Characterization of thermal interface materials using flash diffusivity and infrared microscopy methods

Surface roughness limits the actual contact area between two surfaces and yields a high thermal resistance at the junction. By increasing the actual contact area between surfaces, thermal interface materials (TIMs) reduce the interfacial thermal resistance. However, in many applications, TIMs act as a bottleneck in the removal of heat from the chip to the heat sink. Estimation of the TIM thermal resistance is an important task in thermal management of electronic devices. For thin TIM layers, known as the bondline, high performance particle-laden TIMs have bulk thermal conductivity different from its in-situ value. TIM thermal resistance includes the bulk resistance and the contact resistances at the chip-TIM and TIM-heat sink interfaces. Thermal contact resistance is crucial for a very thin TIM bondline. Presently, TIM characterization techniques like the steady-state thermal gradient stage and flash diffusivity method implicitly determine the contact resistance based on a number of measurements. This suggests a need to develop a method that directly characterizes TIM contact resistance.;The objectives of this study are to (i) develop a methodology based on infrared microscopy for the direct characterization of TIM contact resistance and (ii) compare the flash diffusivity, thermal gradient stage, and infrared microscopy methods for TIM characterization. The flash diffusivity method measures the total TIM resistance, where the TIMs are sandwiched between Si-Si, Si-Al, and Al-Al substrates. A steady-state horizontal thermal gradient stage, aligned with the focal plane of an infrared (IR) microscope, measures the thermal contact resistance. In this method, the IR microscope characterizes the bulk TIM resistance by measuring the temperature drop due to the bulk TIM. Subtracting it from the total TIM resistance, the thermal contact resistance is directly determined in a single experiment. Uncertainties of the measurements are estimated for each TIM characterization technique.;Uncertainties in the TIM thermal resistance were more than 40% due to the large position errors of the thermocouples. Replacing thermocouples with RTDs reduced the uncertainties to less than 15% by a better control of the RTD positions. A parametric study was carried out to examine the effects of bondline thickness, heat flux and surface roughness on the thermal contact resistance. The results show that for the gap filler type TIMs, the TIM contact resistance decreases with the decrease in bondline thickness from 900 to 200 mum. For the bondline thickness of 900 mum, the increase in heat flux across the TIM bondline from 35650 to 44950 W/m2 decreases the thermal contact resistance. By changing the surface roughness of an Al substrate from 589 to 78 nm, it was observed that the thermal contact resistance increases for a bare Al-Al interface for a rougher surface. However, for the condition considered, the thermal contact resistance of the gap filler type TIM decreases for the case of a rougher surface.