Vacuum-side boundary layer

The resistance of the vacuum-side boundary layer ( g) is negligible compared to the others. The gas-filled membrane wall offers much smaller mass transfer resistance across the membrane. The resistance of the membrane has dominated overall mass transfer resistance in the past. This was because the permeability of the membrane was low and because the membrane was thick. The small resistance observed across the new-generation membranes is also the consequence of the choice of a hydrophobic fiber. The water does not wet this fiber, so its pores remain filled with nitrogen gas. Diffusion through the nitrogen gas is fast, making the membrane resistance unimportant. This conclusion would be different had we used a hydrophilic fiber. This implies that the key to the mass transfer is diffusion in the liquid. The overall mass transfer coefficients of oxygen were observed to be dominated by the individual mass transfer coefficient in the liquid film (Tai et al., 1994). 

Prior to the tests, all the samples were dried in a vacuum oven at 80°C for at least 72 h to minimize the moisture effect and then transferred to a desiccator. Measurements were carried out on a cone calorimeter provided by the Dark Star Research Ltd., United Kingdom. To minimize the conduction heat losses to insulation and to provide well-defined boundary conditions for numerical analysis of these tests, a sample holder was constructed as reported in [14] with four layers (each layer is 3 mm thick) of Cotronic ceramic paper at the back of the sample and four layers at the sides. A schematic view of the sample holder is shown in Figure 19.12. Three external heat fluxes (40, 50, and 60kW/m2) were used with duplicated tests at each heat flux. 

---