Thermoelectric efficiency boost is good news for solar

For more than half a century, researchers have been trying to salvage the substantial amounts of waste heat lost in fossil fuel plants and combustion engines. Heat loss throws away 40 per cent of petrol energy through the car's exhaust, and two-thirds of coal energy from coal-fired power plants.
Their putative ability to mop up that lost energy has made thermoelectric materials a perpetual Cinderella technology. The materials use heat to create "free" electricity: current is generated when the temperature difference between the hot side (say, the exhaust) and the cool side (the ambient air) pushes electrons from one side of the material to the other.

 

 

         In practice, however, thermoelectric materials reclaim at best only 5 to 7 per cent of the lost energy. Their efficiency - a material's ability to generate electricity for a given amount of thermal energy - is reflected in a figure called its ZT. For 50 years, researchers have struggled to push that number past 1.
The most straightforward approach is to coax a material to conduct electricity, while preventing the heat from migrating (because efficiency depends on preserving the difference between the material's hot and cold sides). That means decoupling electrons from phonons - a phonon is to heat transfer what an electron is to electricity. Essentially, a phonon is a quasiparticle that can be functionally thought of the vibrations that carry thermal energy.

                In other words, let the electrons flow while stopping the phonons in their tracks. But they have been difficult to decouple. So thermoelectrics been relegated to applications of last resort, such as in space, where the small amount of energy they reclaim is worth the cost of the expensive materials.
                Then, last week, researchers at Northwestern University, in Evanston, Illinois, published a paper in Nature which indicates that they've kicked the ZT from 1 to 2.2. They did it by disturbing the flow of three different wavelengths of the phonons, which allowed the electrons to pass while trapping the phonons in layers, which they compared to a Russian doll. Pushing the ZT to 2.2 bumped the overall efficiency up to 20 per cent.
                 Their work has tantalising implications for a far more consumer-friendly application: solar panels. Normally these can only metabolise the photons from the high-frequency part of the electromagnetic spectrum, meaning most of the sun's rays are lost as waste heat.
Use thermoelectric materials to harness the entire spectrum and your solar panel will get whopping good efficiencies. Charles Stafford, who works on materials that can disrupt phonon flow at the University of Arizona in Tucson, calls the new research "very exciting work".