New Alloy Can Convert Heat Directly Into Electricity

[Imperial528]

@someone_else

Something like this?



I was thinking about what you said and I got this idea.

Basically, I flipped it so the cool part was on top, and put the working part (the material pieces) on one side of the rotor, and a ballast sack on the other side. Note, the rotor spins CCW.

What would happen is that the ballast sack would fall into the heated area, heat up, and rise to the top, which dips the cooled pieces into the hot area. The ballast cools down, falling back down into the hot zone, and raising the pieces up to be cooled again. The ballast sack is made to be off set so the system doesn't reach equilibrium. For ballast it would probably be best to use some sort of liquid coolant that is heated into a gas at the hot area and cooled back to a liquid in the cold.

[Winston Blake]

I think this is interesting and worth investigation, but I don't see any fundamental advantage of these 'multiferroic' materials over relatively conventional magnetocaloric materials. Magnetic refrigeration was invented in 1926 and examples were built in 1933. Since then claims have gone around periodically that its the next big thing, and all of these predictions have failed due to some problem or another. However, it still has good niche applications IIRC.

The use of this technology for domestic refrigerators though is very remote due to the high efficiency of current Vapor-compression refrigeration cycles, which typically achieve performance coefficients of 60% of that of a theoretical ideal Carnot cycle.

I will let myself get excited once a truly working model is made with a decent efficiency, compactness, or cost advantage. I would love to hear that it has some good, reliable advantage. Until then this is another laboratory curiosity being justified by possible future applications, and being trumpeted by overoptimistic tech fans who neither know nor care about its place in the real world and real history. Companies patent a lot of stuff and build a lot of stuff on the off chance it turns out to be gold, while recognising that it usually doesn't.

Don't get me wrong - I expect it's damn good science - I just don't think it's prudent to claim any more for its engineering potential than can be supported. Niche but valuable applications? Great, wonderful. Pouring it onto road surfaces to instantly defeat Peak Oil / Climate Change? Not so fast.

Someone_else and Imperial528 - good ideas! So good, in fact, that it was invented years ago for use with magnetocalorics. A quick googling nets me this.

For some more perspective, as I think many people in this thread already know, 'converting heat directly into electricity' was discovered a long time ago. In 1821 for heat-to-current, and 1834 for current-to-heat. It's in widespread use for thermocouples and is used for powering satellites in RTGs. The efficiency is not good enough to go in your fridge, though.