Refrigerators, boilers, and even lightbulbs continually dump heat into their surroundings. This “waste heat” could—in theory—be turned into electricity, as it is sometimes done with power plants, automobile engines, and other high-heat sources. The problem: These “low-grade” sources give off too little heat for current technology to do the conversion well.
Now, researchers have created a device that uses liquids to efficiently convert low-grade heat to electricity. The advance might one day power energy-scavenging devices that can light up sensors and lights and even charge batteries.
“This is a nice piece of work and a very clever idea,” says Ping Liu, a nanoengineer at the University of California, San Diego, who was not involved with the study.
Scientists have known for nearly 200 years that certain materials can convert heat to electricity, and are being explored for use in providing extra electricity for hybrid vehicles. This job is carried out by specialized semiconductors called thermoelectric materials that are fashioned into tiny devices the size of computer chips. When one side of a thermoelectric is hotter than the other, heat and electrons move from the hot to the cold side. Wiring multiple such chips together allows engineers to generate a steady electric current.
The key to conversion is finding materials that are good at conducting electrons, but not heat, in order to maintain a temperature difference between the two sides. Those that exist are expensive—and work best when the hot and cold sides have a temperature difference of hundreds of degrees Celsius. For low-grade heat sources like refrigerators, they’re all but useless.
To overcome that problem, materials physicist Jun Zhou and colleagues at the Huazhong University of Science and Technology turned to thermocells. These devices use liquid instead of solid materials to conduct charges from a hot side to a cold side. They do so not by shuffling electrons, but by moving charged molecules, or ions.
Thermocells are good at converting small temperature differences into electricity, but they typically produce only tiny currents. That’s in part because ions are more sluggish than electrons. Ions also carry heat through the material (unlike electrons), reducing the temperature difference between the two sides and lowering energy conversion efficiency.
Zhou and colleagues started with a small thermocell: a domino-size chamber with electrodes on the top and bottom. The bottom electrode sat on a hot plate and the top electrode abutted a cooler, maintaining a 50°C temperature difference between the two electrodes. They then filled the chamber with ionically charged liquid called ferricyanide.
Past research has shown that ferricyanide ions next to a hot electrode spontaneously give up an electron, changing from one with a –4 charge, or Fe(CN)6–4, to an ferricyanide with a –3 charge, or Fe(CN)6–3. The electrons then travel through an external circuit to the cold electrode, powering small devices on the way. Once they reach the cold electrode, the electrons combine with Fe(CN)6–3 ions that diffused up from below. This regenerates Fe(CN)6–4 ions, which then diffuse back down to the hot electrode and repeat the cycle.
To reduce the heat carried by these moving ions, Zhou and his colleagues spiked their ferricyanide with a positively charged organic compound called guanidinium. At the cold electrode, guanidinium causes the cold Fe(CN)6–4 ions to crystallize into tiny solid particles. Because solid particles have lower thermal conductivity than liquids, they block some of the heat traveling from the hot to the cold electrode. Gravity then pulls these crystals to the hot electrode, where the extra heat turns the crystals back into a liquid. “This is very clever,” Liu says, as the solid particles helped maintain the temperature gradient between the two electrodes.
It also worked. The thermocell generated five times more power for the same electrode area than previous versions, Zhou and his colleagues report this week in Science. It also more than doubled the efficiency needed to make a viable commercial device. A paperback book–size module of 20 thermocells could run LED lights, power a fan, and charge a mobile phone, the team found.
“This shows you can improve the performance [of these devices] to a very respectable level,” says Gang Chen, a mechanical engineer at the Massachusetts Institute of Technology who was not involved in the research. Whether that will be good enough to make the technology commercially successful remains to be seen, he adds. “Low-grade waste heat is everywhere. But it costs money to collect it.”
The next step for powering real-world devices is to add other, inexpensive materials that absorb as much waste heat as possible from the desired sources while excluding the rest of the ambient environment, Chen says—a task Zhou says his team is already working on. When that happens, we may soon be able to power all sorts of small gadgets with heat that’s all around us.