Feasibility Study of Active Thermal Energy Storage for On-Chip Transient Mitigation

Abstract

Phase-change materials (PCMs) have been widely investigated for their potential application in transiently driven electronic systems. The passive, nonlinear thermal capacitance offered by solid-liquid or solid-solid phase change promises to provide thermal load leveling under dissipation transients. This in turn allows a steady-state thermal management system (TMS) sized to that of the average component power dissipation to be operated at full utilization. The result is a lower size, weight, power, and cost (SWAP-C) of the aggregate PCM/TMS transient thermal solution.One of the challenges in using PCMs becomes limitations in coupling energy into or out of the thermal energy storage (TES) in a timely fashion. Just as liquid-vapor systems must contend with critical heat flux limiting the manageable system power, so too must a static phase change system present sufficient storage power density for the application at hand. For low conductivity PCMs (paraffin waxes, sugar alcohols, salt hydrates) this is often accomplished using conductivity enhancements such as particulate additives, conductive fins, or expanded metal foams. An alternative is to use low melting point metals (LMPM) and alloys, which have higher intrinsic conductivity (and thus storage power density) and comparable volumetric energy storage density. However, even conductivity enhanced LMPM composites will reach a limit to storage power density, leaving very high flux applications un-addressable using a low form-factor, passive solution.This work investigates a new TES paradigm that seeks to extend the storage power density of LMPM composites by active thermal transport. The impact of geometry, materials, and operating condition on performance are assessed. Of particular interest are tradeoffs that result in higher storage power density at a cost of lowered coefficient of performance, which manifests as a lower effective energy storage density. These active systems are compared to baseline passive systems using an effective figure-of-merit as determined from transient finite element simulation.