There are several favorable attributes associated with energy harvesting: it’s a (pick one or more) win-win, free lunch, and “something for nothing.” Those phrases are largely deserved, since a viable harvesting design can capture minute amounts of otherwise wasted energy and provide useful amounts of power.

Yet the reality is that many harvesting applications are hard to implement compared to what abstract theory indicates. One example is use of thermoelectric generators (TEGs) which use body heat in conjunction with the Seebeck effect. Not only is the source temperature fairly modest, but the physical issues of maintaining solid, low-resistance contact with the skin are significant. Ideally, you want something that is in constant contact, yet pliable enough to move with the skin’s normal and unavoidable motion, flexing, and stretching. Conventional TEG materials and contacts are relatively rigid, which is at odds with the user’s needs.

However, work being done by a team at North Carolina State University may make TEG viable for wearables. The team has developed flexible TEG electronics (not to be confused with our widely used flexible circuitry) which they claim have the internal electrical performance of semi-rigid devices, yet can flex as needed (Figure 1). The result is both efficiency – obviously critical in a harvesting situation – as well as user comfort.


Figure 1
This thermoelectric harvester has the material quality of rigid devices inside a flexible package, while the liquid metal in the flexible thermoelectric device allows for self-healing. (Source: North Carolina State University)

To do this, they used a metal comprised of gallium and indium, which is apparently readily available as a standard, non-toxic alloy called EGaIn, to fabricate the “contacts” that connect the multiple solid TEG legs in series (for higher voltage). The EGaIn material has two virtues in this application. First, it has low electrical resistance, which is essential for minimizing losses. Second, the EGaIn liquid-metal interconnects also provide stretchability along with self-healing, so that if the TEG-TEG connection is “broken,” the liquid metal will self-reconnect and restore the current path (this reminds me of the T-1000 Terminator in the 1991 movie Terminator 2).

Their research was reported in a detailed technical paper “Flexible thermoelectric generator using bulk legs and liquid metal interconnects for wearable electronics” which included charts, graphs, images, and tables, but it is behind a paywall. Fortunately, they have posted a detailed 2015 presentation, “Thermoelectric Generators for Body Heat Energy Harvesting,” which actually is a very useful tutorial with considerable insight and information. The presentation points out that low thermal conductivity has more influence than the Seebeck effect here (note that the temperature difference between the body and ambient is only a very modest 10 to 15°C), and includes thermal and physical models (Figure 2), lab details, and fabrication specifics.


Figure 2
This physical and thermal model of a skin-based TEG is the basis for additional analysis of heat sources, losses, and more. (Source: North Carolina State University)

Their approach is still in the R&D stage and far from ready for commercialization, but it reiterates two important power-related points. For advanced technology, including power sources, the path from the lab to a viable, commercial, reliable product at an acceptable cost is very long one. Many promising battery-related technologies and chemistries have not made it to market because issues related to full production were formidable and could not be overcome.

Also, all of our technologies are very dependent on fundamental advances in materials science. Nearly every element of the periodic table has unique attributes and characteristics which make it not only useful in a given situation, but essential for performance of components, coatings, structures, and more.

Have you ever used or wished for truly flexible, liquid, or self-healing contacts, whether for TEGs, wearables, or other uses? How did you work through the application need?

Bill Schweber is an EE who has written three textbooks, hundreds of technical articles, opinion columns, and product features.


Related articles: