Elsevier

Nano Energy

Volume 54, December 2018, Pages 461-476
Nano Energy

Review
Thermoelectricity for IoT – A review

https://doi.org/10.1016/j.nanoen.2018.10.013Get rights and content

Highlights

  • Internet of Things market quantification, future growth and needs identified.

  • Energy harvesting methods hierarchy and their common performance presented.

  • Fossil fuels and heat losses availability, quantification and reserves highlighted.

  • Scientific/commercial thermoelectric topologies, materials and performance reviewed.

  • Performance of energy management circuits for thermo-generators compared.

Abstract

We are witnessing an unprecedented expansion of Internet of Things (IoT) market, whose nodes are already outnumbering human population several times. Despite the huge popularity of IoT, its further expansion is slowed down by a lack of viable power supply methods capable to replace wires or batteries. Due to IoT demand for alternative supply, energy harvesting (EH) gathers attention from scientific groups all around the world. In particular, thermoelectricity (TE) seems to be a natural and intuitive candidate for IoT owing to magnitude and omnipresence of heat losses and amenability to direct, vibrationless, noiseless and reliable conversion. This review provides up-to-date comparison and evaluation of a recent progress in the field of thermoelectricity, resulting primarily from multidisciplinary optimization of materials, topologies and controlling circuitry. The improvement in materials integrates two trends: nanostructural modulation of pre-existing, conventional thermoelectric materials and synthesis of novel ones. Regarding topology, TE responds better and better to miniaturization trend of semiconductor industry, driven by miniaturization trend, by proposing alternatives to conventional π-type topology. And finally, recently developed controlling circuits consume extremely low power while idle, exhibit above-90% efficiency and start-up with ultra-low input voltages. Combined, these improvements position TE closer to marketization than ever before.

Introduction

Urged by the worldwide increasing demand in energy [1], [2] (see Fig. 1a), constrained by the limited reserves of fossil fuels [2], [3], [4], [5] (see Fig. 1a) and facing a problem of global climate change [6], [7], [8], [9] (see Fig. 1b), all innovative solutions contributing to improve the renewable production of energy are playing a strategic role. As presented in Fig. 1a, the natural reserves of fossil fuels are predicted to be exhausted within: next 50 years for oil [5], [10] and more than 100 years for coal [2], [4], gas [2], [3], and nuclear [11]. In this context, searching for alternative energy sources is an urgent necessity. If we target today a fully connected and Internet of Things (IoT) controlled planet, we cannot expose this strategy to a total failure in 50–100 years due to fossil energy shortage.

Meanwhile, an emerging market of IoT expands at a rate never seen before [12]. The energy consumed by an IoT node counts in ~0.86 kW h/annum (supposing 100 μJ per cycle to be repeated every second, see further) – 13 decades behind the world electricity consumption as high as 25.5 TW h in 2017 [2]. Nevertheless, if we take into account the prospected count of IoT nodes (tens of trillions in 50–100 year perspective), the total energy consumption by IoT may also reach the level of 8.6 TW·h/annum, becoming on the same order of magnitude. This clearly indicated the necessity of supplying IoT with harvested rather than fossil energy. IoT principle is to establish communication between devices (Things) omitting human intervention. Basing on this principle countless number of applications are realized e.g. smart city [13], remote health care [14], [15], fully automated production lines [16], [17] etc. Topological particularity of IoT systems lies in their pyramidal structure [13], [18]. A conventional IoT system consists of numerous communicating nodes (also called leafs) which are monitoring particular parameters, such as temperature, presence, light density, traffic intensity etc. measured data are temporarily transmitted to the overriding unit (see Fig. 2a). Supervising unit stores and analyses data and reacts appropriately. Historically, IoT systems first emerged in the military and entered into the civil market owing to a concept of Intelligent buildings [19] and subsequently got popularized [20], [21], [22].

An endless list of IoT applications is reflected by a rapidly rising number of IoT nodes and by growing market penetration by IoT devices (see Fig. 3a). At present, there are ~5 IoT nodes per each human and this proportion will be doubled in the next 4–5 years. It is too early to judge, but in the next 50–100 year perspective, the count of IoT nodes may reach tens of trillions devices. Quickly expanding IoT market has significant societal impact, creating considerable employment growth (see Fig. 3b). At the end of 2018 the number of created jobs is forecasted to reach almost 3 million and is predicted to further rise up to 5 million within next two years. From this perspective, the IoT market is of a huge social importance. Significantly, this sizable employment market built up within less than a decade – a speed that has no precedence.

The IoT expansion on the market could proceed even faster but it is significantly hampered by a lack of economically attractive and energetically efficient alternatives for powering IoT nodes. Production of useful energy from omnipresent or lost energy is called the Energy Harvesting (EH). EH can be a huge relief for IoT allowing its further expansion through construction of wire/battery free devices so needed by the IoT. That is why EH gathers remarkable attention from scientific community, evidenced by clear acceleration in this field (see Fig. 4). Within the period of last 10 years the number of publications per year increased close to 20 times.

Section snippets

New paradigm for energy generation due to IoT

In the field of IoT we more often speak of energy harvesting than on generation. This is due to smaller energy consumption that coms to play. Between the consumption levels in IoT applications (μJ) and those in dwelling/industrial (MJ) we have 12 orders of magnitude difference. Therefore, regarding IoT we rather speak of Energy Harvesting (EH) (minute, free energy from environment) than generation.

Although a typical power consumption of a single IoT node is very small (see Fig. 5), it is

Energy source availability

EH can be considered as a branch of technology developing devices able to convert an ambient energy into a useful energy (usually electric one). Two classes/types of EH can be distinguished: (i) macro-scale harvesting, considered usually as a renewable source of energy [27], [28], [29] and (ii) micro-scale harvesting, mainly converting waste energy [25], [26]. It is worth noting that majority of micro-scale EH techniques have also macro-scale counterparts. Of course we will focus on micro-scale

Energy harvesting – methods

Renewable energy can be harvested in two ways: (i) on macro scale exploiting Nature-provided renewable energy sources e.g. wind, hydroelectric, geothermal etc. and (ii) on micro scale by reusing energy losses resulting from human-driven activities. The macro harvesting is offering access to attractive and sizable reserves and can be a salvation from greenhouse effect and depleting fossils fuels. The micro-scavenging opens a prospect for autonomous, battery/wires-free devices vitally needed for

Thermoelectricity topological aspects

As the most difficult to conserve, the heat energy exhibits the highest losses. Quantitatively, approximately half of heat energy is dissipated to atmosphere [68]. USA industry releases almost 30% of input energy in a form of heat each year [69].

Huge heat losses can be at least partially recovered using the ThermoElectric (TE), TRyboElectric (TRE) or PyRoelectric (PR) [70], [71], [72] effects or by hybridization of one of them with other harvesting methods. Although TE and PR are both

Thermoelectricity – material aspects

Despite of the advantages of TE effect and a widespread availability of heat losses, it took more than 100 years for the first TEG to be installed [106]. This gap between the discovery and its technological exploitation was caused by a lack of thermoelectrically efficient and economically viable materials. TE is a single-step conversion which of performance and attractiveness relies mainly on the material which is executing this conversion. Evaluation of a thermoelectric performance of a given

Subjugate thermoelectric energy

Effective power extraction from TEG requires a use of electronic controllers enabling i.a. voltage adaptation, energy storage and Maximal Power Point Tracking (MPPT) [204], [205], [206], all this functions are integrated in miniaturized microelectronics package as depicted in Fig. 17. In the most usual case, TEG operates under imposed thermal conditions and temperature difference across the TEG. This results in a voltage formation according to the physics of the Seebeck effect (see Fig. 10b).

Market perspectives

Already commercially available TEGs (retrieved from Fig. 16b) can successfully cover total power needed for sensors that can work in IoT, which is graphically illustrated in Fig. 20. Confronting sensor and commercial TEGs power densities related to their footprint leads to a conclusion that TEGs are able to provide sufficient, convenient, reliable DC-power for IoT devices of various type. Numerous applications of TEGs in IoT are currently in study, and the first realizations are commercially

Conclusions

This review compares and evaluates recent progress in the field of thermoelectricity with the emphasis on possible use in the Internet of Things (IoT) devices. IoT creates and builds a market bigger than ever known before. Despite of the fact that typical IoT nodes require very small amount of energy to operate, powering them is problematic, regarding their portability, localization, size and often harsh work environment. At present IoT nodes are usually supplied using batteries or wires, which

Maciej HARAS was born in Gdańsk, Poland in 1984. Received the M.Eng diploma in power electronics from the Gdańsk University of Technology and the Polytechnique de Grenoble (2008). He obtained a Ph.D. in nanotechnology from the Université Lille-1 (2016), investigating silicon-based thermoelectric generators. This work was awarded with the best Ph.D. dissertation prize from the Université Catholique de Lille. He joined STMicroelectronics R&D (2011) working on energy harvesting. In 2014 he joined

References (280)

  • C. Yu et al.

    Thermoelectric automotive waste heat energy recovery using maximum power point tracking

    Energy Convers. Manag.

    (2009)
  • Z.B. Tang et al.

    A research on thermoelectric generator's electrical performance under temperature mismatch conditions for automotive waste heat recovery system

    Case Stud. Therm. Eng.

    (2015)
  • G.S. Hwang et al.

    Micro thermoelectric cooler: planar multistage

    Int. J. Heat. Mass Transf.

    (2009)
  • REN21, Renewables 2018: Global Status Report, Renewable Energy Policy Network for the 21st Century (REN21), Paris,...
  • British Petroleum, 67th Statistical review of World Energy June 2018, British Petroleum, 2018....
  • J. Laherrère, Uncertainty of data and forecasts for fossil fuels, Association for the Study of Peak Oil, Universidad de...
  • J. Schindler, W. Zittel, Coal: Resources and future production, Energy Watch Group, Berlin, 2007....
  • L. de Sousa, World oil and gas production forecasts up to 2100, The Oil Drum. 〈http://europe.theoildrum.com/node/3565〉...
  • F.H. Cocks

    Energy Demand and Climate Change: Issues and Resolutions

    (2009)
  • M.G. Flanner

    Integrating anthropogenic heat flux with global climate models

    Geophys. Res. Lett.

    (2009)
  • NASA Global Climate, Carbon Dioxide, Climate Change: Vital Signs of the Planet....
  • NASA Global Climate, Global surface temperature, Climate Change: Vital Signs of the Planet....
  • I. Dincer et al.

    Advanced Power Generation Systems

    (2014)
  • Cisco, Internet of Everything (IoE) Connections Counter....
  • M. Chen et al.

    Body area networks: a survey

    Mob. Netw. Appl.

    (2011)
  • B. Latré et al.

    A survey on wireless body area networks

    Wirel. Netw.

    (2011)
  • P. Szulewski

    Industrial automation products in the conception of Industry 4.0

    Mechanik

    (2016)
  • P. Nenninger et al.

    Harvest time

    ABB Rev.

    (2011)
  • A. Al-Fuqaha et al.

    Internet of Things: a survey on enabling technologies, protocols, and applications

    IEEE Commun. Surv. Tutor.

    (2015)
  • S. Mclelland

    Intelligent buildings: sensing a need?

    Sens. Rev.

    (1988)
  • C. Perera, Sensing as a Service for Internet of Things: A Roadmap,...
  • C.X. Mavromoustakis, G. Mastorakis, J.M. Batalla, Internet of Things (IoT) in 5G Mobile Technologies....
  • Q.F. Hassan et al.

    Internet of Things: Challenges, Advances, and Applications

    (2017)
  • T. Skotnicki, Energy harvesting for IoT, short course at the 2015 Symposia on VLSI Technology and Circuits, Kyoto,...
  • N. Elvin et al.

    Advances in Energy Harvesting Methods

    (2013)
  • L.B. Kong et al.

    Waste Energy Harvesting

    (2014)
  • T.K. Ghosh et al.

    Renewable Resources

    (2011)
  • T. Suntio, T. Messo, J. Puukko, Wiley-VCH, Power Electronic Converters Dynamics and Control in Conventional and...
  • J.W. Matiko et al.

    Review of the application of energy harvesting in buildings

    Meas. Sci. Technol.

    (2014)
  • H.J. Visser, A.C.F. Reniers, J.A.C. Theeuwes, Ambient RF Energy Scavenging: GSM and WLAN Power Density Measurements,...
  • S. Duangsuwan, T. Leukachorn, S. Chinsawatpan, S. Promwong, Experimental study for polarimetric measurement of RFID...
  • S. Sherrit, H.J. Lee, P. Walkemeyer, T. Winn, L.P. Tosi, T. Colonius, Fluid flow nozzle energy harvesters, in: Sensors...
  • H. Lee et al.

    Piezoelectric energy harvesting in internal fluid flow

    Sensors

    (2015)
  • IEC 61215-1:2016, Terrestrial photovoltaic (PV) modules - Design qualification and type approval - Part 1: Test...
  • D.A. Howey et al.

    Design and performance of a centimetre-scale shrouded wind turbine for energy harvesting

    Smart Mater. Struct.

    (2011)
  • A.S. Holmes, G. Hong, K.R. Pullen, K.R. Buffard, Axial-flow microturbine with electromagnetic generator: design, CFD...
  • ANSI/ASHRAE/IES Standard 90.1-2016, Energy Standard for Buildings Except Low-Rise Residential...
  • Cited by (236)

    View all citing articles on Scopus

    Maciej HARAS was born in Gdańsk, Poland in 1984. Received the M.Eng diploma in power electronics from the Gdańsk University of Technology and the Polytechnique de Grenoble (2008). He obtained a Ph.D. in nanotechnology from the Université Lille-1 (2016), investigating silicon-based thermoelectric generators. This work was awarded with the best Ph.D. dissertation prize from the Université Catholique de Lille. He joined STMicroelectronics R&D (2011) working on energy harvesting. In 2014 he joined IEMN laboratory and engaged there in development of CMOS compatible thermoelectric generators and mechanical characterization of silicon nanowires. Currently he is the principal scientist at the CEZAMAT laboratory.

    Thomas SKOTNICKI was with France Telecom from 1985 till 1999 when he joint STMicroelectronics. He became the first STMicroelectronics Company Fellow and Technical Vice-President. He invented the UTBB FDSOI structure (in production at STMicroelectronics, GF and Samsung). Today he is Director of CEZAMAT Consortium and Professor at Warsaw University of Technology, Poland. He holds 80 patents and has authored close to 400 scientific papers, and several book chapters on CMOS and Energy Harvesting. He is an IEEE Fellow, has supervised 29 PhD theses, served as Editor for IEEE TED, and was on JJ Ebers and Frederik Philips IEEE Award Committees.

    View full text