A difficulty could be that bio processes are meant to create abundent and cheap products. Not quite sure that a drying agent would be cheap enough. At least, it should be recycled by the hydrogen producer, through simple means.
A few special places to collect the condensation water before freezing look more promising. Though, some vapour will remain at 0°C, which will make frost at lower temperatures. So maybe:
- Use a drying agent only there, once most vapour has already condensed
- Let vapour make frost at an acceptable special place, and from time to time, blow reversed warmer air to sublimate the frost.
- Other gasses than water vapour will make ice before 20K! Like methane, carbon dioxide, nitrogen, oxygen, argon... Will you remove each one separately? Or can you design a cold machine that survives some ice of varied nature, which you can filter out of the liquid hydrogen?