Thermoelectrics Go Nano

A piece I wrote for Materials Today – originally appeared here: http://www.materialstoday.com/electronic-properties/comment/thermoelectrics-go-nano/

Thermoelectrics Go Nano

Scientists from Northwestern University have developed a new nanostructured thermoelectric material which they claim converts heat to electricity more efficiently than any existing material.

A thermoelectric (TE) material uses a temperature difference to drive electrons in the material and create a voltage. But, despite impressive efforts to redesign the device structure and the on-going search for new materials, the fundamental issue of the low efficiency of thermoelectric energy conversion has remained.

The ability of a material to convert heat into electrical power is defined by its dimensionless figure-of-merit, ZT (= α2σ/κ, where κ is the thermal conductivity, σ the electrical conductivity and α is the Seebeck coefficient). So, to achieve a high ZT, a material needs to have low thermal conductivity and high electrical conductivity. Evidently, for bulk materials this is a major challenge, as these properties are highly coupled.

But, energy transport in nanostructures differs significantly from macrostructures, because of classical and quantum size effects on energy carriers. So, much of the effort has focussed on nanostructuring the material, with record-high ZT values up to 1.8 at 450 – 650 °C

However, a TE material published by Kanatzidis et al has smashed all of the previous records, reporting a figure-of-merit of 2.2 (at 650 °C). Kanatzidis and his team started with lead telluride (PbTe), a highly-researched thermoelectric material. But, in order to decrease its thermal conductivity, and thus increase its figure-of-merit, its structure was controlled on several length scales. This new material consists of mesoscale grains, nanoscale precipitates of an additive, strontium telluride (SrTe) and trace amounts of sodium provided disorder in the lattice of PbTe.

We know that nanostructuring can scatter a significant portion of the phonon spectrum of TE materials, but this group have taken it beyond nanostructuring. The mesoscale architecture scatters the heat-carrying phonons with long mean free paths. And the use of atomic-scale lattice disorder is known to scatter phonons with short wavelengths. The combination of all of these effects is what results in a material with a record-breaking figure-of-merit.

This work represents a paradigm shift in thermoelectric materials. Their low efficiency has always been quoted as a bottleneck to mass-market. Breaking the ZT = 2 barrier will go a long way to widening that bottleneck, and may led to a more widespread adoption of thermoelectric materials and devices.