Analysis and Design of a Thermoelectric Energy Harvesting System With Reconfigurable Array of Thermoelectric Generators for IoT Applications
In this paper, a novel thermoelectric energy harvesting system with a reconfigurable array of thermoelectric generators (TEGs), which requires neither an inductor nor a flying capacitor, is proposed. The proposed architecture can accomplish maximum power point tracking (MPPT) and voltage conversion simultaneously via the reconfiguration of the TEG array, and demonstrate significantly improved power conversion efficiency over the conventional switching converter and switchedcapacitor architectures. Two systematical scaling approaches— powers-of-two scaling and maximum-factor scaling—are presented and analyzed to serve as the design guideline, catering for reconfigurable TEG arrays of different sizes. In order to optimize the chip area required, a custom-designed, multi-level hierarchy and systematically scalable switch array that requires a reasonable number of switches is also developed to enable the reconfiguration of the TEG array. The 16-node and 12-node versions of the proposed system have been implemented in a standard 0.35-μm CMOS process. Measurement results verify the analysis, and confirm that the proposed system can maintain a higher than 88.8% efficiency over a wide range of temperature gradients.
Research works reported earlier mainly focus on harvesting energy from low input voltage –, or start-up with low input voltage –. However, for the new generation TEGs with a high open-circuit voltage, the requirement for low voltage operation and low voltage start-up is relieved because a much higher voltage can be obtained now with the same temperature difference. In fact, in the field application of thermoelectric energy harvesting systems, the fundamental challenge is not low input voltage operation, but low temperature difference operation instead. For example, in , the conventional TEG in  is used, and the minimum input voltage for this energy harvesting interface is 25 mV, which corresponds to a 1 K temperature difference. If multiple thin film TEGs in  are used to form a TEG array with 92 TEGs which shows a comparable size (10.5 cm2) as the energy source, a 12.88 V open-circuit voltage can be obtained from a 1 K temperature difference. After maximum power point tracking (MPPT), the input voltage of the energy harvesting interface is still as high as 6.44 V, implying that the low input voltage problem is no longer a critical issue for state-of-the-art TEGs. By by-passing the low voltage start-up operation, the ability to achieve high efficiency over a wide input range, and reduce quiescent power consumption become increasingly important. For example, if a cloth or shoe integrated IoT device is powered by body heat, the temperature difference applied across the TEG can vary from <0.5K to >10K, depending on air temperature, movement of the wearer, wind speed and so on . Unfortunately, it is very difficult for existing thermoelectric energy harvesting systems to maintain high efficiency (>80%) over a wide range of temperature gradients –. Although the advanced TEGs show increased power density, the size of the whole system is still limited for body heat energy harvesting, which translates to μW range of total power. Therefore, nW range quiescent power consumption of the system is necessary.
Among the existing approaches, inductive-based and capacitive-based thermoelectric energy harvesting systems are the most commonly used, corresponding to two main types of DC-DC converters–switching converters and charge pumps. The objective of these systems is to harvest the heat energy from the ambient environment and convert it to electrical energy by TEGs to power a load with a regulated output voltage or be stored in storage elements. The common operation flow of these conventional architectures can be briefly divided into three steps. Firstly, the TEGs generate electrical power from heat flow and the open-circuit voltage linearly related to the temperature difference across the TEGs. Then MPPT is performed to extract the maximum available power generated by the TEGs, and the energy is saved in the input capacitor. Finally, voltage conversion is performed to convert the input voltage to the required output voltage. The need for MPPT is the major difference between DC-DC converters used in energy harvesting systems and the ones used in common power conversion applications , . Typically, MPPT is performed by impedance matching, or alternatively, it can be perceived as regulating the input voltage VS (or the actual voltage across the TEG) to be half of the open-circuit voltage VTEG. Since the input voltage is variable depending on the ambient environment, the power converters typically need to handle a wide range of input voltages in energy harvesting applications.
A thermoelectric energy harvesting system with a reconfigurable array of thermoelectric generators is proposed and analyzed. The proposed system with powers-of-two (2n) scaling and maximum-factor scaling can theoretically achieve an arbitrary wide voltage range with >88.8% efficiency, and the latter scaling approach can further achieve higher overall efficiency. The 16-node and 12-node versions of the proposed system are implemented in a standard 0.35-μm CMOS process, and the measurement results show a good match with the analysis and simulation results. With reasonable size for wearable IoT device integration, both versions can achieve consistently high efficiency (>87%) for the operational temperature gradient range of body heat energy harvesting applications. The 12-node RTA can further achieve a 4.5× input voltage range with >94% efficiency, in addition to the >87% efficiency range.
 Tellurex. Tellurex Thermoelectric Energy Harvester-G1- 1.0-127-1.27, accessed on Jan. 2011. [Online]. Available: http://educypedia.karadimov.info/library/termo.pdf
 Micropelt. MPG-D751 Thin Film Thermogenerator. [Online]. Available: http://www.micropelt.com/down/datasheet_mpg_d751.pdf
 E. J. Carlson, K. Strunz, and B. P. Otis, “A 20 mV input boost converter with efficient digital control for thermoelectric energy harvesting,” IEEE J. Solid-State Circuits, vol. 45, no. 4, pp. 741–750, Apr. 2010.
 A. Shrivastava, N. E. Roberts, O. U. Khan, D. D. Wentzloff, and B. H. Calhoun, “A 10 mV-input boost converter with inductor peak current control and zero detection for thermoelectric and solar energy harvesting with 220 mV cold-start and 14.5 dBm, 915 MHz RF kickstart,” IEEE J. Solid-State Circuits, vol. 50, no. 8, pp. 1820–1832, Aug. 2015.
 J. Katic, S. Rodriguez, and A. Rusu, “A dual-output thermoelectric energy harvesting interface with 86.6% peak efficiency at 30 μW and total control power of 160 nW,” IEEE J. Solid-State Circuits, vol. 51, no. 8, pp. 1928–1937, Aug. 2016.
 Y. K. Ramadass and A. P. Chandrakasan, “A battery-less thermoelectric energy harvesting interface circuit with 35 mV startup voltage,” IEEE J. Solid-State Circuits, vol. 46, no. 1, pp. 333–341, Jan. 2011.
 Y. K. Teh and P. K. T. Mok, “Design of transformer-based boost converter for high internal resistance energy harvesting sources with 21 mV self-startup voltage and 74% power efficiency,” IEEE J. Solid- State Circuits, vol. 49, no. 11, pp. 2694–2704, Nov. 2014.
 P.-H. Chen et al., “An 80 mV startup dual-mode boost converter by charge-pumped pulse generator and threshold voltage tuned oscillator with hot carrier injection,” IEEE J. Solid-State Circuits, vol. 47, no. 11, pp. 2554–2562, Nov. 2012.
 V. Leonov, “Thermoelectric energy harvesting of human body heat for wearable sensors,” IEEE Sensors J., vol. 13, no. 6, pp. 2284–2291, Jun. 2013.
 H. Alam and S. Ramakrishna, “A review on the enhancement of figure of merit from bulk to nano-thermoelectric materials,” Nano Energy, vol. 2, no. 2, pp. 190–212, Mar. 2013.