Waste heat surrounds us. Your laptop generates it. So does your refrigerator, your phone, your car. Most of it vanishes into the air, energy squandered. But what if you could catch that heat and convert it back into electricity?
That's the promise of thermoelectric materials—substances that transform temperature differences directly into voltage. The catch? Most high-performers only work well above 300 K, limiting their usefulness for everyday cooling applications. Until now.
Researchers have demonstrated that a weak magnetic field—no stronger than what a common refrigerator magnet produces—can dramatically boost the efficiency of a bismuth-antimony crystal, achieving a thermoelectric figure of merit of 1.7 at just 180 K. That's nearly three times better than the material performs without any magnetic field at all.
The figure of merit, called zT, measures how well a material converts heat to electricity. Higher is better. Values above 1 are considered excellent. This result fills a conspicuous gap in the thermoelectric landscape: until this work, almost nothing performed well below room temperature.
Why does this matter? Thermoelectric cooling is already used in applications from wine coolers to spacecraft thermal management. But the technology has been hobbled by mediocre performance at the temperatures where we actually live and work. A material that achieves high efficiency between 150 K and 300 K could unlock new applications in refrigeration, climate control, and energy recovery from low-grade waste heat.
The secret lies in the material's exotic electronic structure. Bismuth-antimony alloys are topological insulators—quantum materials with a peculiar property. Inside the bulk, they act like insulators. But on their surfaces, electrons flow freely, protected by the laws of topology like hikers on a mountain trail that can't be erased.
These materials harbor what physicists call Dirac electrons. Unlike ordinary electrons, which gain mass as they accelerate, Dirac electrons behave more like photons of light, traveling at constant velocity regardless of their energy. This linear relationship between energy and momentum—the "linear band dispersion"—is crucial.
When a magnetic field is applied, these Dirac electrons begin to spiral. The field doesn't just push them sideways; it fundamentally changes how they carry heat and charge. Electrons at different energies respond differently to the magnetic influence, creating what amounts to an energy filter. Lower-energy electrons get scattered more; higher-energy ones pass through more readily.
This filtering enhances the Seebeck coefficient—a measure of how much voltage a material generates for a given temperature difference. In the bismuth-antimony crystal studied here, the Seebeck coefficient increased substantially under magnetic fields as weak as 0.1 tesla. For context, an MRI machine uses fields of 1.5 to 3 tesla. This worked with far less.
The team tested four different single crystals, each grown with extreme care from ultra-pure bismuth and antimony. The composition settled at approximately 88% bismuth and 12% antimony—a ratio that places the material squarely in topological insulator territory. At that composition, the electronic bands "invert," creating a narrow gap of just 17 millielectronvolts, about a thousandth the energy needed to ionize a hydrogen atom.
What emerged was a material with exceptional charge carrier mobility—how freely electrons move through the crystal lattice. Mobility values reached several hundred thousand square centimeters per volt-second, among the highest known for any solid.
High mobility alone doesn't guarantee thermoelectric excellence. You also need low thermal conductivity, so heat doesn't simply conduct through the material instead of generating voltage. Bismuth and antimony are heavy elements. Their atoms vibrate sluggishly, carrying heat poorly. This is exactly what you want.
The researchers measured how resistivity, Seebeck coefficient, power factor, and thermal conductivity changed as they swept the magnetic field from zero to 2 tesla across a temperature range of 80 K to 300 K. The resistivity increased, as expected—magnetic fields scatter electrons. But the Seebeck coefficient increased even more. And thermal conductivity dropped.
These three effects competed in a delicate balance. Too much magnetic field, and rising resistivity kills performance. Too little, and you don't get the Seebeck boost. The optimal field turned out to be temperature-dependent, following a power law: the colder the material, the weaker the field needed to maximize efficiency.
At 180 K, the sweet spot arrived at just 0.7 tesla. The thermoelectric figure of merit hit 1.7, with an uncertainty of about 20% across the four samples tested. That's comparable to the best thermoelectric materials operating at much higher temperatures, and it beats anything previously reported below 300 K.
Why does the magnetic field have such a profound effect? The answer lies in quantum mechanics. Bismuth-antimony has an unusually large Landé g-factor—a number describing how strongly electron spins respond to magnetic fields. For this material, that number is between 120 and 240, compared to 2 for ordinary electrons.
This giant g-factor means that even a modest magnetic field splits the electron energy levels via the Zeeman effect, creating two distinct populations of charge carriers. One population, occupying lower-energy states, sees its Fermi level drop. The other, at higher energy, sees its Fermi level rise.
Each population contributes to the Seebeck coefficient differently. As the field increases, one contribution grows while the other peaks and then falls. The total Seebeck coefficient reflects this competition, showing a maximum at intermediate field strengths before declining at higher fields. Theoretical modeling captured this behavior, confirming that Zeeman splitting plays an essential role.
The implications extend beyond bismuth-antimony. Recent theoretical work predicts that in topological materials with linear band dispersion, the Seebeck coefficient should increase under magnetic fields and might not saturate even at extreme field strengths. With 88% of all inorganic compounds now known to possess topological electronic bands, the search space for high-performance magneto-thermoelectrics has suddenly become vast.
This approach—using weak magnetic fields to tune thermoelectric parameters synergistically—offers a new strategy for materials design. Traditionally, researchers optimize thermoelectric performance by doping, nanostructuring, or band engineering. Those methods are powerful but constrained by the fundamental trade-offs between electrical and thermal transport.
Magnetic tuning sidesteps some of those trade-offs. It enhances the Seebeck coefficient while simultaneously reducing thermal conductivity, both beneficial for thermoelectric performance. The resistivity does increase, but in materials with high mobility, the gains outweigh the losses.
For practical applications, the question becomes whether the energy cost of generating a 0.7 tesla magnetic field outweighs the efficiency gains. Permanent magnets can produce fields of this strength without ongoing energy input, making the approach viable for static installations. Thermoelectric coolers in scientific instruments, for instance, could incorporate permanent magnets to boost efficiency without added operating costs.
The team also noted that future work should explore p-type variants—materials where positive charge carriers (holes) dominate instead of electrons. Hole pockets in bismuth-antimony alloys show highly dispersive bands, which could yield large magneto-thermoelectric responses if doped appropriately.
Another avenue involves tuning the Fermi level with exquisite precision. Small changes in impurity concentration dramatically affect electron-hole compensation, which in turn governs magneto-resistivity. Purer starting materials, careful deoxidation, and parts-per-million-level doping could push performance even higher.
We tend to think of magnets as simple things—objects that stick to refrigerators or hold up our shopping lists. But in quantum materials, magnetic fields become instruments of transformation, reshaping the electronic landscape in ways that unlock hidden potential. A refrigerator magnet's worth of field strength is all it takes to turn a good thermoelectric material into an exceptional one.
The question is no longer whether we can achieve high thermoelectric efficiency at low temperatures. We can. The question now is how many other quantum materials are waiting for the same magnetic nudge.
Credit & Disclaimer: This article is a popular science summary written to make peer-reviewed research accessible to a broad audience. All scientific facts, findings, and conclusions presented here are drawn directly and accurately from the original research paper. Readers are strongly encouraged to consult the full research article for complete data, methodologies, and scientific detail. The article can be accessed through https://doi.org/10.1038/s41563-024-02059-9
