You could switch in-flight, but you would actually increase the gas costs since you’d need 100% helium for the trip up as well as the trip down. Also, the risks

You could switch in-flight, but you would actually increase the gas costs since you’d need 100% helium for the trip up as well as the trip down. Also, the risks would dramatically increase in exigent circumstances. What if the gas container fails, causing it to make an unexpected descent? Even for a planned descent, there has to be a way to switch back to helium. So now you have to have at least 2 times the helium needed to attain buoyancy, plus mechanisms to generate hydrogen and vent/replace gas. And systems/procedures to handle edge cases. It sounds like a lot of complexity and cost added. However, it might be worth it if extremely high altitude is needed. The altitude ceiling could be increased significantly (not sure how much, but <2x) for a given configuration and payload.