A thermoelectric generator is a solid-state device that converts a temperature difference directly into electricity, with no moving parts, no fluids, and no noise. The underlying physics relies on the Seebeck effect: when one side of certain semiconductor materials is hotter than the other, charge carriers migrate from the hot side to the cold side, producing a voltage. Typical conversion efficiencies sit in the range of about 5 to 10 percent for commercial devices, which sounds modest until you consider that the heat source is often waste energy that would otherwise be lost entirely. That efficiency ceiling, the materials that define it, and the surprisingly wide range of places TEGs show up are all worth understanding.
How the Conversion Works
The Seebeck effect is straightforward in concept. Two dissimilar semiconductor materials, one conducting mainly through negative charge carriers (n-type) and the other through positive ones (p-type), are connected electrically in series and thermally in parallel. Apply heat to one end and keep the other end cool, and charge carriers in each leg drift toward the cold side. That drift generates a voltage across the pair. Stack dozens or hundreds of these pairs into a module, and you get usable electrical output. The whole device has no turbines, no pistons, and nothing that wears out from mechanical friction.
Because there are no moving parts, TEGs are nearly silent and extremely low-maintenance. They can operate for years in remote or harsh environments without servicing, which is why space agencies have relied on radioisotope thermoelectric generators to power deep-space probes for decades. On Earth, the same principle applies wherever steady waste heat exists and conventional generators are impractical.
Why Efficiency Is So Hard to Improve
The performance of a thermoelectric material is captured by a dimensionless number called the figure of merit, usually written as zT. A higher zT means more efficient conversion. The trouble is that zT depends on three properties that fight each other: you want high electrical conductivity so current flows easily, a large Seebeck coefficient so each degree of temperature difference produces a larger voltage, and low thermal conductivity so the hot side stays hot instead of conducting heat straight through to the cold side. Raising one tends to lower another. A material that conducts electricity well almost always conducts heat well too, and boosting the Seebeck coefficient typically comes at the cost of electrical conductivity.
This three-way tug-of-war is the central challenge of thermoelectric research. Most commercial modules hover around a zT of about 1, which translates to that 5 to 10 percent conversion window for real-world systems. Pushing beyond that has occupied materials scientists for years, and the strategies are getting creative.
The Phonon-Glass Electron-Crystal Idea
One of the most compelling strategies for breaking the efficiency deadlock is to engineer a material that conducts electricity like a perfect crystal but blocks heat like a glass. Researchers call this the phonon-glass electron-crystal concept. Heat travels through a solid partly via lattice vibrations called phonons. If you can scatter those phonons without disturbing the flow of electrons, you slash thermal conductivity while keeping electrical conductivity intact.
Achieving this in practice has been rare, but recent work on a material called AgSbTe₂ doped with ytterbium came remarkably close. By tuning the atomic disorder in the crystal, researchers created nanoscale ordered domains just 2 to 4 nanometers across that strongly scattered heat-carrying phonons. The result was a glass-like thermal conductivity that barely changed with temperature, while electron transport remained crystal-like. The material reached a zT of about 2.4 at 300 °C, which is outstanding by current standards.1PubMed. High Thermoelectric Performance in Phonon-Glass Electron-Crystal Like AgSbTe(2)
A separate approach explored entropy-stabilized crystals, specifically a five-element telluride compound. By mixing multiple metals into a single crystal structure, researchers induced local atomic distortions that created anharmonic lattice vibrations. The phonon mean free path dropped so low it approached the distance between individual atoms, which is about as glass-like as a crystalline solid can get, while the average crystal structure remained orderly enough for electrons to move freely.2Journal of the American Chemical Society. Atomic Off-Centering Driven Phonon-Glass Electron-Crystal-like Thermoelectric Transport in Entropy-Stabilized Quinary Telluride
Nanostructuring and Defect Engineering
You don’t always need exotic compositions. Sometimes you can improve a well-known material by introducing nanoscale features that trip up phonons. Work on lead telluride, one of the classic mid-temperature thermoelectric materials, showed that adding small amounts of antimony created nanoscale precipitates within the bulk material. These precipitates introduced misfit dislocations and local elastic strain at the interfaces between the precipitate and the surrounding crystal. Phonons scatter off these tiny defects while electrons pass through largely unbothered, reducing lattice thermal conductivity and improving the overall figure of merit.3PubMed. On the origin of increased phonon scattering in nanostructured PbTe based thermoelectric materials
This kind of defect engineering has become a standard playbook in thermoelectric research. Whether it is grain boundaries in nanostructured bulk materials, embedded nanoparticles, or deliberately introduced dislocations, the goal is always the same: create obstacles that phonons cannot ignore but electrons barely notice.
Materials for Different Temperature Ranges
No single thermoelectric material works well across all temperatures, so device designers choose materials based on where the heat source falls.
- Near room temperature: Bismuth telluride compounds dominate. They perform best from roughly ambient temperature up to about 250 °C and are the backbone of most commercial thermoelectric modules.4Journal of Power Sources. Bismuth telluride-based thermoelectric generators: Advances in synthesis, performance enhancement, and device applications
- Mid-range temperatures: Lead telluride, skutterudites, and half-Heusler alloys operate well from about 300 °C to 600 °C. Cascaded modules that stack skutterudite and half-Heusler stages have been built to cover a wider temperature span in a single device.5PubMed. High-Performance Skutterudite/Half-Heusler Cascaded Thermoelectric Module Using the Transient Liquid Phase Sintering Joining Technique
- High temperatures: Silicon-germanium alloys are used for heat sources around 1000 °C, the kind of temperatures found in industrial furnaces and radioisotope heat sources. Recent additive-manufacturing work produced undoped Si-Ge parts using laser powder bed fusion, though the resulting zT of 0.06 at 400 °C for Si₈₀Ge₂₀ shows how much room remains for improvement in new fabrication methods.6ScienceDirect. Nano/microstructures and thermoelectric properties of silicon germanium manufactured using laser powder bed fusion
- Flexible and organic: Organic thermoelectric materials are attracting attention for wearable applications. They offer low cost, low weight, inherent flexibility, and naturally low thermal conductivity, even though their power output per unit area is still far below inorganic alternatives.7Nano Energy. Organic-based flexible thermoelectric generators: From materials to devices
Cascaded or segmented designs, which use different materials at different points along the temperature gradient within a single device, are one of the more practical ways to extract more power from a large temperature difference. Rather than forcing one material to work across a wide range, each segment operates near its own optimal temperature window.
Recovering Waste Heat from Vehicle Exhaust
Roughly two-thirds of the energy in gasoline leaves a car’s tailpipe as waste heat. That makes automotive exhaust one of the most talked-about applications for thermoelectric generators. The idea is to wrap thermoelectric modules around part of the exhaust system, with coolant on the far side, and let the temperature difference generate electricity that can reduce the load on the alternator or power accessories.
Research groups have built and tested these systems with varying results. One study modeled a TEG system at a highway cruising speed of 120 km/h and predicted an output of about 38 watts with a conversion efficiency of roughly 1.5 percent. After accounting for the added vehicle weight and the power needed to pump coolant, the net useful power dropped to about 24 watts.8Energy. Performance investigation of a thermoelectric generator system applied in automobile exhaust waste heat recovery That is not much on its own, but the appeal lies in the fact that this energy would otherwise simply heat the atmosphere.
Heat exchanger design matters enormously. Experimental setups have compared copper and steel heat exchangers with triangular channels to see how efficiently they transfer exhaust heat to the thermoelectric modules.9Energy Conversion and Management: X. Exhaust heat harvesting of automotive engine using thermoelectric generation technology The thermal contact between the exhaust stream, the heat exchanger surface, and the thermoelectric module’s hot face turns out to be just as important as the thermoelectric material itself. A clever module paired with a poor heat exchanger will disappoint.
There is also a subtlety that simple benchmarks miss. Modeling work has shown that the figure of merit alone is not enough to predict how well a thermoelectric system performs when it is pulling significant energy out of an exhaust stream through a sequence of leg pairs. Adding more thermoelectric elements beyond an optimum point can actually degrade performance, because the exhaust gas cools as it flows along the system and the temperature difference available to downstream modules shrinks.10Applied Energy. Theoretical limits of thermoelectric power generation from exhaust gases System-level engineering, in other words, is where TEGs succeed or fail in real vehicles.
Wearable Generators Powered by Body Heat
Your skin is typically around 33 °C while the surrounding air might be 22 °C. That modest difference of about 10 degrees is enough to generate small amounts of electricity with a wearable thermoelectric device, and that has researchers excited about self-powered health monitors, fitness trackers, and other low-power electronics that could run indefinitely without a battery change.11PubMed Central. Human body heat-driven thermoelectric generators as a sustainable power supply for wearable electronic devices: Recent advances, challenges, and future perspectives
The power output from body heat is tiny, often in the microwatt to low-milliwatt range, because the temperature difference is small and the thermal resistance of human skin limits how much heat actually flows through the device. Making these generators work means miniaturizing the thermoelectric elements and optimizing their geometry. Research on miniaturized TEGs has found that when thermoelectric legs are shorter than about 0.1 millimeters, their shape starts to matter as much as the material. Hollow leg geometries, for instance, proved more efficient than solid ones at that scale because they change how temperature distributes across the device.12PubMed Central. Geometry Optimization for Miniaturized Thermoelectric Generators
Leg geometry also matters at larger scales. Theoretical work on annular (ring-shaped) thermoelectric generators has shown that the optimal leg shape depends on the leg length and on the electrical contact resistance at the junctions, though in practice the performance difference from tweaking the shape is small once contact resistance is accounted for.13Energy Conversion and Management. Influence of leg geometry configuration and contact resistance on the performance of annular thermoelectric generators The takeaway: for body-heat generators, making things smaller is not just about fitting them on a wristband. The physics of heat flow changes at those dimensions, and designs that work at centimeter scale do not simply shrink down.
Pairing TEGs with Solar Panels
Solar panels lose efficiency as they heat up, which is an everyday problem in sunny climates. One solution is to attach thermoelectric modules to the back of the solar panel, using the panel’s excess heat as the hot side and ambient air or a heatsink as the cold side. This hybrid photovoltaic-thermoelectric approach both cools the solar cells, recovering some of their lost efficiency, and generates a small additional stream of electricity from the waste heat.14PubMed Central. Hybrid Photovoltaic/Thermoelectric Systems for Round-the-Clock Energy Harvesting
Reviews of these hybrid systems have found that they consistently outperform standalone solar panels across a range of environmental conditions.15Applied Energy. Advances and challenges in hybrid photovoltaic-thermoelectric systems for renewable energy The gains are not dramatic in percentage terms, but they come essentially for free once the hardware is installed, and the cooling effect on the solar cells can extend their lifespan. Concentrated solar systems, which focus sunlight with mirrors or lenses, create even larger temperature gradients and have shown particularly encouraging results when paired with TEGs, especially when phase-change materials are used to store heat and smooth out temperature fluctuations.16IOP Conference Series: Earth and Environmental Science. Concentrated Solar Thermal-Thermoelectric generator hybrid systems: Review on the Most Recent Developed technologies
TEGs Running in Reverse
The Seebeck effect has a mirror image called the Peltier effect: run current through a thermoelectric module instead of extracting it, and one side gets cold while the other gets hot. This makes thermoelectric devices inherently reversible. The same module that generates electricity from a temperature difference can, with an applied voltage, pump heat and serve as a solid-state cooler or heater. Portable coolers, temperature-controlled car seats, and precision scientific instruments already use this capability.
Performance in cooling mode is generally worse than in heating mode for the same temperature conditions. The internal resistance of the module generates heat that works against you when you are trying to cool but helps you when you are trying to heat, so the heating coefficient of performance is inherently higher.17Applied Energy. Thermoelectric cooling heating unit performance under real conditions This asymmetry is why thermoelectric cooling tends to be limited to small-scale applications where compactness, silence, and precise temperature control matter more than raw energy efficiency.
Reliability Under Thermal Cycling
A TEG might run for years with one side baking at hundreds of degrees while the other stays cool. That sustained temperature gradient creates mechanical stress inside the module. The materials expand at different rates, solder joints flex, and over time these stresses can cause creep, fatigue, and cracking. Understanding and managing this mechanical behavior is critical for long-term reliability, and it is an area where thermoelectric engineering has historically lagged behind the materials science of the thermoelectric compounds themselves.18International Materials Reviews. Mechanical behaviour of thermoelectric materials – a perspective
Bismuth telluride modules, the most common commercial type, are particularly vulnerable at the solder interfaces between the thermoelectric legs and the electrical contacts. Modeling work has simulated how cracks sprout and grow at these interfaces under repeated thermal cycling, using the energy dissipated in the viscoplastic solder layer to predict when a module will fail.19ACS Applied Materials & Interfaces. Interfacial Crack Growth-Based Fatigue Lifetime Prediction of Thermoelectric Modules under Thermal Cycling For applications like automotive exhaust recovery, where the engine cycles between idle and full throttle hundreds of times a day, this kind of fatigue analysis matters as much as the thermoelectric efficiency of the material inside.
Environmental and Economic Considerations
TEGs have an appealing environmental pitch: they recover energy from heat that would otherwise be wasted, they produce no emissions during operation, and they last a long time with zero maintenance. But a full life-cycle view complicates the picture. Many high-performance thermoelectric materials contain tellurium, bismuth, lead, or antimony, all of which raise toxicity and supply-chain concerns. Manufacturing thermoelectric modules is energy-intensive, and production costs remain high because the market has not yet scaled to the point where economies of mass production kick in.20Environmental Impact Assessment Review. Systematic review of life cycle assessment of thermoelectric materials and devices to identify knowledge gaps and sustainability perspectives
On the economics side, TEGs compete with other waste-heat-to-electricity technologies, most notably organic Rankine cycle systems, which use a working fluid and a turbine. Organic Rankine cycles achieve higher conversion efficiencies but bring moving parts, maintenance needs, and minimum viable scale. TEGs, by contrast, work at any scale from milliwatts to tens of kilowatts and require essentially no maintenance. That modularity and simplicity is their strongest commercial argument, even though the per-watt cost is higher and efficiency typically sits around 5 to 10 percent.21ScienceDirect. Comparison between thermoelectric generator and organic Rankine cycle for low to medium temperature heat source: A Techno-economic analysis
The Spin Seebeck Effect and Future Frontiers
Most thermoelectric research focuses on moving charge carriers through semiconductors, but an entirely different branch of physics has opened up a parallel path. The spin Seebeck effect generates a spin current, rather than a charge current, when a temperature gradient is applied to a magnetic material. That spin current can then be converted into a conventional electrical voltage in an adjacent metal layer. The field studying these interactions, called spin caloritronics, sits at the intersection of thermoelectrics and spintronics.22PubMed Central. Transport phenomena in spin caloritronics
The practical power levels from spin Seebeck devices are currently far too small for energy harvesting. But the physics is interesting because spin-based transport obeys different rules than charge-based transport, and it might eventually sidestep some of the trade-offs that limit conventional thermoelectric materials. A temperature gradient in a magnetic insulator, for example, can drive a spin current even though no charge moves through the insulator at all, and the resulting voltage appears in a thin metal film deposited on top.23physica status solidi (b). Thermal‐Spin Conversion: Mechanism, Materials, and Determinants of the Spin Seebeck Effect Whether this will ever compete with conventional TEGs as a power source is an open question, but it represents a genuinely different approach to turning heat into useful energy, and the research is moving quickly.