An ionic thermoelectric material can take strikingly different forms: a liquid, a soft gel, a charged polymer film, or a porous membrane. Yet all four families share one trick. They turn a temperature difference into an electrical signal carried by ions rather than the electrons that drive conventional semiconductors. The science is surprisingly old: Faraday and Nernst laid its foundations in the 19th century, and Soret described the companion thermodiffusion effect in 1880, but after decades of slow progress the field only surged through the 2010s, as new materials pushed thermopowers past 10 mV/K and flexible i-TE gels appeared.
A new review in National Science Review (NSR), coordinated by researchers from nine institutions, takes a bird's-eye view of the field, starting from the fundamentals. I-TEs work through a blend of electrochemistry, thermodynamics and ion transport, which the review sorts into three mechanisms: thermodiffusion (TD) effect, the Soret effect, drives ions along a temperature gradient; thermogalvanic (TG) effect runs on temperature-sensitive redox couples; and thermoextraction (TEx) effect, the youngest, uses active electrodes that intercalate and extract ions like a rechargeable battery. Whichever mechanism is at work, each material pairs an active ion source with a host that shapes how ions move, and what unites them is an unusually high thermopower, ten to a hundred times that of electronic thermoelectrics. That thermopower is what lets i-TEs tap low-grade waste heat, the vast share of the world’s energy too cool to drive a turbine or be harvested efficiently by conventional methods. The materials are built into single- or dual-cell devices: single cells harvest a temperature difference across space, while dual cells need none, drawing on a difference over time so that heat can be recovered and stored.
Their appeal is broader still. An i-TE may hold no clear edge over a semiconductor in raw output power, yet many of its materials, especially the gels, are soft, stretchable and solution-processable, a natural fit for wearable patches and skin-mounted electronics that rigid devices cannot match. The chemistry is unusually versatile, and many of the building blocks are cheap, easy to process and environmentally benign. Turning that promise into hardware, the review stresses, runs into stubborn challenges where electrochemistry, thermodynamics and ion transport are tightly coupled.
On the chemistry side, performance hinges on choices that are hard to optimize together: which thermodiffusion ions or thermogalvanic redox couples to use, how to design the polymer network and the solvation shell around the active species, and how to build mixed ion–electron conductors that escape the usual trade-off between electrical conductivity and thermopower. The physics is just as demanding, governed by how ion size and concentration set transport, how the electric double layer forms and behaves at electrodes, and how the thermal-to-electrical conversion couples to these effects. Together these chemical and physical levers explain why a promising material so often falls short at the device level, and why the review treats them as core problems to solve rather than footnotes.
Those similar mechanisms can be put to strikingly different ends. Depending on design, an i-TE can generate electricity, pump heat for cooling, sense temperature, pressure or humidity, or store energy. These are largely separate design directions, though advanced devices deliberately combine a few: some harvest heat while also storing it, and others go further still, cascading heat into electricity and then into chemical energy, so that a single device can harvest waste heat and generate hydrogen at once. In practice, applications already span an unusually wide range. The most mature is power generation from body heat and ambient warmth, where i-TE fabrics and patches drive low-power sensors, health monitors and wireless transmitters without batteries. Run in reverse, the same redox chemistry enables active cooling that can rival conventional semiconductor coolers: one solvation-engineered electrolyte has reached an electrolyte-specific coefficient of performance of 14.3 at low applied voltages. Their sensitivity also makes them sharp multimodal sensors: a single material can respond to temperature, pressure, humidity and chemical cues at once, and is being explored for smart human–machine interaction. The map extends further still, to i-TE cement for smart buildings, fire-alarm paper chips and energy-harvesting smart windows.
The review closes with a roadmap whose central message is that the field is held back less by physics than by a fragmented research landscape. The rate-limiting steps are no longer record thermopowers but four practical fronts: standardization, cross-scale design, scaling up and reliability for commercialization, and sustainability. The review urges a common classification scheme and a dedicated figure of merit for most i-TE systems. To tie the fronts together, the authors call for “closed-loop” development, in which materials screening, device prototyping, system simulation and field validation continually feed back into one another.
The review was led by co-corresponding authors Shangchao Lin (Shanghai Jiao Tong University), Weishu Liu (Southern University of Science and Technology) and Hulin Zhang (Taiyuan University of Technology). It appears as part of a special topic on ionic thermoelectrics in National Science Review (DOI: 10.1093/nsr/nwag283; CC BY 4.0).
National Science Review
Literature review