To provide a high power density thermal management system for proton exchange membrane (PEM) fuel cell applications, a passively driven thermal management system was assembled to operate in a closed loop two-phase thermosyphon. The system has two major components; a microchannel evaporator plate and a condenser. The microchannel evaporator plate was fabricated with 56 square channels that have a 1 mm × 1 mm cross section and are 115 mm long. Experiments were conducted with a liquid cooled condenser with heat flux as the control variable. Measurements of mass flow rate, temperature field, and pressure drop have been made for the thermosyphon loop. A model is developed to predict the system characteristics such as the temperature and pressure fields, flow rate, flow regime, heat transfer coefficient, and maximum heat flux. When the system is subjected to a heat load that exceeds the maximum heat flux, an unstable flow regime is observed that causes flow reversal and eventual dryout near the evaporator plate wall. This undesirable phenomenon is modeled based on a quasi-steady state assumption, and the model is capable of predicting the heat flux at the onset of instability for quasi-steady two-phase flow.

Another focus of this work is the performance of the condenser portion of the loop, which will be air cooled in practice. The aim is to reduce air side thermal resistance and increase the condenser performance, which is accomplished with extended surfaces. A testing facility is assembled to observe the air side heat transfer performance of three aluminum foam samples and three modified carbon foam samples, used as extended surfaces. The aluminum foam samples have a bulk density of 216 kilograms per cubic meter with pore sizes of 0.5, 1, and 2 mm. The modified carbon foam samples have bulk densities of 284, 317, and 400 kilograms per cubic meter and machined flow passages of 3.2 mm. in diameter. Each sample is observed under forced convection with air velocity as the control variable. Thermocouples and pressure taps are distributed axially along the test section and measurements of pressure and temperature are recorded for air velocities ranging from 1-6 meters per second. Using the Darcy-Forcheimer equation, the porosity is determined for each sample. The volumetric heat transfer coefficient is extracted by means of solving the coupled energy equations of both the solid and fluid respectively. Nusselt number is correlated with Reynolds number. The optimal foam configuration is explored based on a Coefficient of Performance, (COP), Compactness Factor (CF) and Power Density (PD). The COP is the ratio of total heat removed to electrical heat consumption of the blower, CF is the total heat removed per unit volume, and PD is the total heat removed per unit mass. These performance parameters are computed for a hypothetical heat exchanger using each foam sample at various fluid velocities. They are also compared against those for the hypothetical heat exchanger fitted with conventional louvered fins. Given a proper weighting function based on the importance of CF, COP, and PD in the condenser design, an optimal configuration for an air cooled condenser can be obtained for various operating conditions.