Idea: Water battery?

[Demon-Xanth]

I got to thinking again... yes, I know, I should really avoid that...

Fuel cells work on hydrogen and oxygen and create water, electrolysis splits hydrogen and oxygen, so here's my idea:

Take a sealed unit, fill it with water, have two electrodes to split the water into hydrogen and oxygen, the fuel cell will then convert it back without combustion. While this isn't a power generating source, it could operate as a battery that doesn't contain any acids, lead, or any other problem materials. You'd basically have four leads, two to charge it, and two to discharge it. Concievably this can be used on UPSes due to the inherant isolation between the input and outputs.

 

[RadixTwo]

I think that is a great idea, but what kind of lifetime would you get out of those batteries? How big would they be? And what about heat? Would they have a memory (I don't think so, but I could be wrong)?

 

[Mark R]

Nice idea, but there are a few problems with it:

How would you store the hydrogen/oxygen generated? You'd need large tanks or compressors in order to store a sensible amount of energy. You then run into the practical problems of storing hydrogen gas - most large capacity systems use cryogenic storage. There are other problems such as the explosion risk of hydrogen if a tank fails.

You still need acid in order to carry current in the electrolysis cell - so it's not quite as benign as you hoped.

Both fuel cells and electrolytic hydrogen generation are quite inefficient - even with 'high-efficiency' devices, you'd be doing well to get back 50% of the energy input.

Your concept, however, is quite close to an exisitng invention: the flow battery. 'Flow' batteries use an external store of chemical energy together with charging and discharging cells. One commercial example is the vanadium redox flow battery.

This battery consists of a bank of cells and 2 liquid electrolytes - One on each side of the cells. To increase the capacity of the battery you just attach electrolyte storage tanks to the battery and some small pumps to keep the solutions mixed - the only limit to capacity is the size of the tanks. You can have multiple cell stacks in the same electrolyte, e.g. one to charge and one to discharge - they can even have different voltages, and of course, you can charge and discharge simultaneously.

These batteries have been successfully used in large industrial UPSs, and in grid power stabilising systems (e.g. for wind farms). They are highly efficient (better than 80% - and in some installations approaching 90%). One interesting application has been as an alternative to solid-state DC-DC converters in a high-power industrial system. There have been experiments with them in electrical vehicles - charging is easy. Just drain the exhausted electrolytes into your charger, and pour in fresh charged electrolyte from a storage tank.

There are a number of other advantages with these batteries - No damage from overcharge (NiCd and NiMH are damaged by overcharging causing 'memory'). Can be discharged to 0% with no damage (perfect for 'deep-cycle' purposes). They are modular - cells and electrolytes can be replaced (due to failure or for upgrade) independently. Very long life: working batteries have showed virtually no loss of capacity after 10,000 charge/discharge cycles.

 

[PowerEngineer]

If you just store the hydrogen (with the idea of drawing oxygen from the atmosphere later), then you've pretty much arrived at the hydrogen (as a form of energy storage/transportation) concept that many people seem to favor. The problem is, as Mark R has pointed out, that we don't have any efficient way to produce the hydrogen.