Design Considerations for Thermal Energy Harvesting

Article By : Maurizio Di Paolo Emilio

Thermal energy harvesting can provide an autonomous and renewable energy source for a wide range of sensors and electronic devices, enabling them to produce energy from temperature differences.

Energy harvesting refers to the ability to collect from the environment, or from the system itself, the energy needed to power electronics. More specifically, thermal energy harvesting converts the thermal energy collected from a heat source into electricity.

The benefits of thermal energy harvesting include:

• The possibility of eliminating the battery. This advantage is particularly significant for portable devices and for low-power applications.

• The possibility of creating self-sufficient IoT devices from an energy point of view. This element is crucial in the development of standalone and mobile devices that can operate continuously without the need for battery recharging. Thermal energy harvesting enables smart sensing to be applied to remote or difficult-to-reach sites inside metropolitan infrastructure by lowering the need for maintenance and battery replacement.

• The chance to create new wearable solutions for medical and consumer applications

• The development of green energy technology. This will lead to decreased fossil-fuel utilization and greenhouse gas emissions.


This technology could be used to provide an autonomous and renewable energy source for a wide range of sensors and electronic devices, enabling them to produce energy from temperature differences. The introduction of increasingly efficient devices could pave the way for new solutions that take full advantage of thermal energy harvesting.

An interesting method for using thermal energy-harvesting technologies in wearable systems involves creating small electric currents that harness thermal energy as the difference in two temperatures—that of the body and that of the surrounding environment. There are temperature disparities everywhere, in both natural and artificial settings. Thermoelectric energy can be produced by taking advantage of these temperature differences, or gradients.

Thermal energy

The total energy of a system is conserved according to physical rules, with the possibility of change from one form to another. It is possible to get energy from a variety of environmental energy sources.

The environment around us is filled with variations in temperature and heat movement. The heat produced by engine waste, geothermal heat from the soil, heat from cooling water in steelworks, and other industrial operations are typical examples. Using a thermoelectric generator (TEG) and some electronics, thermal energy can be transformed into electrical energy, which can then be saved in a storage device. The TEG works on the basic tenet that heat flux (temperature difference) can be converted into electrical energy. It is perfect for low-power embedded devices because it is typically very small and has no moving parts (solid-state).

Seebeck effect

The Seebeck effect is the process by which an electrical voltage is produced when a temperature gradient exists between two sides of a material. The basic element of a TEG is the p-n junction, which is made up of a single structure of thermoelectric material P and N, each connected electrically in series and doped with impurities like boron (P) and phosphorus (N).

Figure 1: A TEG is basically a Peltier cell with two surfaces: hot (h) and cold (c). (Source: Power Electronics News)

The basic building blocks of a TEG module are several p-n pairs connected in series. The p-n pairs are arranged in parallel in this configuration to produce a voltage proportional to the temperature gradient. The device’s hot (Th) and cold (Tc) sides must be at different temperatures for the process to be effective, and the thermoelectric material’s performance, as measured by the thermoelectric figure of merit ZT, is given by:

where S is the Seebeck coefficient, ρ is the electrical resistivity, and λ is the thermal conductivity, while T is the temperature at which the thermoelectric properties are measured. ZT measures the amount of electrical energy that can be produced at a given temperature gradient: the higher the ZT value of a material, the better its thermoelectric properties are. By increasing the power factor PF = S2 ÷ ρ, or reducing the thermal conductivity λ = λe + λph (λe and λph indicate the electronic and phononic contributions, respectively), the thermoelectric performance of a given material can be improved.

The Seebeck coefficient, electrical resistivity, and thermal conductivity are the three factors that determine how effective this thermal process is. These three distinct physical characteristics, which together make up the figure of excellence, are interdependent. Therefore, it is difficult or impossible to improve one without impairing the other. The only quantity that may be adjusted freely without having an impact on the others is λph(T). Therefore, reducing size is the most promising strategy to boost overall efficiency.


Battery-based solutions are getting much more effective and smaller every day. For some low-power applications, such as IoT sensors, it is not possible to further improve battery life. Hence, those devices will greatly benefit from energy-harvesting technology. Interest in energy harvesting has sparked the development of complementary technologies including ultra-low–power (picowatt) microelectronics and supercondensators.

An excellent thermoelectric material must have a strong Seebeck effect, should conduct electricity as well as possible, and should transport heat as poorly as possible. It is quite difficult to find a material that matches all of these requirements, as electrical conductivity and thermal conductivity usually go hand in hand.

Researchers have recently succeeded in developing a novel material with a ZT value between 5 and 6. Composed of a thin layer of iron, vanadium, tungsten, and aluminum applied to a silicon crystal, this new material could potentially revolutionize the sensor power supply industry, enabling sensors to generate their own power from environmental sources.

Depending on the available temperature gradients, TEGs can generate from 20 µW to 10 mW per square centimeter.

Design hints

On the market, there are several ICs suitable for thermal energy harvesting, including the BQ25570 from Texas Instruments, capable of extracting microwatts to milliwatts from TEGs, as well as e-peas’s AEM10941 and other ICs by Analog Devices Inc. and Renesas Electronics. The BQ25570 integrates a power management system that boosts the voltages by using double circuits while preventing the battery from overcharging or exploding. Harvested energy can be stored to rechargeable Li-ion batteries, thin-film batteries, supercapacitors, or conventional capacitors.

Supercapacitors are the technological prerequisite for energy harvesting to be effectively applied. They are capacitors characterized by very high capacity that simultaneously have the functional characteristics of electrolytic capacitors and rechargeable batteries. However, they can store 10× to 100× more energy per unit volume or mass than an electrolytic capacitor, accumulate electric charges with a speed much higher than that typical of rechargeable batteries, and go through many more charge-discharge cycles than rechargeable batteries.

The process begins when the TEG’s plates have a sufficient temperature difference between them to produce a voltage on their terminals. The BQ25570, which includes a boost charger and a nanopower buck converter (Figure 2), extracts the power, which will vary in size according to the temperature differential, from microwatts to milliwatts. The output voltage is then increased to 3.3 V, with 93% efficiency, thanks to the inbuilt boost converter.

Figure 2: Block diagram of the BQ25570 ultra-low–power harvester PMIC (Source: Texas Instruments)

There are two ways to store incoming power for energy harvesting: Use either capacitors or a battery to hold that charge. When using a conventional capacitor or a supercapacitor, there are some guidelines that help the designer in the selection:

  • Select a capacitator with low ESR (<200 mΩ).
  • Leakage current must be less than 1 μA at 1.2 V.
  • Large capacitors charge slowly but can store high amounts of current. On the other hand, small capacitors charge very fast and increase the startup time.

Depending on the application, the capacitor value can be obtained with the following formula:

C = 15 × VOUT × IOUT × TON

where VOUT is the output voltage of the energy-harvesting sensorIOUT is the average output current from the energy-harvesting sensor, and TON is the IC turn-on time.

If the sensor is not able to provide enough power, the storage capacitors will maintain the system for a certain time.

The power conditioning of a thermoelectric energy harvester is also very important. Even at maximum power operation, the output voltage of a generator is small due to its low voltage. When the energy harvester is recharging a battery, the power-conditioning circuit protects the battery from overcharging. Similarly, when the temperature changes, power conditioning is used to stabilize the output voltage.

Through numerous factors, including input impedance, power control, and filtering, the conditioning circuits play a crucial role in an energy-harvesting system. The transducer (whether it be a thermal, photovoltaic, or vibrational source), power-conditioning circuit, microcontroller, and storage device (supercapacitor) are among the most critical parts.



Di Paolo Emilio, M. (2017). “Microelectronic Circuit Design for Energy Harvesting Systems.” Springer.

This article was originally published on EE Times.

Subscribe to Newsletter

Leave a comment