Thermal management of electronics is becoming more challenging, as billions of transistors are packed in microprocessors leading to increase in the transistor density and heat generation. Due to imperfections at the numerous interfaces in an electronics packaging, the contact area of the mating surfaces is just a few percent of the nominal surface area of either surfaces. Composite materials such as high conductivity filler particles dispersed in a polymer matrix are a popular choice for thermal interface materials (TIMs) that provide efficient heat conduction pathways from the microprocessor to the heat sink. Without TIMs, chip performance eventually deteriorates due to elevated temperatures and this will lead to chip failure. On an industrial assembly line, TIMs are first dispensed on the chip, heat spreader, or the heat sink using a nozzle via an automated process. Various dispense patterns such as dot, line, spiral, serpentine, “X” and star shapes exist. The TIM is then squeezed to spread over the substrate by using the alternate component (i.e., the device, heat spreader, or the heat sink), often followed by curing to form a rigid bond. During squeezing, the particle-laden TIM generally exhibits non-Newtonian behavior [1] and, after squeezing, the particle spatial distribution may be non-uniform [2]. The flow behavior depends on the TIM dispense pattern, squeeze rate, squeeze force/pressure, and the TIM composition (e.g., particle shape, size distribution, volume fraction, matrix viscosity and particle-matrix density ratio). The velocity and applied pressure during squeezing significantly impact the achievable bond line thickness (BLT) and the particle spatial distribution. This can potentially cause the thermal performance of the TIM to deviate from vendor-specified thermal characteristics.

Prior work has focused on (a) macroscopic analysis of TIMs using experiments and models, and (b) particle-level thermal conduction models. However, these works do not consider the effect of the application process on the TIM microstructure. Moreover, experimental investigation of 3D TIM microstructure is lacking and there are open questions regarding the effect of squeezing on the redistribution of particles and its effect on thermal conduction within the particle network. In this work, 3D X-ray micro-computed tomography (XRCT) is used to visualize and quantify the spatial distribution of particles in the TIM after dispensing and squeezing processes. A mock TIM with a target loading of 30 vol% copper microspheres (median diameter 114 um) is created by hand-mixing the particles with a UV-curable epoxy (UV cure 60-7158). Line patterns are dispensed using an in-house designed machine and constant velocity squeezing is performed using an Instron mechanical tester. The squeezed TIM samples are cured using UV light. The 3D XRCT images are processed to extract particle locations and equivalent sphere diameter. Then, microstructural features such as the average particle volume fraction, coordination number and radial distribution function (RDF) are computed to gain insights into the spatial distribution of particles in the TIM. Bulk thermal conductivity is then measured using a miniaturized reference bar method integrated with infrared (IR) microscopy. Thermal imaging enables visualization of heat conduction pathways within the TIM and allows for measurement of bulk material thermal conductivity and contact resistance.