COMBINED EXPERIMENTAL-NUMERICAL INVESTIGATION OF MICROSTRUCTURE AND THERMAL CONDUCTION IN SQUEEZED THERMAL INTERFACE MATERIALS
Thermal management of electronics is one of the biggest engineering challenges of this decade, as billions of transistors are put in each microprocessor. The increasing transistor density leads to significant amount of Joule heating exceeding 100 W/cm^2 and, hence, increased temperatures. Hotspots with heat fluxes in excess of 1 kW/cm^2 locally increase the temperature leading to non-uniform chip temperatures. Thermal interface materials (TIMs) have been developed to efficiently conduct the heat away from the chip to the heat sink for dissipation to the surroundings. The main goal of a TIM is to provide high effective thermal conductivity and minimize contact thermal resistance at a minimal thickness of the material after application. On an industrial assembly line, TIMs are generally dispensed over the substrate at a controlled flow rate and/or quantity and then squeezed to a final pressure to form the desired bond line thickness. This work is focused on investigating the influence of assembly procedures on the TIM final microstructure and thermal performance using a combined experimental-modeling approach. TIMs consisting of high thermal conductivity particles (e.g., ceramic, metallic, or carbon-based) in a polymer matrix that are cured in application are the main focus of this study. Automated procedures for dispensing and constant velocity squeezing are developed in this work to investigate constant velocity squeezing of isolated line dispense patterns of TIMs. Three-dimensional (3D) X-ray micro computed tomography (XRCT) is used to quantify metrics of the TIM microstructure such as the bulk and local particle volume fraction and coordination number. New user-defined functions are integrated into the open-source software Multiphase Flow with Interphase eXchanges (MFIX), authored by the National Energy Technology Laboratory (NETL), to enable discrete element methods (DEM) simulations of TIM squeezing. Further, a finite element (FE) thermal conduction modeling framework is developed based on a particle size reduction method to predict the TIM bulk thermal conductivity.