Thanks for your reply, I've attached some pictures of the mass transfer model here. It's species-mass-transfer, constant sat. pressure, both with Ranz-marshall correlation. ia-symmetric is used for interfacial area. "Liquid" contains a mixture just with liquid water in it, and "air-vapor" has a mixture of water vapor and air.
One other behavior I noticed - when I flipped the sign of the standard state enthalpy difference to be negative, temperatures rose in the vapor region instead of falling, with the same magnitude and profile shown above. I suppose a more relevant question might not be about temperature divergence, but more so why the temperature changes are occurring in the vapor phase instead of the liquid phase.