Engineers usually regard heat as “waste energy” since it is hard to efficiently turn into anything useful. However, a new class of thermoelectric materials could change that after researchers opted to try the exact opposite of the usual approach. A paper in Science Advances explains why, speeding the search for even better versions.
As the name suggests, thermoelectric materials turn heat into electricity, skipping the boiling water stage used in most bulk electricity production. However, cost and inefficiency have kept thermoelectric generators restricted to niche applications, such as powering spacecraft like the Mars Perseverance rover where lightweight, reliable energy production matters more than price.
Thermoelectric materials are too expensive and polluting for more widespread use, but new versions that replace heavier elements with magnesium could change that, opening the door to even better options that could find widespread uses.
Thermoelectric materials work by creating a current between a hot and a cool side. Unfortunately, if the material conducts heat anything like as well as it conducts electricity the temperatures equalize, cutting off the current flow. So materials must be thermally insulating, a trait more often associated with heavy metals than light ones, so scientists working in the field have focused on larger atoms.
When other scientists tried magnesium-based materials, just in case, they were astonished to discover they worked surprisingly well.
Dr Olivier Delaire of Duke University has confirmed these materials, Mg3Sb2 and Mg3Bi2, work three times as well as calcium and ytterbium, elements with more protons and similar chemical properties, which may also explain the unexpected phenomenon.
Magnesium also has the rather significant advantage of being cheap, abundant, and relatively non-polluting. Although it shares these traits with calcium, that's not the case for other materials trialed.
"Traditional thermoelectric materials rely on heavy elements such as lead, bismuth, and tellurium—elements that aren't very environmentally friendly, and they're also not very abundant,” Delaire said in a statement. "These magnesium materials, however, have remarkably low thermoelectric conductivity despite having a low mass density.” Moreover, while high-temperature thermoelectric effects are common, Mg3Sb2 and Mg3Bi2 work well close to room temperature.
Nevertheless, Delaire does not think these specific materials will be the future of thermoelectric generation. Antimony and bismuth are not particularly abundant, and antimony production is quite polluting. However, just as most pharmaceuticals are modifications of a promising but imperfect molecule, Delaire hopes the two magnesium-based materials explored so far, which belong to a class known as Zintls, will open the door to better versions.
"In chemical studies, exploring possibilities for new materials often involves substituting one element for another just to see what happens,” said first author Jingxuan Ding. “Usually we replace them with chemically similar elements in the periodic table, and one of the big advantages to using Zintls is that we can experiment with a lot of different elements and different combinations.”
Although this can be done by trial and error Ding and Delaire hope to shorten the process by identifying why magnesium works so well. They learned a magnesium bond obstructs heat transmission. In its presence, the heat waves that carry vibrations from the warm side of the material to the cooler one interfere with each other, rather than traveling cleanly.