Thermoelectric Energy Conversion

Converts temperature differences directly to electricity via the Seebeck effect in semiconductor junctions, enabling solid-state waste heat recovery and autonomous power for sensors and remote applications.

Fundamental Principle

Natural Asymmetry Exploited

Ultimate Source

Key Physics

The Seebeck effect generates a voltage proportional to the temperature difference:

V = S \cdot \Delta T

Where V is the open-circuit voltage (V), S is the Seebeck coefficient (V/K or μV/K), and ΔT is the temperature difference between hot and cold junctions (K).

The Seebeck coefficient varies by material, typically ranging from a few μV/K for metals to several hundred μV/K for optimized semiconductors. For practical power generation, many thermocouple pairs are connected electrically in series and thermally in parallel.

The efficiency of thermoelectric conversion depends on material properties captured in the dimensionless figure of merit ZT:

ZT = \frac{S^2 \sigma T}{\kappa}

Where S is the Seebeck coefficient, σ is electrical conductivity (S/m), κ is thermal conductivity (W/mK), and T is absolute temperature (K).

The maximum theoretical efficiency is given by:

\eta_{max} = \eta_{Carnot} \cdot \frac{\sqrt{1 + ZT} - 1}{\sqrt{1 + ZT} + T_c/T_h}

Where η_Carnot = (T_h - T_c)/T_h is the Carnot efficiency, T_h and T_c are hot and cold junction temperatures.

This equation reveals that thermoelectric efficiency is fundamentally limited by Carnot efficiency (like all heat engines) and further reduced by a factor depending on ZT. With current materials (ZT ≈ 1), TEGs achieve only about 1/6 of Carnot efficiency. Materials with ZT ≈ 3 would approach 1/3 of Carnot efficiency.

Conversion Mechanism

Energy Capture and Conversion

Physical Processes

1. Heat Input at Hot Junction

Heat enters the thermoelectric element at the hot side, raising the temperature and thermal energy of charge carriers.

2. Carrier Diffusion

Electrons (n-type material) or holes (p-type material) diffuse from the hot region toward the cold region, carrying both charge and thermal energy. This is the Seebeck effect in action.

3. Voltage Generation

The accumulation of charge carriers at the cold end creates an electric field opposing further diffusion. At equilibrium, a voltage (Seebeck voltage) develops across the element proportional to the temperature difference.

4. Current Flow

When connected to an external load, current flows through the circuit, delivering electrical power while simultaneously transporting heat (Peltier effect) and generating Joule heating.

5. Heat Rejection at Cold Junction

Heat exits at the cold junction through the Peltier effect (heat carried by current) and ordinary thermal conduction. Maintaining a large temperature difference requires effective heat sinking.

Competing Effects

Three interrelated phenomena determine TEG performance:

Seebeck effect (beneficial): Generates voltage from temperature difference.

Joule heating (detrimental): I²R losses in the thermoelectric material and contacts reduce efficiency.

Thermal conduction (detrimental): Heat flowing directly through the material (without generating electricity) reduces the temperature gradient and wastes thermal energy.

These effects are coupled through the same material properties, creating fundamental tradeoffs. Good electrical conductors typically have high thermal conductivity. High Seebeck coefficients often correlate with low electrical conductivity. Optimizing ZT requires decoupling these properties, which is the central challenge in thermoelectric materials research.

Module Construction

A practical TEG module consists of:

Multiple p-type and n-type thermoelectric legs (typically Bi₂Te₃ at low temperatures, PbTe at medium, SiGe at high)
Metal interconnects connecting legs electrically in series
Electrically insulating but thermally conducting ceramic plates on hot and cold sides
External leads for power output

Typical commercial modules contain 100-200 thermocouple pairs in a few square centimeters, producing several watts at temperature differences of 100-200°C.

Theoretical Limits

Origin of the Limit

Interdependence of properties:

The Wiedemann-Franz law states that electrical and thermal conductivity are proportional in metals: κ_electronic = LσT, where L is the Lorenz number. This means improving electrical conductivity automatically increases thermal conductivity, limiting ZT gains.

Semiconductors partially circumvent this by having significant lattice (phonon) thermal conductivity separate from electronic conductivity. The strategy for improving ZT is to reduce lattice thermal conductivity while maintaining high electrical conductivity and Seebeck coefficient.

Phonon-glass, electron-crystal paradox:

The ideal thermoelectric would conduct electricity like a crystal (high σ) but conduct heat like a glass (low κ). Achieving this requires nanostructuring, complex crystal structures, or other approaches that scatter phonons more effectively than electrons.

ZT vs. Efficiency

ZT	Efficiency (ΔT = 200K, T_h = 500K)	Fraction of Carnot
0.5	~3-4%	~10%
1.0	~6-7%	~17%
1.5	~9-10%	~25%
2.0	~12-13%	~32%
3.0	~17-18%	~45%
4.0	~21-22%	~55%

Commercial materials (Bi₂Te₃, PbTe) have ZT ≈ 0.8-1.0. Laboratory demonstrations have achieved ZT ≈ 2-2.5 in nanostructured materials. Reports of ZT > 3 exist but have not translated to practical devices.

Practical Performance

Application	Temperature Difference	Typical Efficiency	Notes
Body heat harvesting	5-15°C	0.1-0.5%	Very low ΔT limits efficiency
Low-grade waste heat	50-100°C	2-4%	Industrial exhaust, data centers
Automotive exhaust	200-400°C	4-8%	Exhaust manifold recovery
Industrial high-temp	300-600°C	5-10%	Furnaces, kilns
RTG (radioisotope)	500-1000°C	6-7%	Space missions

Practical Limitations

Material Constraints

The three key material properties (Seebeck coefficient, electrical conductivity, thermal conductivity) are interdependent through fundamental physics:

Increasing carrier concentration improves σ but reduces S
High σ materials have high electronic thermal conductivity
Reducing lattice thermal conductivity often introduces defects that reduce σ

This "thermoelectric trade-off" limits ZT to approximately 1 for bulk materials using conventional approaches. Breaking through this limit requires nanostructuring, band engineering, or fundamentally new materials.

High-performance thermoelectric materials are expensive:

Bismuth telluride (Bi₂Te₃): $500-1000/kg for high-purity material
Lead telluride (PbTe): Contains toxic lead; tellurium is scarce
Skutterudites: Require rare earth elements for optimal performance
Half-Heuslers: Complex ternary compounds requiring precise stoichiometry

Tellurium scarcity is a significant concern. Global tellurium production is only ~640-1,180 tonnes/year (USGS 2024 reports ~640 t; other trackers place recent output above 1,000 t), primarily as a byproduct of copper refining. Massive TEG deployment would strain this supply.

Several high-performance thermoelectric materials pose environmental and health concerns:

Lead telluride: Lead toxicity limits applications, especially wearables
Bismuth telluride: Contains toxic tellurium
Antimony compounds: Antimony toxicity

Research into "green" thermoelectrics (silicides, oxides, organic materials) aims to address these concerns but typically achieves lower ZT.

Efficiency Limitations

The fundamental limitation of thermoelectric generators is their low efficiency compared to other energy conversion technologies:

TEGs: 5-8% typical, 15% maximum demonstrated
Steam turbines: 40-45%
Gas turbines: 35-40%
Internal combustion engines: 25-35%
Photovoltaics: 15-25%

This low efficiency means TEGs are generally not competitive for primary power generation. Their value lies in applications where:

Waste heat is otherwise unused (zero fuel cost)
Simplicity and reliability outweigh efficiency (space, remote sensing)
Small scale and solid-state operation are essential (wearables, IoT)
No moving parts are required (maintenance-free operation)

Thermal Management

TEG efficiency depends critically on maintaining a large temperature difference. This requires effective heat exchangers on both hot and cold sides:

Hot side challenges:

Thermal contact resistance between heat source and TEG
Temperature gradients through interface materials
Thermal expansion mismatch causing mechanical stress

Cold side challenges:

Heat rejection to ambient (air, water, radiation)
Parasitic heat loads reducing ΔT
Condensation in humid environments

In many applications, the cost and complexity of heat exchangers exceed the TEG modules themselves. System efficiency (including pumping/fan power) is typically 30-50% lower than module efficiency.

Temporal Characteristics

Degradation and Lifetime

High-temperature degradation:

Sublimation of volatile components (Te, Sb)
Oxidation of thermoelectric materials
Interdiffusion at contacts
Thermal fatigue from cycling

Contact degradation:

Intermetallic compound formation
Increased contact resistance
Mechanical failure at interfaces

Typical lifetimes:

Automotive (high ΔT, cycling): 5-10 years
Industrial (steady high T): 10-15 years
Low-temperature (wearables, IoT): 15-20 years
Space (RTGs): 20-40+ years (Voyager RTGs still operating after 48+ years)

Scaling Characteristics

Output Scaling Behavior

Viable Scale Range

Power Level	Application	Module Count	Typical ΔT
1-100 μW	Wearable sensors	1 miniature	5-15°C
1-100 mW	Wireless sensors, IoT	1-5 modules	20-50°C
1-10 W	Portable power, camping	10-50 modules	100-200°C
100-1000 W	Small-scale waste heat	50-200 modules	150-300°C
1-10 kW	Automotive, industrial	200-1000 modules	200-400°C
10-100 kW	Large industrial	1000+ modules	300-600°C

The largest demonstrated systems are in the tens of kilowatts range. Scaling to MW would require thousands of modules and extensive heat exchanger infrastructure.

Resource Potential

The global waste heat resource is enormous. Approximately 60-70% of all primary energy consumed becomes waste heat at temperatures ranging from slightly above ambient to over 1000°C.

Waste heat inventory (approximate global):

Industrial processes: ~15-20 EJ/year
Transportation: ~15-20 EJ/year
Power generation: ~30-40 EJ/year
Buildings: ~10-15 EJ/year

At 5% conversion efficiency, capturing just 10% of industrial waste heat could generate ~100 TWh/year of electricity.

Practical constraints:

Much waste heat is low-grade (<100°C), where TEG efficiency is very poor
Distributed sources (millions of vehicles) are difficult to aggregate
Retrofit costs often exceed energy value
Competing technologies (ORC, heat pumps) may be more efficient

Best applications:

High-temperature industrial exhaust where ORC is impractical
Remote/off-grid locations where maintenance-free operation is valuable
Small-scale distributed generation where turbines are impractical
Space applications where no alternatives exist

Cost Considerations

Application	System Cost	Cost per Watt	Payback Period
Commercial TEG module	$5-20$	$5 - 20∣$ 5-15/W	N/A (component)
Waste heat system (industrial)	$10,000-100,000$	$10, 000 - 100, 000∣$ 10-50/W	3-10 years
Automotive TEG	$500-2000$	$500 - 2000∣$ 50-200/W	Typically not economic
Space RTG (MMRTG)	$109 million$	~ $109 mi ll i o n ∣ ⟨ T I L D E ⟩$ 900,000/W	N/A (mission critical)
Wearable TEG	$10-50$	$10 - 50∣$ 100-500/W	N/A (enabling technology)

The high cost per watt compared to other generation technologies ( $1-3/W for solar PV,$ 1-2/W for wind) limits TEG deployment to niche applications where their unique characteristics (solid-state, no maintenance, small scale, works in dark/vacuum) justify the premium.

Current Status

Technology Readiness Level

Application	TRL	Status
Commercial Bi₂Te₃ modules	9	Mature commercial product
Automotive waste heat	6-7	Demonstrated prototypes, limited production
Industrial waste heat	7-8	Commercial systems available
Wearable TEGs	5-6	Research/early commercial
High-ZT materials (lab)	3-5	Research demonstrations
Flexible TEGs	4-5	Research prototypes
Space RTGs	9	Flight-proven, decades of operation

Market Size and Growth

The global thermoelectric generator market is valued at approximately $1 billion (2024-2025) and projected to reach$ 1.4-1.7 billion by 2030, growing at 6-10% CAGR.

Market segments (2024):

Waste heat recovery: ~44%
Automotive: ~29%
Energy harvesting (IoT, wearables): ~22%
Other (aerospace, military): ~5%

Key growth drivers:

Increasing focus on energy efficiency and waste heat recovery
Growth of IoT requiring autonomous power sources
Automotive fuel efficiency regulations
Expansion of data centers with cooling requirements

Major Players and Products

Material/Module manufacturers:

Ferrotec (Japan/US)
Laird Thermal Systems (US)
II-VI Marlow (US)
Kryotherm (Russia)
Thermonamic (China)

System integrators:

Gentherm (automotive thermal management)
Alphabet Energy (industrial waste heat, now defunct)
European Thermodynamics (UK)

Space systems:

Aerojet Rocketdyne/Teledyne (RTGs for NASA)
DOE/INL (fuel and assembly)

Recent Demonstrations

Automotive:

BMW, Ford, GM have all demonstrated TEG exhaust recovery systems producing 200-1000 W
Fuel efficiency improvements of 2-5% demonstrated
Cost and complexity have prevented mass production

Industrial:

15 kW system demonstrated at 80°C temperature difference from coal bed methane plant (2024)
Multiple commercial installations in steel, glass, and cement industries

Wearables:

Flexible Bi₂Te₃ fiber TEGs demonstrated (KAIST, 2024)
Body-powered smartwatch prototypes
Estimated 22-25% of IoT sensors may use TEGs by 2030

Research Frontiers

High-ZT materials:

SnSe single crystals: ZT = 2.6 at 923 K (laboratory)
PbTe nanostructures: ZT = 2.2 at 915 K
Mg₃Sb₂-based materials: ZT > 1.5, tellurium-free
Organic thermoelectrics: ZT = 0.3-0.5, flexible and low-cost

Device innovations:

Segmented legs using different materials for different temperature ranges
Flexible and printed thermoelectric devices
Micro-TEGs for MEMS integration
Hybrid systems combining TEG with thermophotovoltaics

Manufacturing:

Screen printing of thermoelectric materials
Additive manufacturing of complex geometries
Roll-to-roll processing for flexible TEGs

Summary

Key Specifications

Parameter	Low-Temp (Bi₂Te₃)	Mid-Temp (PbTe)	High-Temp (SiGe)
Operating range	200-450 K	450-850 K	850-1300 K
Peak ZT	1.0-1.2	1.5-2.0	0.8-1.0
Typical efficiency	4-6%	6-10%	5-7%
Cost ( $/W)$	$/ W) ∣$ 5-15	$10-30$	$10 - 30∣$ 50-200
Primary applications	Waste heat, wearables	Industrial, automotive	Space, high-temp industrial

Strengths and Limitations

Strengths:

Solid-state: no moving parts, silent, vibration-free
Highly reliable: decades of operation possible (Voyager RTGs)
Scalable: microwatts to kilowatts with same technology
Maintenance-free: no fluids, filters, or mechanical wear
Direct conversion: no intermediate steps
Works in any environment: vacuum, atmosphere, underwater
Instant response: no startup time
Bidirectional: same device can cool (Peltier) or generate power (Seebeck)

Limitations:

Low efficiency: 5-10% typical, fundamentally limited by ZT
High cost per watt: $5-50/W for modules, higher for systems
Material constraints: tellurium scarcity, lead toxicity
Requires temperature difference: no standalone operation
Heat exchanger dependency: system efficiency limited by thermal management
Degradation at high temperatures: limited lifetime in harsh conditions
Poor economics: rarely competitive with ORC, heat pumps, or direct use of heat