A flat, flexible film that turns body heat into electricity - no bending required

Seoul National University College of Engineering

Wearable thermoelectric generators have always faced an awkward contradiction. Make them thin and flexible enough to wear comfortably, and they stop working. The devices need a temperature difference between their hot and cold sides to generate electricity, but when you flatten a thermoelectric film against skin, body heat passes straight through and dissipates into the air. No temperature gradient, no power. It is like trying to run a waterwheel in a river with no slope.

Previous solutions involved bending the device or building three-dimensional pillar structures to create that gradient artificially. These approaches worked electrically but defeated the purpose of having a thin, flexible device in the first place. A team at Seoul National University, led by Professor Jeonghun Kwak, has now found a way to keep the film flat and still generate power.

Redirecting heat instead of fighting physics

The core insight is deceptively simple: if heat insists on traveling straight through a thin film, make it travel sideways instead. Kwak's team accomplished this by engineering a dual thermal conductivity substrate - a single layer of stretchable silicone (PDMS) in which some regions conduct heat well and others do not.

They achieved the contrast by embedding thermally conductive copper nanoparticles into selected areas of the silicone, leaving adjacent areas as plain, poorly conducting polymer. When the device sits on skin, body heat does not simply pass vertically through the film. Instead, it flows laterally along the high-conductivity regions, creating relatively warm and cool zones on the substrate surface. Those zones produce the temperature difference that drives electricity generation.

The team calls the result a pseudo-transverse thermoelectric generator, because it structurally mimics the transverse thermoelectric effect - where heat flows perpendicular to the electrical current - without requiring exotic transverse thermoelectric materials.

Printed like ink, assembled like blocks

The device is fabricated using an ink-based printing process, which gives it inherent flexibility. It can bend and stretch with skin without cracking or losing performance. But the manufacturing approach also offers something potentially more important for practical adoption: scalability.

Because the generator is built from repeating modular units, it can be sized and shaped to fit different applications. Need a small patch for a fingertip sensor? Print a few units. Want to power a more demanding device from a larger skin area? Tile more units together. The researchers compare it to assembling modular blocks - each unit contributes to the total power output, and the geometry can be customized freely.

This flexibility in design distinguishes the approach from rigid thermoelectric devices that come in fixed sizes and cannot easily conform to curved body surfaces.

What the flat configuration actually delivers

The study, published March 18 in Science Advances, demonstrates that the pseudo-transverse generator produces electricity from body heat in a completely flat configuration, without any bending or structural deformation. This is the first demonstration that a thin-film thermoelectric device can maintain a usable temperature difference through substrate engineering alone.

The potential applications span smart clothing, continuous health monitoring sensors, and other wearable electronics that currently depend on batteries. A device that harvests body heat could extend battery life or, for very low-power sensors, eliminate the battery entirely. Temperature monitoring, heart rate tracking, and motion sensing all fall within the power envelope that body-heat harvesting might eventually supply.

The distance between demonstration and your wrist

Several significant hurdles remain between this laboratory demonstration and a product anyone would actually wear. The study establishes proof of concept but does not report long-term durability data. Wearable devices endure sweat, friction, repeated stretching, and washing - conditions that could degrade copper nanoparticle composites or alter thermal conductivity patterns over time.

Power output is another open question. The body's skin-to-air temperature difference is modest under most conditions - typically just a few degrees Celsius - which fundamentally limits how much electricity any thermoelectric device can extract. The study demonstrates that a temperature gradient can be maintained in a flat film, but whether the resulting power is sufficient for practical applications at scale remains to be validated in real-world conditions.

The copper nanoparticle approach also raises questions about biocompatibility and manufacturing cost. While PDMS is well-established as a biocompatible material, the long-term skin contact behavior of copper-doped silicone composites needs separate evaluation. And ink-based printing, while scalable in principle, must still prove cost-competitive with conventional battery technology for the target applications.

Environmental conditions matter too. In hot climates where skin temperature and ambient temperature converge, the available temperature gradient shrinks, and so does power output. In cold environments, the gradient is larger but the wearer is likely wearing insulating clothing that blocks skin access. The sweet spot for body-heat harvesting is narrower than it first appears.

A structural solution to a thermal problem

What makes this work interesting beyond its immediate application is the approach itself. Rather than searching for better thermoelectric materials or more complex device architectures, the SNU team solved a heat-flow problem by engineering the substrate. The thermoelectric semiconductors are conventional. The silicone is conventional. The novelty is entirely in how the substrate directs thermal energy, turning a material design problem into a geometry problem.

That kind of lateral thinking - literally, in this case - tends to be more transferable than solutions tied to specific exotic materials. If the principle holds up at scale, it could be adapted to other thin-film energy harvesting scenarios beyond wearables.