Carnot Battery Technology: A State-Of-The-Art Review

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/344396491
Carnot battery technology: A state-of-the-art review
Article in The Journal of Energy Storage · September 2020

DOI: 10.1016/j.est.2020.101756
CITATIONS READS
7 1,084
6 authors, including:
Olivier Dumont Guido Francesco Frate

University of Liège Università di Pisa
55 PUBLICATIONS 553 CITATIONS 23 PUBLICATIONS 181 CITATIONS
SEE PROFILE SEE PROFILE
Aditya Pillai Steven Lecompte

Ghent University Ghent University
5 PUBLICATIONS 28 CITATIONS 61 PUBLICATIONS 1,299 CITATIONS
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Bricker View project
Special Issue: Organic Rankine Cycle - 'Frontiers In Energy Research' (ISSN 2296-598X) View project
All content following this page was uploaded by Olivier Dumont on 12 October 2020.
The user has requested enhancement of the downloaded file.

Journal of Energy Storage 32 (2020) 101756
Contents lists available at ScienceDirect
Journal of Energy Storage

journal homepage: www.elsevier.com/locate/est
Carnot battery technology: A state-of-the-art review T

a,⁎ b c c,d
Olivier Dumont , Guido Francesco Frate , Aditya Pillai , Steven Lecompte ,
Michel De paepec,d, Vincent Lemorta
a
Université de Liege, University of Liège, Aerospace and Mechanical Engineering Department, Thermodynamics Laboratory, 4000 Liège, Belgium
b
University of Pisa, Department of Energy, Systems, Territory and Constructions Engineering, 56122, Pisa, Italy
c
Ghent University, Faculty of Engineering and Architecture, Department of Electrical Energy, Metals, Mechanical Constructions & Systems, Ghent 9000, Belgium
d
FlandersMake, UGent - Corelab EEDT MP, Leuven 3000, Belgium
ARTICLE INFO ABSTRACT
Keywords: The growth of renewable energy requires flexible, low-cost and efficient electrical storage to balance the mis
Carnot battery match between energy supply and demand. The Carnot battery buffers electrical energy by storing thermal
Pumped thermal energy storage energy (charging cycle mode) from a resistive heater or a heat pump system when the electricity production is
Review higher than the demand. When electricity demand is higher than the production, the Carnot battery generates
Electrical storage
power from the stored thermal energy (power cycle mode). This paper is a review of this emerging and in
novative technology, including a market analysis. First, the different possible technologies and configurations of
Carnot batteries are described. This includes charging cycles, power cycles and thermal energy storage systems.
Furthermore, a state-of-the-art of the existing prototypes in the world is given. The performance indicators for
this technology are unclear, and this paper tries to define objective performance indicators. Finally, all the
described technologies are compared, and conclusions are drawn to help engineers select the optimal technology
for a given case.
1. Introduction specific costs per kW and per kWh [4]. Owing to these differences, each
EES technology has an application niche best suited for it, and several
1.1. Renewable energy and grid flexibility niches have already found their most suited EES. The electric systems
future outlooks [5] predicts that the residual demand curve (commonly
The requirement to grow the share of deployed Renewable Energy known as “duck curve” [6,7]) will be the most affected by the solar PV
Sources (RES) in the electricity grid goes hand in hand with the need of production pattern (i.e. maximum production in the central day hours).
improving the electric system flexibility [1,2]. This challenge may be In order to shift massive amounts of energy from daylight hours to other
tackled from different angles on both the demand and production sides. parts of the day, a storage fleet that can charge at nominal power for
Next to demand-side management (DSM), increasing the energy storage several hours per day will be needed. In other words, grid-scale sys
capacity in the grids is a proven strategy. As is already known, such a tems, mainly oriented towards long duration (from 4 to 8 h), will be
task may be not trivial, as in most of the mature electric systems the required. In this context, as the required power-to-capacity ratios (kW/
easily-exploitable additional capacity for Pumped Hydro Energy Sto kWh) would be very low (from at least 1/4 h-1 and lower), the EES
rage (PHES) is nearly exhausted [3]. PHES is the only grid-scale Electric storage medium and system must be as cheap as possible. For the
Energy Storage (EES) technology that has proven to be technically and moment, this requirement rules out the well-established EES tech
economically viable up to the present day. Now we are looking for al nology based on lithium chemistry batteries. However, these may come
ternative EES technologies, several of which having been recently de back into play as their cost per kWh is falling, which is expected to
veloped, proposed or re-discovered. continue in the future [8]. As for today, some of the largest non-PHES
Different EES technologies are each based on different physical EES installations are actually represented by lithium battery systems,
principles and thus have different characteristic performance in like in Australia, where a 100 MW / 129 MWh system was installed in
dicators, such as power-to-capacity ratios, charge and discharge re 2018 [9]. However, this system is designed for providing ancillary
sponse times, different energy/power-to-volume ratios and different services, as the nominal power-to-capacity ratio is not suited for long-
⁎
Corresponding author.
E-mail address: olivier.dumont@ulg.ac.be (O. Dumont).
https://doi.org/10.1016/j.est.2020.101756
Received 14 May 2020; Received in revised form 31 July 2020; Accepted 7 August 2020
2352-152X/ © 2020 Published by Elsevier Ltd.
O. Dumont, et al. Journal of Energy Storage 32 (2020) 101756
Nomenclature RES Renewable Energy Source

STES Sensible Thermal Energy Storage
E Electrical energy [Wh] TCES Thermochemical Energy Storage
El Electrical TES Thermal Energy Storage
Q Thermal energy [Wh] TI Thermally Integrated
V Volume [m3] TRL Technology Readiness Level
Acronym Greek
CAES Compressed Air Energy Storage ɛ Efficiency [-]

CB Carnot battery η Efficiency [-]
CHEST Compressed Heat Energy Storage γ Compactness
COP Coefficient Of Performance τ Time [h]
EES Electrical Energy Storage
HE Heat Engine Subscripts
HP Heat Pump
HT High temperature ch Charge
HVAC Heating and Ventilation Air Conditionning cv Constant volume
LAES Liquid Air Energy Storage dis Discharge
LCOS Levelized Cost of Storage e Energy
LNG Liquefied Natural Gas ext External
LT Low Temperature in Inlet
ORC Organic Rankine Cycle II Second law
PCM Phase Change Material max Maximum
PEG Polyethylene glycol min Minimum
PHES Pumped Hydro Energy Storage rt Roundtrip
PTES Pumped Thermal Electricity Storage w Power
RC Rankine Cycle 0 Reference
duration load shifting. This confirms that lithium batteries are not fact that, as opposed to CAES, energy is stored as heat and not as
limited by the size they can reach, and it is mainly the cost-per-capacity pressure, even though a slight pressurisation of the reservoirs might be
unit that holds them back from being used for low power-to-capacity required. However, low operating pressures allow the Carnot batteries’
ratio applications. thermal tanks to be artificially built, and any location could be poten
The requirement of using an inexpensive storage medium called tially exploited.
attention to alternative storage concepts, even though in most cases the Among the grid-scale EES technologies, Carnot batteries have the
efficiencies are not comparable with those featured in batteries. The lowest average technology readiness level (TRL), even though they are
technologies that are attracting most attention are [10–13]: Com becoming more and more popular. For this reason, the actual potential
pressed Air Energy Storage (CAES); alternative versions of PHES, based of this heterogeneous technology group is still unclear, despite the re
on using seawater or underground reservoirs; battery technologies levant research currently being carried out. Lately, a significant amount
based on molten sodium salt and liquid electrolytes (i.e. flow batteries); of publications have been dedicated to Carnot batteries. Furthermore,
Liquified Air Energy Storage (LAES) [14] and finally, the technology several prototypes have been developed, or are currently being devel
group named Carnot Batteries. oped, to prove the promising theoretical results which were recently
Carnot batteries include technologies like Pumped Thermal derived in this field (see Section 4.1.2). In this phase, it is relevant to
Electricity Storage (PTES) [11], the systems based on the use of electric collect the main contributions on the topic to provide the reader with
heaters and Rankine or Brayton heat engines and, in extension, also an idea of what is the state-of-the-art for Carnot batteries. This paper
LAES. Including LAES into the Carnot battery group may be seen as a thus contains a discussion on Carnot battery technology, including
controversial choice. Therefore, a detailed argument supporting this storage technologies, a clear definition of Carnot battery performance
choice will be provided further on in Section 1.2. indicators. Furthermore, the existing prototypes are listed and their
In Carnot batteries, electric energy is stored as thermal energy, respective performance parameters are reported. Finally, the financial
which is later recovered during discharge. The charging can be done and market outlook for Carnot batteries are reviewed.
with different heating technologies, whereas the discharging can be
done with different thermal engine technologies [11]. Carnot batteries
are based on several patents dating as far back as 1979 [11], but the 1.2. Definition
original concept was proposed in 1922 by Marguerre [15], whereas
other authors trace the origin of PTES back to the work of John Ericsson Carnot batteries include several technologies and it is difficult to
provide a definition that encompasses all their nuances. To the best of
in 1883 [11].
As Carnot batteries are based on heat pumps and heat engines, they the authors’ knowledge, this is the first attempt to provide a systematic
criterion to decide whether a technology is a Carnot battery or not.
are made up of pumps, compressors, expanders, turbines and heat ex
changers, which are all components that may be easily scaled up. For Therefore, the study may be an essential contribution for future scien
this reason, Carnot batteries might be an alternative to PHES and CAES. tific work and policies.
Compared to these, Carnot batteries might have lower efficiencies, but A Carnot battery is an EES technology. Therefore, there should al
they do not rely on pre-existing reservoirs and caves, which may give ways be at least an electric input and an electric output. A Carnot
them an advantage over PHES and CAES as they can be installed ev battery performance may be improved by using additional thermal
erywhere. Carnot batteries’ geographical independence stems from the energy inputs in the charge or discharge phases, but this should not
change its primary purpose, which is storing electric energy. Similarly, a
2
Carnot battery may produce both electric energy and useful thermal the air vaporisation heat. From the Carnot battery point of view, the
energy. However, the electric output must be comparable with the liquefaction apparatus is just a (very) sophisticated heat pump, which
electric input. In other words, electric heating alone should not be cools down the air. During discharge, the air is compressed, vaporised
considered a Carnot battery. (the required heat may come from the stored compression excess heat,
A comprehensive Carnot battery definition could be as follows: the environment or any additional heat sources) and expanded in a
A Carnot battery is a system primarily used to store electric energy. turbine. In other words, during LAES discharge, an air HE is operated.
In a Carnot battery, the electric energy (input) is used to establish a This operational pattern perfectly reflects the Carnot battery one.
temperature difference between two environments, namely the low Alternatively, for CAES, the energy is mostly stored as mechanical
temperature (LT) and high temperature (HT) reservoirs. In this way, the energy, by compressing the air. In CAES, the compression heat is just
storage is charged, and the electric energy is stored as thermal exergy. rejected into the environment, and, even if it is stored, like in adiabatic
As the heat flows against the thermal gradient, work is spent to charge CAES, it represents the minority of the stored energy. Therefore, the
the storage. In the discharge phase, the thermal exergy is discharged by difference lies in the form under which the electric energy is stored
allowing the heat flowing from the HT to the LT reservoir. The heat (mechanical versus thermal). Even though several Carnot battery
flow powers a heat engine (HE) which converts it into work and dis technologies require a compression step during the charge phase, like
charges the residual heat into the LT reservoir. In this way, a fraction of CAES does, the mechanical energy is immediately recovered, by ex
the electric input is recovered. This definition is illustrated in Fig. 1. panding the fluid to cool it down, which does not happen in CAES.
In practice, the described operation may be realised with several
different technologies. The HT and LT reservoir could be actual physical 1.3. Possible Carnot battery configurations
tanks, filled with gas, liquid, solid or changing-phase materials.
Otherwise, one of the two reservoirs could be missing, and its role may One of the significant subgroups that can be clearly identified
be taken up by the environment. within the heterogeneous group of Carnot batteries is PTES, also called
The absorbed specific work increases, for a fixed amount of charged PHES (Pumped Heat Energy Storage) or CHEST (Compressed Heat
thermal exergy, with the temperature difference between the HT and LT Energy Storage). Here, the “PTES” acronym will be used, as “PHES” is
reservoirs. Similarly, the recovered specific work decreases as the often used for “pumped hydro” in literature. PTES is divided into two
temperature difference between the reservoirs is reduced. However, main branches: the first one is based on direct and inverse Brayton
additional heat sources and heat sinks may be used to reduce, or in systems, whereas the second one is based on direct and inverse Rankine
crease, the operating temperature differences during charge or dis systems [11]. Other cycles and variants are possible, like the Lamm-
charge (i.e. they act as thermal exergy additional sources) (Fig. 1). Honigman process, and are discussed in this Section 1.3.3 (Fig. 2).
Exploiting additional heat sources may improve the Carnot battery In the following overview and in the technical discussions, if the
performance from a purely electric point of view. However, as addi efficiency is discussed, the round trip efficiency, as defined in Section 2,
tional energy sources are exploited, different performance metrics is meant. If this is not the case, this will be clearly stipulated in the text.
should be used.
In the charge, electric energy is used to move the heat from the LT to 1.3.1. Brayton
the HT reservoir. Such a task may be done with a traditional heat pump A Brayton Carnot battery, i.e. a Brayton PTES, is usually comprised
(HP), an electric heater, or any other technology. Likewise, in the dis of a Brayton heat pump and a Brayton heat engine. The heat pump
charge, any heat engine technology may be used, ranging from operation is based on an inverse Brayton cycle with two sensible heat
Rankine, Brayton, or different thermodynamic cycles, to thermoelectric thermal reservoirs (HT and LT). Given this layout, the complete EES
generators. usually contains two thermal reservoirs and four machines (two com
The definition of Carnot batteries is useful to understand why LAES, pressors and two expanders). A more straightforward configuration
for example, is a Carnot battery, whereas other similar technologies, (Fig. 3) is proposed when only two turbomachines are used [16]. An
like CAES, are not. In LAES, the electric input is used to liquefy air. In other possibility is to use a two-piston machine [17]. The principle is
general terms (Linde process), such a task is performed by compressing simple. During the charging mode, the heat is transferred from the LT
the air and by cooling it with a pre-cooling and an expansion until li reservoir to the HT reservoir through the compression of a gas (the
quefaction occurs. The air is not stored under pressure, so no me compressor uses more energy than the expander). However, during the
chanical energy is stored alongside the thermal exergy associated with discharge phase, the pressure difference between the HT and the LT
Fig. 1. Carnot battery definition.
3
Fig. 2. Possible configurations for a Carnot battery.
reservoir is used to entrain the expander (the turbine work is higher common working fluid is argon, even though, with air, a higher effi
than the compressor energy use). It should be noted that the combi ciency may be obtained [11]. Nonetheless, in [21] it is demonstrated
nation of an electrical heater and gas cycle has also been considered, that efficiency is related to the temperature ratio, rather than to the
but with rather low round-trip efficiencies (40–50%) [18]. pressure ratio. Furthermore, pressure must be low, to reduce the storage
Apart from these, due to compression and expansion irreversibility, tanks costs, and argon can reach higher temperatures for equal storage
it is usually impossible to perfectly match compression/expansion pressure ratios. Brayton PTES is most often based on packed bed sen
inlet/outlet temperatures between charge and discharge cycles. In this sible heat storage. It is worth noting that, while round-trip efficiencies
case, heaters and coolers may be used to adjust machine outlet condi of around 60 – 70% are often claimed in the literature, these values are
tions. In this way, the heat generated by irreversibilities is discharged to calculated with very high compressor/turbine polytropic efficiencies,
the environment, and the storage may operate in a cyclic behaviour which are over 90%. The efficiency of Brayton PTES is extremely sen
[18]. sitive to machine polytropic performance, such that if slightly lower
For Brayton PTES, both dynamic [19] and volumetric machines figures are used, a round-trip efficiency of around 27 – 35% can be
[20] may be used. By using a volumetric compressor/expander, a re found [11,17]. To date, the only working Brayton PTES demonstrator
versible machine may be more easily adopted [17,20], which may re reported very low efficiency [20]. Better results are expected for larger
sult in lower global performance, but also in lower capital costs. applications, as in [17], where a conservative theoretical efficiency
A classic Brayton PTES layout can be found in [19]. This system may estimation gave 52% as a result of a 2 MW/16 MWh system.
be considered to be representative of Brayton PTES features: the max Despite the lower efficiency, if compared to batteries or PHES,
imum temperature is 1000°C (HT reservoir), while the minimum tem Brayton PTES systems are interesting due to their high energy density
perature is -70°C (LT reservoir); compression/expansion ratio is low, (up to 200 kWhth/m3) and very low estimated capacity prices (50 – 180
around 4.6, and efficiency is in the range of 60 – 70%. The most €/kWh [22] and 12 – 22 €/kWh) [17], which may result in an
Fig. 3. Reversible Brayton cycle (left: charging, right: discharging).
4
economically feasible application. carefully investigated.

Due to Brayton PTES’ high HT reservoir operating temperatures, Rankine PTES may also be based on different operating fluids. In
which are usually well above 200°C, it may be difficult to integrate literature, several examples of PTES systems based on Water, Ammonia
additional low-grade (i.e. low temperature) thermal energy inputs in or refrigerants can be found. In [30], a cascaded NH3/water vapour
such systems. Therefore, proposed modifications are integration with compression heat pump is used for charging a hybrid sensible/latent
cryogenic waste energy in the LT reservoir [23], or with electric hea heat storage. Given the temperature reached by the system, the dis
ters, as in [24]. The use of electric heaters may fulfil part of the char charge phase is performed with a water-steam cycle. The expected
ging process, raising the operating fluid temperature before the HT round-trip efficiency is around 73%. Similarly to Brayton PTES, Ran
reservoir inlet. In this way, the outlet temperature of the charge phase kine PTES’ round-trip efficiency is strongly affected by compressor and
compressor could be lowered. This may be beneficial from a techno turbine polytropic efficiencies. Thus, values equal to 0.9 were used in
logical point of view, as currently there are few compressor technolo [30] to achieve satisfactory results.
gies able to withstand the high discharge temperature imposed by Other examples of Rankine PTES using natural or synthetic re
classical Brayton PTES architecture [19,25], especially if high poly frigerants may be found in [30–39]. All the systems investigated in
tropic efficiencies are needed. The problem posed by a high compressor these papers are based on the use of vapour compression Heat Pumps
discharge temperature has been recognised by some authors, who (HP) and Organic Rankine Cycles (ORC). Compared to Brayton and
propose a compressor discharge temperature of around 500°C [17,26]. Rankine systems presented previously, the latter uses refrigerants as
Such discharge temperatures are near to the current technical limitation working fluids with more conventional equipment that is readily
and are achieved in the modern aero-derivative gas turbines. available. By directly using available commercial equipment, MW scale
systems may potentially be built without many challenges. For larger
applications, multiple HP and ORC could be used in parallel, given the
1.3.2. Heat pump and Rankine cycle
modular structure of these systems.
1.3.2.1. Classical HP/RC power system. The second branch of PTES
For Rankine PTES, the use of solids as storage media is not cited in
technology is based on Rankine cycles. Rankine PTES could be a valid
the literature. The operating temperatures and the nature of the heat
alternative to the Brayton cycle because it generally achieves higher
absorption and rejection transformations push for the use of phase
energy densities and it stores energy at a much lower temperature. This
change materials, especially for the LT reservoir, where ice storage may
is beneficial for thermal losses and the choice of reservoir/machines
be used [40]. In supercritical CO2 systems, if the heat is absorbed from
materials, and it might allow for the use of phase change materials as a
and rejected to the LT storage at a constant temperature (CO2 eva
storage medium. Maximum efficiencies achieved by the Brayton and
poration and condensation), then a phase change material may be
Rankine systems are similar (62%-65%) [26,27].
suited. On the other hand, if the heat is absorbed from and rejected to
Rankine PTES is mostly based on trans-critical and supercritical CO2
the HT reservoir with a significant temperature glide (CO2 cooling in
cycles [28,29], where an efficiency of 53% can be achieved by storing
gas coolers/heaters), then a sensible liquid heat storage may be used.
energy at 123°C. The basic system is improved in [25], by optimising
Rankine systems that operate with refrigerants are usually subcritical,
the thermal layout of the system. In [26,29], both LT latent storage (ice-
and so phase change materials may be used for both the HT and LT
based) and HT sensible storage (liquid-based) are used. Further im
reservoir. Furthermore, as in vapour compression heat pumps the
provements are found in [28], where a liquid piston is used to allow the
condensation heat load can be both with and without temperature
system to perform a nearly-isothermal compression and expansion. In
glide, a hybrid sensible/latent thermal storage may be used, as in
this way, the compression work is reduced, and the expansion work is
[31,35].
increased. The compression heat in excess is stored for later use during
An interesting concept for improving Rankine PTES performance is
the expansion.
to exploit additional heat sources and heat sinks (as detailed in
A different concept may be found in [29], where LT and HT re
Section 1.3.4). This technique is especially convenient if low-grade, or
servoirs are based on the underground thermal storage concept. In
waste heat sources are used. Integration of low-temperature thermal
other words, geothermal heat exchangers are used to store thermal
energy is easier in Rankine PTES systems due to the operating tem
energy on the ground. Despite the efficiency being around 40–60%,
perature levels, which are often lower than 200°C. A similar integration
depending on the investigated layout, the financial implication of using
is more difficult in Brayton PTES due to the high operating
geothermal heat exchangers, which are usually very costly, should be
Fig. 4. Reversible HP/RC system (left = heat pump mode, right = RC mode) [45].
5
temperatures, as pointed out in Section 1.3.1. closed cycle, [53].

One advantage of the technology is the absence of self-discharge
1.3.2.2. Reversible HP/RC power system. Many similarities were (except for minor heat losses). An in-depth analysis of the process is
observed between a heat pump (HP) and a Rankine Cycle (RC) for missing thus far, although recent studies show the growing interest in
small-scale application (volumetric machines, refrigerants, heat the technology [53].
exchangers). Based on this observation, a system able to work as a
heat pump or as an RC with the same components may be conceived, 1.3.4. Thermally integrated Carnot batteries (waste heat integration)
i.e. a reversible HP/RC system (Fig. 4). Its application for a small-scale Typically, the round-trip efficiency (ɛrt), defined as the electrical
Carnot battery is straightforward and would significantly decrease energy output (discharge) divided by the electrical energy input
investments due to the use of a single machine instead of the (charge), is below 70% for a standard Carnot battery. This low round-
combination of a heat pump and an ORC power system. This system trip efficiency is the reason why it can be helpful to valorise waste heat
presented many possible applications [41] and was studied streams in the system to improve its performance (thermally integrated
theoretically for the automotive sector [42], data centre [40], Carnot battery). Some authors expect more than 100% roundtrip effi
stationary engine [43] and Net Zero Energy Buildings [44,45]. ciency [31,35,38]. When using a thermally integrated Carnot battery,
Finally, several prototypes have been built to prove the feasibility of two different options are possible for the thermal energy storage: the
such technology (Table 1). In this table, inverted architecture refers to HT and the LT configurations (Fig. 6).
the architecture of Fig. 4, where the condenser and evaporator are On the one hand, the HT storage configuration uses a heating
exchanged. In the classical architecture, each heat exchanger conserves system (heat pump in this example) to increase the waste heat tem
its role [46]. perature. This allows the power cycle (RC in this example) to increase
Further information can be found about the inherent constraints of its efficiency by working with a higher temperature difference. On the
the system, the modelling, the optimal sizing and mappings of perfor other hand, the LT storage configuration stores thermal energy at
mance as a function of temperature levels [39,41]. temperatures lower than the ambient (through a vapour compression
cycle in this example). Only a few references discuss this possibility
1.3.2.3. Electrical heater combined with a Rankine cycle power system. It [35,38,40]. Once again, it allows the power cycle to work efficiently
might be interesting to store heat at high temperatures to increase the with a higher temperature difference. From a thermodynamic point of
compactness of a Carnot battery. This may induce a low coefficient of view, it can be shown (analytically or with a constant efficiency model)
performance (down to 1) if a heat pump is used for the charging that the round-trip efficiency is always higher for the HT storage con
process. Therefore, it might be interesting to use an electrical heater, figuration [38]. This does not mean that the LT storage cannot present
which is usually cheaper and simpler than a heat pump. Two options other advantages (a more straightforwrd use of latent energy storage,
have been proposed so far: the use of an electrical resistance [50] or the for example).
use of a rotating heater (asynchronous machine with permanent magnet As will be pointed out in Section 2, the concept of exploiting ad
using induction). The main advantage of the latter option is that the ditional heat sources and sinks is primarily linked to Rankine systems
resistance and the AC/DC converter are not needed compared to the due to the operating temperature range. For Rankine systems, the
former [51]. Three configurations are possible: thermal integration concept can be found in [25,54], and was analysed
in-depth in [31,32,38], where the HT storage integration was in
- A GWh-scale stand-alone solution. In this case, the system would be vestigated for a heat pump/RC system and for several potential oper
able to store and supply electricity, process steam and heat in ating fluids. In these papers, the low-grade heat source is used to
dependently of geographical conditions. eliminate the LT reservoir. A similar idea is explored and expanded
- New flexibility for existing heat cycles. It is attached to a fossil fuel upon in [36, 38, 55,56, where a solar pond represents the heat source.
power plant or an energy-intensive industrial plant. The opposite can also be done, and the heat from solar energy can be
- A second life for power plants. Thermal fossil fuel power stations used to replace the HT reservoir while maintaining the LT one [33,35].
can be transformed into storage plants, combining existing equip Even though the integration of additional heat sources is particu
ment with the new technology [50]. larly suited to Rankine systems, the same concept may also be applied
to Brayton systems, as in [23], where cryogenic waste energy from
1.3.3. Lamm-Honigmann-process LAES plants is recycled in a Brayton PTES system. Furthermore, the
The Lamm-Honigmann process is a thermochemical energy storage same concept has been extensively used in LAES applications, where
invented in the 19th Century [52]. In general, the storage system can be the discharge phase is often powered by a waste heat load [14] to
charged with the input of heat or mechanical work and discharged with improve the electric discharge efficiency.
the release of heat, LT or mechanical work. The discharging process is Two recent papers [38,55] showed the constraints in the sizing of a
achieved through the heating of a solution of water and another liquid thermally integrated Carnot battery. Three constraints have to be taken
(LiBr or NaOH typically) presenting different vapour pressure. Due to into account: the energy density, the round-trip efficiency and the
the difference in the vapour pressure, an expansion machine can be correct exploitation of the heat source. The main parameter to optimise
operated by steam flowing from a water vessel to a solution vessel is the storage temperature lift, i.e. the temperature difference between
(Fig. 5). Recharging can be accomplished with the input of heat to the completely charged thermal storage and the completely discharged
desorb the water out of the aqueous solution. The water vapour will be thermal storage. If the thermal storage lift is high, the round-trip effi
condensed at a lower temperature, to recover the water and maintain a ciency is low, whereas the waste heat exploitation and the energy
Table 1
State-of-the-art reversible HP/ORC systems.
Year Application Architecture El. power Fluid References Status
2015 Solar building classical 4 kW R134a [47] Finished

2016 Stationary engine Inverted 8 kW - [43] Ongoing
2019 Automotive Inverted 1 kW R134a [48] Finished
2019 Carnot battery Inverted 2 kW R1234yf [40] Ongoing
2019 Carnot battery Inverted 1 kW R1233zd [49] Ongoing
6
Fig. 5. Lamm-Honigman process (left: discharge, right: charge).
Fig. 6. Thermally integrated Carnot battery (left = HT storage, right = LT storage). Illustration with a heat pump and a Rankine cycle.
density are high, and vice versa. However, the word “efficiency” will be used in this paper since most of
the literature refers to this name.
2. Performance indicators If the energy balance is applied on the Heat Pump from Fig. 1 (HP
control region - purple dotted line – Eq. 4 and Heat Engine (HE control
Several indicators may be used to characterise a Carnot battery region – orange dotted line – Eq. 5) this gives a relation between the
performance. As an EES technology, the most widespread and useful thermal energy of the system (Q) and the energy of the heat pump and
performance indicator for a Carnot battery is the round-trip efficiency heat engine:
(RTE) εrt. This is defined as the ratio between the net electrical energy Ehe = QHT , he QLT , he (4)
output Ehe and the gross electrical energy input Ehp. The terms “net” and
“gross” refer to the fact that the electrical energy input and output are Ehp = QHT , hp QLT , hp (5)
subjected to losses in electrical components like generators, motors,
The Coefficient Of Performance (COP) of the heat pump and the
converters, inverters, transformers and so forth. The energy input must
efficiency of the heat engine can be calculated via Eqs. 6-7.
be considered before these losses (gross input), whereas the energy
output must be considered after (net output). QHT , hp QLT , hp
COPhp = =1+
By referring to a generic Carnot battery, such as the one represented Ehp Ehp (6)
in Fig. 1, and by applying the energy balance to the closed system of
Carnot Battery (green dotted line), with ΔU the variation of the internal Ehe 1
he = = Q , he
energy of the system at constant volume (CV) (and excluding any ex QHt , he 1 + LT (7)
E he
ternal heat transfer):
During charging, And thus,
Ehp = UCV , charge (1) Ehe QHT , he QLT , he
rt = =
Ehp QHT , hp QLT , hp (8)
During discharging,
The definition in Eq. 8 may be used directly to calculate εrt from the
Ehe = UCV,discharge (2)
heat pump and heat engine thermodynamic cycles, as in [18,20],
The round trip efficiency εrt is defined as the ratio of the delivered among others. However, it could be interesting to relate εrt to heat
work to that put into the CB, pump and heat engine common performance indicators, such as COPhp
Ehe and ηhe.
rt = If fully reversible heat pumps and heat engines are used, and the
Ehp (3)
heat transfer to the storage systems is reversible (thus happening
This ratio can be higher than 1, if waste heat is included in the without temperature difference), so no internal energy is accumulated,
system (as further discussed). It would probably be more suitable and εrt =1.
rigorous to call this “electrical energy ratio” instead of “efficiency”. In real cycles, with all kinds of irreversibilities, εrt <1. This can be
7
linked to the exergy content of heat. Although energy is conserved in ULT = QLT , he tes · Qcold, hp =0 (15)
any conversion process, the work that can be produced by a heat flow is
In Eq. 14, QLT,he and QLT,hp from Eq. 4 and 5 may be used to get to
limited by the exergy of this heat flow.
the following definition of εrt (Eq. 16):
This is true for all the systems without external heat addition. In
case an additional energy (heat) input is provided to the system, this Ehe he
rt = = tes ·(COPhp 1)·
can provide an additional exergy stream that could counterbalance the Ehp 1 he (16)
losses due to irreversibilities.
These equations show that the design specification about where the
These obvious considerations actually have an interesting con
excess heat should be accumulated affects the system performance.
sequence for the Carnot battery efficiency calculation. If again, the
Furthermore, it may be demonstrated how the round-trip efficiency in
control volume around the battery is considered (dotted line, green –
Eq. 14 is always higher than that in Eq. 16, by plugging the same nu
Fig. 1), and taking into account internal energy is stored inside the
merical values into the two equations. This means that accumulating
system the energy balance becomes (Eq. 9):
the heat in the LT reservoir should be preferred.
Ehp = Ehe + U (9) However, some authors choose to do the opposite for practical
reasons. In [18], the heat from irreversibility is accumulated in the HT
where ΔU is a generic internal energy accumulation term.
reservoir. This is done because extracting this heat from the HT re
This internal energy storage can be attributed to the HT an LT re
servoir may be cheaper and easier, as the heat is stored at high tem
servoir so the internal energy can be written as Eq. 10.
peratures. As a matter of fact, extracting the heat from the LT reservoir
U = UHT + ULT (10) may cost additional work, as an auxiliary heat pump may be needed to
From Eqs. 9 and 10,Eq 11 is obtained. move the heat against the thermal gradient between the environment
and the LT reservoir. However, this is a strict necessity only in those
Ehp = Ehe + U = Ehe + UHT + ULT (11) cases in which the LT reservoir is entirely at temperatures lower than the
Two cases may be stated (Eq. 12). environment, as in [26,27,35]. Often the LT reservoir may be designed
in a way that is only partially at temperatures lower than the en
rt =1 Ehp = Ehe U = UHT + ULT = 0 vironment, as in [17,24] and others. In this case, the excess heat may be
rt <1 Ehp > Ehe U = UHT + ULT > 0 (12) extracted from the LT reservoir, in the same way as it would be done for
the HT reservoir.
In other words, if a non-ideal storage is used (i.e. εrt < 1), heat
Finally, it may be observed how both εrt definitions must yield
accumulates as internal energy in the system. This precludes a system
εrt = 1 for ideal systems. In both cases, this happens if the thermal
cyclic behaviour, and the storage system initial conditions cannot be re-
reservoirs are adiabatic (ηtes = 1) and the product COPhp∙ηhe = 1. This
established after discharge. This is a well-known problem, as several
may happen only in the case of perfectly overlapping charge and dis
authors faced this issue in practice. In such cases, two additional
charge thermodynamic cycles. This situation may be achieved only in
components (typically an auxiliary heat pump and a heat exchanger, or
case reversible transformations are performed. This translates into the
two heat exchangers) were added to the EES, the purpose of which was
necessity of having no-heat losses, no-pressure losses and of using
to remove the excess heat from the LT and HT reservoir, thus restoring
isentropic (and adiabatic) machines. In this case, the same thermo
the CB initial conditions before the next charge/discharge cycle (see
dynamic cycle may be followed during charge and, in reverse, during
[18,21,24,27,56], for example). The commonly-proposed idea is to
discharge, discussed by [18]. In this case, the COPhp and ηhe are the
design the system in such a way that all the excess heat is accumulated
exact inverse of one another. Thus, their product yields 1.
in only one of the two heat reservoirs. This allows one of the two ad
Several authors [31–35,37,38,41] propose the use of only one heat
ditional pieces of equipment used to restore the storage's initial con
reservoir. In this case, the choice of which reservoir is emptied to close
ditions to be cut [18]. In this case, it must be decided whether the
the charge/discharge cycle is straightforward and the related equation
excess heat is accumulated in the HT or LT reservoir. This choice im
between Eq. 14 and Eq. 16 should be used for εrt.
pacts on the way in which εrt may be expressed as a function of COPhp
Apart from the round-trip efficiency, several other parameters may
and ηhe, and on the εrt numerical value itself, as one of the two con
be used to characterise the CB system, such as energy and power
figurations is more efficient than the other.
compactness γe/w, i.e. energy and power density, measured in kWh/m3
Let the first configuration be the one in which the heat is accumu
and kW/m3. These may be an important criterion to compare different
lated only in the HT reservoir, i.e. ΔUHT = 0. Without loss of generality,
technologies. The definition may consider the energy effectively dis
the adiabatic reservoir hypothesis may be dropped, to account for CB
charged or charged Eq. 17 and (18):
thermal losses during energy storage period. Thermal losses may be
expressed by means of a Thermal Energy Reservoir (TES) efficiency ηtes. =
Ehp
e, ch Vtes
If the HT reservoir is considered, ηtes accounts for the heat losses to Ehp· rt
Ehe
wards the environment, whereas, if the LT reservoir is considered, ηtes e, dis = Vtes
= Vtes (17)
accounts for the heat leakage from the environment towards the re
servoir. As discussed earlier, a share of the reservoir-stored exergy is =
Ehp
w, ch
lost during the energy conservation period, for both the HT and LT ch·Vtes
Ehp· rt
reservoir.
Ehe
= =
w, dis dis·Vtes dis·Vtes (18)
By using the control volume on the HT storage system (dotted line,
red – Fig. 1), the following equations may be written (Eq. 13): where Vtes is the total TES volume, τch is the nominal charge time in h,
and τdis is the nominal discharge time in h. Charge and discharge
UHT = tes · QHT , hp QHT , he = 0 (13)
nominal time are usually derived from electrical profiles, as shown in
In Eq. 6, QHT,hp and QHT,he from Eq. 4 and 5 may be used to come to [17], which suggests that typical required charging times are around
the following definition of εrt (Eq. 14): 5 h, whereas discharging times are around 3 h.
Ehe All the listed indicators do not take into account the possible in
rt = = tes · COPhp· he tegration of waste heat. It is not always clear if a thermally integrated
Ehp (14)
Carnot battery is more like a storage of electricity or a waste heat re
In the opposite case, i.e. for ΔULT = 0, the following equation may covery system. Of course, this may depend on the ratio between heat
be written (blue dotted line – Fig. 1) (Eq. 15): and electric energy input. Furthermore, the direct conversion of waste
8
heat into electricity, for how low the temperature could be, will always 3. Thermal energy storage
be more efficient than any Carnot battery that can be built upon such
thermal resources. In [31] it is shown how the ratio between CB's first Thermal energy storage (TES) is a crucial component in the overall
law total efficiency and that of direct exploitation of the thermal source Carnot battery system. It is positioned between the power-to-heat and
may range between 1/10 to 1/3, in function of the operating fluids and the heat-to-power system, and as such its discharge and charging pro
of the heat source temperature level. cesses need to be adapted to these systems to achieve optimal opera
However, the purpose of a CB is to store energy, not to produce tion. In this section, different TES technologies are discussed in the
energy, and this cannot be done without efficiency loss. This is true for scope of Carnot batteries. Benefits and drawbacks are highlighted, and
all the storage technologies. current experimental and theoretical research is summarised. For an in-
While it is true that the first-law efficiency of a thermally integrated depth analysis on TES, we refer to the appropriate review papers
CB may be very high, it would be incorrect to consider the heat from the [58–60].
heat source and the electric energy in input as equally valuable. To
consider the difference in thermodynamic quality between the two
3.1. Sensible thermal energy storage
energy streams, a second-law efficiency, i.e. exergy efficiency, may be
defined.
For sensible TES (STES) systems, heat is stored or rejected by using
If we consider that all the waste heat in input is useful energy, we
an increase or a decrease in temperature, respectively. As such, STES
can define the second law efficiency [57], assuming that all the waste
systems use the heat capacity of the filling material to store energy, and
heat energy from high source temperature Tin,source to cold source
the material is always present in a single phase. Typically, this is either
temperature Tout,source = T0 could be converted into electricity through
the solid or the liquid phase. The most straightforward example of STES
an infinite series of infinitesimal Carnot cycles (Eq. 19):
with a liquid medium is water, while for a solid medium this would be
=
Ehe + Qsink · 1 (Tsink, max
T0
Tsink , min )
·ln ( Tsink, max
Tsink, min ) rock type storage. Both have the advantage of being cheap storage
materials. The specific heat capacity of water is roughly four times
( )
II , usable
Ehp + Qsource· 1
T0
·ln
Tin, source higher than that of rock type materials. However, water needs high
(Tin, source T0) T0 (19) pressures to reach temperatures higher than 100°C, while rock type
where Qsink refers to the eventual use of the thermal energy rejected at material can easily go to temperatures of 700°C. A comprehensive list of
the condenser of the ORC or of the HP respectively, in the HT config different materials can be found in several review papers published on
uration (i.e. only HT reservoir), or in the LT configuration (i.e. only LT the topic [58–60].
reservoir). Usually, this thermal energy is wasted, and this term is It has been noted that, for an STES system, the efficiency strongly
omitted in standard configurations. Tsink,max refers to the high tem depends on the efficiency of insulation provided against thermal lea
perature of the sink and Tsink,max is the low temperature of the sink. kages. Depending on this, an STES may reach an efficiency as high as
Another possibility is to consider whether the waste heat not used 90% and as low as 50% [61]. The authors also highlighted the low
by the Carnot battery is still useful for additional waste heat recovery specific energy of the STES system of 10 – 50 Wth/kg, which leads to a
systems or for direct use onsite. In this case, only the thermal input substantial increase in the size of the storage tanks. It has been esti
effectively used by the Carnot battery is taken into account (Eq. 20). mated that the capital costs associated with an STES system are in the
range of 3400 – 4500 $/kW [62], while the price per energy unit stored
=
Ehe + Qsink · 1 (Tsink , max
T0
Tsink, min)
·ln ( Tsink , max
Tsink , min ) lies in the range of 0.1 – 10 $/kWh [11].
The Brayton Carnot battery is most often based on packed bed
· ln ( )
II , used
Ehp + Qsource · 1
T0 Tout , source sensible heat storage. These use high temperatures, and there is a direct
(Tin, source Tout , source ) Tin, source (20) heat transfer with the fluid (e.g. air or an inert gas like argon). Both
increase the efficiency of the Brayton Carnot Battery concept. Packed
An illustration of these performance indicators is shown in Fig. 7.
bed dynamic behaviour has been intensely investigated [20,25,56,63]
Here, the heat source temperature glide (the difference of temperature
to characterise Brayton PTES transient behaviour, losses, energy density
between the high temperature and low-temperature storage) used by
and the relation between efficiency and power output [17,24]. In
the Carnot battery is plotted on the x-axis. In this example, the per
formance indicator of a thermally integrated Carnot battery using an
HT storage is shown. In this example, the combination of a CB based on
a heat pump and a Rankine cycle is considered with a waste heat
temperature of 75°C and an air temperature of 15°C [41].
First, from Fig. 7, the used-to-wasted thermal energy ratio is plotted
against the heat source temperature glide. This ratio is defined as the
ratio between the heat flow absorbed by the Carnot battery divided by
the total available heat flux (from inlet to outlet heat source tempera
tures). Naturally, this ratio increases with the heat source glide across
the Carnot battery. εrt and ηII,used decrease with the heat source glide
since the mean temperature of the heat source decreases. This has a
negative impact on the COPhp, and thus on CB efficiency. However,
ηII,usable increases with the heat source glide until it reaches an optimum.
This is because, at low heat source glides, the waste heat recovery is
minimal, and most of the heat source potential contribution is wasted.
This has a negative impact on ηII,usable and first law efficiency. On the
other hand, i.e. for high heat source glide values, the used-to-wasted
thermal energy ratio is high, but the εrt is low, since it also lowers the
COPhp (the heat must be upgraded over a larger heat pump temperature
lift).
To conclude this section, Table 2 summarises the different perfor Fig. 7. Performance criteria of a thermally integrated Carnot battery using a HT
mance indicators. storage.
9
Table 2 a low-temperature LTES reservoir [40]. As far as the high-temperature

Summary of performance indicators. reservoir, there are some prototypes under development that make use
Nomenclature Full name units Equation of molten salts. In project Malta [16], a phase change occurs at high
temperature, and the LTES is coupled to a Brayton cycle. Low-tem
ɛrt Roundtrip efficiency [-] 3 perature storage systems allow for the inclusion of alternative heat
γe Energy density [Wh/m3] 17
streams like waste heat and solar energy, yet, the maximum achievable
γw Power density [W/m3] 18
ηII Second law efficiency [-] 19–20
performance of the heat engine will be lower. The isothermal eva
poration and condensation in a Rankine cycle, however, makes a good
match with an LTES systems.
optimised designs, the round-trip efficiency and energy density are less Other options include the use of metallic and polymer PCMs. At the
sensitive to losses associated with the packed bed thermal energy sto moment both are not considered for use in Carnot batteries. Metallic
rage than to the expander and compressor performance [56]. However, PCMs operate at high temperature and thus could increase the perfor
the packed bed also pose challenges, such as high-pressure losses and mance of the heat engine. Furthermore, numerical and experimental
brittleness due to thermal stresses caused by the cyclic operation. studies have shown that metallic PCMs are effective in transient high
Furthermore, measuring the state of charge in a packed bed is non- heat flux applications. These applications include the temperature in
trivial and may require direct monitoring of the temperature in several crease in electronic chips [65,66], smartphones [67] and heat transfer
parts of the packed bed. Finally, packed bed thermal behaviour may not in TES systems [68–70]. However, the main drawback is that many
allow for an incomplete charge or discharge, and the residual energy metallic PCMs are corrosive, the storage container should be able to
may be lost due to internal heat exchange and overall temperature sustain this in combination with the high temperature, making this a
reduction. Some pilot demonstrators are based on molten salt sensible challenge [71]. Polymers, on the other hand, operate at too low tem
heat storage, to avoid all these issues [16]. A schematic representation peratures making them not interesting for use in Carnot batteries.
of a double reservoir SHTES PTES system is presented in Fig. 8.
3.2.1. Liquid air energy storage (LAES)

3.2. Latent thermal energy storage Although Liquid Air Energy Storage (LAES) has often been con
sidered merely an advancement of CAES [12,72] proposed to improve
For Carnot batteries using latent TES (LTES) systems, heat is stored energy density, LAES is based on different physical principles. As was
in materials that undergo a phase change during charging or dischar discussed, LAES stores electrical energy as heat, and not as mechanical
ging. The materials used are called phase change materials (PCM). The energy, hence it should be considered a proper Carnot battery. To store
energy released or absorbed during phase change is known as latent electrical energy, LAES exploits the liquefaction of air, which is a
heat, in the context of LTES systems, mostly solid-liquid transitions are convenient way to store latent heat. Liquefied air is produced cryo
taking place [64]. These transitions occur at an approximately constant genically, at -196°C, which is the boiling point of nitrogen. When LAES
temperature, hence facilitating the stabilisation of the temperature over is discharged, the liquid air is pumped, heated and expanded in a tur
which the heat transfer takes place. When heat addition and heat re bine [12].
jection of the energy conversion devices also occur at near isothermal Since liquid air is much denser than compressed air, LAES features
conditions a good match with the LTES can be achieved. As such, ir much higher energy density than CAES [10,72]. Furthermore, the liquid
reversibilities associated to finite temperature heat transfer can be re air storage tank is at low pressure, so it is both more compact and
duced. cheaper, if compared to that of CAES. For all these reasons, the LAES
LTES systems have the advantage of high specific energy (50 – 150 storage can be directly fabricated. Thus the technology is independent
Wthh/kg) when compared to STES systems; the former can be up to 14 from the pre-existence of caves, mines or other suitable geological
times higher than the latter [62]. However, the capital costs required to formations.
establish LTES are in the range of 6000 – 15,000 $/kW and the price per The liquefaction process is the most critical for the LAES operation.
energy unit stored lies in the range of 10 – 50 $/kWh [63], which is Liquefaction can be achieved through a classical vapour compression
significantly higher when compared to STES systems. cycle, through a Linde-Hampson cycle or through more advanced
Only one prototype of Carnot battery is considering melting of ice as cryogenic cycles.
Fig. 8. PTES system with sensible thermal energy storage.
10
successfully commissioned, based on the Claude cycle [76,77]. The

plant operates with a round-trip electrical efficiency of around 8%, but
the projections for a full-scale facility are up to 50%. The pilot plant was
subjected to several dynamic tests to prove its ability to provide an
cillary services. The tests were positive, proving the fast response of
LAES during discharge.
Apart from using more efficient liquefaction cycles, a very effective
technique to improve efficiency is to recycle compression and expan
sion excess thermal energy in order to use it in the next charge/dis
charge phase [31]. In LAES, the most considerable exergy loss is the
turbine outlet [78]. Therefore, several waste heat recovery solutions
were proposed, such as ORC [79–81], Brayton cycle [78] and Absorp
tion cooling [77]. The best strategy, however, could be combining LAES
with other systems that could provide the waste heat. For this task,
thermal power plants [81], or waste cold from LNG regasification fa
cilities [80,82], have been proposed. In particular, the solutions which
exploit the waste cold energy seem to yield the highest efficiencies: 70%
[80] and 88% [82] round trip efficiency.
In an attempt to reduce the energy use of liquefaction, alternative
fluids like CO2, which condenses at a much higher temperature than air,
have been proposed [12]. Liquefied CO2 energy storage is reported to
achieve efficiencies ranging between 40–57% and might have some
advantages for the use of more compact equipment [12].
Fig. 9. Charging and discharging cycles in a PTES system employing a sensible LAES as a technology is considered to be emerging, whereas the
and latent heat storage system. essential components for its construction can be considered mature and
readily available. Therefore, LAES could be a promising alternative
In a Linde cycle, makeup gas mixes with uncondensed gas from the technology for grid-scale storage applications. Thanks to positive fea
cycle. The mixture is compressed by an ideally isothermal compressor, tures like high energy density and independence from geographical
increasing the pressure of the mixture. The temperature is kept constant sites, LAES might have some advantages over CAES and PHES.
by rejecting compression heat to a coolant. The high-pressure gas then However, LAES generally achieves lower efficiency when compared to
enters the heat exchanger where the uncondensed gas cools the gas. At these technologies. LAES efficiency could be improved by constructing
the heat exchanger outlet, the gas is throttled through a valve, expands the LAES facility near to sources of waste heat/cold, such as power
and its temperature decreases, until condensation occurs. The resulting plants and LNG regasification facilities. However, this might cause
vapour-liquid mixture enters a phase separator where liquefied gas is LAES to lose the strategic advantage of being site independent.
obtained [72]. To achieve a higher efficiency, more advanced air li
quefaction cycles may be used. Only basic LAES layouts are based on 3.2.2. Thermochemical energy storage
the Linde cycle, which is notoriously inefficient if compared to lique The process of storing and releasing heat through chemical reaction
faction cycles such as Claude and Kapitza [73]. These more advanced mechanisms is the underlying principle behind thermochemical energy
configurations are supposed to achieve a higher efficiency [31,74,75]. storage (TCES). A chemical material pair can be supplied with heat
After the charge phase, there are four main methods of energy ex energy, resulting in their dissociation into individual components that
traction from cryogens: the direct expansion method, the Rankine cycle, can be separated, thus allowing them to store the provided thermal
the Brayton cycle, and a combination of the methods above. Energy energy. If these separated individual components react with each other,
recovered from the liquefied air using the Rankine cycle can be as high they associate into a compound, thereby releasing the stored thermal
as 36.8%, while energy recovered from the combined cycle could be energy.
increased to 43.3% [72]. TCES systems are primarily employed in space heating applications,
To date, only one LAES pilot plant (300 kW, 2.5 MWh) has been as solar energy provides a high grade of thermal energy which can be
Fig. 10. Pumped thermal energy storage system with a sensible and latent heat thermal energy storage.
11
utilised effectively in such a system. The capital cost required for a now.
TCES system is the lowest among the three types of storage technolo • For applications at temperatures close to 0°C, water can be a very
gies, and is in the range of 1000 – 3000 $/kW, and furthermore, the efficient PCM, while at very low temperatures, liquid air is used
energy density, in the range of 120–250 Wthh/kg, is the highest. (-196°C).
However, the price per energy unit stored is 8–100 $/kWh, making it
the most costly of the three storage technologies [30].
The TCES system possesses advantages when it comes to parameters 4. Industrial state of the art
such as storage capacity and energy density, and it can be stored at
atmospheric conditions without losing thermal energy. However, TCES 4.1. Potential impact on the energy market
systems are not available on a highly commercial scale, as much re
search is required to gain a comprehensive understanding of the prac Emphasis is increasingly being placed on the production of power
tical aspects of the technology before commercial implementation can from renewable energy sources (RES) to decarbonise the electricity
take place [83]. sector. Energy storage serves as one of the ways in which this challenge
can be tackled. It is estimated that an additional electricity storage
3.2.3. Hybrid thermal energy storage capacity of 310 gigawatts that is connected to the grid is required in the
An emerging method for a large-scale energy storage system com United States, Europe, China and India alone [84]. As a means to
bines the latent and sensible thermal energy storage systems. Fig. 9 promote energy storage, governments across Europe, America and Asia
shows the temperature-entropy plot of such a system, along with a are providing support to demonstration projects as an incentive for
charging cycle involving the ORC and a discharging cycle involving the further growth [85]. The federal government of Germany aims to in
heat pump. The layout of such a system is visualised in Fig. 10. In crease the share of electricity produced by renewable energy to 80% by
[43,63], the authors studied the coupling of low-temperature heat 2050 [86]. Furthermore, the energy demand is projected to increase in
sources with PTES. They developed a numerical model for a subcritical the future. In addition, a very significant degree of decarbonisation is
PTES system working with butene. They found that the ratio of supplied aimed at by 2030. The European Commission suggests a structural
electrical power to useful electrical power is 1.25 with a maximum change in the process of power generation to renewable energy sources
exergetic efficiency of 59%, operating between a source and sink tem (RES) to achieve 96 to 99% decarbonisation by 2050 [87].
perature of 100⁰C and 15⁰C respectively. If the thermal energy is not Different countries have different levels of RES penetration in their
utilised, the maximum exergetic efficiency drops to 52%. electricity grid. Curtailment of non-dispatchable RES electricity has
It should be noted that the sensible temperature profiles in Fig. 9 are been increasingly adopted in these countries in times when electricity
only valid at the beginning of charging and discharging, and that a production exceeds demand and cannot be transferred elsewhere due to
stratified storage system is assumed. In reality, temperatures will bottlenecks in the electricity grid. Grid extensions are expensive and
change over time (Fig. 9), leading to a reduced availability to produce may not be able to provide a long term solution. This is a significant
work. Thus, even for systems with a good temperature match for the motivation for transitioning to electrical energy storage (EES) systems
STES, important irreversibilities have to be considered. [87].
If the envisioned RES integration is to be achieved in the forecasted
3.3. Comparison future, it is necessary to perform a financial and technical evaluation of
the currently available EES systems. Most of these studies have been
Among the three types of TES, many types of configurations and based on the utilisation of well-established technologies such as
materials are possible. A summary of the main characteristics is pro Pumped Hydroelectric Storage (PHS) and Compressed Air Energy
posed in Table 3. More details are given in specific literature [62]. Storage (CAES). It is well understood that PHS and CAES systems are
It is interesting to observe which TES is the most suited to the capable of achieving the target costs for ubiquitous applications [84].
Carnot battery. In the literature, some trends appear, although every However, the inherent restrictions of suitable geographical and geolo
combination is theoretically possible: gical requirements have pushed the development of alternative elec
tricity storage technologies.
• For the Brayton Carnot battery and the Rankine cycle using an The Carnot battery system is an electricity storage technology that
does not face the limitations affecting other competitive technologies.
electrical heater, high-temperature levels are expected. Usually, a
packed bed TES consisting of rocks [17] or molten salts [16] are There have been studies evaluating the financial aspects of this tech
considered. Metallic PCM could also be considered. nology when compared to other energy storage technologies.
• For the Rankine Carnot battery, temperature levels are usually low. Smallbone et al. [17] conducted a financial analysis of a Carnot battery
system using the Levelised Cost of Storage (LCOS) method and found
Water is generally used due to its low cost and simplicity (for
temperatures <150°C). Oil could be used for higher temperatures out that the Carnot battery system can be cost-competitive with the
due to its rather low pressure. PCM are also considered due to their other large-scale storage systems. If there is no cost associated with
high compactness, but their costs have been limiting their use up to charging the storage, the LCOS for the Carnot battery system is rela
tively lower than that of the PHS and CAES system. Benato [88] studied
Table 3
Comparison between the three main types of TES [62].
Type Power [MW] Compactness [kWh/t] Max. Temp. [°C] Cost [$/kWh] Storage period Typical materials(Max. temp. [°C])
Sensible 0.1–10 10–50 500 0.1–10 Days-months Water (150)

Rocks (900)
Oil (350)
Latent 0.001–1 50–150 660 10–50 Hours-months Ice (0)
Molten salt (400)
Metallic (500)
Chemical 0.01–1 150–250 180 8–100 Hours-day NaOH (150)
LiCl (100)
Zeolite (180)
12
a packed bed storage Carnot battery system and concluded that the heater and Rankine cycle, Liquid Air Energy Storage and the Lamm-
round-trip efficiency achieved was quite poor but could still compete Honigmann process.
with the PHS and CAES systems against their energy density and spe In terms of maximum energy and power, no limit should be taken as
cific cost. Another study [89] compared the Carnot battery system with a technological constraint since each technology is scalable (up to GW
the Liquid-Air Energy Storage (LAES) system. They found out that the scale).
Carnot battery system is capable of achieving higher round-trip effi In terms of temperature, the systems can be divided into two groups:
ciencies and is competitive when sell-to-buy price ratios are considered, the high-temperature cycles (Brayton and Rankine cycles combined
but has a higher capital cost and a higher levelised cost of storage. One with an electrical heater) and the low-temperature cycles (heat pump
study [90] compares the Brayton and Rankine-based Carnot battery combined with a Rankine cycle, Lamm-Honigmann and LAES). Among
system with other grid-scale EES technologies. It was found that, in case this latter category, it appears that only the HP/ORC and the Lamm-
of energy arbitrage, if both capital costs and operational expenditures Honigmann work with a small temperature difference. It means that the
are considered, Carnot battery systems may compensate for their lower thermal integration of waste heat, for example, is more profitable for
efficiency with a very low initial cost. In this way, Carnot battery may those two technologies.
potentially outperform more efficient, and costlier, systems such as The self-discharge depends on the temperature difference between
molten salt batteries and flow batteries. the storage and the ambient. Therefore, it depends on the specific ap
Although the implementation of EES remains limited at the mo plication. Only the Lamm-Honigmann process presents very low self-
ment, mainly for financial and technical reasons, it is expected to in discharge thanks to its thermochemical reaction.
crease rapidly in the coming years. In this context, an EES technology The price comparison is only qualitative because the numbers pro
may be an attractive solution, as it promises to be cost-effective with a vided in Table 5 are expected to evolve in the future. The prices include
long lifetime (unlike many battery chemistries), and site-independent installation and operational costs. Also, it should be noted that prices
(unlike PHS and CAES). depend on the power (and energy) range. In Table 5, prices refer to the
power ranges considered for the prototypes in Table 4. From the first
economic considerations [17], it appears that the storage would ac
4.2. Prototypes under test or construction
count for a small fraction (8 to 30 %) of the total costs. The electrical
heater combined with a RC should be the cheapest if it can re-use ex
Table 4 presents the different prototypes of Carnot battery that have
isting fossil-fuel power plant. The HP/RC and Lamm-Honigmann cycles
been developed and were reported in the literature. Since the tech
should also be cheap since off-the-shelf components are used for the
nology is emerging, many prototypes are being built and this is the
systems but also the thermal energy storage. Finally, the Brayton cycle
reason why this table is also proposed online with continuous updates
and LAES seem to present similar costs according to the literature.
[91]. Globally, Carnot batteries are not mature yet. Only 2 medium-
Up to now, the technological maturity of an electrical heater com
scale projects have, up to date, been connected to the grid and operated
bined with RC and LAES is already proven (Table 4). Carnot Batteries
successfully. One of them is a LAES of 300 kW [77]. The second one is
based on HP/RC cycles are not as mature as the two aforementioned
an electrical heater combined with a Rankine cycle [50]. However, this
technologies, but the heat pump and RC are already developed world
system presents a relatively low round-trip efficiency, and other con
wide, and some large-scale demonstrators should be built in the coming
figurations are being investigated. A 150 kW Brayton cycle was suc
years (see Table 4). The Lamm-Honingmann process is not mature since
cessfully tested in 2019 [17]. Furthermore, two thermally integrated
only one demonstrator is found in the literature (Table 4).
reversible HP/RC cycles are being tested on the lab-scale [40,49]. Fi
The Lamm-Honigmann process presents the lowest round-trip effi
nally, a 10 kW Carnot battery using a heat pump and a Rankine cycle
ciency (ɛrt), but the latter could probably be improved significantly
has been built [92].
since this technology is arguably the least mature. The electrical and RC
It can be noted from this table that only a few configurations have
technology presents the second-lowest ɛrt because of the low perfor
been tested, and intensive experimental research is necessary in order
mance of an electric heater compared to a thermodynamic cycle. The
to identify the advantages, constraints and costs for each of them.
highest ɛrt are obtained with the Brayton cycle and the HP/RC (parti
cularly with thermal integration). The ɛrt of the LAES is lower than the
4.2.1. Comparison and discussion two aforementioned cycles.
It is essential to identify which technology of Carnot battery is the The Rankine PTES could be a valid alternative to Brayton because it
most appropriate in a given case study. This task is tricky since this generally achieves higher energy densities and it stores energy at much
technology is under development and very few prototypes have been lower temperatures, which is beneficial for thermal losses and the
built and tested up to now (see Section 3). However, the main char choice of reservoir/machines materials. Furthermore, it may allow for
acteristics for each technology are summarised in Table 5, with the the use of phase change materials as a storage medium.
current state-of-the-art (which could potentially evolve rapidly in the A comparison of the Carnot battery with other electrical storage
next decade). The five leading technologies are compared: Brayton, a solutions lies outside the scope of this paper. However, the reader can
combination of heat pump and Rankine cycle, combination of electrical
Table 4
Carnot battery existing prototypes.
Year Type Electrical power [kW] Electrical energy [kWh] Working fluid Storage Temp. [°C] ɛRT [%] Refs
2019 TI Inv. rev. HP/ORC 2 - R1234yf Ice (1m3) 60 100 [40]

2019 TI Inv. Rev. HP/ORC 1 10 R1233zd Water (1m3) 90 100 [49]
2014 Electrical heater+Rankine 700 5 000 Water Rock (40 t) 600 45 [50]
2019 Electrical heater+Rankine 1400 12 000 Water Rock (1000 t) 600 45 [50]
2019 Brayton 150 600 Air Packed bed (9m3) 500 72 [17]
2019 ORC+HP 1000 - R1233zd Sensible+latent 180 - [92]
2011 LAES 300 2500 Air Air - 12 [76]
2018 LAES 5000 15,000 Air Air - - [77]

2020 Rev. HP/ORC 7500 250,000 CO2 Water 150 - [93]
13
Table 5
Comparison of the different technologies.
Cycle Brayton Electrical heater and Rankine Cycle Heat pump and Rankine cycle Liquid Air Lamm-Honigmann
cycle
Power [MW] Up to 100 Up to 100 Up to 10* [10–7800] N/A

Energy [MWh] Up to 400 Up to 400 Up to 40* [50–650] N/A
Temp. [°C] [-70:1000] Up to 750 Up to 150 -196 N/A
Compactness [kW/m3] 25 ~4 [0.05–1.72] [6–46] N/A
Compactness [kWh/m3] 200 ~36 [0.2–207] [32–230] N/A
Self-discharge medium Very low
ɛrt [%] [60–70] [12–55] [30–73] [12–60] [4- N/A]
[70–150]⁎⁎
Price [$/kW] [395–875] ~376 [272–468] [329–3846] N/A
Price [$/kWh] [55–198] ~94 [68–117] [66–666] N/A
Estimated TRL 5 9 7 9 1
Typical fluids Argon, Air Water R1233zd(E),CO2, NH3, water Air H2O/LiBr, H2O/NaOH
References [17,25] [50] [38,92] [82] [53]
⁎
Possible to extend the range by association in series.
⁎⁎
Thermally integrated.
consult other references [38,55]. demonstrated that polytropic efficiency of the cycle components
must be very high (> 0.9) to achieve acceptable round-trip effi
5. Conclusions ciencies. However, both working fluid (argon) and inlet /outlet
machine conditions are non-standard. Furthermore, in the charging
This review of the Carnot battery technology proposes a state-of- phase, Brayton PTES operates with compression at a very high
the-art of this very innovative technology. First, a standard definition is temperature, as the transformation starts at 350 °C – 400 °C and may
proposed to identify which technologies are to be included. theoretically end at 1000 °C. Such operating conditions are currently
In this paper, Rankine, Brayton, LAES and Lamm-Honigmann unfeasible, as modern gas turbine compressor can only operate up to
Carnot batteries are considered. Each of them is described, and the 500 °C. For the listed reasons, the research should focus on the
possible layouts are compared. Furthermore, a review of the Thermal characterisation and design of the machines required for such sys
Energy Storages is proposed and shows which technology should be tems.
used with each application. • Integration with other systems. PTES round-trip efficiencies should
In addition, new performance indicators are proposed as a standard be improved to reduce the losses that a CB battery in mass de
to define the performance of the technology. ployment would cause to the electrical systems. In particular, ther
Finally, existing prototypes and orders of magnitudes of compact mally integrated systems prove to have substantial advantages in
ness, performance, range of power and maturity for the different terms of electric energy loss reduction. However, only a few studies
technologies are given. Based on this background, guidelines are drawn investigated the TI-PTES integration in terms of the actual avail
to select the optimal configuration for a given case study. ability of waste heat/cold energy streams to exploit. In other words,
where, how often and at which cost additional thermal energy
6. Perspectives streams are available? What is their power? Since the available heat
flow is fixed by the upstream system (e.g. an industrial facility, for
As already mentioned, the Carnot battery technology is relatively waste heat), how big the electrical storage system can be, given the
recent, and many aspects have to be studied in details. Here is a list available waste energy? Lastly, given the answer to the questions
with the main challenges to overcome to obtain a complete character above, what is the real technical potential for TI-PTES systems?
isation of the Carnot batteries. • Experimental validation of theoretical studies. As the PTES tech
nology is very novel and many technological breakthroughs are
• Part load performance. Many technologies are in a stage that only expected in the next decade, experimental validation of each con
thermodynamic design has been proposed, so part load calculations cept and architecture is crucial to identify which system present the
should be addressed to provide operators with an idea of how much optimal combination of features depending on the case (power/ca
performance degrades in part-load operation. This is especially in pacity range, additional thermal energy availability, etc.).
teresting since part load also means changing working temperatures. • Carnot battery control. Control strategies should be developed and
This leads to severe issues as the TES temperatures link the charge studied to maximise the selected objective function, depending on
and discharge, so a part load in one of the phases might also in the chosen CB concept, its power/capacity range, the electricity
fluence the other subsequent phase. price fluctuations, the electrical demand and production and the
• Dynamic simulations. Storage sytems must be able to respond eventual integration of waste heat.
quickly. Currently, it is not clear what are the characteristic start-up • Finally, it would be very helpful to provide an accurate and vali
times of the different CB technologies. Start-up times should be dated cost estimation for each type of CB. This would actively help
provided for so-called hot start, warm start and cold start conditions. to promote the PTES technology and to obtain the trust of investors.
This would characterise the CB in a similar way to thermal power
plants. Furthermore, such a dynamic analysis would clarify whether
CB can be considered flexible enough to provide the grid services Declaration of Competing Interest
required for the RES integration and on which time scale. For ex
ample, can a CB provide frequency regulation, or is it useful only for The authors declare that they have no known competing financial
shifting large quantities of energy from a moment of the day to interests or personal relationships that could have appeared to influ
another? ence the work reported in this paper.
• Machine selection and design in Brayton PTES. Preliminary studies
14
Acknowledgements packed-bed thermal reservoirs for energy storage, Appl. Energy 130 (2014)
648–657, https://doi.org/10.1016/j.apenergy.2014.02.071.
[27] M. Morandin, F. Maréchal, M. Mercangöz, F. Buchter, Conceptual design of a
The authors would like to thank Lorenzo Ferrari, Associate Professor thermo-electrical energy storage system based on heat integration of thermo
from the University of Pisa (Italy), for his help in the definition of some dynamic cycles – Part A: Methodology and base case, Energy 45 (2012) 375–385,
performance indicators. https://doi.org/10.1016/j.energy.2012.03.031 a.
[28] Y.M. Kim, D.G. Shin, S.Y. Lee, D. Favrat, Isothermal transcritical CO2 cycles with
TES (thermal energy storage) for electricity storage, Energy 49 (2013) 484–501,
References https://doi.org/10.1016/j.energy.2012.09.057.
[29] M. Mercangöz, J. Hemrle, L. Kaufmann, A. Z'Graggen, C. Ohler, Electrothermal
energy storage with transcritical CO2 cycles, Energy 45 (2012) 407–415, https://
[1] European Commission, Clean Energy for All Europeans, (2016).
doi.org/10.1016/j.energy.2012.03.013.
[2] U.S. Environmental Protection Agency (EPA), Renewable Energy in the Clean
[30] WD. Steinmann, The CHEST (Compressed Heat Energy STorage) concept for facility
Power Plan, (2015).
scale thermo mechanical energy storage, Energy 69 (2014) 543–552, https://doi.
[3] M. Gimeno-Gutiérrez, R. Lacal-Arántegui, Assessment of the European potential for
org/10.1016/j.energy.2014.03.049.
pumped hydropower energy storage based on two existing reservoirs, Renew.
[31] F. Frate G, M. Antonelli, U. Desideri, Pumped thermal electricity storage: an in
Energy 75 (2015) 856–868, https://doi.org/10.1016/j.renene.2014.10.068.
teresting technology for power-to-heat applications, 30th Int. Conf. Effic. Cost,
[4] X. Hu, F. Feng, K. Liu, L. Zhang, J. Xie, B. Liu, State estimation for advanced battery
Optim. Simul. Environ. Impact Energy Syst. ECOS 2017, International Measurement
management: key challenges and future trends, Renew. Sustain. Energy Rev. 114
Confederation, 2017, pp. 1–12.
(2019) 109334.
[32] G.F. Frate, M. Antonelli, U. Desideri, A novel pumped thermal electricity storage
[5] IEA, World Energy Outlook 2018, (2018) Paris.
(PTES) system with thermal integration, Appl. Therm. Eng. 121 (2017) 1051–1058,
[6] Q. Hou, N. Zhang, E. Du, M. Miao, F. Peng, C. Kang, Probabilistic duck curve in high
https://doi.org/10.1016/j.applthermaleng.2017.04.127.
PV penetration power system: concept, modeling, and empirical analysis in China,
[33] S. Henchoz, F. Buchter, D. Favrat, M. Morandin, M. Mercangöz, Thermoeconomic
Appl. Energy 242 (2019) 205–215, https://doi.org/10.1016/j.apenergy.2019.03.
analysis of a solar enhanced energy storage concept based on thermodynamic cy
067.
cles, Energy 45 (2012) 358–365, https://doi.org/10.1016/j.energy.2012.02.010.
[7] M. Obi, R. Bass, Trends and challenges of grid-connected photovoltaic systems - a
[34] H. Jockenhöfer, W.D. Steinmann, D. Bauer, Detailed numerical investigation of a
review, Renew. Sustain Energy Rev. 58 (2016) 1082–1094, https://doi.org/10.
pumped thermal energy storage with low temperature heat integration, Energy 145
1016/j.rser.2015.12.289.
(2018) 665–676, https://doi.org/10.1016/j.energy.2017.12.087.
[8] Few S., Schmidt O., Offer G.J., Brandon N., Nelson J., Gambhir A.. Prospective
[35] RB. Peterson, A concept for storing utility-scale electrical energy in the form of
improvements in cost and cycle life of off-grid lithium-ion battery packs: an analysis
latent heat, Energy 36 (2011) 6098–6109, https://doi.org/10.1016/j.energy.2011.
informed by expert eli.
08.003.
[9] The Guardian, South Australia's Tesla Battery on Track to Make Back a Third of Cost
[36] D. Roskosch, B. Atakan, Pumped heat electricity storage: potential analysis and orc
in a Year, (2018) https://www.theguardian.com/technology/2018/sep/27/south-
requirements, Energy Procedia 129 (2017) 1026–1033, https://doi.org/10.1016/j.
australias-tesla-battery-on-track-to-make-back-a-third-of-cost-in-a-year (accessed
egypro.2017.09.235.
July 20, 2019).
[37] S. Staub, P. Bazan, K. Braimakis, D. Müller, C. Regensburger, D. Scharrer, et al.,
[10] M.C. Argyrou, P. Christodoulides, S.A. Kalogirou, Energy storage for electricity
Reversible heat pump-organic rankine cycle systems for the storage of renewable
generation and related processes: technologies appraisal and grid scale applications,
electricity, Energies (2018) 11, https://doi.org/10.3390/en11061352.
Renew. Sustain. Energy Rev. 94 (2018) 804–821, https://doi.org/10.1016/j.rser.
[38] O. Dumont, V. Lemort, Mapping of performance of pumped thermal energy storage
2018.06.044.
(Carnot battery) using waste heat recovery, Energy (2020) in press.
[11] A. Benato, A. Stoppato, Pumped thermal electricity storage: a technology overview,
[39] Pillai A., Kaya A., De Paepe M., Lecompte S., Performance analysis of an organic
Therm. Sci. Eng. Prog. 6 (2018) 301–315, https://doi.org/10.1016/j.tsep.2018.01.
rankine cycle for integration in a carnot battery. 5th International Seminar on ORC
017.
Power Systems.
[12] X. Luo, J. Wang, M. Dooner, J. Clarke, Overview of current development in elec
[40] AES, 2019. https://www.activeenergysystems.com/, consulted on the 03/06/2019.
trical energy storage technologies and the application potential in power system
[41] O. Dumont, V. Lemort, Thermo-technical approach to characterise the performance
operation, Appl. Energy 137 (2015) 511–536, https://doi.org/10.1016/j.apenergy.
of a reversible heat pump/organic Rankine cycle power system depending on its
2014.09.081.
operational conditions, Proceedings of the ECOS 2019 – The 32st International
[13] X. Luo, J. Wang, M. Dooner, J. Clarke, C. Krupke, Overview of current development
Conference on Efficiency, Cost, Optimisation, Simulation and Environmental
in compressed air energy storage technology, Energy Procedia 62 (2014) 603–611,
Impact of Energy Systems, 2019.
https://doi.org/10.1016/j.egypro.2014.12.423.
[42] Olivier, G., 2006. Système et Procédé de Gestion d’énergie d'un Véhicule. French
[14] C. Damak, D. Leducq, H.M. Hoang, D. Negro, A. Delahaye, Liquid Air Energy
patent FR29132117.
Storage(LAES) as a large-scale storage technology for renewable energy integration
[43] B. Peris, O. Dumont, S. Quoilin, J. Navarro-Esbría, Internal combustion engines
- a review of investigation studies and near perspectives of LAES, Int. J. Refrig.
cooling water valorisation through invertible HTHP/ORC systems, Symposium
(2019), https://doi.org/10.1016/J.IJREFRIG.2019.11.009.
Waste Heat Valorisation in Industrial Processes, Kortrijk, Belgium, 2016.
[15] WD. Steinmann, Thermo-mechanical concepts for bulk energy storage, Renew.
[44] S. Schimpf, K. Uitz, R. Span, Simulation of a solar assisted combined heat pump-
Sustain. Energy Rev. 75 (2017) 205–219, https://doi.org/10.1016/j.rser.2016.10.
organic Rankine cycle system, Proceedings of World renewable Energy Congress,
065.
2011, Sweden, 2011.
[16] Malta, 2019. https://x.company/projects/malta/ consulted on the 14/10/2019.
[45] O. Dumont, C. Carmo, V. Fontaine, F. Randaxhe, S. Quoilin, V. Lemort,
[17] A. Smallbone, V. Jülch, R. Wardle, A.P. Roskilly, Levelised cost of storage for
B. Elmegaard, M.P. Nielsen, Performance of a reversible heat pump / organic
pumped heat energy storage in comparison with other energy storage technologies,
Rankine cycle unit coupled with a passive house to get a Positive Energy Building, J.
Energy Convers. Manag. 152 (2017) 221–228, https://doi.org/10.1016/j.
Build. Perform. Simul. (2016) 11–2018.
enconman.2017.09.047.
[46] O. Dumont, Investigation of a Heat Pump Reversible into an Organic Rankine Cycle
[18] K. Attonaty, P. Stouffs, J. Pouvreau, J. Oriol, A. Deydier, Thermodynamic analysis
and its Application in the Building Sector, (2017).
of a 200MWh electricity storage system based on high temperature thermal energy
[47] O. Dumont, S. Quoilin, V. Lemort, Experimental investigation of a reversible heat
storage, Energy 172 (2019) 1132–1143.
pump/organic Rankine cycle unit designed to be coupled with a passive house to
[19] T. Desrues, J. Ruer, P. Marty, J.F. Fourmigué, A thermal energy storage process for
get a Net Zero Energy Building, Int. J. Refrig. 54 (2015) 190–203, https://doi.org/
large scale electric applications, Appl. Therm. Eng. 30 (2010) 425–432, https://doi.
10.1016/j.ijrefrig.2015.03.008.
org/10.1016/j.applthermaleng.2009.10.002.
[48] D.I. Caraino, W. Nader, F. Breque, M. Nemer, Assesing fuel consumption reduction
[20] J. Howes, Concept and Development of a Pumped Heat Electricity Storage Device,
of revercycle: a reversible mobile air conditioning/organic Rankine cycle system,
IEEE, 2012, https://doi.org/10.1109/JPROC.2011.2174529 0018-9219.
Proceeding of ORC conference, 2019.
[21] A. White, G. Parks, C.N. Markides, Thermodynamic analysis of pumped thermal
[49] O. Dumont, V. Lemort, First Experimental Results of a Thermally Integrated Carnot
electricity storage, Appl. Therm. Eng. 53 (2013) 291–298, https://doi.org/10.
Battery Using a Reversible Heat Pump / Organic Rankine Cycle, 2nd international
1016/j.applthermaleng.2012.03.030.
workshop on Carnot batteries, 2020 . http://hdl.handle.net/2268/250924 .
[22] J. Guo, L. Cai, J. Chen, Y. Zhou, Performance evaluation and parametric choice
[50] Siemens, 2019. https://www.siemensgamesa.com/products-and-services/hybrid-
criteria of a Brayton pumped thermal electricity storage system, Energy 113 (2016)
and-storage/thermal-energy-storage-with-etes consulted on the 15/09/2017.
693–701, https://doi.org/10.1016/j.energy.2016.07.080.
Consulted on the 09/09/2019.
[23] P. Farres-Antunez, H. Xue, AJ. White, Thermodynamic analysis and optimisation of
[51] T. Okazaki, Y. Yasuyuki, N. Taketsune, Concept study of wind power utilising direct
a combined liquid air and pumped thermal energy storage cycle. J. Energy Storage
thermal energy conversion and thermal energy storage, Renew. Energy 83 (2015)
18 (August) (2018) 90–102, https://doi.org/10.1016/j.est.2018.04.016 citations.
332–338, https://doi.org/10.1016/j.renene.2015.04.027.
Energy Policy 2018;114:578–9010.1016/j.enpol.2017.12.033.
[52] Honigmann M.Utilization of exhaust-steam. Specification forming part of Reissued
[24] A. Benato, A. Stoppato, Heat transfer fluid and material selection for an innovative
Patent No. 10,675, dated 22 December 1885. Original No. 287,937, dated 6
pumped thermal electricity storage system, Energy 147 (2018) 155–168, https://
November 1883. Application for reissue filed 6 October 1884, Serial No. 144,891.
doi.org/10.1016/j.energy.2018.01.045.
United States Patent Office.
[25] F. Ni, HS. Caram, Analysis of pumped heat electricity storage process using ex
[53] A. Jahnke, L. Strenge, V. Fleßner, N. Wolf, T. Jungnickel, F. Ziegler, First cycle
ponential matrix solutions, Appl. Therm. Eng. 84 (2015) 34–44, https://doi.org/10.
simulations of the Honigmann process with LiBr/H2O and NaOH/H2O as working
1016/j.applthermaleng.2015.02.046.
fluid pairs as a thermochemical energy storage, Int. J. Low-Carbon Technol. 8
[26] A. White, J. McTigue, C. Markides, Wave propagation and thermodynamic losses in
(2013) i55–i61.
15
[54] W. Steinmann, The CHEST (Compressed Heat Energy STorage) concept for facility energy storage systems, Int. J. Heat Mass Transf. 30 (1987) 43–57.
scale thermo mechanical energy storage, Energy (2014) 543–552. [75] M.C. Lott, Technology Roadmap: Energy storage, International Energy Agency,
[55] Frate, G.F., Ferrari, L., Desideri, U., Multi-criteria investigation of a Pumped Paris, France, 2014.
Thermal Electricity Storage (PTES) system with thermal integration and sensible [76] “Highview Power.” 2019. 2019. https://www.highviewpower.com/technology/.
heat storage, Energy Convers. Manag. 208, doi:10.1016/j.enconman.2020.112530. [77] R. Morgan, S. Nelmes, E. Gibson, G. Brett, Liquid air energy storage - analysis and
[56] J.D. McTigue, A.J. White, C.N. Markides, Parametric studies and optimisation of first results from a pilot scale demonstration plant, Appl. Energy 137 (January)
pumped thermal electricity storage, Appl. Energy 137 (2015) 800–811, https://doi. (2015) 845–853, https://doi.org/10.1016/j.apenergy.2014.07.109.
org/10.1016/j.apenergy.2014.08.039. [78] M. Antonelli, S. Barsali, U. Desideri, R. Giglioli, F. Paganucci, G. Pasini, Liquid air
[57] T.J. Kotas, Basic exergy concepts, The Exergy Method of Thermal Plant Analysis, energy storage: potential and challenges of hybrid power plants, Appl. Energy 194
Elsevier, 1985, pp. 29–56, https://doi.org/10.1016/B978-0-408-01350-5.50009-X. (May) (2017) 522–529, https://doi.org/10.1016/j.apenergy.2016.11.091.
[58] I.a. Sarbu, A comprehensive review of thermal energy storage, Sustainability (2018) [79] A. Tafone, E. Borri, G. Comodi, M. van denBroek, A. Romagnoli, Liquid air energy
191. storage performance enhancement by means of organic rankine cycle and absorp
[59] S. Hasnain, Review on sustainable thermal energy storage technologies, Part I: heat tion chiller, Appl. Energy 228 (October) (2018) 1810–1821, https://doi.org/10.
storage materials and techniques, Energy Convers. Manag. (1998) 1127–1138. 1016/j.apenergy.2018.06.133.
[60] S. Furbo, Using water for heat storage in thermal energy storage (TES) systems, in: [80] I. Lee, F. You, Systems design and analysis of liquid air energy storage from li
S. Furbo (Ed.), Advances in Thermal Energy Storage Systems, Elsevier, 2015, pp. quefied natural gas cold energy, Appl. Energy 242 (May) (2019) 168–180, https://
31–47. doi.org/10.1016/j.apenergy.2019.03.087.
[61] A.B. Gallo, J.R. Simões-Moreira, Energy storage in the energy transition context: a [81] Lecompte S., Gusev S., Vanslambrouck B., De Paepe M., Experimental results of a
technology review, Renew. Sustain. Energy Rev. (2016) 800–822. small-scale organic Rankine cycle: steady state identification and application to off-
[62] S. Kalaiselvam, R. Parameshwaran, Thermal energy storage technologies for sus design model validation. Appl. Energy 226, 82–106 10.1016/j.apenergy.2018.05.
tainability systems design, Assess. Appl. (2014). 103.
[63] L. Wang, X. Lin, L. Chai, L. Peng, D. Yu, J. Liu, et al., Unbalanced mass flow rate of [82] S. Peng, A. Fuchs, Polymeric phase change composites for thermal energy storage, J.
packed bed thermal energy storage and its influence on the Joule-Brayton based Appl. Polym. Sci. (2004) 1240–1251.
Pumped Thermal Electricity Storage, Energy Convers. Manag. 185 (2019) 593–602, [83] H. WG., Compact Thermal Energy Storage: Material Development and System
https://doi.org/10.1016/j.enconman.2019.02.022. Integration, IEA, 2015.
[64] H. Mehling, LF. Cabeza, Heat and Cold Storage with PCM, Springer, Berlin, 2008. [84] IEA, Technology Roadmap, Energy Storage, OECD/IEA, 2014.
[65] L.R. Shao, Figure-of-merit for phase-change materials used in thermal management, [85] B. BMWi, Energiekonzept für eine Umweltschonende, Zuverlässige und Bezahlbare
Int. J. Heat Mass Transf. (2016) 764–771. Energieversorgung, Bundesministerium für Wirtschaft und Technologie (BMWi),
[66] S.J. Gurrum, Thermal management of high temperature pulsed electronics using Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit (BMU), Berlin,
metallic phase change materials, Numer. Heat Transf. Part A (2002) 777–790. 2010.
[67] H.a. Ge, Keeping smartphones cool with gallium phase change material, J. Heat [86] E. Commission, Energy Roadmap 2050, European Commission, Brussels, 2011.
Transf. (2013) 054503. [87] M. Fursch, S. Hagspiel, C. Jagemann, S. Nagl, D. Lindenberger, E. Troster, The role
[68] D.a. Pal, Application of phase change materials to thermal control of electronic of grid extensions in a cost-efficient transformation of the European electricity
modules: a computational study, J. Electron. Packag. (1997) 40–50. system until 2050, Appl. Energy 104 (2013) 642–652.
[69] X.T. Yang, Evaluation and optimisation of low melting point metal PCM heat sink [88] A. Benato, Performance and cost evaluation of an innovative Pumped Thermal
against ultra-high thermal shock, Appl. Therm. Eng. (2017) 34–41. Electricity Storage power system, Energy 138 (2017) 419–436, https://doi.org/10.
[70] X.T. Yang, Finned heat pipe assisted low melting point metal PCM heat sink against 1016/j.energy.2017.07.066.
extremely high power thermal shock, Energy Convers. Manag. (2018) 467–476. [89] S.N. Georgiou, A thermo-economic analysis and comparison of pumped-thermal
[71] G. Muñoz, A. Solé, C. Barreneche, L. Cabeza, Corrosion of metal containers for use and liquid-air electricity storage systems, Appl. energy (2018) 1119–1133.
in PCM energy storage, Renew, Energy 76 (2015) 465–469, https://doi.org/10. [90] G.F. Frate, L. Ferrari, U. Desideri, Critical review and economic feasibility analysis
1016/j.renene.2014.11.036. of electric energy storage technologies suited for grid scale applications, E3S Web
[72] Y. Li, H. Cao, S. Wang, Yi Jin, D. Li, X. Wang, Y. Ding, Load Shifting of nuclear Conf 137 (2019) 01037, https://doi.org/10.1051/e3sconf/201913701037.
power plants using cryogenic energy storage technology, Appl, Energy 113 [91] https://users.ugent.be/~slecompt/carnot_battery.htm.
(January) (2014) 1710–1716, https://doi.org/10.1016/j.apenergy.2013.08.077. [92] Chester, 2019. https://www.chester-project.eu/, consulted on the 10/10/2019.
[73] H. WG., Compact Thermal Energy Storage: Material Development and System [93] https://www.man-es.com/docs/default-source/energy-storage/etes/etes-energy-
Integration, IEA, 2015. storage.pdf?sfvrsn=54340d66_4consulted on the 10/10/2020.
[74] R.J. Krane, A Second Law analysis of the optimum design and operation of thermal
16
View publication stats

Carnot Battery Technology: A State-Of-The-Art Review

Uploaded by

Copyright:

Available Formats

Carnot Battery Technology: A State-Of-The-Art Review

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Carnot Battery Technology: A State-Of-The-Art Review

Uploaded by

Copyright:

Available Formats

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

Carnot battery technology: A state-of-the-art review

Article in The Journal of Energy Storage · September 2020

Olivier Dumont Guido Francesco Frate

SEE PROFILE SEE PROFILE

Aditya Pillai Steven Lecompte

SEE PROFILE SEE PROFILE

Bricker View project

The user has requested enhancement of the downloaded file.

Contents lists available at ScienceDirect

Journal of Energy Storage

Carnot battery technology: A state-of-the-art review T

ARTICLE INFO ABSTRACT

Nomenclature RES Renewable Energy Source

CAES Compressed Air Energy Storage ɛ Efficiency [-]

Fig. 1. Carnot battery definition.

Fig. 2. Possible configurations for a Carnot battery.

Fig. 3. Reversible Brayton cycle (left: charging, right: discharging).

economically feasible application. carefully investigated.

temperatures, as pointed out in Section 1.3.1. closed cycle, [53].

2015 Solar building classical 4 kW R134a [47] Finished

Fig. 5. Lamm-Honigman process (left: discharge, right: charge).

Table 2 a low-temperature LTES reservoir [40]. As far as the high-temperature

3.2.1. Liquid air energy storage (LAES)

Fig. 8. PTES system with sensible thermal energy storage.

successfully commissioned, based on the Claude cycle [76,77]. The

Sensible 0.1–10 10–50 500 0.1–10 Days-months Water (150)

2019 TI Inv. rev. HP/ORC 2 - R1234yf Ice (1m3) 60 100 [40]

2018 LAES 5000 15,000 Air Air - - [77]

Power [MW] Up to 100 Up to 100 Up to 10* [10–7800] N/A

View publication stats

You might also like