CLEMSON, South Carolina – On Sept. 15, NASA plunged its Cassini spacecraft into the atmosphere of Saturn, ending a two-decade-long mission to study the planet’s signature rings and moons.

cassini spacecraft flies over Saturn's rings

Cassini was powered by three radioisotope thermoelectric generators (RTGs).
Image Credit: Wikimedia Commons

Over the course of those 20 years – seven of which were spent traveling from Earth to Saturn – Cassini was powered by three radioisotope thermoelectric generators (RTGs), a technological device that College of Science physicist Terry Tritt has studied since 1993.

Now, Tritt and physics and astronomy colleague Jian He have published an invited review on the state of thermoelectric energy in the high-impact journal Science.

The publication brings Tritt full circle, given that his first paper as a Clemson faculty member was a two-page summary for Science in 1996 titled “Thermoelectrics Run Hot and Cold.”

The technology of thermoelectrics is based on coupling two compound semiconductors, which are materials that have the electrical conductivity about halfway between that of insulators, like glass, and metals. In the first semiconductor, called an N-type, electrons dominate as they flow in one direction. The second semiconductor, a P-type, has an abundance of positive charges, or holes, that act as placeholders for electrons to move into.

“If we take these two compound semiconductors – one N-type and one P-type – and put them in series, connect them electrically and install a circuit where current can flow, it will pull heat out of one end, thus cooling it, and release the heat at the other end,” Tritt said.

diagram of thermoelectric couple

In a thermoelectric couple made of N-type and P-type thermoelectric materials, electron holes move in the P-type, while electrons move in the N-type.
Image Credit: Department of Physics and Astronomy

The electrons and holes – collectively referred to as charge carriers – are the reasons thermoelectric devices can generate energy, as they impart the ability to cool things from temperatures as high as 200 degrees Celsius down to well below freezing.

These “charge carrier coolants” are necessary in everything from portable camping coolers to X-ray detectors in hospitals to infrared detectors in night vision goggles. A scientist’s lab might also have many instruments operating on the thermoelectric effect. Thermal cyclers that amplify DNA, scanning electron microscopes that magnify minuscule biological samples and spectrometers that break chemicals into their components all make use of thermoelectric cooling.

Rather than using electricity to cool – or even heat, in some circumstances – thermoelectric energy can also do the opposite by turning waste heat into electricity.

“This idea was a resurgence in the field about 12 years ago because we had an energy issue. The country needed renewable energy – like solar, hydroelectric or wind – but on the small scale. Well, thermoelectric devices can take heat and turn it into electricity, all the way from milliwatts to kilowatts,” Tritt said. “For example, you could have a little thermoelectric device that you can take with you camping – say, a can that’s heating your water – but you can also use it to charge your cell phone. Or you could take waste heat from your automobile to act as an added power source.”

Both the refrigeration and power-generation modes of thermoelectrics operate on solid-state devices, in which charge carriers are contained entirely within the materials that make the device. Given that thermoelectric devices have no moving parts, they are ideal for deep space applications, like Cassini, where power needs to be sustained for many years.

For He and Tritt’s latest review, published online on Sept. 29, the pair wanted to take a different approach when summarizing some of the latest findings in the field of thermoelectrics.

“I’ve written a number of review articles and review articles are mainly, ‘here are the materials that are out there, and here are the properties of those materials.’ But Jian and I didn’t want to do what’s already been done,” Tritt said. “So we decided that we wanted to talk about the mechanisms that make a good thermoelectric material or not. These are the so-called ‘tuning knobs’ or ‘design parameters’ that you can vary to manipulate a material’s properties to make it have the qualities or properties that you desire.”

A “reasonably good electrical conductor with a high thermoelectric power” is the leading quality of an ideal thermoelectric material, according to Tritt. However, these materials also have very low thermal conductivity, where heat passes through slowly but generates substantial voltage because of the temperature difference.

If these properties sound contradictory to one another, it’s because they are.

“These kinds of properties are only seen in a very small set of materials because you have to break symmetry and break down some of these mechanisms to be able to dissociate the electrical and thermal transport properties,” Tritt said. “So we thought that was the cool approach to write this paper. How do you manipulate these properties to maximize the material’s performance?”

Terry and Jian in physics office

Terry Tritt (left) and Jian He worked together to review thermoelectric for the journal Science.
Image Credit: College of Science

The measure of a material’s performance is referred to as the figure of merit or ZT – a number that summarizes the main parameters of a thermoelectric material. By manipulating a material’s “tuning knobs,” such as its charge or composition of electrons, researchers can increase the ZT, resulting in a more efficient thermoelectric material.

The “holy grail of thermoelectrics,” in Tritt’s words, is an inexpensive, bulk material with a ZT between 2.5 and 3.

“When researchers at Oak Ridge published in 1996 about cobalt antimonide – filled skutterudites – it had a ZT of about 1. Currently, that material is at about 1.5 to 1.7, so even though it hasn’t come far, we’re already more than halfway there,” Tritt said. “At the holy-grail level of a ZT, and if we had the Earth-abundant materials with low-cost processing, then thermoelectrics would become a major renewable energy source because the cost effectiveness would be huge.”

ZT’s of 2.5 and above have been found in many nanomaterials, like graphene and chalcogenides, but these thin films aren’t able to carry much heat, nor are they available in bulk – two factors that limit nanotechnology’s applications in thermoelectrics.

“The problem, too, is that this is an area that has been heavily investigated for over 20 years. To continue putting money into fundamental thermoelectrics research is a struggle when scientific funding is currently so hard to come by,” Tritt said.

Although scarce in the United States, countries such as Germany, South Korea, Japan, China and Saudi Arabia are rich in funding for materials and thermoelectric research. Germany, specifically, is studying thermoelectrics to turn waste heat from a car’s exhaust system into electricity to power the car.

“These places know that they need energy other than oil. They know that if they’re going to make an impact, they’ve got to look at Earth-abundant materials,” Tritt said. “Eventually, the U.S. is going to have to invest more into it, too, as well as many other renewables.”

Tritt, who has authored three of the small handful of papers on thermoelectrics in Science, plans to retire at the end of the 2017-18 academic year.

“This is neat for me because my first paper after I had accepted the position to come to Clemson was reviewing thermoelectric materials, and it really set the tone for my research for the next 21 years,” Tritt said. “Now, this is going to be a really important paper for the rest of Jian’s career, and nobody deserves that more than him. He and I would like to acknowledge all the extraordinary students and colleagues who have contributed to this research over the past 20 years.”