An experimental investigation on the fabrication of W-Cu composite through hot-press

In this paper the feasibility of fabricating W-Cu composite by hot-press has been studied and the best parameters for hot-pressing were acquired. W/20 vol%Cu composite was successfully prepared by this process, in which at first the powder of tungsten and copper, containing 20% vol copper were mixed in a ball mill for 3 h at rotation speed of 200 rpm. Then the mixed powders were hot-pressed for 3 h at a compact pressure of 30 MPa and temperatures of 1,250°C, 1,350°C, and 1,450°C. The composites made have been investigated and revealed the making W-Cu composite with good properties included density, hardness, modules of elasticity, flexural strength, and microstructure. Also, the sample made at 1,450°C possesses better properties, and its microstructure showed the tungsten matrix and the copper reinforcement separately; the X-ray diffraction patterns showed that no compound has been formed between W and Cu.

Composite materials are engineered or naturally occurring materials made from two or more constituent materials with significantly different physical or chemical properties which remain separate and distinct at the macroscopic or microscopic scale within the finished structure. W-Cu composites are made by mixing a refractory phase called tungsten, which possess high strength and low coefficient of thermal expansion, and a Cu-phase having high thermal and electrical conductivity [1]. The combination of these elements optimizes some properties such as ductility, strength, and corrosion and wear resistance. In the recent years, W-Cu composites have gained great importance in automotive, electrical, and military industry because of their high thermal conductivity, low thermal expansions, high wear resistance and excellent electrical conductivity. They are used as electrical contacts, resistance welding electrodes, heat sinks, electro-forging dies, packaging materials, and so on [2,3].

The fabrication of a full density W-Cu composite is very difficult. Because of the big difference between melting points of tungsten (approximately 3,400°C) and copper (approximately 1,083°C), there is no overlap of sintering temperature ranges. Also, tungsten-copper system has no mutual solubility which means poor sinter ability [1]. The infiltration of a porous-sintered tungsten skeleton by liquid copper is one of the most common methods for producing W-Cu composite. However, in this method defects like pores, copper lakes, and tungsten agglomerates form easily, so this technique results in a poor quality [2-4]. Other methods to fabricate W-Cu composites include the following: thermal-mechanical process [2], metal powder injection molding [5], hot extrusion [6], liquid sintering and hot-hydrostatic extrusion [7-9], microwave sintering [10], resistance sintering [11], vacuum plasma spray [12,13], mechanical alloying, and other new methods [14-18].

Although one of the most important methods in producing composites is hot-press, in which the mixed powders will be subjected to high pressure and high temperature at the same time, this method has not been used as sole way for making W-Cu composite. Of all the advantages that this method contains, increasing the properties of row and sintered material such as density, strength, and fatigue properties is the main one. In this experiment, we use hot-press to produce W-Cu composite because of its simplicity and low cost.

Physical properties

The bulk density of pieces made at 30 MPa and temperatures of 1,250°C, 1,350°C and 1,450°C was measured about 15.21, 15.76, and 16.25 g/cm³ respectively. Also, the porosity of specimens can be measured through this method. For example, the apparent porosity which does not include the volume of the sealed pores was evaluated about 0.78% for composite made at 1,450°C.

Since in Brinell hardness test method, the diameter of ball indenter is big and pressure is applied to the bulk of sample, this method was used to calculate the hardness of the work pieces. According to the results, the hardnesses of the samples made at 30 MPa and temperatures of 1,250°C, 1,350°C and 1,450°C were equal to 131, 149, and 163 HB respectively.

For calculating modulus of elasticity in accordance with ASTM C769 test standard (ASTM International, West Conshohocken, PA, USA), the speed of sound in the sample must be calculated first. In this way, a 2.5 × 2.5 cm sample is made and then an ultrasonic pulse with frequency of 4 MHz will be sent to the surface of samples and then will be received. By recording the return time of sound and sample thickness, the speed of sound (v) passing the sample was obtained; by using the following relations, modulus of elasticity (E) in GPa was obtained:

Thus according to calculated density for the composite, the modulus of elasticity are 312.5, 325.4, and 341.8 GPa for samples made at 1,250°C, 1,350°C and 1,450°C, respectively.

Flexural strength of samples is obtained based on the ASTM C1161 test standard. According to this test standard, specimens with dimensions of 3 × 4 × 45 mm are made and the surfaces are prepared by polishing. Then the load was applied by flexural strength machine with speed of 0.5 mm/min and distance (l) of 40 mm between two bases of applying load. By drawing the force-strain diagram, the maximum fracture force (F) is recorded. Using the following relation, flexural strength or modulus of rupture (MOR) is calculated in MPa,

where F is fracture force (N); B is sample width (mm); D is thickness of sample (mm); and L is the distance between two bases of applying load (mm).

Correspondingly, the flexural strength of composite made at 1,250˚, 1350˚ and 1450°C is 381.2, 350.7 and 343.8 MPa.

The study of microstructure

Since it is clear that the best result was obtained from the sample made at 30 MPa pressure and temperatures of 1,450°C, we perform the microstructure study just for this sample. After preparation and polishing, the parts as a result were seen under optical microscopy and scanning electron microscopy. The etching of work pieces is done according to ASTM 98 C standard, in which 10 g of ferrocyanide potassium (K₃Fe(CN)₆) and 10 g of NaOH is added to 10 ml distilled water. As shown in Figure 1, the optical micrographs of W-Cu composite made in 1,450°C and 30 MPa copper and tungsten phases are visible. The porosity in the composite is also seen, one of the main defects that exists in hot-press.

Also by secondary scanning electron microscopy (SEM) and back scattered electron (BSE) the morphology of samples and their particle size and shape are studied. Figure 2 shows the SEM, and Figure 3 shows the BSE micrographs of the W/20 vol%Cu composite prepared by hot-press at 1,450°C and 30 MPa.

In Figure 4, the elemental analysis map of composite can be seen. It is clear in the picture that the composite is made of tungsten, copper, and nickel.

On the other hand, as it is clear in the back-scattered electron images, the composite is made of dark phase and bright phase. According to the feature of back-scattered electron images, the bright phase is related to heavier elements because heavier elements reflect more light. However, to prove this issue, energy dispersive X-ray spectroscopy (EDX)-line scan and elemental analysis of both phases can be used. As shown, the dominant bright phase is about 80% volume and the dark phase of copper can also be seen in the image. To prove that bright phase belongs to tungsten and dark phase belongs to copper, an EDX-line scan from a bright grain and EDX-line scan from the dark grains are taken and shown in Figures 5 and 6.

Also, the fact that the bright phase belongs to tungsten and dark phase belongs to copper can be shown by elemental analysis of the bright and dark phases. Figure 7 shows the elemental analysis of bright and dark phases.

Figure 8 shows the XRD patterns of starting tungsten and copper powders before production of composite. Also the XRD result of W/20%volCu is present in the image. As clearly seen in the pictures for the pure powders before making composite, only peaks related to copper and tungsten are seen. However, in the XRD patterns of composite made from tungsten and copper, there are only peaks related to these two elements, which show that no compound between these two elements has been formed. Furthermore, because of the low amount of copper in the composite, about 10 weight percent, peaks related to copper have a low intensity in comparison with peaks related to tungsten.

The elemental tungsten powder and copper powder were used as raw materials. In order to make W/20vol%Cu composite, weight percent of powders must be calculated first. For this purpose, the weight percent of copper and tungsten in the desired composite by Equation 3 is calculated, according to density and volume percent of Cu and W, into percent Wt_Cu = 10.34 and percent Wt_W = 89.66.

On the other hand, the theoretical density of composite is easily calculated by adding up the mass of each component:

Where f is the volume fraction of fibers or the second phase, ρ is density, m is mass, V is volume, m_f is mass of second phase and m_m is mass of matrix. According to these equations, the theoretical density of composite is equal to ρ = 17.22 g/cm³.

The required amount of powders is calculated considering the density of composite and size of the piece. In this experiment, we attempt to produce a piece with dimensions of 12 × 12 × 0.4 cm, so the volume will be 57.6 cm³ and m = 991.87 g. Hence, we need 991.872 g of tungsten-copper powder to make the piece, and the value of each of these two components is calculated as follows:

Thus to fabricate a piece with aforementioned dimensions from W/20% volCu composite, it must be seen how much tungsten and copper powder is required. Therefore at first stage, the required amount of each powder was weighed and then mixed for 3 h in a ball mill, with rotation speed of 200 rpm. A 2%wt of nickel powder was added to the composite as well as to reduce the composite density, considering the low density of nickel (approximately 8.9 g/cm³), in comparison with the composite (approximately 17.22 g/cm³). The mixed powders, after being dried, were performed at room temperature under 80 MPa pressure and then hot-pressed for 3 h with the compaction pressure of 30 MPa at three different temperatures of 1,250°C; 1,350°C; and 1,450°C to obtain the optimum temperature. The heating rate was chosen at 10°C/min. The sintering process is shown in Figure 9. The hot-pressing was also done under high-purity argon (99.99%) atmosphere to avoid the powder from being oxidized.

The density of composite was measured through Archimedes test, according to ASTM B311 standard. This test standard is mostly used to determine the density of powder metallurgy parts and is based on the water displacement method. The hardness was evaluated in accordance with Brinell hardness test method ASTM E10. In addition, the bending strength of the composite has been measured by bending test according to ASTM C1161, and the modulus of elasticity was measured as well by ASTM C769 test standard method, in which the modulus of elasticity is evaluated by the help of sound velocity in the work piece. Lastly, the microstructure of made specimens was investigated by optical microscopy and scanning electron microscopy.

In this study, the feasibility of fabricating W/Cu composite via hot-press was studied. The W/20 vol%Cu composites were made at 30 MPa pressure and temperatures of 1,250°C, 1,350°C and 1,450°C showed good properties, including density, hardness, modulus of elasticity, and flexural strength. The microstructure showed the tungsten matrix and the copper reinforcement distinctly, and the XRD patterns showed that no compound has been formed between W and Cu. The best results were obtained for the sample made at temperatures of 1,450°C.

The authors declare that they have no competing interests.

MR, the first author, carried out making the composite and testing it. HB, participated in the design of tests and coordination . HA, conceived of the study, and helped to draft the manuscript. All authors read and approved the final manuscript.

The authors would like to express their gratitude to Dr. Ebrahim Afshari for providing language help and useful suggestion in writing the article.

1. Dongdong G, Shen Y (2008) Mater Sci Engin. 489:169–177 Publisher Full Text

2. Ho PW, Li QF, Fuh JYH (2008) Mater Sci Engin. 485:657–663 Publisher Full Text

3. Amirjan M, Zangeneh-Madar K, Parvin N (2008) Int J Refractory Metals Hard Mater (in press) :–

4. Yunping L, Yu S (2008) Int J Refract Metals Hard Mater. 26:540–548 Publisher Full Text

5. Ibrahim A, Abdallah M, Mostafa SF (2009) Abousree Hegazy A. Mater Design. 30:1398–1403 Publisher Full Text

6. Li D, Liu Z, Yu Y, Wang E (2008) J Alloys Comp. 462:94–98 Publisher Full Text

7. Li D, Liu Z, Yu Y, Wang E (2008) Int J Refractory Metals Hard Mater. 26:286–289 Publisher Full Text

8. Liu BB, Xie JX, Qu XH (2008) Compos Sci Technol. 68:1539–1547 Publisher Full Text

9. Luo S, Yi J, Guo Y, Peng Y, Li L (2009) J Ran J J Alloys Comp. 473:L5–L9 Publisher Full Text

10. Maneshian MH, Simchi A (2008) J Alloys Comp. 463:153–159 Publisher Full Text

11. Junbo W, Wen'ge C, Bingjun D (2006) J Wuhan Univ Technol-Mater Sci 21:32–35

12. Li DR, Liu ZY, Yu Y, Wang ED (2009) Mater Sci Engin. 499:118–122 Publisher Full Text

13. Zhou Z, Du J, Song S, Zhong Z, Ge C (2007) J Alloys Comp. 428:146–150 Publisher Full Text

14. Lee J (2008) Int J Refractory Metals Hard Mater. 26:290–294 Publisher Full Text

15. Pintsuk IG, Smid J, Doring E, Hohenauer W, Linke J (2007) J Mater Sci. 42:30–39 Publisher Full Text

16. Kim S, Kim Y, Suk M (2007) Metals Mater Int. 13(5):391–394 Publisher Full Text

17. Kang H, Kang SB (2004) J ThermalSprayTechnology 13(2):223–228

18. Hong SH, Kim BK (2003) Mater Lett. 57:2761–2767 Publisher Full Text