Electronics Guide

Articles in This Category

Heat Sources and Scales

Thermal harvesting spans an enormous range of scales, from millikelvin fluctuations in ambient air to the incandescent interiors of combustion chambers. The available temperature difference sets the ceiling on efficiency, and the available heat flux sets the ceiling on power. Characterizing both, over the full operating cycle rather than at a single convenient moment, is the first step in any design.

Body Heat

Skin sits a few degrees above room air, but skin is also a poor thermal conductor and the air film against it resists heat rejection, so only a fraction of that difference appears across a device worn on the body. Wearable thermoelectric generators consequently deliver single-digit microwatts per square centimeter when the wearer is seated indoors, rising to roughly ten to twenty microwatts per square centimeter during walking, when convection over the heat sink improves. Research prototypes with aggressive heat sinking or forced airflow have reported substantially more, but such conditions rarely persist in normal wear. Practical wearable systems therefore accumulate energy over long intervals and spend it in short bursts.

Machinery and Process Equipment

Motor housings, gearboxes, steam and condensate lines, boiler skins, and heat exchangers hold surfaces tens to hundreds of degrees above their surroundings for months at a time. A clamp-on or magnetically mounted thermoelectric module on such a surface typically yields milliwatts, which is ample for a duty-cycled wireless sensor node that wakes, measures vibration or temperature, transmits a short packet, and sleeps. The steadiness of these sources is their chief virtue: unlike vibration or indoor light, process heat is often continuous and predictable.

Combustion and Concentrated Solar

Burners, flares, kilns, and concentrated solar receivers reach well above one thousand degrees Celsius. At these temperatures radiative transfer dominates, which favors thermophotovoltaic conversion, and the large temperature difference lifts the Carnot ceiling high enough that meaningful fractions of the heat can be recovered. High-temperature thermoelectric materials such as silicon-germanium and skutterudites also operate in this range, though thermal stress, oxidation, and sublimation complicate long-term service.

Environmental and Geothermal Gradients

Soil, water bodies, and building structures store heat and release it with a lag, producing gradients between the surface and the subsurface and between the interior and exterior of a wall. These differences are small, often only a few kelvin, but they persist without human intervention and reverse predictably with the day-night cycle. Devices exploiting them favor large collection areas over high power density, and pyroelectric or phase-change approaches suit the cyclic reversal better than a fixed-polarity thermoelectric module.

Conversion Mechanisms

Thermoelectric Conversion

Thermoelectric generators rely on the Seebeck effect, in which a temperature gradient across a semiconductor produces a voltage as charge carriers diffuse from the hot side toward the cold side. A module pairs p-type and n-type legs connected electrically in series and thermally in parallel, so their contributions add. Because a single couple of bismuth telluride contributes only a few hundred microvolts per kelvin, modules stack many couples: a conventional module of roughly 127 couples produces on the order of fifty millivolts per kelvin of temperature difference. Even so, a wearable harvester operating across a few kelvin delivers only tens to hundreds of millivolts, which is why start-up voltage dominates the design of the accompanying electronics.

Module efficiency is modest. Commercial bismuth telluride generators convert roughly five percent of the heat passing through them when driven by a temperature difference of a couple hundred kelvin, and far less at small differences. The multi-mission radioisotope thermoelectric generators aboard the Curiosity and Perseverance rovers illustrate the practical ceiling for a mature high-temperature system: about 2,000 watts of decay heat from plutonium-238 yields roughly 110 watts of electricity at the start of the mission, an efficiency near six percent, declining by a few percent per year as the fuel decays and the couples degrade.

Pyroelectric Conversion

Pyroelectric materials generate charge in response to a change in temperature over time rather than to a static gradient. The current is proportional to the pyroelectric coefficient, the electrode area, and the rate of temperature change, so a pyroelectric element produces nothing at thermal equilibrium no matter how hot it is. This makes the mechanism a natural match for cyclic environments: reciprocating machinery, intermittent burners, chopped infrared radiation, and day-night thermal swings.

Simply connecting a resistor across a pyroelectric element extracts very little of the available energy. Thermodynamic cycles recover far more by coordinating heating and cooling with applied electric fields. In the Olsen cycle, the element is charged at low temperature under a high field, heated while polarized, discharged at high temperature, and cooled again, tracing a loop in the electric-displacement-versus-field plane whose area is the harvested energy. Laboratory demonstrations on relaxor ferroelectrics such as lead magnesium niobate-lead titanate have reported roughly one hundred millijoules per cubic centimeter per cycle, and multilayer lead scandium tantalate capacitors driven across large field and temperature spans have reached several joules per cubic centimeter. Cycle rate, not energy per cycle, usually limits the average power, because heating and cooling a solid element takes time.

Thermophotovoltaic Conversion

Thermophotovoltaic systems interpose radiation between the heat source and the converter. A hot emitter radiates infrared light onto photovoltaic cells whose bandgaps are matched to the emitter spectrum, and a highly reflective back-surface reflector returns sub-bandgap photons to the emitter instead of wasting them as heat in the cell. This photon recycling, rather than any improvement in the cell itself, accounts for much of the recent progress. A two-junction III-V device with 1.4 and 1.2 electron-volt junctions reached an efficiency of 41.1 percent at an emitter temperature of 2,400 degrees Celsius and a power density of about 2.4 watts per square centimeter, with a companion 1.2 and 1.0 electron-volt device reaching 39.3 percent near 2,127 degrees Celsius. Those figures place solid-state thermophotovoltaic converters in the same efficiency class as turbine-based heat engines, which is what makes them interesting for high-temperature thermal storage on the grid.

The trade-off is temperature. Thermophotovoltaic conversion needs an emitter far hotter than most waste-heat streams, so it belongs to combustion, nuclear, and concentrated-solar applications rather than to the low-grade heat that thermoelectric harvesters address.

Thermionic Conversion

A fourth direct mechanism, thermionic conversion, has no subcategory of its own here because its temperature range places it outside ordinary harvesting practice. A thermionic converter emits electrons from a hot cathode across a narrow gap to a cooler collector, producing current with no moving parts, but appreciable emission requires emitter temperatures well above one thousand degrees Celsius. Space charge accumulating in the gap repels further emission; suppressing it calls for either cesium vapor, which also lowers the emitter work function, or gaps narrowed to micrometer scale. Those constraints confine the method to combustion chambers, concentrated solar receivers, and nuclear sources, and High-Temperature Energy Harvesting treats it alongside the other converters that survive such sources.

Materials and Figures of Merit

Thermoelectric materials are ranked by the dimensionless figure of merit ZT, which combines the Seebeck coefficient, electrical conductivity, absolute temperature, and thermal conductivity. Raising ZT is difficult because the three transport properties are coupled: doping a material to raise its electrical conductivity generally lowers its Seebeck coefficient and raises its electronic thermal conductivity. Most progress over the past two decades has come from suppressing lattice thermal conductivity through nanostructuring, alloy scattering, and complex crystal structures, rather than from improving the electronic properties.

Each material family occupies a temperature window. Bismuth telluride alloys, with ZT near one, dominate near room temperature and account for essentially all commercial modules. Lead telluride, skutterudites, and half-Heusler compounds serve the mid-range of roughly 500 to 800 kelvin, the regime of exhaust and process heat. Silicon-germanium alloys tolerate temperatures above 1,000 kelvin and have flown on deep-space radioisotope generators for decades. Laboratory materials reach higher: single-crystal tin selenide has been reported with ZT above two near 900 kelvin, though brittleness, anisotropy, and contact metallization stand between such results and manufacturable modules.

Thermophotovoltaic and pyroelectric devices are judged differently. Thermophotovoltaic cells are selected for bandgap match to the emitter temperature and for sub-bandgap reflectance, since a reflector that returns 90 percent of unusable photons wastes ten times more than one that returns 99 percent. Pyroelectric materials are selected for a large pyroelectric coefficient, a high breakdown field, low dielectric loss, and a Curie temperature placed just above the operating range, where the coefficient peaks.

Thermal System Design

The converter is one resistance in a series thermal circuit that runs from the heat source, through interfaces and spreaders, across the device, and out to ambient. Every other resistance in that chain steals part of the temperature difference. In low-grade harvesting the external resistances usually dominate, and the device sees far less than the nominal source-to-ambient difference.

Thermal Resistance Matching

Maximum electrical power is obtained when the thermal resistance of the converter roughly matches the total external thermal resistance, an analogue of electrical impedance matching. A module that is too conductive short-circuits the thermal path and collapses its own temperature difference; a module that is too resistive holds a large difference but passes almost no heat. Designers adjust leg length, leg cross-section, and fill factor to place the module near this optimum for the intended heat sink, which is why a module optimized for a finned sink in still air performs poorly on the same source under forced convection.

Heat Collection and Rejection

On the hot side, conformal mounting, clamping pressure, and spreaders reduce the resistance from an irregular surface into the module. On the cold side, heat sinks are usually the largest and heaviest part of the assembly, and their size, not the module's, sets the practical power ceiling in wearable and clamp-on products. Thermal interface materials matter at both boundaries: a poorly applied grease layer or an air gap of a fraction of a millimeter can consume more of the temperature difference than the semiconductor itself.

Transients and Duty Cycling

Sources that switch on and off create transients that a steady-state analysis misses. Thermal mass on the hot side smooths short interruptions and keeps a generator producing through them, while thermal mass on the cold side delays saturation of the heat sink. Pyroelectric harvesters invert this logic and want low thermal mass, because they depend on rapid temperature change. Phase-change materials placed on one side of a thermoelectric module exploit the same idea from the other direction, holding one face near a fixed melting point while the other follows the environment, so that an ambient swing yields a usable difference.

Power Conditioning

Thermal harvesters produce low voltage rather than low current, which makes the interface electronics unusually demanding. A wearable generator may present only twenty to two hundred millivolts, well below the threshold at which ordinary switching regulators start.

Specialized converters solve the cold-start problem with transformer-coupled or charge-pump front ends: a widely used converter starts from inputs as low as twenty millivolts when paired with a step-up transformer of about one-to-one-hundred turns ratio, then bootstraps a conventional switching stage once its output rail is alive. Other harvesting controllers cold-start in the hundreds of millivolts and continue operating well below that threshold once running. Conversion efficiency at these input levels is poor, often only a fraction of what the same converter achieves at a volt, so the front-end choice can matter more than the module choice.

Because a thermoelectric module's optimal operating point sits near half its open-circuit voltage, maximum power point tracking for thermal sources is simpler than for photovoltaics: a fractional open-circuit-voltage method, sampling the unloaded voltage periodically, tracks the optimum closely with negligible overhead. Harvested energy is then buffered in a supercapacitor or a small rechargeable cell, and the load runs on a duty cycle set by the accumulation rate. Designing that duty cycle honestly, against the worst-case temperature difference rather than the nominal one, is the difference between a system that survives a cold night and one that resets.

Applications

Thermal harvesting enables autonomous, maintenance-free operation wherever a temperature difference persists. Its appeal is rarely raw energy cost; it is the elimination of battery replacement in places where access is expensive, hazardous, or impossible.

Wearable and Medical Devices

Body-heat generators supplement or replace batteries in fitness monitors, hearing aids, and continuous physiological sensors, and implantable versions draw on the small but reliable difference between deep tissue and surrounding structures. The dominant constraints are comfort and heat rejection: a device thin and flexible enough to wear cannot carry a large heat sink, so achievable power stays in the microwatt range and the electronics must be designed around it.

Industrial Condition Monitoring

Surface-mounted thermoelectric harvesters run vibration, temperature, and corrosion sensors on motors, pumps, boilers, steam traps, and pipelines. A wireless node powered this way avoids both the wiring cost of a powered installation and the recurring labor of battery changes across hundreds of monitoring points, which is often the decisive economic argument in a plant.

Transportation and Waste Heat Recovery

Automotive programs have demonstrated exhaust and exhaust-gas-recirculation generators producing on the order of a few hundred watts in test vehicles. None reached volume production: the added mass, exhaust backpressure, cost of tellurium-bearing materials, and durability under thermal cycling outweighed the fuel savings, and the shift toward electrified powertrains reduced the quantity of exhaust heat available. Stationary industrial recovery faces gentler constraints and remains the more plausible near-term market for mid-temperature thermoelectrics.

Space and Remote Power

Radioisotope thermoelectric generators have powered outer-planet probes and Mars rovers for decades, valued not for efficiency but for producing power continuously for many years with no moving parts and no dependence on sunlight. On Earth, the same reasoning applies to remote pipeline cathodic protection, navigation aids, and instrumentation on sites where a maintenance visit costs more than the hardware.

Self-Powered Sensing

Pyroelectric elements serve simultaneously as sensor and power source. A passive infrared detector that generates its own signal charge needs no bias supply, and the same principle extends to flame detection, thermal-motion sensing, and infrared imaging, where a self-powered element simplifies the front end and reduces standby consumption to near zero.

Limitations and Trade-offs

Honest expectations matter more in thermal harvesting than in most fields, because the Carnot ceiling is unforgiving at small temperature differences and because published device results are often measured under conditions that a deployed system will never see. Three limits recur.

The first is thermodynamic. Low-grade heat is abundant precisely because it is close to ambient, and closeness to ambient is exactly what caps the efficiency. No material advance changes this; it only closes the gap to a ceiling that remains low.

The second is thermal-mechanical. Modules spend their service lives under thermal cycling, and differential expansion fatigues solder joints, metallizations, and interface layers. Sublimation, oxidation, and diffusion degrade high-temperature materials further. Lifetime under cycling, not initial efficiency, usually decides whether an installation succeeds.

The third is economic and material. Tellurium is scarce, bismuth telluride modules are costly per watt, and lead-bearing mid-temperature materials face regulatory pressure. These constraints push thermal harvesting toward applications where the value lies in eliminating maintenance rather than in generating cheap energy, and they motivate research into abundant-element alternatives such as magnesium silicide, higher manganese silicides, and organic and hybrid thermoelectrics.

Future Directions

Several lines of work are changing the practical picture. Flexible and printed thermoelectric materials, including conducting polymers and composite pastes, trade peak performance for conformability and low-cost fabrication over large areas, which suits wearable and building-surface deployment. Nanostructured and hierarchically engineered bulk materials continue to lower lattice thermal conductivity without destroying carrier mobility. Near-field thermophotovoltaics, in which emitter and cell are separated by sub-wavelength gaps, promises radiative transfer far above the blackbody limit, though maintaining nanometer-scale gaps in service remains unsolved.

System-level integration is advancing in parallel. Hybrid devices that combine pyroelectric and piezoelectric or thermoelectric and photovoltaic conversion extract more from environments that offer several ambient sources at once. Receiver electronics continue to lower their start-up thresholds and quiescent currents, and microcontrollers and radios continue to lower their energy per operation, so the same harvested microwatts accomplish more each product generation. In many applications the load, rather than the harvester, has been the limiting factor, and progress on that side has quietly widened the range of viable thermal harvesting deployments.

About This Category

This category examines the physics, materials, and engineering principles that make practical thermal energy conversion possible. The subcategories above address thermoelectric, pyroelectric, and thermophotovoltaic approaches in turn, together providing the foundation for selecting a technology, sizing a converter to its heat source, and designing the thermal and electrical systems that turn waste heat into reliable power.

Related Topics