Scientists propose quantum wells as high-power, easy-to-make energy harvesters

October 23, 2013 by Lisa Zyga feature
This artistic drawing of the proposed quantum well energy harvester shows the central hot region in red, the quantum wells in green, and the cold electrodes in blue. Credit: Bjӧrn Sothmann, et al.

( —By collecting heat energy from the environment and transforming it into electrical power, thermoelectric energy harvesters have the potential to provide energy for a variety of small electronic devices. Currently, the biggest challenge in developing thermoelectric energy harvesters is to make systems that are both powerful and efficient at the same time.

One material that scientists have experimented with for making thermoelectric energy harvesters is , nano-sized crystals with semiconducting properties. Due to their sharp, discrete energy levels, quantum dots are good energy filters, which is an important property for thermoelectric devices.

In a new study published in the New Journal of Physics, a team of researchers from Switzerland, Spain, and the US has investigated a thermoelectric energy harvester design based on quantum wells. Although quantum wells are also made of semiconducting materials, they have different structures and energy-filtering properties than quantum dots.

"We have shown that quantum wells can be used as powerful and efficient energy harvesters," said coauthor Björn Sothmann, a physicist at the University of Geneva in Switzerland. "Compared to our previous proposal based on quantum dots, quantum wells are easier to fabricate and offer the potential to be operated at room temperature."

The energy harvester design that the researchers investigated here consists of a central cavity connected via quantum wells to two electronic reservoirs. The central cavity is kept at a hotter temperature than the two electronic reservoirs, and the quantum wells act as filters that allow electrons of certain energies to pass through. In general, the greater the temperature difference between the central cavity and the reservoirs, the greater the electron flow and output power.

In their analysis, the researchers found that the quantum well energy harvester delivers an output power of about 0.18 W/cm2 for a temperature difference of 1 K, which is nearly double the power of a quantum dot energy harvester. This increased power is due to the ability of quantum wells to deliver larger currents compared to quantum dots as a result of their extra degrees of freedom.

Although the quantum well energy harvester has a good efficiency, the efficiency is slightly lower than that of energy harvesters based on quantum dots. The researchers explain that this difference occurs because of the difference in energy filtering: quantum wells transmit electrons of any energy above a certain level, while quantum dots are more selective and let only electrons of a specific energy pass through. As a result, quantum wells are less efficient energy filters.

Quantum well energy harvesters appear promising for applications. For one thing, they may be easier to fabricate than energy harvesters that use quantum dots, since quantum dots are required to have similar properties in order to achieve good performance, and there is no such requirement for . In addition, the fact that they can operate at room temperature may make quantum well energy harvesters suitable for a variety of applications, such as electric circuits.

"The harvester can be used to convert waste heat from electric circuits, e.g. in computer chips, back into electricity," Sothmann said. "This way, one can reduce both the consumed power as well as the need for cooling the chip."

The researchers hope that their work encourages experimental groups to build and test the device.

Explore further: A quantum dot energy harvester: Turning waste heat into electricity on the nanoscale

More information: Bjӧrn Sothmann, et al. "Powerful energy harvester based on resonant-tunneling quantum wells." New Journal of Physics, 15 (2013) 095021. DOI: 10.1088/1367-2630/15/9/095021

Related Stories

New nanomaterial increases yield of solar cells

August 26, 2013

Researchers from the FOM Foundation, Delft University of Technology, Toyota Motor Europe and the University of California have developed a nanostructure with which they can make solar cells highly efficient. The researchers ...

Team demonstrates quantum dots that assemble themselves

April 16, 2013

( —Scientists from the U.S. Department of Energy's National Renewable Energy Laboratory and other labs have demonstrated a process whereby quantum dots can self-assemble at optimal locations in nanowires, a breakthrough ...

Recommended for you

Scientists update four key fundamental constants

October 23, 2017

Paving the way for transforming the world's measurement system, an international task force has determined updated values for four fundamental constants of nature. The updated values comprise the last scientific piece of ...

Scientists discover superconductor with bounce

October 23, 2017

The U.S. Department of Energy's Ames Laboratory has discovered extreme "bounce," or super-elastic shape-memory properties in a material that could be applied for use as an actuator in the harshest of conditions, such as outer ...

Experiment provides deeper look into the nature of neutrinos

October 23, 2017

The first glimpse of data from the full array of a deeply chilled particle detector operating beneath a mountain in Italy sets the most precise limits yet on where scientists might find a theorized process to help explain ...


Adjust slider to filter visible comments by rank

Display comments: newest first

1.8 / 5 (8) Oct 23, 2013
Well there you go. As discussed in this thread:

-The total solar energy including thermal, as converted to electricity, may be enough to charge a vehicles batteries in a reasonable amount of time. Benefits could include sucking the heat out of the interior.
1.8 / 5 (9) Oct 23, 2013
Couldn't you just put these on the back side of a piece of black painted sheet metal to use them as solar panels? and put it on the roof of a small building with no walls (for easy ventilation) so that a huge heat difference forms between the surface and the chip.

If they produce that much energy as is stated in the article, they'd be over 16 times as good as solar panels.

In the scenario I just described you could easily get much more than 1K difference between the sheet metal and the air below it.
4 / 5 (1) Oct 24, 2013
Wait, isn't 0.18 W/cm2 the same as 1800 W / m2? That would be some serious amount of power. I must be missing something here.
1.2 / 5 (6) Oct 25, 2013
Wait, isn't 0.18 W/cm2 the same as 1800 W / m2? That would be some serious amount of power. I must be missing something here.

No, its 18 W/m2/1K.

So let's run the numbers. A core I7's exterior surface area for contact is about 18.9cm2. The maximum possible temperature gradient, assuming the processor is completely pegged, at room temperature of 70f, and a 100% efficient way to keep the other side of the quantum well at room temperature(obviously not possible), is 43.889K, providing a total of about 150 watts. The processor would be consuming 84 watts of power in this scenario, leaving a fairly healthy margin of about 65 watts of wiggle room to account for inefficiency in maintaining the gradient(or perhaps even an active solution) BEFORE you reach break even.

Interesting. I don't know what type of gradient is actually realistic to maintain in said conditions, but it appears the deck is surprisingly well stacked for this solution to be feasible and worthwhile.
2.5 / 5 (2) Oct 26, 2013
@Requiem - there are 100x100 cm2 in a square meter, so 1800 W/m2 is correct.

@Lurker - no, you couldn't maintain a 1 degree temperature difference just in normal sunlight because this has really low thermal resistance. There is no getting around the laws of thermodynamics - with a 1K temperature difference at room temperature, it can't be more than 0.33% efficient, so you'd need at least 54W/cm2 of heat flow through it (and probably more like 100 W/cm2 of heat flow through it). So you'd need to concentrate the sunlight >540x (and probably >1000x) to keep a 1K temperature difference.
To get that much heat flow at a 1K difference its thermal resistance must at most be equivalent to 0.75 mm of copper.
1 / 5 (3) Nov 02, 2013
I imagine this device coupled with diffraction grated photon could net background information as power.
1 / 5 (5) Nov 08, 2013
Use the stuff in pebble bed reactors and build small reactors incorporating this design in small towns all over the nation.

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

Click here to reset your password.
Sign in to get notified via email when new comments are made.