Material may help autos turn heat into electricity

Researchers have invented a new material that will make cars even more efficient, by converting heat wasted through engine exhaust into electricity. In the current issue of the journal Science, they describe a material with twice the efficiency of anything currently on the market.

The same technology could work in power generators and heat pumps, said project leader Joseph Heremans, Ohio Eminent Scholar in Nanotechnology at Ohio State University.

Scientists call such materials thermoelectric materials, and they rate the materials' efficiency based on how much heat they can convert into electricity at a given temperature.

Previously, the most efficient material used commercially in thermoelectric power generators was an alloy called sodium-doped lead telluride, which had a rating of 0.71. The new material, thallium-doped lead telluride, has a rating of 1.5 -- more than twice that of the previous leader.

What's more important to Heremans is that the new material is most effective between 450 and 950 degrees Fahrenheit -- a typical temperature range for power systems such as automobile engines.

Some experts argue that only about 25 percent of the energy produced by a typical gasoline engine is used to move a car or power its accessories, and nearly 60 percent is lost through waste heat -- much of which escapes in engine exhaust.

A thermoelectric (TE) device can capture some of that waste heat, Heremans said. It would also make a practical addition to an automobile, because it has no moving parts to wear out or break down.

"The material does all the work. It produces electrical power just like conventional heat engines -- steam engines, gas or diesel engines -- that are coupled to electrical generators, but it uses electrons as the working fluids instead of water or gases, and makes electricity directly."

"Thermoelectrics are also very small," he added. "I like to say that TE converters compare to other heat engines like the transistor compares to the vacuum tube."

The engineers took a unique strategy to design this new material.

To maximize the amount of electricity produced by a TE material, engineers would normally try to limit the amount of heat that can pass through it without being captured and converted to electricity. So the typical strategy for making a good thermoelectric material is to lower its thermal conductivity.

In Heremans' lab, he used to work to lower the thermal conductivity by building nanometer-sized structures such as nanowires into materials. A nanometer is one billionth of a meter.

Those nanostructured materials are not very stable, are very difficult to make in large quantities, and are difficult to connect with conventional electronic circuits and external heat sources.

For this new material, he and his colleagues took a different strategy: they left out the fancy nanostructures, and instead focused on how to convert the maximum amount of heat that was trapped in the material naturally.

To do this, they took advantage of some new ideas in quantum mechanics.

Heremans pointed to a 2006 paper published by other researchers in the journal Physical Review Letters, which suggested that elements such as thallium and tellurium could interact on a quantum-mechanical level to create a resonance between the thallium electrons and those in the host lead telluride thermoelectric material, depending on the bonds between the atoms.

"It comes down to a peculiar behavior of an electron in a thallium atom when it has tellurium neighbors," he said. "We'd been working for 10 years to engineer this kind of behavior using different kinds of nanostructured materials, but with limited success. Then I saw this paper, and I knew we could do the same thing we'd been trying to do with nanostructures, but with this bulk semiconductor instead."

Heremans designed the new material with Vladimir Jovovic, who did this work for his doctoral thesis in the Department of Mechanical Engineering at Ohio State. Researchers at Osaka University -- Ken Kurosaki, Anek Charoenphakdee, and Shinsuke Yamanaka -- created samples of the material for testing. Then researchers at the California Institute of Technology -- G. Jeffrey Snyder, Eric S. Toberer, and Ali Saramat -- tested the material at high temperatures. Heremans and Jovovic tested it at low temperatures and provided experimental proof that the physical mechanism they postulated was indeed at work.

The team found that near 450 degrees Fahrenheit, the material converted heat to electricity with an efficiency rating of about 0.75 -- close to that of sodium doped telluride. But as the temperature rose, so did the efficiency of the new material. It peaked at 950 degrees Fahrenheit, with a rating of 1.5.

Heremans' team is continuing to work on this patent-pending technology.

"We hope to go much further. I think it should be quite possible to apply other lessons learned from thermoelectric nanotechnology to boost the rating by another factor of two -- that's what we're shooting for now," he said.

Source: Ohio State University

Explore further

High thermoelectric performance in low-cost SnS0.91Se0.09 crystals

Citation: Material may help autos turn heat into electricity (2008, July 24) retrieved 19 October 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Jul 24, 2008
What does the rating mean? At what efficiency % does the heat create electricity? I suppose that such a measure would be problematic, but if you're going to use a different measure it is silly to not at least tell us what it means.

Jul 24, 2008
add this breakthrough to the following

Jul 24, 2008
Hmm interesting. All sorts of applications. Put it on an exaust and feed the energy back into a battery, make it small enough and drill deep enough and one should be able to power a house anywhere.

Jul 24, 2008
I'm calling bullpooh. I'm betting they are using quantum physics for the electrical interaction and newtonian physics for the thermal interactions. Phonons (not phoTons) are quantum particles and need to be treated as such.

It's practically impossible to seperate thermal and electrical conductivity when you want to.

Jul 25, 2008
if that 1.5 rating is an efficiency rating then perhaps this stuff converts 1.5% of possible energy to electricity?

In any case nice improvement, wonder if they also used the techniques described in article http://www.physor...802.html to maximize the overall effect.

Jul 25, 2008
"the new material is most effective between 450 and 950 degrees Fahrenheit -- a typical temperature range for power systems such as automobile engines" 450 degrees!!? According to my steam table, water at 450 degrees will have a pressure of 405 PSI and at 490 degrees the pressure is 600 PSI. My steam table doesn't go any higher than that. The steam locomotive I am a fireman on lifts the first safty valve at 192 pounds at a temperature of 385 degrees. I want to know just WHAT automoble engine operates at those temperatures? An auto radiator with a 15 pound pressure cap will have a working temp. of 250 deg.

Jul 25, 2008
Interesting but doesn't the alternator already do that job? Good idea if they can develop it for useful and required applications like generating power for homes.

imo, the car industry would be more efficient if every car made had a turbo in it. Turbo's work in a similar way except they create horsepower from the exhaust and not electricity. If every car was fitted with a turbo, cars would require smaller engines for the same power and that would surely boost efficiency?

Jul 25, 2008
The temperature range of 450-900F is actually quite low for an internal combustion engine. Doggone, you may be referring to the coolant temperature of the engine and not that of the exhaust. Exhaust gas temperatures routinely run 1200-2000F right as they exit the head. It's not uncommon for the exhaust manifold or headers to get white hot during dynamometer runs.

The value 1.5 is not an efficiency. It is the thermoelectric figure of merit, Z. Engineers like to use figures of merit to express how good something is without giving you a concrete number. Thermoelectric figure of merit is the ratio of (alpha^2*sigma)/lambda. Alpha is the Seebeck coefficient in microvolts per Kelvin. Sigma is the electrical conductivity. Lambda is the thermal conductivity.

Increasing this figure of merit can be done by finding a material with high Seebeck coefficient, high electrical conductivity, and low thermal conductivity. Unfortunately, most things with high electrical conductivity have high thermal conductivity as well.

You can't postulate an efficiency because the output voltage will be dependent on the temperature drop across the device and many other system dependent factors. A car with this installed on its exhaust would make less power sitting in traffic than at full highway speed. This is because of the better ability to shed heat with airflow. Honda and a few other companies have been working on thermoelectric regenerators for cars in the past few years. Keep in mind that there has to be a temperature difference across the thermoelectric junction. This could not simply be dropped in a hot hole in the ground and expected to produce power. The opposite side of the junction needs to be cooled. That could be done with water, forced air, a big thermal rod to the surface or any number of things.

Egnite is correct that the alternator does that job already and is extremely efficient at converting rotational motion into electricity. Most electrical generators/motors are in the high 90%'s for efficiency. However, they do cause drag on the engine lowering available horsepower and increasing fuel consumption along with the subsequent emissions. By capturing the waste heat and turning that into power, overall engine efficiency can be improved.

Jul 25, 2008
thallium-doped lead telluride

Jul 25, 2008
thallium-doped lead telluride - Anyone notice that it is likely to be quite toxic since it is made with both lead and thallium? RoHS and most US environmental laws will limit it's use to primarily laboratory work and possibly high end scientific and space equipment. It's unlikely that it will be produced in any type of large scale. The whole concept it quite interesting and has been looked at for several years. Hopefully they can find a family of coupounds that don't contain lead that can offer similar output.

Oh, and sorry about the previous comment, apparently you can't make arrows in physorg's comment boxes.

Jul 25, 2008
There is a chart here that shows how ZT converts to efficiency of waste heat recapture. A 500 degree K difference in heat (which would be possible at the 950F high end) would mean 15% conversion of heat to electricity. There is industrial waste heat in the USA of about 7 quads. Get 15% of that is 1.0 quad. This is about equal to all of the solar and wind power generated in the USA from 2002 to 2006.

They say they can double the ZT to 3. This would be 23% energy recapture at the 500K temperature difference. Which would 1.5 quad applied to the available industrial waste heat.


Jul 25, 2008
Many different ways to capture the waste heat in cars.

Summary here:

Not just thermoelectrics could be used for capturing waste heat in cars but :
%u2022 Electrical Turbo-Compounding (Caterpillar) : 3 to 10% announced fuel economy
%u2022 Mechanical Turbo-Compounding : 5 to 10% announced fuel economy
%u2022 TIGERS : Turbo-generator Integrated Gas Energy Recovery System : 6% announced fuel economy
%u2022 Thermo-electricity : 20% announced fuel economy
%u2022 Stirling Cycle in co-generation : up to 40% announced fuel economy but at too low specific power
%u2022 Rankine Cycle : Turbosteamer : 17% announced fuel economy
%u2022 Organic Rankine Cycle (ORC) : up to 60% announced fuel economy
%u2022 Thermo-acoustics : low specific power

Details at heat2power:

Some spots to place systems are under the chassis to take the exhaust from the engine.

Or replacing cylinders in the engine.

Loboy - be on your way for about 30 to 100 miles. Range is still a problem for all electric cars.

Jul 25, 2008
Remember the First Law of Thermodynamics, "You can't win."

I thought the First Law of Thermodynamics was you do not talk about Thermodynamics?

Jul 26, 2008
Just forget about 15% or any significant efficiency by thermoelectric converters. They're all in the 1% range or even lower.

They're still useful, and better materials are welcome, but thinking of car engines is just the wrong choice.

The only non-lab use I know it for radioisotopic generators on deep space probes. Saving 50% - or even 10% - of the scarce 238Pu would be very appreciated by space engineers.

For cars and other sensible uses, a 1% efficiency doesn't justify any move. For more exotic uses like solar power, efficiency tells not use gases or vacuum instead of solids.

And then you have semi-lab application, like cooling CCD retinas in telescopes, which would welcome a less inefficient element. On a 3-layer stack, saving a factor of 2 per layer gains a factor of 8 at the end: 8 times less power, 8 times smaller heatsink.

Jul 28, 2008
Sounds like a band-aid for the internal combustion engine.

Remove the engine, remove the gas tank, remove the carburator, remove the exhaust, and remove the alternator. Then install some batteries and an electric drive in your car, and be on your way.

You have my vote on that one!

Jul 28, 2008
What a stupid interpretation, reflecting major ignorance. "You can't win" is NOT a law of anything except for losers. Just because energy can't be created or destroyed does NOT mean you can't win, idiot... I would say the last 6,000 years of technological history are pretty good proof of that fact... Of course you can't make more energy than you consume but that is a LONG way from "not winning" since the energy output of our single tiny little nothing star is enough to make us all winners a billion times over and we have hardly even begun to tap what is available. Loser..
Remember the First Law of Thermodynamics, "You can't win."

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more