Abstract
Quantifying the resistance to heat flow across well-bonded, planar interfaces is critical in modern electronics packaging architectures, particularly as device length scales are reduced and power demands continue to grow unabated. However, very few experimental techniques are capable of measuring the thermal resistance across such interfaces due to limitations in the required measurement resolution provided by the characterization technique (i.e., Rth < 0.1 mm2·K/W in steady-state configurations) and restrictions on the thermal penetration depth that can be achieved as a result of the heating event that is typically imposed on a sample’s surface (for optical pump-probe thermoreflectance techniques). A recent numerical fitting routine for Frequency-domain Thermoreflectance (FDTR) developed by the authors1 offers a potential avenue to rectify these issues if the transducer’s geometry can be confined. This work utilizes numerical simulations to evaluate the sensitivity of FDTR to a range of thermal boundary resistance (TBR) values as a function of the thermal resistance of adjacent material layers. Experimental measurements are performed across a handful of different material systems to validate our computational results and to demonstrate the the extent to which confined transducer geometries can improve our sensitivyt to the TBR across so-called “buried” interfaces when characterized with FDTR.
