# Would a hydrogen/oxygen fuel cell be a good choice for energy storage and generation in H2 rich atmospheric environments?

So imagine an atmospheric probe sent to enter the atmosphere of a giant gas planet. There would be no need to store hydrogen since a compressor or an inlet with sufficient dynamic pressure would allow sufficient $\ce{H2}$ to enter the system in the upper atmosphere. The output would be basically limited by dynamic pressure and max output of the cells. $\ce{O2}$ would be delivered from a pressure vessel or cryogenic tank (seems possible, it's cold in the shade in space) and valve regulated.

Besides low operation temperature, what would be the drawbacks?

## 2 Answers

An "inverted" $\ce{H2/O2}$ fuel cell should work well, and be easier to maintain than one in an $\ce{O2}$ atmosphere, because there are a plethora of membranes to pass pure $\ce{H2}$ (e.g. hot platinum, plastics, ceramics...). That means that the cell would be less likely to be poisoned by trace gases.

It would also be possible to use methane, common on gas giants, in a fuel cell.

That said, carrying the oxygen in the probe from Earth to the target would take energy. A fuel cell might be useful for a short-lifetime probe, such as one penetrating deep into a gas giant's atmosphere, only to be crushed in a few days. For a longer lifetime, nuclear thermoelectric power (e.g. MMRTG) is more practical.

Good point. Using answer mode because did not fit in comments.

But when you wrote about trace gases I thought that there are indeed traces gases in the Jovian atmosphere and a descending probe has some probability of hitting a cloud deck made of NH4, then NH4SH, NH42S, and finally water clouds. Are these sulfur compounds nasty for the majority of membranes? But I worry too about temperature, surely the hydrogen supply would need to be preheated, and oxygen exiting the tank would lose temperature too due to pressure drop. Or are there any fuel cells that operate well at low temperatures?

FYI the possible design would be a quadcopter assisted by an infrared balloon (derived from https://www2.jpl.nasa.gov/adv_tech/balloons/outer_jupisat.htm) The quadcopter would need at least 6kw of power for a 10kg descent module to just hover in free fall conditions due to high g (in the case there is a balloon failure/malfunction). That's quite a lot of power. So the cell stack would need to be lightweight and compact (less or equal to 1kg and 1l volume would be ideal) A 2L oxygen tank would provide the fuel cell up to 12kwh of energy at 80% efficiency. that would be enough for 2 hours of hovering, so the best would be to have it assisted by a solar infrared heated balloon. (a big one, pumped with hydrogen taken from the atmosphere, and may be preheated before pumping)

The thermodynamic equilibrium temperature could be modeled taking into account, heat gain from the fuel cell, heat gain from joule losses and payload, heat loss from gas input temperature and heat loss from the enclosure. As the probe descends, the problem would be reversed, heat gain would come from Jupiter atmosphere, and H2 inlet temperature and pressure would be significantly higher and I think the efficiency of the cell would be better. At some point however the valve would need to be closed to prevent pressure and temperature damage. Maybe an empty lightweight tank could be filled just before to make some reserves for the late descent. At this point, the probe is on final reserves and then secondary battery power (presumably not much capacity, only for science payload and critical systems at this point) and is sinking. With that design, the probe could stay operational in the atmosphere for 72 to 96 hours, maybe.

• The problem, however, seems that even with current state of the art technology, fuel cells are notoriously heavy. A 10kW fuel cell weighing 100kg is at least two orders of magnitude too heavy for this kind of application. Maybe when (if) technology matures it would be worth it. – user3239774 Jun 17 '18 at 8:39