See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/343950613

Waste Heat to Power: Technologies, Current Applications and Future Potential

Article in Energy Technology · September 2020

DOI: 10.1002/ente.202000413

CITATIONS READS

53 1,417

5 authors, including:

Erik Garofalo Matteo Bevione

RINA Consulting S.p.A. École Polytechnique Fédérale de Lausanne
21 PUBLICATIONS 260 CITATIONS 13 PUBLICATIONS 113 CITATIONS

SEE PROFILE SEE PROFILE

Luca Cecchini Fabio Mattiussi

Università degli Studi di Trento Istituto Italiano di Tecnologia
18 PUBLICATIONS 184 CITATIONS 5 PUBLICATIONS 95 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Matteo Bevione on 22 June 2023.

The user has requested enhancement of the downloaded file.

REVIEW
www.entechnol.de

Waste Heat to Power: Technologies, Current Applications,

and Future Potential
Erik Garofalo, Matteo Bevione, Luca Cecchini, Fabio Mattiussi,
and Alessandro Chiolerio*

earlier. Three key examples of this are

Energy consumption, environmental impact, and sustainability have risen fast 1) the International Energy Outlook 2019
through the ranks, achieving the first places in driving investments, policies, delivered by the US Energy Information
and concerns of all countries at any developmental stage. Energy transformation, Administration, 2) the independent assess-
ment issued by the UK Committee on
though, must cope with nonunitary efficiency of devices and processes, which
Climate Change by request of the UK gov-
results in a distributed production of waste heat. A reduction of emissions, ernment, and 3) the European Green Deal
implying a conversion of waste heat to more noble forms of energy and a for a climate neutral EU within 2050, with
concurrent increase in efficiency of the same devices and processes, is of par- a plan to invest 1 trillion EUR. As regards
amount importance. In view of the enthalpy content and distribution of the the UK, the national target defined by the
different sources of waste heat, low-grade/low-enthalpy sources below 200
C government is the achievement of more
than a 20% increase in industrial energy effi-
are considered the most fertile field for research and development, with an
ciency by 2030. Action plans are therefore
impressive industrial growth rate. Thermodynamic cycles and thermal conver- required by virtuous governments to sup-
sion devices based on the most relevant physical effects are herein introduced port innovation and make it easier to access
and briefly described, including both solutions that already achieved industrial financial resources.[1] As an example, the
maturity and less developed systems and devices whose study is still in progress. world’s largest fund manager, BlackRock
A specific focus on three application domains, selected due to their economic (close to 2.64 trillion EUR), announced
to the US Securities and Exchange
relevance, is done: industrial processes for the vast energy and capital availability,
Commission the launch of a money-market
automotive sector for its permeation, and wearable devices for the market size.
fund, primarily investing in developed envi-
Limits and opportunities are critically discussed. ronmental practices.[2]
According to a study conducted by
Smil,[3] the world energy consumption
has followed a monotonically increasing
trend since the beginning of the first industrial era (1840). As
1. Introduction
shown in Figure 1, the speed of this growth has recently
1.1. An Ecological Overview increased (Data source: Global Energy Statistical Yearbook web-
site, the most recent available (2018)). Energy consumption is
Energy-related issues, including the location and exploitation an excellent indicator for the maturity level of economics, in
of resources, the costs of extraction, transformation, and distribu- particular growth/recession. In the first period of industrializa-
tion, accessibility and demand, as well as awareness about use and tion, energy consumption experienced a constantly increasing
consumption, are of paramount importance at this moment in superlinear trend. At a global scale, when the period of postin-
time. The governments of every developed nation are allocating dustrialization began, economic growth started to stabilize and
increasingly wider funds to evaluate problems such as those listed ultimately decrease, and the associated energy consumption

E. Garofalo, M. Bevione, L. Cecchini, F. Mattiussi, Prof. A. Chiolerio L. Cecchini

Istituto Italiano di Tecnologia Scuola Superiore Sant’Anna
Center for Sustainable Future Technologies Piazza Martiri della Libertà 33, 56127 Pisa, Italy
Via Livorno 60, 10144 Torino, Italy
E-mail: [email protected] F. Mattiussi
SPEA S.p.A.
E. Garofalo, L. Cecchini, F. Mattiussi Via Torino 16, 10088 Volpiano (TO), Italy
Department of Electronics and Telecommunications
Politecnico di Torino
Corso Duca degli Abruzzi 24, 10129 Torino, Italy
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/ente.202000413.
DOI: 10.1002/ente.202000413

Energy Technol. 2020, 2000413 2000413 (1 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

104 World Energy Consumption environment also have the side effect of locally increasing the
18
atmospheric temperature, especially in urban environments,
16 slightly moving the solution equilibria of several gaseous species.
Focusing on the main green house gases (GHG) and listing them
14 in order of importance, according to the Intergovernmental Panel
Energy Consumption, TWh

on Climate Change (IPCC),[7,8] they are the following.

12
Water vapour is the major contributor of all since it provides
10 from 30% to 70% of the overall greenhouse effect.[9,10] The rea-
son for its potential resides in the large amount of this chemical
8 species in the atmosphere, enabling a positive feedback loop.
Carbon dioxide is the principal gas monitored, because, even if
6
it absorbs less heat per molecule than others, it is more abundant
4 and has a longer lifetime in air (thousands of years with respect
to the decades of methane). Furthermore, it plays an important
2 Vaclal Smil data role in regulating the pH of oceans, as these represent the main
GESY data CO2 reserve, and its increase provokes acidification. Another pos-
0
1880 1900 1920 1940 1960 1980 2000 2020 itive feedback effect can be found here: a temperature increase
Year leads the oceans to release CO2 into the atmosphere because of
the decrease in solubility.
Figure 1. Global energy consumption trend during the last 140 years. The
Methane: The impact of this gas on the greenhouse effect is
energy sources that have been taken into account are coal, crude oil, nat-
60 times that of CO2 and a monotonic increase in the past dec-
ural gas, hydroelectricity, nuclear electricity, and biofuels for data until
1990. The data provided by the Global Energy Statistical Yearbook (red) ades has been registered. However, its level and lifetime are low
are more precise and complete, and for this reason the two curves are enough to keep its contribution under control.
slightly mismatched at the 1990 entry. Nitrous oxide: Its capability to absorb heat is 300 times that of
CO2 and around 40% of total N2O emissions are caused by
human activity. Moreover, its lifetime in atmosphere is over
followed a fairly similar trend. CO2 emissions are also directly 110 years, making it one of the most persistent GHGs.
correlated with economic growth.[4] Energy is the fundamental Ozone is naturally present in the upper level of the atmo-
source transformed by our everyday life processes, spanning sphere and, in contrast with the previously analyzed gases,
from the residential, to mobility, to the industrial sectors, includ- mainly absorbs the radiation coming from the Sun instead of
ing food as the basic source of our physiological activity. those reflected by the Earth’s surface.
Due to global warming, increasing energy costs, and green To estimate the amount of waste heat Q [J] released, for exam-
politics, the search for sustainable, responsible, clean resources ple, from a continuous hot vapour volume into the environment,
represents a highly impacting and exponentially growing sector. it is possible to use Equation (1)
According to a review of energy consumption at the global level,
it is evident that most of our energy is derived from fossil fuels, Q ¼ c m ΔT (1)
whereas renewables account for less than one-quarter of global
consumption worldwide.[5] where m [kg] is the mass of the hot vapor volume, c [Jkg1 K1] is
The over-exploitation of energy sources may result in critical- the heat capacity of the vapor, and ΔT [K] is the temperature dif-
ities, at the political, social, and logistic level. It is therefore of ference between the hot vapor and the ambient environment. In
fundamental importance to increase energy efficiency at the scav- addition, the heat coming from phase transformations (latent
enging/harvesting/transformation level.[4] Regarding the opera- heat) must be taken into account, which is released after the con-
tion of industrial and power plants, dilemmas arise on how to densation of a gas or the solidification of a liquid. Thus, the
increase efficiency, how to better convert energy between its higher the mass flow and ΔT between hot and cold reservoirs,
forms, and how to utilize waste heat arising from thermodynamic the greater the waste heat amount. To understand the abundance
inefficiencies (e.g., due to friction, thermal dispersion, electromag- of this source, its economic value, and the importance of convert-
netic induction, etc.) or practical needs. The creation of a common ing it into more noble energy forms, as proposed by waste heat to
approach with a new attitude toward our fragile natural environ- power (WHP) applications, the average efficiency of industrial
ment is required. Alongside an increase in the consumption of processes and thermal machines should be considered: between
primary energy, the amount of waste heat increases, therefore rep- 30% and 60% of the overall energy consumed is wasted into the
resenting a greater economic and ecological potential for utiliza- environment as heat. Other fundamental aspects also play a role
tion. Although waste heat may be exploited, ideally, it should be and must be considered in a multivariable optimization model,
avoided: this is done through the energy optimization of processes before any technology is adopted.
by needs-based controls, process operation and design, improved Time availability: The temporal distribution of energy avail-
insulation, the use of highly efficient electric drive systems, and a ability throughout a day, week, and season, where a more con-
change toward more energy-efficient production methods. tinuous flow is always preferred is considered. To increase heat
However, even after energy optimization, significant potential stability in time, heat storage could be implemented, which is
remains.[6] Enormous amounts of waste heat injected into the another huge field under development.

Energy Technol. 2020, 2000413 2000413 (2 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

Thermoeconomic analysis: This aspect takes into account the more noble forms of energy. The outline is as follows. In
Return on Investment (ROI), usually comprising between 5 and Section 2, thermodynamic cycles and thermoelectric (TE) devices
10 years, to pay back the installation. that have already achieved industrial maturity are described. In
Environmental sustainability should also be evaluated through Section 3, thermodynamic cycles and conversion devices based
a full life cycle assessment (LCA) of plant/device life expectancy, on less known physical effects that have shown profitability and
development costs, and decommissioning. are currently available at research or pilot scale are described. In
It is almost impossible to find studies addressing all the three Section 4, a specific focus on the most relevant application
fundamental aspects evidenced earlier, with rare exceptions.[11] To domain for capital intensity, that of industrial production plants,
give a greater breadth to the thermodynamic analysis, in particular is given, followed by Section 5, presenting a focus on the auto-
dealing with heat, a form of energy that intrinsically contains entro- motive sector for its permeation and intermediate characteristics
pic (dispersive) terms, it is worth recalling the concept of exergy, by of portability and complexity. It is finally followed by Section 6 for
definition: “the total amount of energy that can be extracted from the potential market size of portable WHP applications. The last
a physical system given its position in an external environment.” Section 7 provides a final comparison between technologies
The concept of exergy brings together two aspects of an energy- and anticipates, in our critical vision, future developments in
transformation process: its “quality” and its “quantity.” Ideal ther- the field.
modynamics is conceived as a far-from-reality analysis, where
processes are deconstructed as an integral of infinite processes
in quasiequilibrium. Exergy analysis is fundamental for waste heat 2. Current Technologies
recovery (WHR) processes, as its outcomes take into account the
sources of irreversibility which are intrinsic to real processes.[12,13] In this work, a large number of WHR methods have been
described. They are divided into thermodynamic approaches on
1.2. WHP Economic Potential one side and cross-thermal effects on the other one. Examples
of the first category are the Brayton cycle (BC), first introduced
As can be noticed, taking into consideration the total amount of and quite common in airplanes and gas power plants for its mar-
waste heat generated, 72% of global energy consumption is lost ginal maintenance costs; the Stirling cycle (SC), operating homon-
in the form of heat,[14] totalling 12 PWh year1. Around 66% of ymous engines, which feature good efficiency, lower pollution,
this heat is available at low-temperature (LT) conditions below silent operation, and are simply configured, reliable, and with mul-
200
C, which is the focus of our Review Article, totalling a tifuel capability; the organic Rankine cycle (ORC); the Kalina cycle
bit less than 8 PWh year1; 25% is available at medium tempera- (KC); and the carbon dioxide transcritical cycle (CDTCC). On the
ture (MT), between 200 and 500
C, totalling 3 PWh year1; the other hand, examples of the second category are TE devices and
last 1 PWh year1 is available at a high temperature (HT), with pyroelectric (PE) devices. At the nanoscale, mixed effects are typi-
conditions above 500
C. The average heat to power efficiency for cally enhanced, thanks to both thermal and other physical proper-
conversion in the two temperature/entalpy ranges covered by this ties (i.e., electronic, magnetic, etc.), enabling the direct conversion
Review is 15% (LT) and 45% (MT), as highlighted in Section 7. of temperature gradient. Although this field is still mainly unex-
Therefore, excluding HT from the analysis, a possible amount of plored, it is very promising.[20]
noble (i.e. electric) energy recovered from waste heat and injected
into the grid or used locally is estimated to be around 2.1. Thermodynamic Machines
2.5 PWh year1, generating a turnover of more than 300 billion
EUR if the electricity is sold to industries (at 0.125 EUR kWh1 2.1.1. Brayton Cycle
which is the 2015 cost toward industrial consumption) or more
than 550 billion EUR if sold to households (at 0.221 EUR kWh1 The BC is characterized by a process in which air is compressed
which is the 2015 cost toward households consumption).[15–17] isentropically by a compressor. Then it receives heat from
A bulk of 2.5 PWh year1 represents 1.5% of the global energy exhaust gases at constant pressure and finally it expands isentro-
production. However, the cost per kWh generated could be pically in a turbine to generate electrical power. At this point, the
much higher, considering the installation costs for WHR/WHP air is discharged back to the atmosphere and the process is
solutions, and must benefit from intelligent politics aimed at repeated. BC represents an optimal choice for WHR systems
sustaining research, development, and production of highly working at mild temperatures in the industrial sector. This
efficient integrated processes and investment in equipment approach has been explored in the 1980s to convert waste heat
manufacturing. On the other side, a more economical way of from adiabatic engines, enabling a reduction of running costs on
recovering waste heat is through direct use of the heat as it one side and a reasonable additional capital investment on the
is.[18] According to market reports estimates for the WHP market other. Recently, Zhang et al. proposed a BC WHR system for
size, a significant value of 13.2 billion EUR has been pointed out the blast furnace slag in the iron and steel industry.[21] Song
for 2018 and, expecting a current annual growth rate (CAGR) of et al. integrated a BC WHR system into a diesel engine, connect-
13%, its value in 2025 will approach 26.4 billion EUR.[19] ing the turbocharger compressor with the compressor that
operates on the BC. Results showed that the fuel economy of
1.3. Review Outline the diesel engine can be improved by 2.6% at fast rotation
regimes and 4.6% at lower regimes, at full load.[22] Galindo
The following sections provide a complete review of technologies et al. investigated WHR for internal combustion engine (ICE)
available at present day to recover thermal waste and convert it to exhaust gases using a BC machine.[23] Gequn et al. (2015)

Energy Technol. 2020, 2000413 2000413 (3 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

proposed a CO2-based BC to recover the engine exhaust heat and components and for the low flammability, corrosion, and toxicity
compared its performances with an air-based BC. When com- of working fluids. Nowadays, several ORC-based plants are
pared with a BC based on air, the results show that the CO2 cycle installed in different countries (Italy, Austria, Germany, the
provides better results, including net output power, thermal effi- Netherlands, Sweden, U.S.A., Canada, etc.) and the number
ciency, and recovery efficiency.[24] of plants is still growing. The adaptability and the scalability
of this technology permit generating power from different heat
2.1.2. Stirling Cycle sources (from industrial to domestic applications), selecting the
dimension of the components and matching them with the
In the SC, the gas medium (usually air, helium, or hydrogen) is choice of the working fluid. In recent years, this technology
compressed isothermally. It follows a passage through a regenera- has been widely used also in the transportation sector.[27]
tor (or heat exchanger) that operates at constant volume, where an The working principle of ORC is shown in Figure 2a and its
amount of heat is absorbed by the gas, which raises its pressure temperature–entropy (TS) diagram is shown in Figure 2b: a
and temperature. Then, the gas is submitted to an isothermal pump pressurizes the working fluid in liquid form (from point
expansion and finally the hot gas goes back through the regenera- 1 to point 2); heat is then moved from the heat source to the
tor, releasing an amount of heat. The Stirling engine (SE) has sev- refrigerant (working fluid) through a superheater and an evapo-
eral features that make interesting its application, especially in rator, where the organic fluid gets vaporized (from point 2 to
conjunction with gas turbine (GT) cycles where waste heath is gen- point 3); then, the high-pressure working fluid flows into an
erated: it is silent and its working fluid can operate without any expansion turbine, connected to the load (electrical power gener-
mixed-phase operation. Furthermore, it is simple to power up ator) (from point 3 to point 4); at the end, the cooling source con-
and is sealed only toward the cold section.[25] Quite recently the trols the organic fluid returning back into liquid form in the
combined operation of GT and SE has been studied from a ther- condenser (from point 4 to point 1). There are many factors that
modynamic point of view. Hou et al. introduced a double-action can affect the ORCs’ performances. In case the available energy
thermoacoustic SE-based electrical generator capable of recovering is small and the available source is low grade, to safeguard the
liquified natural gas (LNG) cold exergy, converting the external overall performance of ORCs, attention is given to expanders
thermal energy into acoustic work (the operating temperature (centrifugal or axial-flow turbines, allowing the expansion of
is between LNG and ambient temperature).[26] the high-pressure gas) as they play a key role.[1] Bademlioglu
et al. provide a statistical analysis on system efficiency, using
the Taguchi method.[28] In an ORC system, the main parameters
2.1.3. Organic Rankine Cycle
affecting the design are choice of the working fluid, pinch point,
Based on Rankine cycle (RC), ORCs are steam-like systems that superheating temperature, evaporator and condenser tempera-
use organic working fluids instead of water, allowing to harvest ture, heat exchanger effectiveness, pump and turbine efficiency,
waste heat at temperatures up to 300
C.[27] In ORC machines, installation, and running costs.
hydrocarbons, siloxanes, refrigerants, and CO2 can be exploited
instead of water, thanks to their advantageous properties in LT 2.1.4. Kalina Cycle
applications: lower boiling and critical points, lower specific vol-
umes, as well as lower viscosities, higher vapour pressures, and KC is an absorption-based power generation cycle, based on RC.
higher molecular masses with respect to water-based fluids.[27] The working fluid enabling KC systems is a mixture of ammonia
Furthermore, ORCs are very much appreciated in current and water, to increase the recovery efficiency. Condensed water–
research for the availability and the simplicity of their ammonia mixture is compressed isentropically and then it is

(a) ORC schematic (b) Example of ORC TS diagram

Figure 2. a) a typical schematic of an ORC system where all the components and thermodynamic transformations are shown. b) An example of the
TS diagram using R245fa as the organic working fluid.

Energy Technol. 2020, 2000413 2000413 (4 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

heated at constant pressure while it evaporates (first the ammo- expanded, and finally condensed, releasing heat at constant pres-
nia and then the water); subsequently, it is isentropically sure. Cayer et al. applied four different methodologies on a
expanded into a turbine and finally condensed (first the water CDTCC power cycle with a low-grade energy heat source and
and then the ammonia) at constant pressure releasing heat. conducted parametric optimization.[34] Li et al. theoretically com-
The operating temperature lies between and 90 and 500
C. pared a CDTCC power cycle and a R245fa ORC, operating on a
With respect to ORC and supercritical cycles, KC shows superior low-grade heat reservoir, and concluded that exergy efficiency
performances. As the temperature difference between the work- would benefit from using a recuperator.[32,35]
ing fluid and the heat source is reduced, reversibility is increased
and dissipation limited. With respect to RC, where a considerable
amount of heat is lost during the isothermal vaporization of 2.2. TE Systems
water to steam, the binary mixture in the KC vaporizes noniso-
thermally, improving the efficiency of the cycle.[29] Moreover, KC TE devices also belong to WHP/WHR technologies, but have
has one more degree of freedom compared with RC, as it is pos- rather different features with respect to thermodynamic cycles.
sible to act on the NH3:H2O ratio in the working fluid and the First of all, TE systems are solid-state devices, which transform
system pressure levels. In fact, using a mixture as the working heat into electricity by means of the Seebeck effect in several
fluid permits to manipulate the pressure in the system by varying application domains such as factories, power plants, computers,
the composition. In this way, the thermal performances are vehicles, stoves, and wearable devices with a compact design.[27]
maximized.[30] At given cooling conditions the pressure in the The Seebeck effect derives from the capability of a material to
condenser can be reduced by diminishing the ammonia concen- produce current when submitted to a temperature gradient.
tration. Increasing the ammonia concentration may increase the The tendency of a material to manifest this effect is represented
evaporation pressure.[29] While KC allows a higher thermody- by the Seebeck coefficient S [VK1] which is a nonlinear function
namic efficiency, the split ammonia and water streams add com- of the temperature, and it is specific for each different material
plexity and may require an additional pumping system. For this and crystalline configuration. In Figure 3, the typical apparatus of
reason, plant operators and developers have different opinions a thermocouple (the first device exploiting the Seebeck effect) is
about the potential of KC to overcome Rankine-type power plant shown. TE materials are currently commercialized for integra-
performance.[31] Another relevant feature of NH3:H2O mixtures tion into high-performance cooling systems needed to realize
is the extremely low amount of oxygen dissolved in the fluid, that high heat fluxes to very LTs at precise rates.[27]
severely reduces oxidation likelihood and running costs.[1] This In general, the materials can be divided into three categories
process allows for flexibility in choosing boiling points which can depending on the temperature of operation: low (below 250
C),
compensate for temperature fluctuations in the heat flow. medium (250–600
C), and high (above 600
C) working temper-
Nevertheless, it is a technically complex system, so it is much more ature. Since the first device until now, bismuth is the most widely
capital intensive and far less tested than ORC modules. In the used material for LT applications. The Seebeck coefficient is
same range of temperatures it is known that ORCs are 15–25% deeply dependent on the effective mass and mobility of the car-
less efficient than KCs, with the tangible outcome that power riers.[32] For this reason, antimony–telluride Sb2Te2 or its alloys
plants featuring hot fluid below 150
C are increasingly based are used for p-type composition instead of bismuth–telluride,
on this new technology.[27] Several specific applications have pro- because of their higher carrier mobility. One important property
duced adaptations of KC systems, with more than 30 solutions of these alloys is the possibility to adjust the carrier concentration
which have been introduced so far. For example, in the cement and, hence, finely tune the material-specific figure of merit (FoM)
industry, superheated NH3:H2O vapor is generated, harvesting (The term figure of merit is referred to a parameter or a physical
low-grade heat; it is then expanded in a turbine, condensed in quantity used to characterize the material performances with
regenerative heat exchangers, diluted and fed into a low-pressure respect to other alternative elements.) Moreover, Figure 4 shows
condenser, and finally injected into the high-pressure condenser why both Bi2Te3 and Sb2Te2 are, at LTs, the most suitable
before reaching the vapor generators again.[1]

2.1.5. Carbon Dioxide Trans-Critical Cycle

As mentioned earlier, to recover or convert waste heat by means

of thermal machines, organic and natural fluids are often used. It
is worth noting that CDTCC technology, based on CO2, provides
nontoxic, nonflammable, noncorrosive operations. CO2 reaches
the supercritical state at 7.38 MPa and 31.1
C and is therefore
widely applied to low-grade heat recovery systems.[32] By the intel-
ligent design of a CDTCC system, heat is rejected below the crit-
ical point and absorbed above, having therefore a working fluid
in the superheated state under supercritical pressures, resulting
in maximized performances.[33] The working principle is the Figure 3. Thermoelectric effects: In the left figure the typical representa-
same described as that in the ORC, but in this case, CO2 is isen- tion of a thermocouple based on Seebeck effect is reported, whereas in the
tropically compressed, heated at constant pressure, isentropically right one, the Peltier effect is represented.

Energy Technol. 2020, 2000413 2000413 (5 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

Figure 4. The material FoM of different n- and p-type materials are plotted with respect to the temperature in (a) and (b), respectively. For the low-grade
WHP, thus, T ≤ 200
C, Bi2Te3 and Sb2Te3 show the highest ZT value. Reproduced with permission.[36] Copyright 2008, Springer Nature.

materials to be used, as they feature the highest FoM in this tem- observation has brought attention on PE devices which promise a
perature range. better efficiency and ease in installation. The main parameter
Recent studies by Pacific Northwest National Laboratory con- defining PE performances is the PE coefficient [Cm2 K1],
sidered some well-known industrial activities, such as aluminum described as the change in the spontaneous polarization vector
smelting, glass manufacturing, and cement production, and con- with respect to temperature. Several kinds of PE materials have
cluded that small ICEs could be profitably replaced by external been studied, trying to find the best compromise to enhance
combustion TE engines, that are vibration free and less impact- efficiency in the realization of PE generators (PEGs). Lead
ing from the acoustics point of view. TE engines can be driven zirconate titanate (Pb,Zr)TiO3 (PZT) or barium titanate
using several fuels, for example, propane, butane, liquefied nat- BaTiO3 (BT) features the highest-known PE coefficients (268
ural gas, bioalcohols, and not necessarily ones based on fossil and 200 μCm2 K1, respectively). A collection of information
sources.[37] Both US and Japan Governments have introduced about the most popular PE materials has recently been pro-
regulations to help companies introduce TE devices in trucks duced.[40] In the work developed by Sebald et al. a comparison
and cars to partially convert waste heat from the exhausts into between the performances of TEGs and PEGs is reported.[41]
electricity, powering steering, brakes, water pumps, turbocharg- It is shown that linear materials, i.e., those having the dielectric
ers, etc. Although efficiency estimates depend on the degree of constant which increases linearly with the applied electric field,
system integration and driving conditions, diesel consumption are able to produce just few μWcm3 for temperature variations
(nowadays highly diffused among developed countries to enable of 20
C with a frequency of 102 Hz.
on-site power generation) could be reduced by 5–10%.[37] Several However, using nonlinear materials, the performance can be
leading automotive manufactures such as Volvo, Volkswagen, strongly increased by adopting particular thermodynamic path
BMW, and Ford are introducing WHR systems able to provide cycles like the following.
around 1 kW of electrical power and increase consequently the Ericsson cycle consists of two isobaric transitions, isoelectric
thermal engine overall efficiency using TE generators (TEGs). in the polarization-E plane[42] and two isothermal transitions, a
Hsu et al. invented a WHR system for automotive exhaust heat compression and an expansion.[43] It is represented in Figure 5a.
recovery composed of 24 TE modules able to provide 12 W Olsen cycle that is formed by two isoelectric processes and two
(electrical) with a ΔT of 30
C.[38] Sano et al. developed a very isothermal ones. Several experiments have been conducted with
efficient TE module (15% efficiency).[39] Furthermore, solar ther- different materials under different crystalline forms (e.g. single
moelectric generators (STEGs) have been developed efficiently, crystals and thin films).[45–47] It is shown in Figure 5b.
coupling thermal solar collectors with TE devices, as the Lenoir cycle, a cycle formed by three transitions: isochoric heat
absorbed heat is conveyed on the TE devices by means of a dia- injection, isoentropic expansion, and isobaric heat rejection.
thermal fluid. Mohammadi et al. showed that this cycle in comparison with
the Ericsson one results in a much bigger amount of harvested
2.3. Pyroelectric Systems energy in the latter.[48] It is shown in Figure 5c.
In the work of Sebald et al. the results obtained testing differ-
Another effect that can be exploited in energy harvesting is pyro- ent materials are reported, and the increase in power output is
electricity, through the so-called pyroelectric (PE) effect. Unlike quite evident, moving from a few μWcm3 to hundreds of
thermoelectricity, where a stable (stationary) temperature gradi- μWcm3.[41] From the analysis of heat exchanged and other prop-
ent is needed, this effect allows to displace electrical charges from erties of TE and PE effects, it turns out that even though the
temporal temperature changes across a suitable material or power generated by PEGs is low, their efficiency is much higher
device. PE allows to avoid the main issue of TEGs, i.e., the need than that of TEGs. A complete analysis of PE efficiency for
of huge heat flows to sustain a large temperature gradient. This each of the cycles here introduced can be found in the work

Energy Technol. 2020, 2000413 2000413 (6 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

Figure 5. a) The typical trend of an Ericsson cycle. b) The typical Olsen cycle where the polarization is represented as a function of the electric field E.[44]
c) the Lenoir cycle. The colour map indicates the temperature variation. Reproduced under the and conditions of the a Creative Commons Attribution CC
BY 4.0 license.[44] Copyright 2019, The Authors. Published by Springer Nature.

of Batra et al.[49] A more detailed explanation of the phenomena granting excellent energy extraction at HT and increasing the
and its exploitation in some PE nanogenerators can be found in energy per kg/per m3 of the installed plant, one order of magni-
the work of Thakre et al.[50] tude below STs/GTs. SCO2 systems are mainly used in the HT
end of the thermal spectrum, for direct WHR, driving an electri-
cal power plant for both uses. Considering that the heat rejection
3. Emerging Technologies also occurs at HTs, other systems such as a properly matched
ORC could be cascaded. The most typical application is energy
Several advanced, emerging technologies in the WHP/WHR generation from geothermal sources, accounting for almost 75%
field have not been included in the previous section, for example, of world installed capacity. As can be noticed, taking into account
trilateral flash cycle (TFC), supercritical CO2 (SCO2) BC, magne- the amount of installed systems, the diffusion of SCO2 is quite
tocaloric and thermomagnetic hydrodynamic machines. Other limited, due to the capital-intensive requirements that make rea-
solid-state emerging technologies such as thermophotovoltaic sonable its use in multi-MW power plants. The free market fea-
devices (TPV) and thermionic devices (TI) have been discarded, tures a few technology providers (ORMAT, owning about 75% of
as their description is beyond the purposes of the present Review. installations, Exergy with 13%, TAS with 6%, and Turboden with
2%), whereas a number of smaller providers cover the niche of
3.1. Trilateral Flash Cycle small WHR plants ranging from 10 to 150 kW. A relevant frac-
tion of these units is very small (<4 kW) plants installed by
TFC is a promising technology, having a huge recovery potential ORMAT for valve operation and cathodic protection along pipe-
in comparison with ORCs, as the energy recovered can roughly lines in remote areas.[51]
double over the same temperature difference; an economic out-
come of this efficiency is related to the savings generated by avoid- 3.3. Magnetocaloric Machines
ing any cooling/heat rejection system in a power plant. A TFC
consists of a modified ORC where the organic working fluid is A magnetic colloid (a suspension of magnetic nanoparticles in a
heated up to the saturation temperature under high pressure liquid carrier) is a system where magnetization is thermodynam-
rather than evaporated, implying that the heat transfer is optimal ically coupled with temperature, at equilibrium. This aspect is
when no fluid is brought across its boiling point. TFC-based sys- at the basis of the so-called magnetocaloric effect (MCE): at rest
tems can substitute more conventional ORC units installed in MT and at a specific temperature, in a magnetic material, some
70 and 200
C processes, and they are able to follow any tempera- energy is transferred from phonons to the magnetic domain
ture variation on both the inlet and outlet phases of the cycle.[4,51] structure; if an external magnetic field is applied adiabatically,
the magnetic domains align and reduce their number (entropy
3.2. Supercritical CO2 BC is consumed), and when the field is removed, in the absence of
energy exchanges with the environment, the material cools
The SCO2 BC is a standard thermodynamic cycle operating with down. This effect has been discovered for the first time by
CO2 as working fluid. It is the power-conversion system which Warburg[52] and is an intrinsic property of all materials.
combines the advantages of both steam RC and GT systems. In Consider a sample of ferromagnetic material, for example,
other words, the fluid is compressed in the incompressible MnCr2S4 crystals, whose magnetic specific heat is shown in
region and the higher turbine inlet temperature can be utilized Figure 6a. The following refrigeration process, called adiabatic
with no material-related issues in comparison with the SRC. As demagnetization, shown in Figure 6b, can be applied. Starting
known, CO2 is able to sustain dramatic density variations as a from the top left, the sample is at thermal equilibrium with
consequence of slight temperature and pressure fluctuations, the surrounding environment and the magnetic dipoles are

Energy Technol. 2020, 2000413 2000413 (7 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

Figure 6. a) The specific magnetic heat for a ferromagnetic material, in this case MnCr2S4 crystals. Reproduced under the terms of the RNP/20/APR/
024834 license.[53] Copyright 2020, American Physical Society. b) The magnetic-refrigerating cycle related to magnetocaloric materials in general.
Reproduced under the terms of the 1028944-1 license.[54] Copyright 2020, IOP Publishing. c) The thermomagnetic generation cycle, featuring magneti-
zation (1), heating (2), demagnetization (3), and cooling (4). Reproduced under the terms of the 4810850356977 license.[55] Copyright 2020,
AIP Publishing.

randomly oriented, as no magnetic field is present; then, a mag- which induces sedimentation of the colloidal magnetic particles
netic field is switched on, and the sample starts heating up, by the effect of fluid density stratification, useful configurations
whereas the dipoles become aligned in the direction of the arise. As said, in the presence of an external magnetic field and a
applied field. Subsequently heat is transported to the surround- thermal gradient, FF shows the so-called thermomagnetic advec-
ing environment by means of a heat-transfer medium (the tion which induces different magnetic body forces (the cooler
dipoles are still aligned), but when the external magnetic field particles present higher magnetization than the warmer ones).
is removed, the dipoles’ orientations randomize, which translates In literature, several methods to harvest energy using FF are
into a further temperature reduction of the sample (below the described, in particular from a thermal gradient.[58] A way to pro-
temperature of the thermal bath). Possible heat-transfer materi- duce electrical power by exploiting the FF motion activated by
als are water or for very LTs, liquid He.[56] Furthermore, the (rem- convection currents generated by a thermal gradient is described.
anent) magnetization of a ferromagnetic material is a function of Furthermore, it is possible to generate solitonic density waves
temperature. Under static magnetic field, cycling the tempera- inside a closed-loop system wrapped with copper wire coils, to
ture produces cycling of magnetization, in turn. The interaction extract electromagnetic forces and electric power, giving birth
between fluctuating magnetization and static magnetic field pro- to the first generation of colloidal energy harvesters.[59] In all
duces magnetic forces acting on the ferromagnetic (or superpar- the cases where the gravitational field is parallel to both the ther-
amagnetic) particles that can be profitably converted, for mal gradient and the external magnetic field, it is possible to
example, using extraction coils and induction principle, into elec- exploit the Benard–Marangoni convection and extract energy
tromotive force for electric powering.[55] A representative ther- through the use of coils, as reported by Kemkar et al.[60]
modynamic cycle for active thermomagnetic generation is
shown in Figure 6c.
4. Industrial Applications
3.4. Thermomagnetic Hydrodynamic Machines 4.1. Introduction

Magnetic colloids, such as FerroFluids (FFs), are nanomaterials According to the IPCC, nowadays, one-third of the global energy
that present high magnetic susceptibility (some orders of mag- consumption is used in the industry and this amount corre-
nitude higher than other nonmagnetic natural substances).[57] sponds approximately to over 40% of CO2 emissions world-
In fact, when no external magnetic field is present, magnetic wide.[61,62] UK data show that industries absorb about
moments are randomly oriented inside the carrier fluid, and 3200 TWh year1, which represents about 26% of the total con-
the resulting net magnetization is equal to zero. Otherwise, if sumption in EU (Eurostat, 2015). Germany, Italy, France, UK,
external magnetic field is present, the dipoles align along the and Spain, respectively, as expected, account for the highest
direction of the field, and a nonzero net magnetization appears. industrial energy consumption, totalling about 60% of all EU
FFs can be used to harvest energy, giving the possibility to imple- numbers.[63] Agathokleus et al. analyzed the energy consumption
ment a low-power microgenerator, especially suitable for elec- of the industrial sector of the EU28 (member states of the EU
tronics where battery replacement has higher costs and is before Brexit)[51]: chemical and steel industries are the main
quite difficult to be performed. energy-demanding sectors, followed by paper processing, non-
Furthermore, magnetic colloids show variations in the ther- metallic minerals treatment, and food, accounting for about
mal convection due to competition among the magnetic field 65% of the total energy consumption in the industry.
and the gravitational one. In particular, when magnetic forces To manage the waste heat problem the most adopted approach
act in a direction transverse with respect to the gravitational field, is WHR. In a WHR device heat is exchanged between the

Energy Technol. 2020, 2000413 2000413 (8 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

diathermal material (which is cooled) and another intermediate mapping of the potential of waste heat considering EU countries
fluid (which is heated). Such intermediate fluid is never dumped is shown in Figure 8.[63]
but recovered and used to produce utilities such as steam or
power.[51] Industrial WHR addresses temperatures ranging from 4.2. Technical Potential of the Waste Heat from Industrial
60
C (cleaning processes) to more than 1700
C (iron and steel, Activities
cement, glass, and ceramic processing industries). In the chemi-
cal industry, the typical temperature range is from 100 to 500
C Among the various industrial sectors, the amount of waste heat
(in some cases up to 1000
C).[6] WHR technologies suffer from fraction (with respect to the total energy consumption) can vary
long payback periods, higher investment requirements, several significantly. According to Agathokleus et al., the largest amount
additional requirements in terms of materials and design due of waste heat is found in food and tobacco processing, pulp and
to chemical activity, and corrosion, and as such they have been paper processing, basic metal industries, chemical industry, and
introduced in specific situations without wider diffusion. nonmetallic mineral processing, and 50% of waste heat is in the
Nevertheless, developing countries (hosting more than 4 bn peo- temperature range between 300 and 350
C.[51] Metallurgy and
ple) are experiencing an ongoing industrialization process and nonmetallic mineral transformation industries in the USA are
can take advantage of experiences and technologies developed wasting 20–50% of the energy used. McKenna and Norman con-
in more advanced countries. For sure, the industrial WHR/ clude that the recovery potential for energy-intensive industries in
WHP will definitely be under spotlight for the next years.[4] the UK is around 10% of the total heat quota.[64] Waste heat poten-
In Figure 7, several industrial processes are associated with tial in the EU has been estimated to lie above 350 TWh year1, the
the corresponding temperature range, whereas the complete greatest part falling in the 100 and 200
C range. Waste heat below

Figure 7. Temperature ranges for different industrial heating processes.[5]

Figure 8. Waste heat potential in each EU country per year in all industries. Reproduced under the terms of the 4811241236352 license.[63] Copyright
2020, Elsevier.

Energy Technol. 2020, 2000413 2000413 (9 of 22) © 2020 Wiley-VCH GmbH

www.advancedsciencenews.com www.entechnol.de

Table 1. Waste heat potential percentage per industry sector in Europe.[63] which has been hosting an ORC plant since 1996, updated in
2018, and producing 690 kW; Toscelik Hot Strip Mill in
Industrial sector Heat consumption [TWh] Waste heat [%] Turkey producing 1 MW since 2011; NatSteel in Singapore with
Iron and steel 580 11.4 an installed plant producing 555 kW since 2013; Elbe-Stahlwerke
Feralpi in Germany producing 2.7 MW since 2013; ORI Martin
Chemical and petrochemical 600 11.0
in Italy producing 1.9 MW since 2016; the Munksjö pulp mill
Nonferrous metal 115 9.6
in Sweden producing 750 kW since 2010; and the glass
Nonmetallic minerals 390 11.4 manufacturing plants of Sisecamin in Ostellato, Porto Nogaro,
Food and tobacco 345 8.6 and Manfredonia in Italy, sharing a capacity of 5 MW since
Paper pulp and print 395 10.6 2016 for each of the locations. A cement plant in Lengfurt,
Wood and products 100 6.0 Germany, burns around 3150 tons of clinkers per day in a rotary
Textile and leather 50 11.0
furnace at flame temperatures of about 2000
C. This generates
hot furnace exhaust gases (350
C, heat flux at around 8 MW) and
Other 660 10.4
waste heat from the clinker cooler (275
C, about 60 MW). Part of
this heat is injected back into the furnace. Until the ORC system
was not installed, the remaining part of about 30% was wasted
100
C represents a marginal portion, whereas within the 200 and into the atmosphere. In Figure 9, the schematic of an example of
500
C range more important quantities are found.[63] This is an ORC installation is shown.
important amount of energy compared with the 3218 TWh energy KCs can recover industrial waste heat in the range from 80 to
consumption of 2016.[63] Papapetrou et al. developed statistics 400
C for power generation. Existing industrial WHR applica-
related to the waste heat potential in percentages per industry sec- tions include Sumitomo plant in Japan with an installed plant
tor for all the former countries belonging to EU28.[63] The results producing 3.5 MW since 1999; Husavik, Iceland, plant convert-
are shown in Table 1. ing 2.0 MW since 2000; Fuji Oil plant in Japan converting
4.0 MW since 2005; DG KHAN plant in Pakistan converting
4.3. Technology Overview 8.6 MW since 2013; and Star Cement plant in Dubai producing
4.75 MW since 2013. KCs are limited in their industrial applica-
4.3.1. Thermodynamic Machines tions for some technical issues that have not been solved yet, in
particular, for fine tuning the boiler evaporation ratio and the
To date, there has been a consistent number of ORC manufac- early condensation of the NH3–H2O mixture.[1]
turers (Turboden, Opcon Powerbox, Orcan, and EXERGY) and Solid oxide fuel cells (SOFCs) are considered emerging tech-
commercial applications in industry sectors. Leaving to the nologies for both small and large power plants for high-efficiency
reader the disclosure of the manufacturer through the references power generation with low emissions. A comparative analysis in
given as examples about ORC units operated by iron and steel terms of energy and exergy has been conducted, to use ORC and
foundries. there are the Fonderia di Torbole (Brescia) in Italy, KC for WHR from hybrid SOFC/GT systems. The results show

Figure 9. Schematic of ORC installation in an industrial plant.[65]

Energy Technol. 2020, 2000413 2000413 (10 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

good motivation for using ORC or KC for WHR to increase the and complexity, vast scalability, robustness, rapid response time,
global energy-conversion efficiency. The results indicate the lack of toxic operational emissions, long reliability and lifetime,
superiority of ORC over KC.[66] A comparison of performances inertness, and lack of moving components.[27,37] Many countries,
between KC and transcritical ORC for WHR in different ICE such as the USA, UK, Australia, Ukraine, Japan, and South
working conditions is done. Compared with KC, the transcritical Korea, are focusing their attention on this field[70,71] and several
ORC shows evident advantages on the overall thermodynamic researches have been already accomplished to characterize the
efficiency, low operation pressure, and simple component con- TEGs’ performances under various heat sources like diathermal
figuration with exhaust temperature from the engine of over oil heaters, cook stoves, and waste heat from industrial process.
220
C. The optimal thermal performance of the transcritical The Aerospace, Mechanical and Manufacturing Engineering
ORC is in the range 295–345
C. Nevertheless, when moving School at the RMIT University of Melbourne, Australia, intro-
over or under the optimal temperature drop, the performances duced the concept of the hybrid power-generation system.[72]
decrease considerably. Moreover, the extremely high expansion This is composed of one TE module held by two copper layers
ratio of the turbine requires a complex multistage design and with four heat pipes generating the temperature difference
large dimensions.[67] Another comparison between the thermo- between the two TEG sides.[72] Another application of TEG mod-
dynamic performances of KC and ORC has been conducted for ules is in support of the industrial wireless sensor networks
the case of heat recovery from two diesel engines, each one with (IWSNs), clusters of wireless sensor nodes in industrial plants
an electrical power of 8900 kWe. Supposing a mean temperature and commercial areas. They consist of a high number of sensor
difference in the heat recovery exchanger of 50
C, a net electric nodes, including sensor units and control electronics, communi-
power of 1615 kW and of 1603 kW can be generated, respectively, cation, and power supply. Most of these are powered by means
for KC and ORC. Although the output levels are almost equal, KC of batteries having strong limitations in terms of energy because
requires a much higher maximum pressure to obtain high per- of the constrained node and battery size. Therefore, several
formances. Then, at least for a low power level and MT and HT researchers explore the possibility of coupling IWSNs with
thermal sources, KC appears to be unjustified because the TEGs to harvest energy and extend batteries’ lifetimes, achieving
increase in performances is very small (with respect to a properly higher power densities with respect to other energy harvesters. It
optimized ORC) and must be obtained with a more complex has been concluded that with 0.6 m s1 of air flow on the TEGs
plant scheme, larger surface heat exchangers, and particular and without using any heat sink, the hot side being at 50
C and
plants featuring pressure and corrosion resistance.[68] the cold one at 20
C, the output power is around 80 μW. Dalola
In the past decade, the development of innovative thermody- et al. used a TE component to supply an autonomous sensing
namic cycles has been fundamental for the efficient utilization node for T measurement.[73] Bonin et al. produced and charac-
of LT heat sources such as solar, geothermal, and waste heat terized small TEGs used to power systems for environmental
sources and many researchers have started to explore these monitoring, in the order of 10 mW for a temperature difference
technologies. Padilla et al. analyzed a hybrid power/cooling of 10
C.[74] In another application domain, 52% of Australian
cycle, which combines RC and absorption refrigeration cycles, households’ hot water heaters are powered using natural gas
using ammonia–water mixture and generating power and cool- and liquified petroleum gas (LPG) systems. By exploiting the
ing at the same time. Also known as the Goswami cycle, it can thermal gradient generated between hot exhaust gas exiting
be used as a bottoming cycle using waste heat from the conven- and cold water from ducts, it would be possible to generate power
tional power cycle or as an independent cycle using solar or geo- by means of TEGs. Through the testing conducted, a power gen-
thermal energy. The results show that with a heat source eration unit including 60 TE cells would be able to produce more
temperature between 90 and 170
C and the absorber tempera- than 40 W and more than 20 W in case hot water is around
ture of 30
C, the maximum First Law efficiency (ratio between boiling temperature and 80
C, respectively. In the latter case,
the net output power and the heat absorbed from the HT source the efficiency of the system is estimated to fall in the range
to produce it) and exergy efficiency are estimated to be 20% and 0.37–1.03%.[75]
72%, respectively.[69] Furthermore, some hybrid concept and devices have been ana-
Nowadays, the number of commercial applications is very lim- lyzed. Gholamian et al. focused on enhancing the performances
ited, even though industrial waste heat is available in abundance of a geothermal-based ORC and proposed two novel systems in
and the concept of utilization or recovery is not new. The reasons which, in the first case, part of the waste heat is recovered using a
are mainly related to resource constraints and the insufficiency of TEG for power generation whereas in the second for hydrogen
regulatory, organizational, and business plans.[1] production (by means of a proton-exchange membrane electro-
lyzer). Results indicated that the proposed system has exergy effi-
4.3.2. TE Devices ciencies higher than that of the basic ORC by 21.9% and 12.7%,
respectively. They also reveal that the specific product cost for the
Several industrial companies have focused on TE technologies as proposed solutions is lower than that for the basic ORC, despite a
these devices can compete with fluid-based systems, like com- higher total cost rate.[76] Zare et al. studied the possibility of using
pressors or heat pumps. Furthermore, solid-state energy conver- TEG to recover waste heat from a KC and the results revealed an
sion is more appealing because of its simplicity, if compared with improvement of around 7.3% of the net output power and higher
compressing or expanding two-phase fluid systems. Although energy and exergy efficiencies. Furthermore, they evaluated eco-
their limited efficiency, several commercial uses have been real- nomically the integration of TEG with the KC, indicating modest
ized. The reasons behind this success are the relatively small size profitability.[77]

Energy Technol. 2020, 2000413 2000413 (11 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

4.3.3. Novel Approaches 5. Automotive Applications

Recently, the innovative method for low-grade WHR based on 5.1. Introduction
nanogenerators has been deeply investigated and, even though
a large margin for improvements is still present, some interest- Road freight vehicles and passenger cars play an essential role in
ing results have been reached. To be mentioned is the work of today’s global economic activity, energy consumption, and envi-
Raouadi et al.,[78] where a hybridized pyro-vortex device, realized ronmental impact. According to the International Energy Agency
in polyvinylidene fluoride (PVDF), was realized, with an area of (IEA), transport accounts for more than half of the oil global
2 cm2. The idea behind this work is to exploit the energy of the demand, totalling around 52 million barrels of oil per day
wind to realize an inexpensive and efficient harvester providing (mb d1) (Considering Barrel of Equivalent Oil (BOE), it is pos-
an uninterrupted energy output. In particular, the vortexes gen- sible to say that the transport sector has an energy demand of
erated by this give rise to two conversion mechanisms: molecular 88 TWh day1).[83] Furthermore, IEA estimates that the global
transport, proportional to the gradients of the averaged velocity of fuel consumption for high-duty vehicles is increasing more rap-
the wind and heat fluxes, and convective transport, which gives idly than any other transport mode, especially in emerging
rise to turbulent stresses caused by the momentum transfer pro- markets such as India, China, and Latin America.[84] In view
duced by velocity fluctuations. In fact, this device allows to pro- of the Paris Agreement, countries have been considering differ-
duce a stable output current, reaching a maximum power density ent solutions to secure improvements in transport efficiency
of 2.82 μW cm2. and emissions reductions, both for light-duty vehicles (LDVs)
The flexible device developed by Chen et al.[79] is based more and heavy-duty vehicles (HDVs). In response to this energy and
on mechanical solicitation instead of temperature variation. In environmental challenges, the automotive industry is focusing
this case, the mechanisms of conversion are piezoelectric and on high-efficiency vehicles. According to the Global Fuel
triboelectric, i.e., relying on deformation and friction processes, Economy Initiative objective, within 2030, LDVs should improve
respectively. The energy to be harvested can be associated with the fuel economy by 50% and save 0.5 Gt of CO2 emissions per
different sources and the device results as one suitable for many year.[83] Nowadays, the principal path toward sustainable trans-
portation is the electrification and hybridization of automotive
applications. As an example, it can be thought as a device to be
powertrains, while the industry is also focusing on different tech-
inserted in the floor or, thanks to its dimensions, for biomedical
nical solutions to increase the fuel efficiency of ICE vehicles.
applications supplying sensors for health monitoring, converting
the energy coming from the motion in electricity, e.g., placing it
under the shoes. The maximum power produced in this case 5.2. Mechanical and Thermodynamic Analysis
reached 0.834 μW cm2.
A combination of PE and triboelectric hybrid energy harvest- Sovran and Blaser[84] proposed a complete physical analysis of
ers has been also proposed in the work of Jiang et al.,[80] where fuel consumption in an automotive system, indicating that
the idea is to harvest useful energy from low-grade waste fluids. 58% of the traction power at the wheel is lost due to vehicle iner-
In practice, drops of fluid can be made to fall onto the device tia in an urban driving condition. On the contrary, aerodynamic
surface and energy can be recovered, exploiting both their ther- losses become dominant in highway driving conditions (50% of
mal and kinetic energy. In fact, the drop hitting the surface will the traction power at the wheel). The US EPA City and Highway
release its energy on it, provoking friction and thus triboelectri- Driving Cycles provide an energy loss analysis based on energy
fication to take place, but also have a certain temperature that will utilization in different driving conditions and different size-mass
provoke temperature variation, allowing pyroelectrification. The vehicles (Figure 10). It is interesting to note a loss in fuel energy,
maximum power produced per unit area is similar to the one of around 35% and 30% in exhaust and coolant[85,86] gases,
proposed by Rouadi, lying around 2.6 μW cm2. Interesting is respectively, representing more than half the losses in wasted
to notice that all these devices have been realized in PVDF, which heat form.
seems to be a promising material in LT waste heat harvesting, Nowadays, engine manufacturers concentrate on two different
and are in the solid state. ways to increase the thermal efficiency of an ICE. The first one
Two innovative works propose a liquid-state low-grade waste focuses on improving the thermodynamic cycle of the engine,
heat harvester. The idea is to combine solid and liquid properties, operating on the choice of fuel, combustion system, and air path.
realizing a stable suspension of nanoparticles in a carrier fluid, For example, downsized engines deployed in production vehicles
and move this mixture by means of the combination of thermal improve the fuel economy by 8–10% for LDVs in urban path-
and magnetic flux gradient. This thermomagnetic motion will ways.[87] The drawback of this approach consists of a reduction
allow the nanoparticles to crawl on the walls of the container, of available power and an increase of fuel consumption, espe-
provoking charge accumulation by means of a triboelectric pro- cially in highway conditions.
cess,[81] and experience a temperature variation in time, provok- The second trend concerns WHR and WHP. WHR has the
ing pyroelectrification.[82] The device has been tested only in the important feature to not affect the engine design and technical
laboratory and still needs some improvements to be reproduced features of the vehicle, as it uses the heat rejected by combustion.
in a large scale. However, the results are encouraging and the Therefore, it is suitable for different driving conditions and it can
possible applications are various as the device, until now, is pro- be easily integrated. However, the key point of this technology is
posed in two shapes, i.e., toroidal or planar, and can be applied on to guarantee competing requirements of costs, engine durability,
whatever hot surface. and emissions.[88]

Energy Technol. 2020, 2000413 2000413 (12 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

Figure 10. A Sankey diagram showing the fuel energy utilization of an ICE.

By considering detailed models and simulations of the ther- area of investigation to exploit LT recovery.[95] The low mass flow
modynamic behaviour in an ICE,[89,90] it is possible to conclude rate of the working fluid can guarantee the realization of smaller-
that the exergy loss (or availability loss) mainly originates by irre- sized recovery systems.
versible combustion processes and the energy is mainly con-
veyed by exhaust gases. This suggests that the best option to 5.3.1. Organic Rankine Cycle
maximize energy recovery is to install the recovery plant in close
contact with the exhaust gas ducts or the cooling system. In the automotive research field, all the bottoming cycles
A further important feature is thermal availability or rather described in the previous section (ORC, SC, KC, and BC) are
entropy maximization, representing the maximum amount of under investigation. In particular, ORC is the one most econom-
work necessary to establish equilibrium in the system.[91] This ically advantageous for WHR when the thermal source is below
concept is closely related to heat grade (temperature). While about 150
C.[96] It is also very attractive for mass production,
HT and MT conditions are widely used for practical power gen- because of the low cost, the easy integration with the cooling sys-
eration, LT sources represent today’s challenge in the automotive tem (where the working fluid temperature ranges from 60 to
industry. LT sources are mainly localized in the cooling systems 90
C), and the low impact on engine design. The industrial
(radiator and air conditioning) and in exhaust gases exiting other groups that are currently investing in research and development
recovery devices.[92] in this area are BMW, General Electrics, Cummins Engine, and
United Technologies Corporation.
5.3. Technological Overview An ORC permits exploiting LT sources in a very efficient
way.[97] In Figure 11 it is possible to observe a standard configu-
From a historical point of view, the first approach adopted in ration for automotive applications. The design challenge consists
WHR for the automotive field is the mechanical turbocompound- of the miniaturization of the expander machine[99] which oper-
ing (MTC). This technique uses a turbine to extract energy from ates with organic fluids.[100] This is important to reduce the
HT/MT thermal energy (from 400 to 600
C[93]) using the com- weight-to-power ratio of the dynamic machine,[101] decreasing
bustion gas streams. At present, Detroit Diesel, Iveco, Volvo, and the inertia of the vehicle. The operation conditions of the
Scania truck manufacturers produce engines that make use of ORC can be divided into in direct and indirect. In the direct
this recovery technique. MTC technology is essentially based recovery method, it is possible to use the engine block as the heat
on the concept of additional heat by a second exhaust turbine exchanger, avoiding specific heat conversion parts of the cooling
from a turbocharger: instead of ejecting the excess of energy system.[102] This means that a higher T can be exploited in the
via the exhaust pipe, it is possible to extract additional heat using WHR system; hence, it is possible to use more heat rejected by
a second exhaust turbine. The rotational motion passes through a the ICE. Other studies show how the heat dissipated by the
gear system and a hydraulic coupling, boosting the torque avail- engine block can be coupled with the vaporizer cooling system.
able on the crankshaft. This permits having extra driving force Doing that, it is possible to recover about 3% of output energy of
without additional fuel expenditure. Several studies show that the engine.[103] In indirect harvesting procedures, heat is trans-
MTC improves the thermal efficiency from 3% to 5%.[94] The ferred from the coolant to the ORC working fluid using an extra
main disadvantage of this technology is that MTC design expects heat exchanger. Several studies[104–107] show how the heat dissi-
the interaction of the recovery system with the engine, generat- pated by the coolant fluid can be used for the evaporation system
ing an exhaust back pressure. This makes it suitable only for the or as a preheater for a HT Rankine–Hirn cycle. In this second
HDV industry. The ICE cooling system is the most promising case, it is possible to reach higher efficiencies,[108] but the system

Energy Technol. 2020, 2000413 2000413 (13 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

Figure 11. Model of ORC in an automotive system. Reproduced under the terms of the Attribution 4.0 International license.[98] Copyright 2017,
Springer Nature.

is heavy and cumbersome and therefore economically unfavour- emissions. BMW, Nissan, Jaguar, Land Rover, and Porsche
able for automotive applications.[93] Other approaches[109] involve are some of the manufacturers investing in this field. Exhaust
the use of the heat rejected by the cooling system for the conflu- TEG (ETEG) is the most promising technology, because it has
ent cascade expansions of the ORC system. no moving parts, it requires less maintenance in comparison
The matching conditions between working fluid and ORC sys- with ORC and MTC systems, and it is silent in operation.
tems is investigated to make the cooling system absorb the whole Despite the light weight of ETEG and its versatility, technology
waste heat. In LT recovery systems organic fluids with a low boil- must address a number of issues such as low efficiency (5–10%),
ing temperature and low freezing temperature are used. In fact, high cost, and large temperature differences required for consid-
these characteristics allow the absorption of the low T heat in the erable energy conversion. In particular, the cold side of the ETEG
RC system. However, it is difficult to keep a high mass flow, as module should be at a temperature as low as possible to maxi-
organic fluids have inherently a low stability at HTs and low spe- mize efficiency. Therefore cooling mechanisms are required to
cific heat. One option to increase the efficiency is sizing a larger provide the maximum temperature gradient.
recovery system (as in industrial plants). Another solution could be In a typical ICE, the ETEG is composed of four units: a TEG
using high-specific-heat fluids like water. Other important features module, a heat sink, a heat exchanger, and a power conditioning
that affect the recovery efficiency are thermal fluctuations, due to unit (PCU). An example is shown in Figure 12. Exhaust gases
changeable driving conditions. Jimenez-Arreola et al. give a gen- pass through the heat exchanger placed on the exhaust pipe,
eral overview on these effect. This phenomenon negatively affects causing an increase in temperature on the warm TEG module
the WHR systems, limiting the operational range and inducing an side. The other side can be refrigerated using a liquid or an
efficiency drop during partial load. A possible solution is repre- air-cooled heat tank system. The generated electric power due
sented by stream control, used to ensure a close and safe optimal to the temperature difference established across the TEG module
point operation.[110] Pili et al. analyzed the economic impact of is stored in a battery, matched with the vehicular electrical system
ORC technology in the transportation sector. They considered through a PCU.[113,114] This latter is essential to guarantee the
the relationship between WHR that impacts the net fuel economy maximum power transfer in different driving conditions (idle
of the vehicle and the costs of additional volume that the ORC or operational).
system occupies, affecting the vehicle transportation capacity.[111] Komatsu (Japan), Marlow (USA), and HiZ (USA) produce and
commercialize TEG modules using Bi2Te3 as active material. In
5.3.2. Thermoelectric Generators particular HiZ produces several devices that work with a maxi-
mum temperature of 250
C and ΔT of 200
C, producing from
Another possible way to improve vehicle efficiency using the 2 to 20 W of electrical DC power. The cost of these devices can
WHR system is to exploit the TE effect. TEGs allow to recover range from 18 EUR to 80 EUR per module, depending on dimen-
heat from ICE to reduce fuel consumption, without increasing sions and on purpose.

Energy Technol. 2020, 2000413 2000413 (14 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

healthcare[121] and self monitoring of activities and vital func-

tions,[122,123] the production of electric energy from human body
waste heat has become significantly important because of the rise
of ultralow-power (ULP) devices.[124] Despite the development of
new supply technologies, researchers have started to explore a
possible definitive solution, battery-less devices integrating an
energy-harvesting system where mechanical load, vibrations,
temperature gradients, heat, light, salinity gradients, wind, etc.
are scavenged and converted to obtain relatively small amounts
of power. Examples of these kinds of devices are thermo/tribo/
piezo/pyroelectric effects. Of course wearable devices are not
comparable with industrial or automotive applications in terms
of energy produced, but still they are part of categories of tech-
nologies able to convert waste heat into power, which, moreover,
can be explored in the sports and space application field, where
Figure 12. Example of ETEG design. Reproduced with permission.[112] small amounts of power are enough to power smart devices.
Copyright 2010, General Motors Global Research & Development.

6.1. Thermal Energy Harvesters

It is important also to focus on the installation position of the
TEG device, as the heat exchange efficiency plays a fundamental According to a previous study[125] the energy requirements
role. In general, it is possible to adapt both size and shape of the of an average person under 60 years are in the range of 8.8–
TEG module according to the recovery application, for example, 17.6 MJ day1 (man) and between 7.1 and 14.2 MJ day1
in the case of exhaust pipes, a square configuration can be chosen (woman). (The numbers reported here are obtained for a subject
for flat surfaces or a linearly shaped one for circular pipes. Liu at rest (basal metabolism rate].) Therefore, assuming that on
et al. tested several positions of the ETEG, finding that for LT average the daily energy need is about 10 MJ and the correspond-
applications the optimal choice is positioning modules after ing power is 120 W, considering to radiate a large part of this
the catalytic and filtering blocks of the exhaust system.[115] energy in the form of heat, a thermal source of some tens of
An interesting experimental study[116] shows that while using Watts, at rest, is available.[126] Therefore harvesting even a small
six Bi2Te3 TEG modules on a 3696 c.c. gasoline engine that has part of this energy would enable powering wearables such as
hot and cold sides at 225 and 255
C, respectively, it is possible to smartwatches that have a few tens of mW to operate.[127]
generate an output power of more than 40 W. In accordance with Assuming a normal condition, that is, a body temperature of
other researches[117,118] the fuel economy of a vehicle can be 36.6
C[128] and a room temperature of 27
C, it is possible to esti-
improved up to 2% using LT exhaust gases. mate a Carnot efficiency of 3%, which makes immediately clear
A different approach for WHR is proposed by Park, Yoo, and the constraints for the application on a human body. However,
Kim (2010).[119] As in the ORC case, they recover energy from the even 1% of a total power in the order of 100 W still represents an
engine coolant, substituting the radiator system with a system amount of energy sufficient to power wearables.
of 72 TE modules (Bi2Te2 with an area of 4 mm  4 mm) and
128 heat pipes. The coolant flows inside a chamber where the
6.1.1. Rigid and Flexible TE Devices
squared modules are installed, transferring heat to the hot sur-
face of the material and reaching 95
C. To keep the temperature
The wearable TEGs (WTEGs) for LT waste heat scavenging do
of the cold side at 25
C, a standard air-cooling system has been
not use bulky material but rather thin layers of material. For this
used. Results show that it is possible to generate almost 30 W of
reason they are typically referred with the name μ-WTEG.
electrical DC power during the idle condition and almost 75 W
Initially, rigid substrate TEGs have been the focus of research.
during the driving condition. This approach allows to recover
Leonov et al. published a study on TEGs, aiming at finding a
0.4% of wasted heat.
smart way to maximize the energy scavenged, resulting in the
At present, the cost of commercialized TEGs is in the range of
employment of a multistage thermopile.[129] Torfs et al. proposed
2500–3500 EUR kW1. To be competitive, the cost per energy
a completely self-powered device to measure the oxygen content
capacity has to decrease at 395 EUR kW1.[120] Nowadays only
in blood.[130] It consists of an oximeter for the measurement of a
HDVs can benefit from TEG systems, as it is easier to depreciate
wristwatch, disguising the TEG and providing the energy needed
installation cost, considering both the driving conditions (long
(an explicative scheme of the whole apparatus is shown in
driving distances) and the higher amount of energy from exhaust
gases with respect to LDVs. Figure 13).
The maximum power extracted is more than 100 μW during
daily activities. A year later, the same research group proposed an
6. Wearable Applications electroencephalography (EEG) system.[132] A battery-less device
specific for EEG, electrocardiography, and electromyography data
During the past decades, technology has performed steps in transfer was investigated by Zhang et al. few years later.[133]
the direction of an extreme miniaturization of electronic More recently, the work presented by Wahbah et al. has opened
components and portable devices. In addition to devices for a new way toward self-powered systems.[134] They characterized

Energy Technol. 2020, 2000413 2000413 (15 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

Figure 13. Fully powered wireless pulse oximeter:[131] on the forefinger a

commercial sensor has been placed. The wristwatch TEG charges a stor- Figure 14. Scheme of the transition from the rigid to flexible device substi-
age capacitor able to provide 0.7–2 V as output voltage. A DC/DC con- tuting the ceramic plates with liquid bendable connection. Notice that the
verter to ensure stable and higher Voc is mounted after the capacitor legs are the same of the rigid device. The final step consists of an encap-
itself. A V ¼ 2.05 V is obtained, which is enough to supply the wireless sulation in polydimethylsiloxane (PDMS). Reproduced with permis-
sensor. Reproduced under the terms of the 4877660286134 license.[131] sion.[136] Copyright 2017, Elsevier.
Copyright 2020, AIP Publishing.

6.1.2. PE Energy Harvester

two different devices, a TE and piezoelectric energy harvester, to As already pointed out, heat constantly flows out from our
find the best way to supply the system. They showed that a 9 cm2 body because of self-thermal regulation, i.e., the homeostasis
TEG system is able to generate up to 20 μW at RT with respect to process through which the body is able to maintain always an
3.7 μW of a 5 cm2 piezoelectric one. By properly setting the param- external temperature around 36
C, in a range of about 1
C.
eters, they reached V o ¼ 12 mV and V oc ¼ 24 mV for a ΔT of Moreover, the physiological rate of temperature changes is
0.5
C, i.e., output and open-circuit voltages respectively. The main not so fast. Thus, it is clear that the efficiency of the devices
drawback of rigid WTEGs is the incapability of following the cur- exploiting PE is so low which makes it unsuitable for body
vature of human skin, with concurrent heat transfer losses. WHR applications.
Moreover, the integration of rigid WTEGs into clothes is not easy Nevertheless, Yu et al.[139] conducted a study on the possibility
and makes them uncomfortable. For these reasons flexible devices of using nanogenerator technology as a self-driven power supply
have substituted the rigid ones as the focus of research. for wearable devices. Here, it resulted in a combination of more
Concerning wearable applications, the problem of comfort is effects, in particular triboelectric and PE that could be a good
not the only one. The LT difference deeply affects the overall effi- solution for solving the problem of powers supply. The perform-
ciency of the devices. Thus, to solve this issue, an interesting way ances of such hybridized nanogenerators have been investigated
to increase was explored by Jung et al.[135] They proposed using a in a previous study.[140]
local absorber placed below the TEG legs (BiTe and SbTe) onto Potnuru et al. proposed a mouse supplied using human heat,
such layers. The temperature difference passes from the usual exploiting just the PE effect.[141] They found that the current gen-
2 ÷ 4 to 20
C (Figure 14). erated by the PE effect is so low that it is completely dominated by
A prototype consisting of ten p–n couples develops an output the one due to the piezoelectric effect, which is also intrinsic of
of around 4.44 μW and an open-circuit output voltage of those materials that have a PE effect, provoked by the weight of
55.15 mV. However, it is possible that such a local increase in the user’s hand. Sultana et al.[142] investigated a commercial PE
temperature could result to be uncomfortable for the user but device which is composed of a piezoceramic foil placed between
open doors for possible future implementation in WTEGs. two conductive electrodes. They prove the possibility of using
It is important also to mention the work proposed by Qian water vapour to drive the device obtaining an output voltage
et al.[137] that tried to look for another way to integrate the harvest- higher (1.6 V) than that in other conditions (0.4 V), thanks to
ing system directly into the clothes’ fibers. They have analyzed the increase in T difference. The maximum output power
closely the textile TE material from both the application (moni- obtained is 0.034 μW cm2.
toring, computing, WHR) and production (fabric finishing and Hybrid systems exploiting piezo and pyroelectric effects have
coating or fibers/yarns coating technologies) points of view. been studied.[143] A very interesting hybrid device was proposed
These can be very good alternatives to other kinds of devices, recently by Ding et al. implementing a thermocell and a PE har-
thanks to the ease of integration, comfort, and air permeability. vester which is used to recover heat by both stable gradient and
Its structure is different as it does not rely on a sandwich design temporal variation of temperature.[144] This device consists of two
but on a planar one. This implies that the temperature difference layers of poly (vinylidene difluoride) (PVDF), coated with CNT/
is applied along the length and not the thickness of the leg for CNC nanocomposites, that serve as the electrode and absorber,
stability reasons. However, they can be arranged in either a 2D and a block of polyurethane (PU), which is placed under the two
(Figure 15) structure or 3D one (Figure 16). PVDF films. The former deals with heat recovery from T

Energy Technol. 2020, 2000413 2000413 (16 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

Figure 15. The arrangement of 2D textile thermoelectric device. a) A multilayer arrangement proposed by Hewitt et al. is reported. It uses n- and p-type
carbon nanotubes (CNT) separated by insulating PVDF films. b) A yarn structure proposed by Ryan et al. is shown. This is a silk yarn made of PEDOT:PSS
which ensures stability in performances also after washing and drying. Reproduced under the terms and conditions of a Creative Commons Attribution CC
BY 3.0 license.[138] Copyright 2018, The Authors. Published by IntechOpen.

fluoride (PVDF), of dimensions 35  35 mm and thickness

30 μm, in a mask. Exploiting the heat released by the breadth,
the maximum temperature variation registered on the harvester
was 12
C and the maximum power extracted, with an external
load of 50 MΩ, reached 0.68 ΩW cm2.
Finally, it is worth to mention a work proposed by Cha et al.
dealing with liquid-based PEGs. They proposed a harvester based
on the PE effect but using a liquid-based switchable inter-
face.[146,147] The structure of their device is shown in Figure 17.

Figure 16. Arrangement for a 3D textile thermoelectric generator

7. Conclusions
proposed by Kim et al. This kind of structure allows to sustain a tempera-
ture gradient along the thickness contrary to 2D structures.[138] Intech
The identified technologies have been collected and sorted
Open under Creative Commons Attribution License by Qian et al.[137] according to their typical application and optimal temperature
The images has been originally published in and therein references. range in Figure 18. A comparison in terms of efficiencies,
Available from: http://dx.doi.org/10.5772/intechopen.75474. post-installation costs per kW, and typical installed power output
is shown in Table 2.
In the earlier table, data relative to CDTCC are not given at MT
variation, which is the PE part, whereas, once the temperature due to the intrinsic properties of CO2, implying that the cycle
gradient is established in the thermocell, a redox of ferrous/ferric operates at cryogenic temperatures. The TE efficiencies have been
chloride acid develops and converts it into useful electric energy. computed using conventional physical models and assuming rea-
Moreover, an analysis on performance versus the content of sonable temperature ranges in operating conditions (55
C of ΔT
CNTs in CNCs is performed, resulting in the verification of for LT and 150
C of ΔT for MT) and an average ZT FoM of 1.5
the crucial role played by CNCs. The thermogalvanic cell has for LT and 2.5 for MT (on the basis of the previously reported
shown a current density production of 1.2 Am2 for a ΔT of analysis). For PE efficiency, data have been derived by a previous
100
C. Assuming an illumination of AM 1.5G, the PVDF film study[43] and the same temperature ranges as above have been
generates up to 200 V of output voltage under illumination and considered. [148]a [149]b [150]c [151]d [152]e [151]f [33]g [154]h
120 V once the illumination is removed. The importance of this [155]i [156]j [157]k [158]l [159]m [39]n.
device lies mainly in its versatility as it can harvest energy in dif- The reader should never forget that closed-loop thermodynamic
ferent conditions, thus being able to act as a supply 24/7. The cycles based on any specific fluid feature intrinsic installation costs
hybrid has shown a maximum output power produced, of that strongly limit their power output lower bound, whereas purely
1.86 and 0.9 mWm2, for the thermogalvanic and PE compo- solid-state systems such as TEGs and PEGs suffer from the spec-
nents, respectively. ular limit, toward the upper bound. Once the available WHP tech-
Recently, a wearable PE nanogenerator based on the breathing nologies have been analyzed, the question arises on whether they
process has been proposed in the work of Xue et al.,[145] where could represent a valid approach to increase the quota of “green”
they successfully introduced a thin film of polyvinylidene installed power, working in conjunction with both fossil fuel-based

Energy Technol. 2020, 2000413 2000413 (17 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

processes and renewable resources-based processes. Considering

their costs per kW and comparing them with photovoltaic systems
(on average the cost per kW is 1.4 kEUR kW1 as of December
2018), two technologies are already competitive: BC and ORC.
On the other side, the higher production costs of wind turbines
(on average 1.4 kEUR kW1 as of 2017 for onshore plants
and 3.5 kEUR kW1 for offshore ones), of biomass-based power
plants (3.3 kEUR kW1), and of hydroelectric power plants
(3.6 kEUR kW1) give already a strong evidence of the bright
future this sector will experience, as at least some, if not all, of
the technologies here described could be potentially competitive.
Parallel to industrial development and technology readiness
level, the most statistically relevant results in all ranges of
reviewed current solutions mainly involve optimization through
thermoeconomic, exergetic, and LCA analyses, as well as
material development. A totally different approach is seen for
emerging technologies, where basic studies and theoretical
developments are frequent.
In conclusion, WHR/WHP technologies have reached a matu-
rity level that could significantly boost their mass application in
close relationship with industrial processes and wherever waste
heat is generated, such as in the automotive field, regardless of
the policies enforced at the local/global level. Scientific research,
Figure 17. Image of the liquid-based thermal interface device structure
proposed by Cha et al. in a previous study.[146] Reproduced with permis- on the other side, is rapidly expanding and casting light on
sion.[146] Copyright 2012, Elsevier. less explored multidisciplinary fields, where combined effects
(such as pyro þ thermo þ magneto) have been shown to greatly

Figure 18. Most of the technologies are analyzed and grouped according to their application field and optimal temperature range of operation. Note that
this graphical summary is conceived only to give an immediate, qualitative glimpse on the statistics of technology application. For this reason, the reader
should focus on the temperature ranges rather than on the specific temperature values.

Table 2. Efficiency and installation cost/kW comparison between the mentioned technologies.

BC SC KC CDTCC ORC TE PE

η @ LT (T Hot ¼ 20  200 °C) 0.40a 0.13 0.10d 0.15f 0.12g 0.037 0.15
η @ MT (T Hot ¼ 200  500 °C) 0.40a 0.20 0.15d – 0.20g 0.073 0.13
1
Cost per kW [kEUR kW ] 1b 12–2.5 6.5  2.5e N.A. 2.8  1.2h 2.5 N.A.
Installed power [MW] 7.8i 0.0015j 0.4k 0.0046l 0.35m 0.012n N.A.

Energy Technol. 2020, 2000413 2000413 (18 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

enhance conversion efficiencies and volume/mass specific [17] O. Gavaldà, V. Depoorter, T. Oppelt, K. van Ginderdeuren, RenewIT
power/energy. In particular, the most promising directions to FP7 Deliverable D4.1 – Report of Different Options for Renewable Energy
explore are 1) magnetocaloric and thermomagnetic hydrody- Supply in Data Centres in Europe, 2014.
namic machines on the one side and 2) TE, triboelectric, and [18] J. Holman, IDC Energy Insights: Renewable Energy Strategies:
PE devices on the other side. The former group is suitable for Perspective, 2011, pp. 1–11.
[19] A. Gupta, A. Singh Bais, Waste Heat to Power Market Size By Product
small-scale systems and the latter suitable for wearables.
(Steam Rankine Cycle, Organic Rankine Cycle, Kalina Cycle), By
Application (Petroleum Refining, Cement Industry, Heavy Metal
Production, Chemical Industry, Pulp Paper, Food Beverage, Glass
Acknowledgements Industry, Others), Industry Analysis Report, Regional Outlook (U.S.,
Canada, Mexico, Germany, UK, Italy, France, Belgium, Spain, Russia,
The support of Fondazione Istituto Italiano di Tecnologia is gratefully
China, Australia, India, Japan, South Korea, Philippines, Thailand,
acknowledged.
Vietnam, UAE, Saudi Arabia, South Africa, Brazil, Argentina),
Application Potential, Competitive Market Share & Forecast,
2019–2025, 2019.
Conflict of Interest [20] A. Chiolerio, M. B. Quadrelli, Energy Technol. 2019, 7, 1800580.
[21] Z. Zhang, L. Chen, B. Yang, Y. Ge, F. Sun, Appl. Therm. Eng. 2015,
The authors declare no conflict of interest.
90, 742.
[22] B. Song, W. Zhuge, R. Zhao, X. Zheng, Y. Zhang, Y. Yin, Y. Zhao,
J. Mech. Sci. Technol. 2013, 27, 1721.
[23] J. Galindo, J. R. Serrano, V. Dolz, P. Kleut, Adv. Mech. Eng. 2015, 7, 1.
Keywords [24] S. Gequn, L. Lina, T. Hua, W. Haiqiao and Y. Guopeng, Trans. Tianjin
Kalina cycle, organic Rankine cycle, thermoelectric devices, waste heat Univ. 2015, 21, 193.
recoveries, waste heat to power [25] A. Entezari, A. Manizadeh, R. Ahmadi, Energy Convers. Manage. 2018,
159, 189.
Received: May 5, 2020 [26] M. Hou, Z. Wu, G. Yu, J. Hu, E. Luo, Appl. Energy 2018, 226, 389.
Revised: July 31, 2020 [27] K. Zeb, S. M. Ali, B. Khan, C. A. Mehmood, N. Tareen, W. Din,
Published online: U. Farid, A. Haider, Renew. Sustain. Energy Rev. 2016, 75, 1142.
[28] A. H. Bademlioglu, A. S. Canbolat, N. Yamankaradeniz, O. Kaynakli,
Appl. Therm. Eng. 2018, 145, 221.
[1] J. Ling-Chin, H. Bao, Z. Ma, W. T. Roskilly, A. Paul, in Energy [29] D. K. Sarkar, Thermal Power Plant – Design and Operation, Elsevier,
Conversion – Current Technologies and Future Trends, Vol. 1, Amsterdam 2015, pp. 1–37.
IntechOpen 2018, pp. 56–74. [30] S. M. Besarati, D. Y. Goswami, Advances in Concentrating Solar
[2] T. Hunnicutt, BlackRock Plans Environmentally Conscious Money Thermal Research and Technology, Woodhead Publishing Series in
Market Fund, https://urly.it/326sg (accessed: April 2020). Energy, Vol. 8, Woodhead Publishing, Duxford, pp. 157–178.
[3] V. Smil, Energy Transitions – History, Requirements, Prospects, 1st ed., [31] S. C. Bhatia, Advanced Renewable Energy Systems, Woodhead
Praege, Santa Barbara 2010, p. 190. Publishing, New Delhi 2014, pp. 334–388.
[4] F. Huang, J. Zheng, J. M. Baleynaud, J. Lu, J. Energy Inst. 2016, 90, 951. [32] A. Naseri, M. Bidi, M. H. Ahmadi, R. Saidur, J. Cleaner Prod. 2017,
[5] G. Boštjan, S. Erlih, in Conf. on Exploitation of Waste Heat Potentials in 158, 165.
Europe, Dubrovnik, Croatia 2017. [33] A. Baheta, M. Emad, ARPN J. Eng. Appl. Sci. 2015, 10, 10169.
[6] M. Pehnt, J. Bödeker, M. Arens, E. Jochem, F. Idrissova, in ECEE 2011 [34] E. Cayer, N. Galanis, M. Desilets, H. Nesreddine, P. Roy, Appl. Energy
Summer Study, Germany 2011, pp. 691–700. 2009, 86, 1055.
[7] G. Myhre, D. Shindell, F. M. Bréon, W. Collins, in Climate Change [35] L. Li, Y. T. Ge, X. Luo, S. A. Tassou, Appl. Therm. Eng. 2017, 115, 815.
2013: The Physical Science Basis. Working Group I Contribution to [36] G. J. Snyder, E. S. Toberer, Nat. Mater. 2008, 7, 105.
the Fifth Assessment, Report of the Intergovernmental Panel on [37] L. E. Bell, Science 2008, 321, 1457.
Climate Change, 2013, pp. 659–740. [38] C.-T. Hsu, G.-Y. Huang, H.-S. Chu, B. Yu, D.-J. Yao, Applied Energy
[8] D. Ehhalt, M. Prather, in Climate Change 2001: The Scientific Basis, The 2011, 88, 1291.
Press Syndicate of the University Of Cambridge 2001, pp. 239–287. [39] S. Sano, H. Mizukami, H. Kaibe, Komatsu Techn. Rep. 2003, 49, 1.
[9] G. Stephens, S. Tjemkes, Aust. J. Phys. 1993, 46, 149. [40] M. Zhou, M. S. H. Al-Furjan, J. Zou, W. Liu, Renew. Sustain. Energy
[10] NOAA National Centers for Environmental Information, State of the Rev. 2018, 82, 3582.
Climate: Global Climate Report for Annual 2019, https://www.ncdc. [41] G. Sebald, D. Guyomar, A. Agbossou, Smart Mater. Struct. 2009,
noaa.gov/sotc/global/201913 (accessed: August 2020). 18, 125006.
[11] G. G. Ochoa, J. C. Gutierrez, J. D. Forero, Resources 2020, 9, 2. [42] A. Khodayari, S. Pruvost, G. Sebald, D. Guyomar, S. Mohammadi,
[12] O. Kaska, Energy Convers. Manage. 2014, 77, 108. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2009, 56, 693.
[13] A. Naeimia, M. Bidia, M. Ahmadib, R. Kumarc, M. Sadeghzadehd, [43] G. Sebald, S. Pruvost, D. Guyomar, Smart Mater. Struct. 2008,
M. Nazari, Therm. Sci. Eng. Prog. 2019, 9, 299. 17, 015012.
[14] C. Forman, I. Muritala, R. Pardemann, B. Meyer, Renew. Sustain. [44] S. Pandya, G. Velarde, L. Zhang, J. D. Wilbur, A. Smith, B. Hanrahan,
Energy Rev. 2016, 57, 1598. C. Dames, L. W. Martin, NPG Asia Mater. 2019, 11, 26.
[15] United States Department of Energy, Waste Heat Recovery: Technology [45] F. Lee, S. Goljahi, I. McKinley, C. S. Lynch, L. Pilon, in Proc. ASME
and Opportunities in the U.S. Industry, United States Department of 2012 3rd Micro/Nanoscale Heat and Mass Transfer Int. Conf.,
Energy, Washington, DC 2008. ASME, Atlanta, GA 2012, p. 597.
[16] A. Elson, R. Tidball, A. Hampson, Waste Heat to Power Market [46] I. M. McKinley, R. Kandilian, L. Pilon, Smart Mater. Struct. 2012, 21, 5.
Assessment, Oak Ridge National Laboratory, Oak Ridge 2015. [47] A. Navid, L. Pilon, Smart Mater. Struct. 2011, 20, 025012.

Energy Technol. 2020, 2000413 2000413 (19 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

[48] S. Mohammadi, A. Khodayari, Smart Mater. Res. 2012, 2012, 160956. [79] C. Y. Chen, C. Y. Tsai, M. H. Xu, C. T. Wu, C. Y. Huang, T. H. Lee,
[49] A. Batra, M. Aggarwal, Pyroelectric Materials: Infrared Detectors, Particle Y. K. Fuh, Express Polym. Lett. 2019, 13, 533.
Accelerators and Energy Harvesters, SPIE Press, Bellingham 2013, [80] D. Jiang, Y. Su, K. Wang, Y. Wang, M. Xu, M. Dong, G. Chen, Nano
p. 202. Energy 2020, 70.
[50] B. Thakre, A. Kumar, H.-C. Song, D.-Y. Jeong, J. Ryu, Sensors 2019, [81] E. Garofalo, L. Cecchini, M. Bevione, A. Chiolerio, Nanomaterials
19, 2170. 2020, 10, 1181.
[51] R. Agathokleous, G. Bianchi, G. Panayiotou, L. Arestia, M. C. Argyrou, [82] M. Bevione, E. Garofalo, L. Cecchini, A. Chiolerio, MRS Energy
G. S. Georgiou, S. A. Tassou, H. Jouhara, S. A. Kalogirou, Sustain. 2020.
G. A. Florides, P. Christodoulides, Energy Procedia 2019, 161, 489. [83] IEA, International Comparison of Light-Duty Vehicle Fuel Economy
[52] E. Warburg, Annalen der Physik 1881, 249, 141. 2005–2015, International Energy Agency, 2017, pp. 1–133.
[53] V. Tsurkan, M. Mücksch, V. Fritsch, J. Hemberger, M. Klemm, [84] D. Sovran, B. Gino, SAE Technical Paper, SAE International,
S. Klimm, S. Körner, H. A. Krug von Nidda, D. Samusi, New York 2003.
E. W. Scheidt, A. Loidl, S. Horn, R. Tidecks, Phys. Rev. B: Condens. [85] J. A. Caton, SAE Technical Paper Series, SAE International, New York
Matter Mater. Phys. 2003, 68, 134434. 2010.
[54] E. Bruck, J. Phys. D: Appl. Phys. 2005, 38, 23. [86] H. H. Pang, C. J. Brace, Proc. Inst. Mech. Eng. D 2004, 218, 1209.
[55] A. Post, C. Knight, E. Kisi, J. Appl. Phys. 2013, 114, 033915. [87] L. Guzzella, U. Wenger, R. Martin, SAE Technical Paper Series, Vol. 1,
[56] T. Tegusi, PhD Thesis, Van der Waals-Zeeman Institute (WZI), 2003. SAE International, New York 2010.
[57] A. Bozhko, G. Putin, Microgravity Sci. Technol. 2009, 21, 89. [88] S. Rajoo, A. Romagnoli, R. Martinez-Botas, A. Pesiridis,
[58] P. Bajpai, A. Varshney, Adv. Res. Electr. Electron. Eng. 2016, 3, 157. C. Copeland, A. Bin Mamat, in Automotive Exhaust Emissions and
[59] A. Chiolerio, E. Garofalo, F. Mattiussi, M. Crepaldi, G. Fortunato, Energy Recovery, Nova Science Publishers, New York 2014.
M. Iovieno, Appl. Energy 2020, 277, 115591. [89] A. C. Alkidas, J. Eng. Gas Turbines Power 2016, 110, 462.
[60] S. Kemkar, M. Vaidya, D. Pinjari, C. Holkar, S. Kemkar, S. Nanaware, [90] J. T. Farrell, J. G. Stevens, W. Weissman, SAE Technical Paper Series,
S. Kemkar, Int. J. Eng. Res. Appl. 2017, 7, 01. Vol. 1, SAE International, New York 2006.
[61] M. Fischedick, J. Roy, A. Abdel-Aziz, A. Acquaye Ghana, J. Allwood, [91] P. Perrot, A to Z of Thermodynamics, Oxford University Press, Oxford
G. Baiocchi, R. Clift, V. Nenov, M. Yetano Roche Spain, J.-P. Ceron, 1998.
Y. Geng, H. Kheshgi, A. Lanza, D. Perczyk, L. Price, E. Santalla, [92] US Department of Energy, Industrial Technologies Program, US
C. Sheinbaum, K. Tanaka, R. Pichs-Madruga, Y. Sokona, Department of Energy, Washington, DC 2008, pp. 1–112.
E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, [93] H. Aghaali, H. E. Angstrom, Renew. Sustain. Energy Rev. 2015,
P. Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer, C. von 49, 813.
Stechow, T. Zwickel, J. Minx, in Climate Change 2014: Mitigation of [94] A. E. Teo Sheng Jye, A. Pesiridis, S. Rajoo, SAE Technical Paper Series,
Climate Change. IPCC Working Group III Contribution to AR5, Vol. 1, SAE International, New York 2013.
Cambridge University Press, New York 2014. [95] B. Mashadi, A. Kakaee, A. Jafari Horestani, Energy Convers. Manage.
[62] E. Worrell, L. Bernstein, J. Roy, L. Price, J. Harnisch, Energy Efficiency 2019, 182, 451.
2008, 2, 109. [96] M. Yari, A. S. Mehr, V. Zare, S. M. Mahmoudi, M. A. Rosen, Energy
[63] M. Papapetrou, G. Kosmadakis, A. Cipollina, U. La Commare, 2015, 83, 712.
G. Micale, Appl. Therm. Eng. 2018, 138, 207. [97] K. Ebrahimi, G. F. Jones, A. S. Fleischer, Appl. Therm. Eng. 2017,
[64] R. C. McKenna, J. B. Norman, Energy Policy 2010, 38, 5878. 126, 393.
[65] Exergy ORC, A waste heat recovery cycle explained, http://exergy-orc. [98] S. Lion, C. N. Michos, I. Vlaskos, R. Taccani, Int. J. Energy Environ.
com/application/heat-recovery-from-industrial-process (accessed: Eng. 2017, 8, 81.
June 2019). [99] R. Cipollone, D. Di Battista, F. Bettoja, Energy Procedia 2017,
[66] V. Zare, E. Gholamian, Energy Convers. Manage. 2016, 117, 150. 129, 770.
[67] C. Yue, D. Han, W. Pu, E. He, Energy Convers. Manage. 2015, [100] E. Macchi, A. Perdichizzi, J. Eng. Power 1981, 103, 718.
89, 764. [101] R. Capata, G. Hernandez, Energies 2014, 7, 7067.
[68] P. Bombarda, C. M. Invernizzi, C. Pietra, Appl. Therm. Eng. 2010, [102] F. Zhou, E. Dede, S. Joshi, SAE Int. J. Mater. Manuf. 2016, 9, 224.
30, 212. [103] H. Oomori, S. Ogino, SAE Technical Paper Series, Vol. 1, SAE
[69] R. V. Padilla, G. Demirkaya, D. Y. Goswami, E. Stefanakos, International, New York 2010.
M. M. Rahman, Energy 2010, 35, 4649. [104] Z. Ge, J. Li, Q. Liu, Y. Duan, Z. Yang, Energy Convers. Manage. 2018,
[70] D. Rowe, Thermoelectrics Handbook, Taylor & Francis Group, Boca 166, 201.
Raton, FL 2006. [105] G. Shu, L. Liu, H. Tian, H. Wei, G. Yu, Appl. Energy 2014, 113, 1188.
[71] T. Kajikawa, in 15th Int. Conf. on Thermoelectrics, IEEE, Piscataway, NJ [106] M. H. Yang, R. H. Yeh, Energy Convers. Manage. 2014, 88, 999.
1996, pp. 343–351. [107] G. Shu, L. Liu, H. Tian, H. Wei, Y. Liang, Energy Convers. Manage.
[72] M. F. Remeli, L. Kiatbodin, B. Singh, K. Verojporn, A. Date, 2013, 76, 234.
A. Akbarzadeh, Energy Procedia 2015, 75, 645. [108] E. Wang, Z. Yu, C. Peter, Z. Hongguang, Y. Fubin, C. Bei, in Heat
[73] S. Dalola, M. Ferrari, V. Ferrari, M. Guizzetti, D. Marioli, A. Taroni, Powered Cycles Conf. 2016, Nottingham, UK 2016, pp. 27–29.
IEEE Trans. Instrum. Meas. 2009, 58, 99. [109] V. Chintala, S. Kumar, J. K. Pandey, Renew. Sustain. Energy Rev. 2018,
[74] R. Bonin, D. Boero, M. Chiaberge, A. Tonoli, Energy Convers. Manage. 81, 493.
2013, 73, 340. [110] M. Jiménez-Arreola, R. Pili, F. Dal Magro, C. Wieland, S. Rajoo,
[75] L. C. Ding, N. Meyerheinrich, L. Tan, K. Rahaoui, R. Jain, A. Romagnoli, Appl. Therm. Eng. 2018, 134, 576.
A. Akbarzadeh, Energy Procedia, 2017, 110, 32. [111] R. Pili, A. Romagnoli, K. Kamossa, A. Schuster, H. Spliethoff,
[76] E. Gholamian, A. Habibollahzadea, V. Zare, Energy Convers. Manage. C. Wieland, Appl. Energy 2017, 204, 1188.
2018, 174, 112. [112] G. P. Meisner, in General Motors Global Research Development 16th
[77] V. Zare, V. Palideh, Appl. Therm. Eng. 2017, 130, 418. Directions in Engine Efficiency and Emissions Research (DEER) Conf.,
[78] M. Raouadi, O. Touayar, Sens. Actuators A 2018, 273, 42. 2010.

Energy Technol. 2020, 2000413 2000413 (20 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

[113] M. Saqr, K. M. Mansour, M. Musa, Int. J. Autom. Technol. 2008, [135] Y. S. Jung, D. H. Jeong, S. B. Kang, F. Kim, M. H. Jeong, K. S. Lee,
9, 155. J. S. Son, J. M. Baik, J. S. Kim, K. J. Choi, Nano Energy 2017, 40, 663.
[114] J. Yang, F. R. Stabler, J. Electron. Mater. 2009, 38, 1245. [136] F. Suarez, D. P. Parekh, C. Ladd, D. Vashaee, M. D. Dickey,
[115] X. Liu, Y. D. Deng, S. Chen, W. S. Wang, Y. Xu, C. Q. Su, Case Studies M. C. Öztürk, Appl. Energy 2017, 202, 736.
in Thermal Engineering, Vol. 2, Elsevier, Amsterdam 2014, pp. 62–66. [137] W. Qian, H. Jinlian, IntechOpen 2018, 1, 23.
[116] J. G. Haidar, J. I. Ghojel, in ICT Proc., IEEE, Beijing, China 2001, [138] Q. Wu, J. Hu, Bringing Thermoelectricity into Reality, IntechOpen,
pp. 413–418. London 2018.
[117] S. K. Kim, B. C. Won, S. H. Rhi, S. H. Kim, J. H. Yoo, J. C. Jang, [139] J.-Y. Yu, L. Liu, The Prospect and Analysis of Nanogenerator for
J. Electron. Mater. 2011, 40, 778. Wearable Devices, Springer, Cham 2018, pp. 75–91.
[118] L. M. Goncalves, J. Martins, J. Antunes, R. Rocha, F. P. Brito, in Proc. [140] K. Zhang, Y. Wang, Y. Yang, Adv. Funct. Mater. 2019, 29, 1901730.
of the 2010 Int. Mechanical Engineering Congress & Exposition, Vol. 5, [141] A. Potnuru, Y. Tadesse, Integr. Ferroelectr. 2014, 150, 23.
American Society of Mechanical Engineers, New York 2010, [142] A. Sultana, M. M. Alam, T. R. Middya, D. Mandal, Appl. Energy 2018,
pp. 1387–1396. 221, 299.
[119] S. Park, J. Yoo, S. Kim, in 2010 Int. Conf. Electrical Machines and [143] M. Kang, E. M. Yeatman, J. Phys. Conf. Ser. 2016, 773, 012073.
Systems (ICEMS), IEEE, Incheon, South Korea 2010, pp. 2–5. [144] T. Ding, L. Zhu, X. Q. Wang, K. H. Chan, X. Lu, Y. Cheng, G. W. Ho,
[120] K. Smith, M. Thornton, in 23rd Int. Electric Vehicle Symp., USA 2009, Adv. Energy Mater. 2018, 8, 1802397.
p. 11. [145] H. Xue, Q. Yang, D. Wang, W. Luo, W. Wang, M. Lin, D. Liang,
[121] O. Aquilina, Images Paediatr. Cardiol. 2006, 8, 17. Q. Luo, Nano Energy 2017, 38, 147.
[122] M. Swan, J. Sens. Actuator Netw. 2012, 1, 217. [146] G. Cha, Y. Jia, Y. S. Ju, in Proc. of the IEEE Int. Conf. on Micro Electro
[123] K. Aizawa, D. Tancharoen, S. Kawasaki, T. Y. Yamasaki, in Mechanical Systems (MEMS), IEEE 2012, pp. 1241–1244.
CARPE’04: Proc. of the the 1st ACM Workshop on Continuous [147] G. Cha, Y. S. Ju, Sens. Actuators A 2013, 189, 100.
Archival and Retrieval of Personal Experiences, Association for [148] M. Prieto, N. Solomakhina, P. Alvarez de Uribarri, Urban Energy
Computing Machinery, New York 2005, pp. 22–31. Systems for Low-Carbon Cities, Vol. 5, Elsevier, San Diego 2019,
[124] D. J. I. Alper Erturk, Piezoelectric Energy Harvesting, 1st ed., Wiley, pp. 181–239.
Hoboken, NJ 2011, p. 402. [149] EIA, https://www.eia.gov/electricity/generatorcosts (accessed:
[125] FAO/WHO/UNU Expert Consultation, Public Health Nutrition, March 2020).
Vol. 8, FAO/WHO/UNU Expert Consultation, Rome 2001, p. 103. [150] M. Ebrahimi, A. Keshavarz, CCHP Technology. Combined Cooling,
[126] T. Starner, IBM Syst. J. 1996, 35, 12. Heating and Power, Elsevier, Boston 2015, pp. 35–91.
[127] X. Liu, T. Chen, F. Qian, Z. Guo, F. X. Lin, X. Wang, K. Chen, in Proc. [151] P. Valdimarsson, Lecture at Washington State University,
of the 15th Annu. Int. Conf. on Mobile Systems, Applications, and Washington, DC 2003.
Services, MobiSys 2017, pp. 385–398. [152] Kalina Power Limited whitepaper, shorturl.at/BXZ45 (accessed:
[128] Z. Obermeyer, J. K. Samra, S. Mullainathan, BMJ 2017, 359, 5468. April 2020).
[129] V. Leonov, T. Torfs, P. Fiorini, C. Van Hoof, IEEE Sens. J. 2007, [153] L. Yunfei, Theses and Dissertations, Lehigh University, 2013.
7, 650. [154] P. Arvay, M. Muller, V. Ramdeen, G. Cunningham, in ACEEE
[130] T. Torfs, V. Leonov, R. J. Vullers, Sens. Transd. J. 2007, 80, 1230. Summer Study on Energy Efficiency in Industry, American Council
[131] V. Leonov, R. J. M. Vullers, J. Renew. Sustain. Energy 2009, 1, 062701. for an Energy Efficient Economy, Washington, DC 2011, pp. 1–12.
[132] T. Torfs, V. Leonov, R. F. Yazicioglu, P. Merken, C. Van Hoof, [155] K. Mohammadi, J. McGowan, Energy Convers. Manage. 2019, 185, 920.
R. J. M. Vullers, B. Gyselinckx, Proc. IEEE Sens. 2008, 1269. [156] M. Guven, H. Bedir, G. Anlas, Energy Convers. Manage. 2019, 180,
[133] Y. Zhang, F. Zhang, Y. Shakhsheer, J. D. Silver, A. Klinefelter, 411.
M. Nagaraju, J. Boley, J. Pandey, A. Shrivastava, E. J. Carlson, [157] S. Ogrisek, Appl. Therm. Eng. 2009, 29, 2843.
A. Wood, B. H. Calhoun, B. P. Otis, IEEE J. Solid State Circuits [158] X. Li, H. Tian, G. Shu, M. Zhao, C. N. Markides, C. Hu, Appl. Energy
2013, 48, 199. 2019, 250, 1581.
[134] M. Wahbah, M. Alhawari, B. Mohammad, H. Saleh, M. Ismail, IEEE [159] K. Zeb, A. M. Ali, B. Khan, C. A. Mehmood, N. K. Tareen, U. Din,
J. Emerg. Sel. Top. Circuits Syst. 2014, 4, 354. W. U. Farid, A. Haider, Renew. Sustain. Energy Rev. 2017, 75, 1142.

Erik Garofalo received his Master’s degree in Aerospace Engineering in 2017 from Politecnico Di Torino.
In 2016, he was visiting student at the California Polytechnic State University to write a thesis about two-
phase flow analysis in hybrid rocket nozzle cooling, using experimental and computational fluid dynamics
methods. Currently, he is a Ph.D. candidate at Istituto Italiano di Tecnologia within the Colloidal
EneRgEtic System (CERES) research project, aiming for the development of the first waste heat to power
energy-harvesting device entirely at the liquid state, suitable for applications ranging in industrial,
domestic, automotive, wearable, and space exploration fields.

Energy Technol. 2020, 2000413 2000413 (21 of 22) © 2020 Wiley-VCH GmbH
www.advancedsciencenews.com www.entechnol.de

Matteo Bevione obtained his Master’s degree in Nanotechnology for ICTs in 2019 from Politecnico di
Torino. He developed his experimental thesis at Istituto Italiano di Tecnologia, involving harvesting of
low-grade waste heat by means of pyroelectric materials, within Colloidal EneRgEtic System (CERES)
project. He is currently fellow researcher at the Italian Institute of Technology in Turin within the SMART
3D project concerning additive manufacturing in the field of industry 4.0. Since 2019, he has also been
an active member of the DRones Autonomous Flight Team (DRAFT) of the Politecnico di Torino, as a
squad leader in the Computer Vision group.

Luca Cecchini obtained a Bachelor’s degree in Electronic Engineering from the University of Roma Tor
Vergata in 2017, working on layered double hydroxide nanostructures for gas-sensing applications. In
2019, he received a Master’s degree in Nanotechnologies for ICTs from Politecnico di Torino, developing
his thesis on the theoretical and experimental analysis of superparamagnetic and triboelectric colloidal
solutions for waste heat to power applications. Currently, he is a research fellow at Scuola Superiore
Sant’Anna in Pisa, working on an ESA project, with the aim of developing an intra- and extra-satellite
optical wireless communication.

Alessandro Chiolerio holds a Ph.D. (Physics Department of Politecnico di Torino, 2009), a Full
Professorship (Solid State Physics), and an Associate Professorship (Electronics) since 2017. He currently
leads the development of liquid cybernetic systems at Istituto Italiano di Tecnologia, including the
Colloidal EneRgEtic System (CERES) project.

Energy Technol. 2020, 2000413 2000413 (22 of 22) © 2020 Wiley-VCH GmbH

View publication stats