Best guess is to use DEFINE_DPM_HEAT_MASS or DEFINE_DPM_SOURCE along with multicomponent droplets and a species mixture for the fluid phase. Most of the default examples are for evaporation, if you set a negative rate it ought to condense too (I don't know for sure). The rate of transfer is going to be fun too: it's possibly concentration rather than temperature based. I suspect the latter may be easier to control.
We're not permitted to get too specific on here, so I can't give too much more advice on the macros. I do advise breaking the model down, start with looking at whether you have good fluid mixing and particle distribution.