How would that energy be extracted? Pleiter-elements?
You mean my heat engine working off body heat vs LOX? It could be anything capable of working between those two temperatures; maybe a Stirling engine. Thermocouples and Peltier devices would probably be too inefficient. NASA Glenn has been working on a Stirling engine for converting Pu-238 RTG heat to electricity that's much (4-5x) more efficient than the thermocouples used on current RTGs. Since it has moving parts and is considerably more complex than the thermocouples, they're only doing this because of the critical shortage of Pu-238.
The Carnot limit is a theoretical limit on conversion efficiency that applies to any heat engine, set only by the ratio of the hot side and cold side temperatures. As always it's difficult to approach the theoretical limits in practice.
It's easy to see intuitively how the Carnot limit works. Kelvin is an absolute temperature scale so there's no heat energy at all in something at 0K. So if you start with heat at, say 310K and dump waste heat at 90K, you are not extracting 90/310 = 29% of the heat input, leaving 71% that you can extract.
An interesting point about my heat engine idea is that if you really could produce a heat engine with 71% efficiency, only 29% of the astronaut's body heat would reach the LOX, the rest having been converted to useful work. That would reduce the LOX boil rate by over a factor of 3. So you could reduce your LOX consumption rate back down to that required to maintain a 1% CO
2 concentration or even higher.
Edited to add: Actually it seems more complicated than that. Any electricity generated from a heat engine working between body temperature and LOX temperature would be used to power systems within the PLSS like pumps, fans and radios. They would in turn generate additional waste heat that would have to be removed, e.g., by boiling LOX, but some of that waste heat could again be used to generate more elecricity! The only energy that wouldn't have to be removed is that radiated by radio transmitters, which would be a small fraction of the total.
But you'd still eliminate the need for a battery and all the
extra waste heat the PLSS systems would generate from that battery power.