Abstract

Accurately estimating the effective thermal conductivity of membranes is critical to describing process performance in systems with simultaneous heat and mass transport. Unfortunately, existing approaches for modeling the effective thermal conductivity poorly represent real membrane morphologies. Meanwhile, existing measurement techniques are subject to uncertainties from thermal contact resistances, are prone to large systematic errors with low thermal conductivity samples, or require measurement of multiple additional parameters that each introduce additional sources of error. This work introduces the use of comparative infrared microscopy for directly measuring membrane thermal conductivity in highly porous membrane materials. Comparative infrared microscopy negates the need for absolute measurements of heat flow, additional properties such as heat capacity, or any prior assumptions regarding thermal contact resistances, thus overcoming the shortcomings of prior experimental methods. We demonstrate the use of comparative infrared microscopy on three chemically and morphologically distinct membrane distillation membranes. Our results for a specific PVDF membrane are approximately 30% higher than previously reported values measured with a Lees’ disc apparatus, likely due to the influence of additional contact resistances in the Lees’ disc measurement. Our measurements confirm that membrane morphology plays a significant role in effective membrane thermal conductivity and suggest that morphology can guide the selection of theoretical models for approximating membrane thermal conductivity when direct measurements are not possible. The fibrous stretched PTFE membrane is best represented by the series conduction model, while the phase inversion PP and PVDF membranes are better represented by the Maxwell-Eucken model. We conclude with recommendations for further refining thermal conductivity models of structurally heterogeneous membrane materials.