Did you know that the Lunar Module was originally designed to be electrically powered by fuel cells?
After being awarded the contact to build the then-named Lunar Excursion Module in 1963, Grumman subcontracted Pratt & Whitney for developing three fuel cells to power the spacecraft. Not coincidentally, the Gemini program’s subcontractor to McDonnell Aircraft for its fuel cells, General Electric, were having serious developmental problems in making the fuel cell work as the primary supply for the Gemini. It was not until 1965 when GE overcame their problems to have Gemini 5 fly and complete an 8-day mission using the fuel cell as their primary power source.
The equipment bay of the LM ascent stage was meant to house one or more of the fuel cells, among other things.
Pratt & Whitney was burning the developmental candle at both ends, having the subcontract not only for the LM but for the Service Module’s fuel cells through North American Aviation. Delays forced the LM power design from three cells to two cells. By late 1964, NASA (likely fearing the Gemini fuel cell development’s delays would be mirrored in the more-important Apollo) told Grumman (who also asked for the change) to scrap the LM fuel cells in favor of silver-zinc batteries.
These batteries, lighter than lead-acid ones, would be placed throughout the spacecraft, two in the ascent stage and five in the descent stage, providing enough power for a two-day stay. Later LM designs would add additional batteries for J-missions for up to three days.
So you might ask yourself, how would’ve the Apollo 13 incident played out if the LM had its own fuel-cell system? This answer isn’t as intuitive as you might think.
The LM H-mission design (Apollo 11 – 14) was built to power itself for 45 days, including life support. Fuel cells needed oxygen, so additional oxygen would be needed for the cells and the ECLSS. You had to carry enough hydrogen for the cells, too, adding to the vehicle weight and, overall, the total resources that could be carried. They’d be one or two auxiliary batteries for abort emergencies but that’s it. With three men aboard, the total amount of reactants might’ve been enough to power the vehicle but oxygen supplies for ECLSS for three men may have been a serious problem.
A fuel cell design did not have a charging circuit between the Command/Service Module and the LM to recharge batteries from either vehicle. Apollo 13’s three batteries in the Command Module, designed to power the spacecraft during reentry, were partially depleted as the last close-down of Odyssey’s systems was made, two hours after the accident. A fuel-cell powered LM could not have tapped off the batteries in the CM, and it is more likely that Odyssey could not have been restarted properly prior to re-entry, or would have lost vital power needed during re-entry and be lost.
(January 29 2018) Grumman lead engineer for the LM, Tom Kelly, mentioned in his book, Moon Lander, that the fuel cell system was the last major revision of the LM in development stages before construction began.
Lots of background reading for more. The link from the NSF was a thread I created that explored this very question.