In the near future I will be tasked with modeling an ultrasonic wave nebulizer, a device which uses a piezoelectric diaphragm to convert liquid water into mist. I'm wondering what would be the best way to tackle this in software.

My initial idea is to model it as a mixture of two gasses: air and an artificial "mist gas" fluid that would represent the slightly denser mist. After I have a steady-state solution I would go back and do a particle analysis as well. I'm curious if anybody has any insight or advice on this matter. For one thing I am still not sure what kind of properties the "mist gas" will have. From preliminary experiments I have seen that it flows much more viscously than air, but I do not have any derived thermodynamic quantities to work with. Additionally I have seen that water droplets which are dropped onto the nebulizer cause a mist to rapidly form and expand, almost as if they were flash boiling. Again the problem is the numbers, I am not sure how to define a mass flow rate (i.e. how much air is entrained into the mist) or an average velocity.