How to Learn Thermoelectricity and Thermoelectric Materials: From the Seebeck and Peltier Effects to zT, Cooling and Waste-Heat Power

Wait, What? The Best Electrical Conductor Is Usually a Bad Thermoelectric

Suppose you want heat at one end of a solid to generate useful voltage. Choosing the best electrical conductor sounds sensible—until that conductor also carries heat so efficiently that the temperature difference disappears.

Now choose a thermal insulator and the gradient survives, but charge may barely move.

a useful thermoelectric must conduct charge well while blocking heat as independently as possible

The One-Sentence Answer

Learn thermoelectricity by first understanding how a temperature gradient redistributes charge carriers, then connect the Seebeck and Peltier effects to electrical resistance and heat conduction before using zT to see why useful materials must transport charge efficiently while blocking heat as independently as possible.

Stage 1: Begin With a Temperature Gradient

Charge carriers at the hot side of a material have a broader energy distribution than those at the cold side. Their unequal diffusion can create an electric potential.

Stage 2: The Seebeck Effect Converts Temperature Difference Into Voltage

A common definition is S = -ΔV/ΔT. Its units are V/K, often μV/K in practical materials.

Stage 3: Seebeck Coefficient Is Not Power Output

A high Seebeck coefficient can create a large open-circuit voltage but poor power if electrical resistance is high.

Stage 4: Metals Usually Have Small Seebeck Coefficients

Metals contain abundant carriers and their transport changes only modestly with thermal gradient. Semiconductors allow far stronger tuning of carrier concentration and energy dependence.

Stage 5: Carrier Type Changes the Sign

In many conventional semiconductors, p-type transport gives positive S and n-type transport negative S. Thermoelectric modules often pair p-type and n-type legs so their voltages add.

Stage 6: The Peltier Effect Runs the Conversion the Other Way

Pass current through a junction of dissimilar thermoelectric materials and heat can be absorbed at one junction and released at another.

Stage 7: A Peltier Cooler Is a Solid-State Heat Pump

Electrical current moves heat from cold side to hot side. The hot side still needs a heat sink; the device does not make heat disappear.

Stage 8: Joule Heating Competes With Cooling

Electrical resistance produces P = I²R heat inside the thermoelectric legs. Increasing current too far can overwhelm useful Peltier cooling.

Stage 9: Fourier Heat Conduction Competes Too

Heat naturally leaks from hot side to cold side. High thermal conductivity therefore collapses the very gradient the thermoelectric effect needs.

Stage 10: The Thomson Effect Completes the Classical Trio

If current flows through a material that also has a temperature gradient, reversible heat can be absorbed or released along the conductor. Seebeck, Peltier and Thomson effects are thermodynamically related.

Stage 11: Kelvin Relations Connect Generator and Cooler Physics

Under appropriate assumptions, the Peltier coefficient and Seebeck coefficient satisfy Π = ST.

Stage 12: Electrical Conductivity Must Be High

Useful current requires high electrical conductivity σ. But increasing carrier concentration to improve σ usually reduces S.

Stage 13: Power Factor Combines Seebeck and Conductivity

PF = S²σ. Power factor rewards materials that preserve voltage response while allowing strong charge transport.

Stage 14: High Power Factor Is Still Not Enough

A high-power-factor material can conduct heat so strongly that little temperature difference remains. Thermal transport must be included.

Stage 15: zT Combines the Main Material Properties

zT = S²σT/κ, where κ is total thermal conductivity. Large zT is generally favourable.

Stage 16: zT Is Not Device Efficiency

Actual module performance also depends on hot/cold temperatures, contacts, geometry, interfaces and how properties vary with temperature.

Stage 17: Thermal Conductivity Has Electronic and Lattice Parts

A useful decomposition is κ = κe + κL. Electrons carry some heat; phonons carry lattice heat.

Stage 18: Electrons Carry Heat as Well as Charge

Increasing electrical conductivity often raises electronic thermal conductivity. Charge and heat transport cannot usually be tuned independently without limit.

Stage 19: Phonons Carry Lattice Heat

Lattice vibrations transport thermal energy. Defects, alloy disorder, interfaces and complex crystal structures can scatter phonons and reduce κL.

Stage 20: “Phonon Glass, Electron Crystal” Is a Design Ideal

The dream material lets electrons move as though through an ordered crystal while making phonons scatter as though through a glass.

Stage 21: Carrier Concentration Has an Optimum

Too few carriers gives high S but poor conductivity. Too many carriers gives high conductivity but reduced S and larger electronic heat transport.

Stage 22: The Pisarenko Relationship Exposes the Trade-Off

Plots of S versus carrier concentration help test whether a material follows simple semiconductor models or contains more unusual electronic structure.

Stage 23: Effective Mass Influences Seebeck Response

Large density-of-states effective mass can increase S but can also reduce mobility. Improvement in one electronic parameter can damage another.

Stage 24: Band Convergence Can Increase Useful Electronic States

Bringing several electronic valleys close in energy can increase the number of states contributing to transport without an equally severe mobility penalty.

Stage 25: Resonant Levels Can Reshape the Density of States

Selected dopants can create sharp electronic features near the Fermi level, potentially improving Seebeck response while also introducing scattering.

Stage 26: Alloying Scatters Phonons

Substituting atoms of different mass or size disrupts lattice vibrations and lowers κL, but excessive disorder can also reduce carrier mobility.

Stage 27: Nanostructuring Creates More Interfaces

Nanoscale grains and inclusions can scatter heat-carrying phonons more strongly than charge carriers under selected conditions.

Stage 28: Grain Boundaries Help and Hurt

Grain boundaries scatter phonons but can also scatter carriers, trap charge and weaken mechanical integrity.

Stage 29: Bismuth Telluride Dominates Near Room Temperature

Bi₂Te₃-based systems remain important for cooling and near-room-temperature generation because they combine strong electronic transport with relatively low lattice thermal conductivity.

Stage 30: Durability Matters as Much as Peak zT

A January 2026 Energy & Environmental Science study used nanotwin engineering in p-type bismuth telluride to improve thermal stability and retain device output across repeated cycling.

Stage 31: Lead Telluride Works at Higher Temperature

PbTe-based materials perform strongly at elevated temperatures, but toxicity and materials sustainability motivate alternatives.

Stage 32: Skutterudites Use Structural Cages

Skutterudites can host guest atoms whose local motion scatters phonons while electronic bands remain tunable. A February 2026 study combined filling, doping and defect engineering to improve performance.

Stage 33: Half-Heuslers Offer Mechanical Strength

Half-Heuslers attract interest for high-temperature systems because they can combine thermal stability and mechanical robustness with useful thermoelectric performance.

Stage 34: Mg₃Sb₂-Based Materials Expanded n-Type Options

Mg₃Sb₂ families use relatively abundant elements and can achieve strong performance, though grain boundaries and defect chemistry are crucial.

Stage 35: SnSe Shows Why Direction Matters

Tin selenide has strongly anisotropic bonding. Electrical and thermal transport vary by crystal direction, so one scalar property can hide important physics.

Stage 36: Oxides Trade Efficiency for Robustness

Oxide thermoelectrics can tolerate high temperatures and oxidising conditions even when zT is lower than leading chalcogenides.

Stage 37: Organic Thermoelectrics Add Flexibility

Conducting polymers and molecular systems can be lightweight, printable and flexible. Their advantage is often mechanical and manufacturing-related rather than record zT.

Stage 38: Wearables Work With Small Temperature Differences

Body-heat harvesting operates with modest ΔT. Useful power therefore depends strongly on area, thermal contact, module geometry and ambient conditions.

Stage 39: Generators Need a Maintained Heat Gradient

A thermoelectric generator cannot produce continuous power from one uniform temperature. It needs sustained heat flow from hot reservoir to cold reservoir.

Stage 40: Maximum Power and Maximum Efficiency Differ

The electrical load that gives the most watts is not necessarily the load that gives the highest conversion efficiency.

Stage 41: Contact Resistance Can Erase Material Gains

Electrical contacts add Joule loss and thermal interfaces change where the temperature drop occurs. A high-zT leg inside poor contacts can form a mediocre module.

Stage 42: Diffusion Barriers Protect Interfaces

At high temperature, atoms can migrate between thermoelectric material, solder and electrodes. Interface reactions change resistance and mechanical strength.

Stage 43: Segmented Legs Match Materials to Temperature Range

One material may perform best at the hot side and another at the cold side. Segmented legs exploit this, but every interface adds complexity.

Stage 44: Thermoelectric Coolers Have No Moving Mechanical Parts

This makes them attractive for electronics, optical detectors and compact temperature control, though their coefficient of performance is usually below efficient compressor refrigeration for large loads.

Stage 45: Waste Heat Is Attractive but Difficult

Industrial exhaust and engines contain recoverable heat, but module performance depends on heat exchangers, thermal cycling, cost and interfaces as much as bulk material zT.

Stage 46: Radioisotope Thermoelectric Generators Use Nuclear Heat

NASA RTGs use heat from radioactive decay and thermoelectric couples to convert part of that heat flow to electricity. The thermoelectric legs are not undergoing fission.

Stage 47: RTGs Trade Efficiency for Reliability

Deep-space missions value long life, predictability and no moving conversion parts. A device can be valuable with modest efficiency when reliability is the dominant receiver.

Stage 48: Thermoelectric Measurement Is Harder Than Plugging Numbers Into zT

Reliable zT needs independent measurements of S, σ, κ and temperature. Uncertainty in any one term propagates into the final figure of merit.

Stage 49: NIST Built Reference Standards for This Reason

NIST thermoelectric-property standards and Seebeck reference materials exist to improve interlaboratory comparability. Record performance is meaningful only when another laboratory can reproduce it.

Stage 50: Seebeck Measurement Needs Correct Temperature Difference

If temperature probes do not represent the actual voltage-contact positions, systematic error appears even when voltage is measured perfectly.

Stage 51: Thermal Conductivity Is Often the Hardest zT Input

Thermal conductivity may be measured directly or inferred from thermal diffusivity, density and heat capacity. Different methods can disagree.

Stage 52: Hall Measurements Add Carrier Mechanism

Hall-effect data estimate carrier concentration and mobility, helping explain why S and σ changed instead of merely reporting performance.

Stage 53: Temperature-Dependent Properties Matter

S, σ and κ vary across a temperature gradient. A record zT at one temperature is not the constant value of a whole working leg.

Stage 54: Professional Thermoelectric Science Is a Coupled-Transport Problem

Which electronic structure creates the observed Seebeck response, which carriers and phonons dominate electrical and thermal transport, and do independent measurements plus device-level cycling show that the zT improvement survives contacts, temperature gradients and real operating time?

Evidence: How Do We Know a Material Is Genuinely Better?

Strong evidence combines independently measured Seebeck coefficient, electrical conductivity and thermal conductivity with Hall data, structural characterisation, repeated thermal cycling, module tests and interlaboratory reproduction.

Misconceptions Worth Hunting

The Seebeck effect creates energy from temperature alone.
The best electrical conductor is automatically the best thermoelectric.
High Seebeck coefficient guarantees high power.
zT is the same as device efficiency.
Lower thermal conductivity is always better even if electrical transport collapses.
Peltier coolers destroy heat.
More current always gives more cooling.
A record material automatically makes a record module.
One zT measurement proves a breakthrough.

Transfer Check

Material A has twice the Seebeck coefficient but 100 times the electrical resistance. Is A automatically better? No.

A cooler pumps heat from the cold side but has no heat sink on the hot side. Will it remain cold indefinitely? No.

A sample has low lattice thermal conductivity but poor mobility. Does low κ guarantee high zT? No.

How We Know the Learning Has Held

A learner should be able to explain Seebeck, Peltier and Thomson effects; power factor; zT; electronic and lattice thermal conductivity; carrier-concentration optimisation; phonon scattering; major thermoelectric material families; module interfaces; cooling and generation; and why metrology is central to performance claims.

Model Limits

zT compresses spatially varying device physics. Wiedemann–Franz assumptions can fail in unusual materials. Nanostructuring alters both electrons and phonons. Contact and thermal-interface losses are not contained fully in bulk zT. Keep Seebeck response + carrier concentration + mobility + electronic heat + phonon heat + contact resistance + temperature range + measurement uncertainty visible.

Teaching Guide

Teach in this order: temperature gradient → carrier diffusion → Seebeck → Peltier → Joule heating → heat conduction → power factor → zT → electrons vs phonons → doping → band engineering → nanostructuring → material families → module → metrology.

Begin with: “Why can copper conduct electricity brilliantly yet be a poor thermoelectric material?”

Connect This to the eduKate Learning Estate

The Quiet Ending

The beginner asks, “How can heat make electricity?” The developing materials scientist asks, “Which carriers are creating this Seebeck voltage?” The advanced learner asks, “Did we improve electronic transport, reduce phonon transport, or merely move the trade-off?”

Which independently measured transport properties, interface losses and thermal-cycle tests show that the thermoelectric improvement survives the transition from a material number to a working device?