Page 137 - Introduction to Transfer Phenomena in PEM Fuel Cells

126 Introduction to Transfer Phenomena in PEM Fuel Cells
Several models reflecting the heat transfer phenomenon in a PEMFC
have emerged; initial efforts have been developed by Nguyen and White
[NGU 93]. A PEMFC is built and regulated in such a way that its
temperature is as uniform as possible to avoid creating hot spots in its core.
The presence of these hot spots could have adverse effects on the materials,
in particular on the polymer membrane. However, it is difficult to keep the
temperature of the cell uniform, as a temperature gradient exists between the
inlet and the outlet of the cooling circuit [COL 08]. This gradient could be
used to boost the transport of water and thus the electrical performance of
the battery. Strong thermal gradients, for example, would lead to the
transport of strong water flows and this would limit waterlogging [BRA 06].

The simulation of these thermal phenomena by thermal models is
therefore essential in order to be able to correctly estimate how a cell works
when its temperature field is not uniform.

Thermal models have only recently appeared in the literature, with the
exception of the article by Fuller and Newman [FUL 93] who were already
interested in the thermal effects in the core of fuel cells and who stress the
importance of coupled modeling of heat and mass transfers in the cell core to
correctly describe the phenomenon of water sorption in the membrane and,
consequently, its hydration. Nguyen and white [NGU 93] developed a two-
dimensional (2D) model of a PEMFC in which they demonstrated a one-
dimensional (1D) heat transfer model in the direction of flow. This model
considers only the phase change of the water in the flow channels as the only
heat source allowing convective heat transfer between the gas and the solid.
Fuller and Newman [FUL 93] developed a pseudo-2D thermal model with a
mass transfer (1D) through the membrane, and a heat transfer (1D) in the
direction of flow.

More recently, the effects of temperature on the diffusion of water in the
membrane have been modeled by Yan et al. [YAN 04].

The risks of membrane dehydration (in particular, at the anode) at high
operating temperatures or at high current densities were shown through
coupled modeling of heat and mass transfers in the membrane alone. Yi and
Nguyen [YI 99] used Nguyen and White’s model [NGU 93] to introduce the
heat due to entropy and irreversibility at the same time as the heat due to
phase change. This model allows the temperature of the solid phase to vary
in the direction of flow only, assuming a uniform temperature in the