Preparation and Property Modulation of Multi-Grit Diamond/Aluminum Composites Based on Interfacial Strategy

Abstract

The development of electronic devices has a tendency to become more complicated in structure, more integrated in function, and smaller in size. The heat flow density of components continues to escalate, which urgently requires the development of heat sink materials with high thermal conductivity and a low coefficient of expansion. Diamond/aluminum composites have become the research hotspot of thermal management materials with excellent thermophysical and mechanical properties, taking into account the advantages of light weight. In this paper, diamond/Al composites are prepared by combining aluminum as matrix and diamond reinforcement through the discharge plasma sintering (SPS) method. The micro-interfacial bonding state of diamond and aluminum is changed by adjusting the particle size of diamond, and the macroscopic morphology performance of the composites is regulated. Through this, the flexible design of diamond/Al performance can be achieved. As a result, when 150 μm diamond powder and A1-12Si powder were used for the composite, the thermal conductivity of the obtained specimens was up to 660.1 W/mK, and the coefficient of thermal expansion was 5.63 × 10⁻⁶/K, which was a good match for the semiconductor material. At the same time, the bending strength is 304.6 MPa, which can satisfy the performance requirements of heat-sinking materials in the field of electronic packaging.

1. Introduction

The continuous development of electronic and information technologies, including 5G communications and artificial intelligence, has brought enormous heat loss challenges. Together with the application requirements of miniaturization and lightweighting of electronic devices, the heat flow density of core components has been rising [1,2,3]. The continuous accumulation of heat will seriously affect the reliability and shorten the service life of the devices [4,5], which requires the development of high thermal conductivity (TC) heat sink materials. During the use of heat sinks, they will experience frequent temperature changes together with electronic components, and the matching of the two coefficients of thermal expansion (CTE) will help to reduce the possibility of failure caused by thermal stress problems [6]. In addition, in order to provide reliable support and protection for components, the mechanical reliability of heat sinks is also an extremely important consideration in the process of electronic packaging, and many use the bending strength value of the material as a reference indicator [7]. The development of heat sink materials with high thermal conductivity, low expansion, and high mechanical reliability is imminent [1,8].

With ultra-high thermal conductivity (1200~2000 W/mK) [9] and an ultra-low coefficient of thermal expansion (1.2 × 10⁻⁶/K), diamond has a unique potential for applications in thermal management [10]. By taking it as the reinforcing phase and choosing typical high thermal conductivity metals (e.g., Ag, Cu, Al, etc.) as the matrix phase, diamond-reinforced metal matrix composites combining both high thermal conductivity and adjustable coefficients of thermal expansion can be obtained, which is considered to be a perfect choice for the next-generation heat-sinking materials [11]. Among them, aluminum with low density, good wettability, and easy accessibility, the optimization of properties and mechanism study of diamond/aluminum composites using it as matrix, has attracted more and more attention. Diamond/aluminum composites can be formed by various methods, such as powder metallurgy [12], squeeze casting [13], gas pressure infiltration [14], and SPS [15,16]. Among them, the SPS method has high heating speed and high preparation efficiency, which can reduce the sintering temperature, shorten the holding time, and effectively inhibit particle recrystallization and grain coarsening processes. In addition, it can improve the mechanical properties of composites by generating heterogeneous (or bimodal) nanostructured materials [17].

The interfacial bonding state of heterogeneous components is a key factor affecting the thermal and mechanical performance of diamond and metal composites. Among them, changes in the crystal quality, filling amount [18], and particle size [19] of diamond particles will have a significant effect on the number and distribution area of the interface and consequently change the overall performance of the composites. Tan et al. [20] had prepared diamond/aluminum composites with a diamond filling amount of 40% by using the vacuum hot pressing method, and the diamond/aluminum composites with a diamond filling amount of 40% were gradually increased in the range of 30 μm~200 μm diamond particle size. The thermal conductivity of the diamond/aluminum composites increased from 313 W/mK to 475 W/mK during the process of increasing the diamond grain size. However, when the diamond grains were too large, the densities of the materials decreased from 99% to 97%, and the bending strengths of the materials also decreased. Chu et al. [21] used a combination of experimental results and model predictions to establish the correlation relationship between the grain sizes and the interfacial structures, as well as to investigate the relationship between different grain sizes and the interfacial structures of different diamond grains. The correlation between grain size and interfacial structure was established to analyze and predict the thermal conductivity of diamond/Al composites made of diamond with different grain sizes. When the diamond grain size is 70 μm, the TC value of the composite material is the highest, which can reach 325 W/mK. The overall performance of the material will have a significant impact, and its mechanism has been analyzed theoretically and verified experimentally. However, the thermal conductivity of the composites prepared so far still has a large room for improvement, and the great thermal conductivity potential of the diamond phase has not been fully utilized. Meanwhile, how to prepare diamond/Al composites by the SPS method and carry out the research on the effect of particle size on performance still needs to be supported by more experimental data.

In this study, we prepared diamond/Al composites by mixing diamond powders of different particle sizes with aluminum alloy powders using the SPS molding method. Starting from the correlation between the size of the raw materials and the heterogeneous interface, the effect of the reinforcing particle adjustment on the properties of the composites was investigated and analyzed. In addition to thermal conductivity, coefficient of thermal expansion, and other important thermal performance indicators, this study also focused on the mechanical properties of composites during the change process, with the bending strength value as the main investigation index for the change in diamond particle size on the mechanical properties of the mechanism of analysis and research. From the practical application requirements of thermal management devices, we developed micro-macro multi-scale and thermal-mechanical multi-dimensional performance studies to optimize the performance of diamond/Al composites and analyze the mechanism. The research could provide new solution ideas for obtaining diamond/Al heat sink materials with high thermal conductivity, a low thermal expansion coefficient, and high mechanical properties.

2. Materials and Methods

2.1. Materials

The aluminum metal matrix of the composite was selected from Al-Si alloy powder containing 12 wt% Si with an average particle size of 10 μm. MBD4-type synthetic diamond particles with diameters ranging from 50 μm to 200 μm were purchased from the Yellow River Hurricane Company of Henan, China. The SPS equipment used was produced by Sumitomo Coal Mining Co., Ltd. (Tokyo, Japan) for composite sintering of Al and diamond powders to prepare diamond/Al composites.

2.2. Preparation of Diamond/Al Composites

The diamond and Al-12Si alloy powders were proportioned according to the 3:2 volume ratio and then fully mixed using a V-type high-efficiency asymmetric mixer. Among them, the thermal conductivity of Al-12Si powder was about 180 W/mK. The mixed powders were loaded into graphite molds of specific specifications and prepared by sintering using an SPS plasma sintering furnace. Prior to heating, the vacuum of the sample chamber of the SPS sintering device was evacuated to below 10 Pa; subsequently, the following heating and pressurization program was set up. The sintering reaction was carried out after heating up to 510 °C at a rate of 50 °C/min, and the uniaxial pressure was maintained at a constant value of 50 MPa. The SPS sintering process was held for 5 min, and after the pressure release and cooling stage, the diamond/Al composites were produced.

2.3. Characterization

The material composition and interfacial physical phase of composites have a great influence on their properties. Using a Rigaku Ultima IV X diffractometer, X-ray diffraction (XRD) tests were performed to characterize the lattice structure information of diamond and Al samples. The X-ray source was Cu-Kα1 (λ = 1.5418 Å), the test range (2Theta) was 10–90°, and the scanning speed was 5°/min. The tube voltage was 40 kV, and the tube current was 100 mA.

The actual density $\mathsf{\rho }$_c of the samples was measured by the Dahometer DH-300 Solid Density Tester (Shenzhen, China) according to the principle of the Archimedes Drainage Method. The theoretical density values of the composites were calculated and derived as per the following equation:

$${\mathsf{\rho }}_0={\mathsf{\rho }}_{\mathrm{Diamond}}\times {\mathrm{V}}_{\mathrm{Diamond}}+{\mathsf{\rho }}_{\mathrm{Al}}\times {\mathrm{V}}_{\mathrm{Al}}$$

(1)

Furthermore, the relative density (RD) value was calculated according to Equation (2), which can reflect the macroscopic denseness of the sample. The value closer to 1 indicates a higher degree of internal densification and a lower number of defects, such as microscopic pores.

$$\mathrm{RD}=\frac{{\mathsf{\rho }}_{\mathrm{c}}}{{\mathsf{\rho }}_{\mathrm{o}}}\times 100\%$$

(2)

The laser flash method (LINSEIS-LFA thermal conductivity tester (Shanghai, China) was applied for the test. The specimen size for testing the thermal diffusivity was Φ12.7 mm × 3 mm, and the thermal diffusivity value ${\mathsf{\alpha }}_{\mathrm{c}}$ and the specific heat capacity value ${\mathrm{C}}_{\mathsf{\rho }}$ measured directly by the instrument need to be further calculated to obtain the thermal conductivity ${\mathsf{\lambda }}_{\mathrm{c}}$ of the material [22]:

$${\mathsf{\lambda }}_{\mathrm{c}}={\mathsf{\alpha }}_{\mathrm{c}}{\mathsf{\rho }}_{\mathrm{c}}{\mathrm{C}}_{\mathsf{\rho }}$$

(3)

where, ${\mathsf{\rho }}_{\mathrm{c}}$ represents the actual measured density of the specimen.

In order to minimize the thermal stress during the use of heat sink materials, the coefficient of thermal expansion of the composites is required to be monitored. Similarly, the specimen for testing the coefficient of thermal expansion was a rectangular shape of 4 mm × 5 mm × 25 mm, and the mixed powder was loaded into a graphite mold of this size prior to preparation for subsequent molding and preparation operations. The specimens were exposed to the LINSEIS thermal expansion coefficient tester (Shanghai, China) for measurement. The constant heating rate of 3 °C/min was applied during the test, and the temperature range was 25 °C to 100 °C. Ar was used as the protective atmosphere during the heating and cooling process to avoid oxidization of the specimen surface. Before the test, the instrument was calibrated with the standard sample of alumina.

In order to ensure that the heat sink material can provide better protection and support for the chip in the process of use, the mechanical properties of the composite material need to be tested and evaluated. The specifications of the specimen for bending strength testing are 4 mm × 3 mm × 25 mm rectangular. The mixed powders were packed into graphite molds of this size prior to preparation for subsequent molding preparation operations. The specimens were placed in the Ligao computerized tensile testing machine (Suzhou, China) for bending performance testing. The micro-morphology of the diamond/aluminum composite at the fracture, the bonding state of the diamond-aluminum heterogeneous interface at the fracture section, and the cause of the fracture were also visualized using a field emission electron scanning microscope (S-4800, HITACHI, Tokyo, Japan).

3. Results and Discussion

In diamond/Al composites, diamond as a reinforcement significantly boosts the mechanical properties of the single aluminum matrix. At the same time, it also assumes the primary function of thermal conductivity due to its unusually high thermal conductivity. For diamond/Al composites, the heat transfer within the diamond is mainly by lattice waves of lattice vibrations. Generally speaking, the quantum excited by the lattice vibration, known as phonons, is the main transport carrier of heat. Within the Al metal matrix, however, it relies mainly on electrons to achieve heat transfer. The difference in heat carriers leads to unavoidable phonon-phonon and phonon-electron coupling processes at the interface formed by the two components. These thermal conducting processes are closely related to the bonding state of the interface, the distribution of interfacial thermal resistance, the coverage area of the thermal network, and so on, and directly determine the thermodynamic performance of the composite. The particle size of diamond is a key parameter affecting the microscopic interfacial state, which is of great research significance.

In order to determine the material composition of the diamond/Al composites, the samples were subjected to XRD tests. Further, in order to make as many diffraction peaks as possible to be clearly shown in the figure, the XRD test results of the diamond/Al composite prepared with a diamond particle size of 150 μm were shown in Figure 1. Evidently, the diffraction peaks of diamond were very sharp and of high intensity, indicating that the material was highly crystallized and was most prominent in its (111) plane. The intensity of this diffraction peak was so high that it even masked part of the diffraction peaks of other substances to a large extent. As the matrix in the composite, the diffraction peaks of Al are also more obvious. Here, the characteristic peaks of Al can mainly correspond to the three crystal planes (111), (200), and (220) in PDF #04-0787. Compared to the two main constituent phases, the diffraction peaks of the product Al₄C₃ are of low intensity, are largely masked by the diamond characteristic peaks, and do not stand out in the figure. For this reason, the left figure was enlarged locally, and the diffraction peaks of the substance were observed at about 31.8°, which proves the existence of the substance.

The overall macroscopic performance is the integration and reflection of the microscopic organization and structure. In order to visually analyze the combination of the two heterogeneous materials in the diamond/Al composites, SEM tests were carried out on the fracture surfaces, as is shown in Figure 2. It could be seen that the diamond particles were wrapped around the metal matrix, and no interfacial adhesion or toughness fossa fracture were found at the fracture. The fracture of the composite was dominated by the toughness of the matrix. The through-crystal fracture phenomenon of diamond was found, indicating that the interfacial bond strength was relatively high in this case. Besides, on the surface of the diamond particles, a certain degree of chemical reaction with the Al matrix was also observed in the form of white carbide particles. The existence of these carbides demonstrated the formation of stronger chemical bonding connections between the diamond and aluminum through chemical reactions, enhancing the interfacial bond. It was also equivalent to pinning to the surface of the matrix. When the composite material was subjected to external forces, the pinning effect formed between the carbide particles and the matrix could inhibit the occurrence of deformation to a certain extent and play a certain buffering effect. A variety of factors synergistically enhance materials and improve their mechanical properties. These carbide particles are able to develop the strong pinning effect with the aluminum matrix, which further improves the mechanical properties of the composites [23].

The relative densities of diamond/Al composites prepared from various diamond sizes were different. Figure 3 demonstrates the relative density values of the four samples. The highest relative density value of up to 99.5% was obtained when the diamond particle was 150 μm. Such high densities were the basis and prerequisite for achieving excellent performance in composites. High densities tend to imply low porosity within the material. The presence of pores introduced a large amount of air, which has a very low thermal conductivity and poor mechanical capacity. The transfer processes of heat and external forces within the material would be hindered or even interrupted. As a result, composites with high porosity usually exhibit correspondingly low thermal conductivity and low mechanical properties when confronted with processes such as heat transfer or the application of external forces. This indicated that the interface between diamond and aluminum showed comparatively low voids and porosity, forming a relatively tight bond between the two components.

The lower relative density at a particle size of 50 μm was primarily due to the selective bonding situation at the interface between the two components. The double-bonded surface atoms on the (100) face of diamond are susceptible to solvation and participate in carbide formation, resulting in strong bonding to the aluminum matrix [24]. In contrast, due to the high chemical stability caused by the triple bonding of its C atoms, the interfaces are difficult to bond with metallic aluminum. That could be prone to the debonding phenomenon, which leads to the existence of gaps with the matrix [25,26]. Under the same sintering process, the energy received during the material formation process is constant. Therefore, as the diamond particle size gradually increases, the number of interfaces decreases, and the bonding process with the aluminum matrix receives a more adequate energy supply. The (111) crystalline surface of the diamond was gradually bonded with the matrix to increase the overall interfacial bond strength of the composite. While the relative density decreased when the particle size was over 200 μm, could result from the smaller interfacial area, longer and deeper reaction time with the substrate under the same infiltration process, and more interfacially reactive carbonation products (Al₄C₃) generated with the substrate, which reduced the actual and relative densities of the material.

Figure 4 demonstrates the TC of diamond/Al composites prepared with different diamond particles. The thermal conductivity values of them presented a trend of increasing and subsequently decreasing with increasing diamond particle size. The highest value of thermal conductivity of the samples was exhibited when the diamond grain size was 150 μm, which showed 660.1 W/mK, while the lowest was when the diamond grain size was 50 μm, which was only 484.4 W/mK. From a microscopic point of view, the formation of interfacial thermal resistance has an extremely important influence on the thermal conductivity of materials. During the generation of diamond/Al composites, a large number of diamond-diamond homogeneous interfaces and diamond-aluminum heterogeneous interfaces were formed, which introduced interfacial thermal resistance and hindered the transfer of heat carriers. Herein, when the particle size of diamond particles was too small, the total surface area was too large, which introduced a large amount of interfacial thermal resistance and reacted to generate more Al₄C₃ carbides upon contact with the aluminum matrix. The excessive generation of Al₄C₃ at the interface not only significantly reduced the TC of the composite but also affected the stability of the diamond/Al composite due to the easily hydrolyzed nature of Al₄C₃ [27]. At the same time, the macro-scale TC of a material was largely determined by the degree of perfection and distribution range of the micro-scale thermal conductivity network, as well as the number of heterogeneous interfaces and the quality of bonding. In particular, defects such as holes, voids, and cracks were often present at heterogeneous interfaces due to the chemical inertness that exists between the reinforced phase and the matrix. At defects, phonons and electrons were not only scattered, but the coupling process between them was also blocked, reducing the efficiency of heat transfer. At this point, the increase in particle size significantly favors the increase in thermal conductivity. However, when the diamond particles was too large, the TC cannot consistently increase but rather decreases slightly. There were two reasons for this: The interface between the diamond and aluminum was poorer, and the filling density decreased when the particle size was too large; on the other hand, the interface between the large particles and matrix was smaller, which resulted in the generation of more Al₄C₃ in the same infiltration process. Through chemical bonding connections, the Al₄C₃ interfacial layer could promote the tightness of interfacial bonding to some extent. However, the excess also reduced the overall thermal conductivity value of the composites due to their low TC value and susceptibility to hygroscopicity [28,29,30].

In order to investigate the effect of phonons and electrons on heat transfer in diamond/Al composites, theoretical calculations of their thermal conductivity were carried out using the e differential medium model (DEM) model. At the metal-diamond junction, processes such as phonon transport and phonon-electron coupling would be mainly involved. Therefore, we first introduced the acoustic mismatch model (AMM) to theoretically calculate the interfacial thermal resistance of the composites, combined it with the DEM model, and analyzed the final results in comparison with the actual test values.

The AMM model assumes that all phonons interact with the interface, i.e., no scattering of phonons occurs at the interface. It was used to describe the thermal transport of phonons at the interface and to calculate the penetration probability of a phonon hitting the interface using methods from the theory of acoustics of continuous media. The interfacial thermal conductivity of the diamond/Al composite was determined by the diamond and aluminum matrix on both sides of the interface. At this time, it was assumed that no scattering of phonons occurs at the interface, according to the Debye model of phonon mismatch theory [27,31]:

$$h=\frac{1}{4}\times {\rho }_{in}\times C_{in}\times {\upsilon }_{in}\times \eta$$

(4)

where, ${\rho }_{in}$ was the density of the material on the side of the phonon incident direction, $C_{in}$ was the specific heat capacity of the material on the side of the phonon incident direction, ${\upsilon }_{in}$ referred to the speed of the phonon, and $\eta$ represented the probability of the phonon penetrating at the interface. Combined with the intrinsic properties of diamond/Al composites at the interface, the thermal conductivity of diamond/Al composites at the interface could be calculated using the AMM model.

The DEM involved the embedding of reinforcing phase particles of volume, particle size, and thermal conductivity into an Al matrix of thermal conductivity to obtain a homogeneously mixed new material, and the following mathematical model was obtained [32,33]:

$$(1-V_r){\left(\frac{K_C}{K_m}\right)}^{\frac{1}{3}}=\frac{K_r^{eff}-K_C}{K_r^{eff}-K_m}$$

(5)

$$K_r^{eff}=\frac{K_r}{1+\frac{K_r}{h_{c}a}}$$

(6)

where, $K_C$ and $K_m$ were the thermal conductivity of the composite and the metal matrix, respectively, $V_r$ was the volume fraction of the reinforcing phase, $K_r$ was the thermal conductivity of the reinforcing phase, $h_c$ was the interfacial thermal conductivity between the two phases, and $a$ was the average size of the reinforcing phase. As shown in the figure, there was still a deviation between the theoretically calculated and actually measured values of thermal conductivity due to the presence of a small number of microscopic pores in the actual samples. However, the higher densities allowed the difference to be controlled within a small range, with the highest only occurring at a particle size of 200 μm, which was 10.32%. Compared with the Maxwell-Eucken model and the H-J model, the DEM took into account the effects of diamond volume content, particle size, and interfacial thermal resistance on the thermal conductivity of the composites when predicting. Especially, the prediction of the thermal conductivity at higher volume content was of great theoretical guidance [34]. According to the DEM model, for a given diamond particle and Al, with a determined TC, the thermal conductivity of diamond/Al composites will have an increasing trend with the increase in diamond particle sizes.

By combining the results obtained from the AMM tests with the DEM model, the theoretical thermal conductivity values of diamond/Al composites corresponding to different diamond grain sizes can be further calculated. The values of the relevant parameters are shown in

Table 1. Basic parameters of different materials, adapted from Ref. [31].

Materials	Density (kg/m3)	TC (W/mK)	Specific Heat (J/kg·K)	Phonon Velocity (m/s)
Diamond	3520	1800	512	13,430
Al	2700	237	895	3620

[31].

According to the comparison results, it can be concluded that there is still a certain gap between the theoretical and experimental values. This was due to the fact that the DEM model did not take into account the presence of defects such as pores at the interface in the actual samples and that the lattice mismatch and interfacial bonding also contribute to the interfacial thermal conductivity in the actual application process [31]. However, it was found that the deviation of the theoretical thermal conductivity values calculated by the model from the actual values is small, and the match between the experimental values and the predicted values is relatively high due to the relatively high diamond filling amount.

As a material for electronic packaging applications, diamond/Al composites should have a CTE that matches the semiconductor material to which they are attached in order to minimize cracks and failures caused by thermal stresses during the operation of electronic devices and to prolong their service life. Figure 5 demonstrated the values of the coefficient of thermal expansion for the samples, which gradually decreased from 7.76 × 10⁻⁶/K to the minimum of 5.63 × 10⁻⁶/K and then slightly increased to 5.90 × 10⁻⁶/K as the diamond particle size increased. This tendency was likewise related to the differences in the number and area of the introduced heterogeneous micro-interfaces.

During the reaction process of SPS, the temperature increased, and the molecules and atoms absorbed energy and moved more actively. At this time, aluminum tended to undergo a rapid expansion in volume due to its large coefficient of thermal expansion (about 23.6 × 10⁻⁶/K). Diamond, on the other hand, with a lower coefficient of thermal expansion (only about 1.0 × 10⁻⁶/K) [35], exerted some compressive stress on the thermally expanding aluminum matrix [18]. If the heterogeneous interface was well bonded, the internal stresses in the Al matrix could be quickly transferred from the interface to the inside of the diamond particles. The diamond itself had a very high modulus of elasticity, which could counteract the internal stress effect above, thus playing a significant role in limiting the expansion process of Al. On the contrary, if the interfacial bond was weak, the transfer process of thermal stresses to the interior of the diamond was hindered, and accumulation occurred at the interface, the restraining effect of the diamond on thermal expansion would be significantly weakened. The generation and transfer of this effect mainly came from the heterogeneous interface formed by the contact and bonding of diamond and Al. Therefore, the difference in the bonding state of the interface, as well as in the degree of force applied during changes in the external temperature, resulted in the difference in the CTE. According to this theory, as the diamond grain size increased, the number and area covered of interfaces gradually decreased, with the bonding force and restraining effect to thermal expansion decreasing accordingly, which would manifest as a corresponding increase in CTE [36]. However, in the range of 50 to 150 μm, the CTE of the composites instead decreased slightly. As mentioned above, when the diamond particles were too small, the metal “infiltration” process would be subjected to more resistance, and the “infiltration” would be more difficult. Under the same insulation and pressure holding process, it would be more likely to have incomplete encapsulation of the matrix in the reinforcing phase or a poor reaction, resulting in weakened bonding strength. Additionally, when the element content in the aluminum alloy is certain, the carbide content generated by the reaction with diamond is also basically the same. When the diamond particles were too small, the surface area increased significantly. If the interface was to be fully bonded, more carbides needed to be introduced to achieve complete coverage of the diamond by the metal, which was clearly contradictory to reality. As the particle size increased, this phenomenon was gradually alleviated, and the interfacial bonding strength was significantly improved. The combined effect of the two factors causes the CTE of the composite material to show a tendency to decrease and then increase.

In order to play a certain role in supporting and protecting the chip, the mechanical properties of the heat sink are also one of the performance parameters worth paying attention to. The strengthening mechanism of diamond-reinforced metal matrix composites is similar to that of the diffusion strengthening of alloys. When the composite is subjected to an external force, the load is transferred from the plastic aluminum matrix to the rigid diamond particles without being sufficient to initiate through-crystal fracture of the high-strength particles, thus improving the strength of the mechanical properties of the composite [37]. Figure 6 illustrates the flexural strength values of diamond/Al composites prepared from diamond particles of different sizes. Similar to the change in TC, the four bending strength values also showed the same trend of increasing first and then gradually decreasing with the increase in diamond size. Combined with the analysis of the microscopic interfacial state at the fracture, when the particle size was 150 μm, the interfacial bonding between the matrix phase and the reinforcing phase was stronger, and diamond through-crystal fracture occurred during the force fracture process, which tended to predict considerable flexural strength. As mentioned earlier, when the diamond grains were too small, defects such as incomplete matrix coverage and poor interfacial bonding were likely to occur within the composite, which would significantly weaken the mechanical performance of the material. However, when the particle size continued to increase, it also led to a weakening of the interfacial strength due to the smaller number of interfaces where the reaction occurred [9]. Meanwhile, the surface activity of diamonds with too large particle sizes was low, which made it difficult to closely bond with the metal matrix. When subjected to external force, it was difficult for diamond to deform with the matrix, and then the composite was prone to fracture due to the debonding phenomenon of diamond from the matrix [38].

Figure 7 demonstrates the performance comparison of diamond/aluminum composites prepared by different molding processes and different parameters, and the diamond/aluminum composite samples prepared by the SPS sintering method investigated in this paper have obvious advantages in terms of thermal conductivity, coefficient of thermal expansion, and other performance manifestations. In conclusion, we obtained diamond/Al samples with high densities by optimizing the preparation parameters of the SPS method. The large amount of air introduction in cases of high porosity was avoided, which is an important prerequisite for the excellent performance of the composites. On this basis, the purpose of multiscale modulation of diamond/Al micro-interfacial bonding and macroscopic property performance was realized only by adjusting the particle size of diamond particles. Compared with other methods, the modulation is more flexible, easier to operate, and more effective. The high densities of the samples, coupled with the continuous optimization of the heterogeneous interfacial states, achieve the excellent performance of the diamond/Al composite samples in this paper. In addition, it can lay an important experimental foundation for other process parameter studies.

4. Conclusions

In summary, in diamond/Al composites, the difference in diamond particle size changes the number and coverage area of the microscopic heterogeneous interfaces, which in turn affects the magnitude of interfacial bonding and interfacial thermal resistance. This manifests itself at the macroscopic scale as a change in the overall thermal and mechanical property performance of the composite. Based on this, aluminum alloy powder with Al-12 wt% Si and diamond powder with an average particle size of 150 μm were selected to produce diamond/Al composites by the SPS sintering method under the conditions of 510 °C, 50 MPa, and a holding time of 5 min. Due to the low interfacial thermal resistance and good interfacial bonding strength, the resulting samples can reach a relative density of 99.5% with dense organization and low porosity and can achieve a TC of up to 660.1 W/mK. The low CTE of the composite (5.63 × 10⁻⁶/K) facilitates the formation of a good expansion match with the semiconductor material in practical applications. At the same time, it also has a high flexural strength value of 304.6 MPa, realizing a comprehensive improvement in multi-dimensional performance parameters. This diamond/Al composite can better meet the application requirements in electronic packaging and provides a solution for efficient heat dissipation and the safe application of high-heat flow density electronic devices.