnson J, German R. Phase equilibria effects on the enhanced liquid phase sintering of tungsten–copper. Metall Mater Trans A 1993;24:2369–77. [13] Boonyongmaneerat Y. Effects of low-content activators on low-temperature sintering of tungsten. J Mater Process Technol 2009;209:4084–7. [14] Hong S-H, Kim B-K. Fabrication of W–20 wt.% Cu composite nanopowder and sintered alloy with high thermal conductivity. Mater Lett 2003;57:2761–7. [15] Chen P, Shen Q, Luo G, Li M, Zhang L. The mechanical properties of W–Cu composite by activated sintering. Int J Refract Met Hard Mater 2013;36:220–4. [16] Kim J-C, Moon I-H. Sintering of nanostructured W–Cu alloys prepared by mechanical alloying. Nanostruct Mater 1998;10:283–90. [17] Raghu T, Sundaresan R, Ramakrishnan P, Rama Mohan TR. Synthesis of nanocrystalline copper–tungsten alloys by mechanical alloying. Mater Sci Eng A 2001;304–306:438–41. [18] Pintsuk G, Smid I, Döring J-E, Hohenauer W, Linke J. Fabrication and characterization of vacuum plasma sprayed W/Cu-composites for extreme thermal conditions. J Mater Sci 2007;42:30–9. [19] Hamidi AG, Arabi H, Rastegari S. A feasibility study of W–Cu composites production by high pressure compression of tungsten powder. Int J Refract Met Hard Mater 2011;29:123–7. [20] Wang CP, Lin LC, Xu LS, Xu WW, Song JP, Liu XJ, et al. Effect of blue tungsten oxide on skeleton sintering and infiltration of W–Cu composites. Int J Refract Met Hard Mater 2013;41:236–40. [21] Abbaszadeh H, Masoudi A, Safabinesh H, Takestani M. Investigation on the characteristics of micro- and nano-structured W–15wt.%Cu composites prepared by powder metallurgy route. Int J Refract Met Hard Mater 2012;30:145–51. [22] Hashempour M, Razavizadeh H, Rezaie H. Investigation on wear mechanism of thermochemically fabricated W–Cu composites. Wear 2010;269:405–15. [23] Mohammed KS, Rahmat A, Ahmad KR. Sintering behavior and microstructure evolution of mechanically alloyed W–bronze composite powders by two-step ball milling process. J Mater Sci Technol 2013;29:59–69. [24] Johnson JL, Brezovsky JJ, German RM. Effects of tungsten particle size and copper content on densification of liquid-phase-sintered W–Cu. Metall Mater Trans A 2005;36:2807–14. [25] Das J, Chakraborty A, Bagchi TP, Sarma B. Improvement of machinability of tungsten by copper infiltration technique. Int J Refract Met Hard Mater 2008;26:530–9. [26] Ho PW, Li QF, Fuh JYH. Evaluation of W–Cu metal matrix composites produced by powder injection molding and liquid infiltration. Mater Sci Eng A 2008;485:657–63. [27] Yang N, Wang Z, Chen L, Wang Y, Zhu YB. A new process for fabricating W–15wt.%Cu sheet by sintering, cold rolling and re-sintering. Int J Refract Met Hard Mater 2010;28:198–200. [28] Chen P, Luo G, Shen Q, Li M, Zhang L. Thermal and electrical properties of W–Cu composite produced by activated sintering. Mater Des 2013;46:101–5.