MIT finds new way to harvest energy from heat


May 23, 2014

Waste heat could be harnessed more effectively, using the thermogalvanic effect (Photo: Shutterstock)

Waste heat could be harnessed more effectively, using the thermogalvanic effect (Photo: Shutterstock)

Image Gallery (2 images)

Researchers at MIT and Stanford have found a new way to transform waste heat into electricity, particularly in situations where the temperature gradient is small, below 100º C (180° F). The technology uses widely available materials, and could be used to recycle the large amounts of wasted heat generated in industrial processes and electric power plants.

According to the US Environment Protection Agency (EPA), one third of the industrial energy consumption in the United States is wasted as heat. If we could get even a small portion of it back, we'd be able to decrease power consumption significantly.

Scientists have turned to thermoelectric materials, which can generate electricity from a temperature gradient, as a possible solution. However, there are two problems with that approach: firstly, they are mostly made from rare materials, meaning they are expensive and can't be scaled up for industrial applications; and secondly, when the waste heat is only tens of degrees hotter than the surrounding environment, which is very common, their efficiency of thermoelectric materials is of only about 0.5 percent.

Researchers at MIT and Stanford led by Prof. Gang Chen have devised an alternative approach that takes advantage of the thermogalvanic effect, which describes a peculiar relationship between the temperature of a battery and the voltage at which it can effectively be charged up.

The scientists built a system that allows waste heat to first raise the temperature of a battery. Because of the thermogalvanic effect, the battery can now be charged at a lower voltage than would normally be required. The battery is then allowed to cool down, and at this point its lower temperature allows it to be discharged at a higher voltage, releasing more energy than was put into it through the electric grid. The difference in energy was gathered from waste heat.

The idea to harness the thermogalvanic effect to generate electricity was first proposed in the 50s, where it demonstrated efficiencies of up to 50 percent of the Carnot limit.

Compared to the original system, Chen and colleagues have achieved the capability to harness much smaller differences in temperature with relatively high efficiency (a difference of 50° C (90° F), with 5.7 percent efficiency) and the use of commonly available materials, such as copper, that could easily work at scale. Finally, their system could be manufactured quite easily, as it fits very well into the existing production chains of the battery industry.

The researchers will now turn to tackle the remaining challenges – namely, the low power density of 1.2 W/kg compared to thermoelectric materials, the speed of battery cycling of about one hour, and the need for extensive testing to ensure a useful operating life.

A paper describing the results appears in the journal Nature Communications.

Source: MIT

About the Author
Dario Borghino Dario studied software engineering at the Polytechnic University of Turin. When he isn't writing for Gizmag he is usually traveling the world on a whim, working on an AI-guided automated trading system, or chasing his dream to become the next European thumbwrestling champion. All articles by Dario Borghino

how about some actual data ... like starting with an empty 100 amph battery, heating it up, charging it with x amount of energy, letting it cool down and then extract Y amount of energy from the battery ... let us see the difference between X and Y ... otherwise it pie in the sky ...

until then this is just a grant funding press release ...

Jeffrey Carlson

When batteries are already needed for load-leveling or backup, this could be economical. I'd think that "flow batteries", utility-scale batteries which store the charged and discharged electrolytes separately from the electrodes, would be the best type of battery for this purpose since they would just need to insulate the tanks and heat/cool the electrolyte rather than the whole battery.


I'll just use a Stirling cycle.


Jeffrey Carlson: did you happen to go so far as to click on the link to Nature Comm.? Surely you don't expect Gizmag to provide that data for you.

Whether this proves to be practical or not remains to be seen; nothing done in a university lab, no matter how much data accompanies it, can be considered commercially practical until further demonstrated. Your comment is immature.

EH: the electrolyte temperature is not part of the process; rather it is the temperature dependence of the electrochemical reaction (which takes place at the electrodes). "Practicality" depends, to be sure, on the thermal stability of these materials (which I imagine is surely part of the reason why they chose the specific copper-based materials, which are not often seen in commercial batteries). Still this is a genuinely novel (from the commercial perspective) twist which has a real chance; it's not based on impractical theoretical grounds or claims about material systems that have long ago been explored.

Jayna Sheats
Post a Comment

Login with your Gizmag account:

Related Articles
Looking for something? Search our articles