Research Article  Open Access
Effect of Titanium Addition on the Thermal Properties of Diamond/CuTi Composites Fabricated by Pressureless LiquidPhase Sintering Technique
Abstract
In this study, minoraddition elements such as Si, Co, Cr, W, Mo, and Ti were added to matrix to improve the wettability between the diamonds and Cu matrix. The pressureless liquidphase sintering technique adopted in this study provides a lowcost method for producing diamond/Cu composites with high potential for industrial mass production. Thermal properties of the diamond/CuTi composites fabricated by pressureless liquidphase sintering at 1373 K with variation in Ti contents were thoroughly investigated. XRD and TEM analysis show that TiC layer formed in the interface between Cu and diamond. The composites exhibited thermal conductivity as high as 620 W/m·K for 50 vol% diamond/Cu0.6 at % Ti composite with diamond particle size of 300 µm. This value comes up to 85% of the thermal conductivity calculated by the Hasselman and Johnson (HJ) theoretical analysis. Under these conditions, a suitable coefficient of thermal expansion of 6.9 ppm/K was obtained.
1. Introduction
Conventional thermal management materials with thermal conductivities (above 200 W/m·K) and coefficients of thermal expansion (4–8 ppm/K) were summarized by Zweben [1]. However, owing to the increase in device power and integration levels [2], the demand for novel thermal management materials with high thermal conductivities is on the rise.
Due to their excellent thermal properties, synthetic diamonds with high thermal conductivity (1200–2000 W/m·K) are the most attractive thermal spreading material. However, diamond films utilized to spread heat are too thin to spread efficiently and it is very difficult to form bulk synthetic diamond. Therefore, high thermal conductive metals, such as Al, Ag, and Cu, were utilized as binders to form bulk composites [3–7] reinforced with diamond to match this demand. However, due to the poor wettability between diamond and metal, the interfaces between diamond and metal are not in well contact and this will be detrimental for heat conduction between the interfaces. There are two alternative methods to improve the contact between diamond and metal. The first method is to fabricate composites under high pressure (above 1 GPa) and high temperature (above the melting point of metal) process conditions which requires expensive and complex equipment [8]. The second method can be achieved through carbideforming elements addition to form a thin carbide layer at the interface and to improve the contact and wettability between diamond and metal [9–16].
In recent years, high thermal conductive diamond/metal composites used as heat sink have been fabricated by various techniques with high pressure, expensive, and complex equipment, such as hot pressing [17], gas pressure infiltration [18–20], and spark plasma sintering [21]. In this study, diamond/Cu composites were fabricated by pressureless liquidphase sintering technique with minoraddition elements to improve the contact and wettability between the diamond and Cu matrix. The results in this study demonstrate that pressureless liquidphase sintering technique is a very promising process for fabricating diamond/CuTi composites with the potential for mass production.
2. Materials and Methods
Saw diamond powders with thermal conductivity of 1800 W/m·K and a particle size of 300 μm (YK9E, Fine Abrasives Taiwan Company) were used as reinforcements. Cu powders with thermal conductivity of 400 W/m·K and with a particle size of 5 μm (purity in 99.9 wt%, Artc1000 A, Advance Research Technology Corporation) and minoraddition element powders (Si, Co, Cr, W, Mo, and Ti) with a particle size of 100 μm (purity in 99.9 wt%, Alfa Aesar) were mixed as matrix materials.
Diamond, copper, and minoraddition element powders were mechanically mixed and compacted under 700 MPa for 10 minutes to fabricate diamond/Cu composites with a diameter of 12.9 mm and height of 3 mm and then sintered in tube furnace with H_{2} (100 sccm) gas flow under vacuum (1.8 torr) at 1373 K for 30 minutes.
Micrographs and element compositions of diamond/CuTi composites were analyzed by a scanning electron microscope (SEM, JEOL5410) and an Xray diffractometer (XRD, Rigaku_D/DMAXIIB). Thermal conductivity was measured via laserflash analyzer (LFA, NetzschLFA 447). Coefficient of thermal expansion (CTE) was measured by thermal analysis apparatus (TMA, SeikoSSC5200) with a heating rate of 3 K/min, ranging from 373 K to 473 K. The Archimedes method was applied to measure the densities of the composites and the relative densities were obtained by comparing measured densities with the theoretical densities. Focused ion beam (FIB, FEINova 200) technology was used for the preparation of transmission electron microscope (TEM, JEOLJEM3000F) samples. The samples were then observed under TEM to reveal the boundary regions of the diamond/Cu interface.
3. Results and Discussion
3.1. Diamond/Cu Composites with Varying MinorAddition Elements
The sample images of 50 vol% diamond/Cu composites with 3 at % varying minoraddition elements are shown in Figure 1. It can be seen that only W, Mo, and Ti addition can form integralshape composites without matrix repelling from the composites. Of all the chosen minoraddition elements, W, Mo, and Ti all belong to the carbideforming elements. The other minoraddition elements added to the composites are incapable of forming integralshape composites due to matrix repelling from the composites during the sintering process.
(a)
(b)
(c)
(d)
(e)
(f)
Figure 2 shows the surface SEM images of the 50 vol% diamond/Cu composites with 3 at % (a) W, (b) Mo, and (c) Ti additions. Figures 2(a) and 2(b) show apparent interfacial pores at the boundaries between the diamonds and Cu matrix. It is considered that W and Mo additions can form integralshape composites but cannot improve the contact between the diamonds and Cu matrix efficiently. In Figure 2(c), it can be seen that the diamonds and Cu matrix are in well contact at the boundaries and the Cu matrix can cover the diamond surfaces entirely. Furthermore, the thermal conductivities of the composites are 209, 257, and 348 W/m·K for 3 at % W, Mo, and Ti additions, respectively. The highest value of thermal conductivity was obtained with Ti addition in this study. Therefore, Ti is regarded as the most suitable carbideforming element in this study to improve the contact and wettability at the boundaries between the diamonds and Cu matrix.
(a)
(b)
(c)
3.2. Microstructure and Composition Analysis of Diamond/CuTi Composites
The sample images of the 50 vol% diamond composites with varying Ti addition (0.2, 0.3, and 0.6 at %) in the Cu matrix are shown in Figure 3. It can be seen that the addition of 0.2 at % Ti (Figure 3(a)) is not enough to form an integralshape composite without matrix repelling. The integralshape composites can be achieved only above 0.3 at % Ti addition. Composite with 0.6 at % Ti addition has a good integralshape as shown in Figure 3(c).
(a)
(b)
(c)
Figure 4 shows the surface SEM images of the 50 vol% diamond composites with varying Ti addition in the Cu matrix. Figure 4(a) shows apparent interfacial pores at the boundary between the diamonds and the matrix appearing in Cu0.3 at % Ti composites. With increasing Ti content, fewer interfacial pores were found in the contact regions between the diamonds and the Cu matrix. When the Ti content increased to 0.6 and 0.9 a t%, there were no apparent interfacial pores and the diamond surface was entirely covered by the copper matrix as shown in Figures 4(c) and 4(d). These phenomena indicate that increasing Ti content improves the wettability and contact between Cu matrix and diamond surface effectively, and therefore less interfacial pores were produced during the process.
(a)
(b)
(c)
(d)
The Xray diffraction patterns of the diamond/CuTi composites, shown in Figure 5, indicate that the peaks only belonged to diamond, Cu, and TiC. It can be deduced that TiC forms during the sintering process. The peak intensity grows stronger with increasing Ti content, indicating greater amount of TiC formation.
Figure 6 shows the TEM bright field image and EDS interface linescan signals of a 50 vol% diamond/Cu0.6 at % Ti composite interface. The interface between Cu matrix and diamond is composed of a columnar structure near Cu and an irregular layer near diamond. The C, Ti, and Cu EDS line scans were used to determine the distribution of element across the interface, as shown in Figure 6. It is clear that the scan path consists of four zones: (a) Cu matrix region, (b) Tirich columnar region, (c) CuC irregular region, and (d) diamond region. The insert in Figure 6 is the diffraction pattern of circle region in Tirich columnar structure and indicates that this columnar structure is a TiC crystal structure. This is consistent with the Xray diffraction analysis as shown in Figure 5. This columnar TiC layer is about 270 nm in thickness. The CuC irregular layer is comprised of Cu and C which may have resulted from the reaction during sintering. The structure of this irregular layer is composed of Cu and C. However, the mechanism for the formation of this interfacial structure still requires further investigation.
3.3. Thermal Conductivity and Coefficient of Thermal Expansion (CTE)
Figure 7 shows the curves of relative density, thermal conductivity, and coefficient of thermal expansion versus Ti content with (a) 50 vol% diamond and (b) 60 vol% diamond. It can be seen that the relative densities decrease with increasing diamond volume fraction, due to the increase in interfacial areas and the more bridging effect between diamonds.
(a)
(b)
Figure 7 indicates that initially the thermal conductivity values increase with increasing Ti addition, upon reaching a maximum, and then the thermal conductivities begin to show apparent decrease with further addition. The Ti addition enhanced the wettability between diamond and Cu, thus promoting the interfacial contact between the matrix and diamond during pressureless liquidphase sintering and increasing the thermal conductivities of the composites. Thermal conductivity of the highest 620 W/m·K value can be achieved for 50 vol% diamond/Cu0.6 at% Ti composite. However, upon exceeding 0.7 at% Ti addition, the thermal conductivity decreases drastically.
It is considered that varying amounts of diamond surface areas need varying Ti addition to improve the wettability of the interface and then have the highest thermal conductivity in each diamond volume fraction condition. By assuming diamonds as spheres with a diameter of 300 μm for simplified calculation, the diamond surface areas of the 50 vol% condition (the volume of the composite is m^{3}) are m^{2} and 60 vol% condition (the volume of the composite is m^{3}) is m^{2}. Considering that the highest thermal conductivity of 50 vol% condition is obtained with 0.6 at % Ti addition and comparing the surface areas values of 50 vol% and 60 vol%, it can be calculated that the optimum Ti addition in 60 vol% condition is 0.92 at%. The experimental results verify this calculation, as shown in Figure 7(b). The 60 vol% diamond composites exhibit the highest thermal conductivity up to 579 W/m·K with 0.92 at% Ti content. However, upon exceeding this optimum 0.92 at% Ti addition, the thermal conductivity decreases drastically as the trend in the 50 vol% diamond composites.
The above results indicate that suitable Ti addition and TiC thickness strengthen the interfacial bonding to improve thermal conductivity. However, the thermal conductivities of the composites later decreased owing to the excess formation of the TiC interphase (Figure 6) with low thermal conductivity during the sintering process. Thermal conductivities of 60 vol% diamond composites are lower than those of 50 vol% diamond composites. These can be attributed to lower relative densities of the composite under 60 vol% conditions. It is considered that the suitable Ti addition in each diamond volume fraction condition is proportional to the amounts of diamond surface areas.
The CTE values in this study are in the range of 5 to 8 ppm/K, which are the adequate values for thermal management materials that are to be utilized in electronic devices [1]. The CTE values of the composites decrease monotonously from 7.7 to 6.5 ppm/K (for composites with a diamond content of 50 vol% and 0.3–0.9 at % Ti) or from 5.9 to 5.1 ppm/K (for composites with a diamond content of 60 vol% and 0.52–1.02 at% Ti) with increasing Ti content. This can be explained by the improvement in bonding at the interface with higher content of Ti addition. The CTE values of 60 vol% diamond composites are lower than those of 50 vol%. This is due to the fact that the low CTE value of diamonds (1.3 ppm/K) provides greater contribution to the CTE values of composites comprised of higher diamond content, thus reducing the final CTE value outcomes of the composites.
In recent years, high thermal conductive Cu/diamond composites with thermal conductivities of 700, 657, 493, and 615 W/m·K fabricated by highcost techniques have been published. The composites produced in this study demonstrate thermal conductivities comparable with findings in previous literature made by other authors [18, 21–23]. However, in comparison with these techniques, the fabrication technique applied in this study provides a lowcost, simple alternative to fabricate diamond/Cu composites with great potential for industrial mass production.
3.4. Theoretical Analysis of Thermal Conductivity
In order to fully understand the thermal conduction behavior in diamond/CuTi composites, it is crucial to compare the experimental results with theoretical predictions. The Hasselman and Johnson (HJ) model applied in combination with effective medium approximation (EMA) scheme is the most popular predictive method, in which the Kapitza resistance effect and particle size are taken into consideration [23]: where is the thermal conductivity, is the volume fraction of reinforcement, and the subscripts , , and refer to the composite, matrix, and reinforcement particles, respectively. The effective thermal conductivity of reinforcement particles, , is defined as [23]: where is the average diameter of reinforcements and is the interfacial resistance (Kapitza resistance). However, the model was made under the assumption that the interface between the matrix and reinforcements is perfect and sharp [23, 24]. In the case of this study, interphases formed at the interface, which contributed additional thermal resistance to the initially assumed diamond/metal sharp interface. To obtain an accurate model for thermal conductivity prediction, the thermal resistance of the interphases should be considered. In order to simplify the calculation, the CuTiC irregular layer is still neglected. The interfacial thermal resistance can be expressed by where , , and are the thermal resistance of Cu/TiC interface, TiC interphase, and TiC/Diamond interface, respectively, and is the thermal conductivity of TiC.
The thermal resistance, and , can be described by diffuse mismatch model (DMM). This interphase thermal resistance, , based on DMM model is related to the phonon velocities () in the phase and the 3 phase, as well as the absolute temperature , can be expressed in the following form [23]: where the subscripts and refer to the two adjacent phases on each side of the interface. The subscript represents the mode of phonon velocity (longitudinal or transverse).
In this modeling, the effect of porosity was taken into account by reassigning an effective thermal conductivity to the composite. The effective thermal conductivity of the composite was derived from combining the thermal conductivity of composite and air according to the rule of mixture (ROM): where represents the thermal conductivity of composites containing pores. is the result of thermal conductivity of the composite calculated by (1)–(4). and are the volume fraction of composite and porosity, respectively.
Table 1 shows the longitudinal, transverse phonon velocities [24, 25] and thermal conductivities of Cu, TiC, and diamond in theoretical values. Combining the above equations (1)–(5) and theoretical values, the values of the interfacial resistance and theoretical thermal conductivity can be calculated as shown in Table 2.


The calculated theoretical thermal conductivity is 725 W/m·K and the experimental value is 620 W/m·K as shown in Table 2 and Figure 7(a), respectively. It appears that the experimental value comes up to 85% of the theoretical value. The good comparison results between the theoretical analysis and experimental measurements support the feasibility of incorporating the consideration of additional interfacial thermal resistance effect caused by the interphase into HJ model.
4. Conclusions
Diamond/CuTi composites with high thermal conductivities can be successfully fabricated via pressureless liquidphase sintering method. XRD and TEM analysis show that the TiC layer formed at the interface between Cu and diamond. Under the condition of 50 vol% diamond/Cu0.6 at % Ti, the composite shows the highest value of thermal conductivity of 620 W/m·K and a low CTE value of 6.9 ppm/K, which comes up to 85% of the thermal conductivity calculated by the Hasselman and Johnson (HJ) theoretical analysis. Applying pressureless liquidphase sintering method to fabricate high thermal conductive composites provides an easy and costeffective alternative method to produce diamond/CuTi composites with high thermal conductivities for thermal management applications.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgment
The authors would like to thank the National Science Council, Taiwan, for supporting this research through Project no. NSC1022221E007046MY3.
References
 C. Zweben, “Advanced materials for optoelectronic packaging,” Electronic Packaging and Production, vol. 42, no. 9, pp. 37–40, 2002. View at: Google Scholar
 D. B. Miracle, “Metal matrix composites—from science to technological significance,” Composites Science and Technology, vol. 65, no. 1516, pp. 2526–2540, 2005. View at: Publisher Site  Google Scholar
 J. H. Wu, H. L. Zhang, Y. Zhang, J. W. Li, and X. T. Wang, “Effect of copper content on the thermal conductivity and thermal expansion of AlCu/diamond composites,” Materials and Design, vol. 39, pp. 87–92, 2012. View at: Publisher Site  Google Scholar
 W. B. Johnson and B. Sonuparlak, “Diamond/Al metal matrix composites formed by the pressureless metal infiltration process,” Journal of Materials Research, vol. 8, no. 5, pp. 1169–1173, 1993. View at: Google Scholar
 K. Hanada, K. Matsuzaki, and T. Sano, “Thermal properties of diamond particledispersed Cu composites,” Journal of Materials Processing Technology, vol. 153154, no. 1–3, pp. 514–518, 2004. View at: Publisher Site  Google Scholar
 O. Beffort, F. A. Khalid, L. Weber et al., “Interface formation in infiltrated Al(Si)/diamond composites,” Diamond and Related Materials, vol. 15, no. 9, pp. 1250–1260, 2006. View at: Publisher Site  Google Scholar
 R. Tavangar, J. M. Molina, and L. Weber, “Assessing predictive schemes for thermal conductivity against diamondreinforced silver matrix composites at intermediate phase contrast,” Scripta Materialia, vol. 56, no. 5, pp. 357–360, 2007. View at: Publisher Site  Google Scholar
 K. Yoshida and H. Morigami, “Thermal properties of diamond/copper composite material,” Microelectronics Reliability, vol. 44, no. 2, pp. 303–308, 2004. View at: Publisher Site  Google Scholar
 Z. Q. Tan, Z. Q. Li, G. L. Fan et al., “Enhanced thermal conductivity in diamond/aluminum composites with a tungsten interface nanolayer,” Materials and Design, vol. 47, pp. 160–166, 2013. View at: Google Scholar
 P. Xiao and B. Derby, “Wetting of titanium nitride and titanium carbide by liquid metals,” Acta Materialia, vol. 44, no. 1, pp. 307–314, 1996. View at: Publisher Site  Google Scholar
 A. M. Abyzov, S. V. Kidalov, and F. M. Shakhov, “High thermal conductivity composites consisting of diamond filler with tungsten coating and copper (silver) matrix,” Journal of Materials Science, vol. 46, no. 5, pp. 1424–1438, 2011. View at: Publisher Site  Google Scholar
 T. Schubert, B. Trindade, T. Weißgärber, and B. Kieback, “Interfacial design of Cubased composites prepared by powder metallurgy for heat sink applications,” Materials Science and Engineering A, vol. 475, no. 12, pp. 39–44, 2008. View at: Publisher Site  Google Scholar
 H. Bai, N. G. Ma, J. Lang, and C. X. Zhu, “Effect of a new pretreatment on the microstructure and thermal conductivity of Cu/diamond composites,” Journal of Alloys and Compounds, vol. 580, pp. 382–385, 2013. View at: Google Scholar
 Q. P. Kang, X. B. He, S. B. Ren et al., “Effect of molybdenum carbide intermediate layers on thermal properties of copperdiamond composites,” Journal of Alloys and Compounds, vol. 576, pp. 380–385, 2013. View at: Google Scholar
 X. Y. Shen, X. B. He, S. B. Ren, H. M. Zhang, and X. H. Qu, “Effect of molybdenum as interfacial element on the thermal conductivity of diamond/Cu composites,” Journal of Alloys and Compounds, vol. 529, pp. 134–139, 2012. View at: Publisher Site  Google Scholar
 K. Chu, Z. Liu, C. Jia et al., “Thermal conductivity of SPS consolidated Cu/diamond composites with Crcoated diamond particles,” Journal of Alloys and Compounds, vol. 490, no. 12, pp. 453–458, 2010. View at: Publisher Site  Google Scholar
 B. B. Liu, J. X. Xie, and X. H. Qu, “Fabrication of WCu functionally graded materials with high density by particle size adjustment and solid state hot press,” Composites Science and Technology, vol. 68, no. 6, pp. 1539–1547, 2008. View at: Publisher Site  Google Scholar
 L. Weber and R. Tavangar, “On the influence of active element content on the thermal conductivity and thermal expansion of CuX (X = Cr, B) diamond composites,” Scripta Materialia, vol. 57, no. 11, pp. 988–991, 2007. View at: Publisher Site  Google Scholar
 I. E. Monje, E. Louis, and J. M. Molina, “Optimizing thermal conductivity in gaspressure infiltrated aluminum/diamond composites by precise processing control,” Composites A, vol. 48, pp. 9–14, 2013. View at: Google Scholar
 P. W. Ruch, O. Beffort, S. Kleiner, L. Weber, and P. J. Uggowitzer, “Selective interfacial bonding in Al(Si)diamond composites and its effect on thermal conductivity,” Composites Science and Technology, vol. 66, no. 15, pp. 2677–2685, 2006. View at: Publisher Site  Google Scholar
 S. Ren, X. Shen, C. Guo et al., “Effect of coating on the microstructure and thermal conductivities of diamondCu composites prepared by powder metallurgy,” Composites Science and Technology, vol. 71, no. 13, pp. 1550–1555, 2011. View at: Publisher Site  Google Scholar
 Y. Zhang, H. L. Zhang, J. H. Wu, and X. T. Wang, “Enhanced thermal conductivity in copper matrix composites reinforced with titaniumcoated diamond particles,” Scripta Materialia, vol. 65, no. 12, pp. 1097–1100, 2011. View at: Publisher Site  Google Scholar
 K. Chu, C. C. Jia, H. Guo, and W. S. Li, “On the thermal conductivity of CuZr/diamond composites,” Materials and Design, vol. 45, pp. 36–42, 2013. View at: Google Scholar
 E. T. Swartz and R. O. Pohl, “Thermal boundary resistance,” Reviews of Modern Physics, vol. 61, no. 3, pp. 605–668, 1989. View at: Publisher Site  Google Scholar
 B. Klein, N. Frage, E. Zaretsky, and M. P. Dariel, “Dynamic response of titanium carbidesteel, ceramicmetal composites,” AIP Conference Proceedings, vol. 620, pp. 1119–1122, 2002. View at: Google Scholar
Copyright
Copyright © 2014 ChihYu Chung et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.