*2.4. TES Systems for CB*

Lastly, as many TES systems are developed separately but their application is specifically suggested for CB, they are summarized in this chapter. These are mostly high temperature technologies, where the heat might be utilized directly in the industry, but also drive a steam turbine, eventually in some high temperature TES systems also in gas turbines. As it can be seen from the summary in Table 4, various concepts and principles are employed. The review includes only entire TES systems (technical solutions) which comprise auxiliary systems for heat transfer and connection to heat demand and source; considering also separate development of suitable materials would be a topic for a separate extensive review work.


**Table 4.** Overview of the TES systems for Carnot battery applications.

Starting from the systems using sensible heat storage, a specifically developed high temperature energy storage concrete named Heatcrete from EnergyNest [162] works on the principle of having HTF in pipes embedded into the cylindrical high temperature concrete blocks and together coupled as multiple blocks creating a thermal battery module. Suggested applications are primarily industrial heat, CSP, but CB with steam turbine is also suggested. Such is the situation for most of the TES systems listed here. The modules have undergone various pilot testing [163]. The Thermal Battery pilot has been tested at Masdar Institute Solar Platform (MISP) in Abu Dhabi, UAE with 2 × 500 kWhth thermal capacity at temperatures up to 380 ◦C over a period of more than 20 months. The measured demonstrator behaved as was predicted from numerical simulations and multi-cycle operation have proved the integrity and operational feasibility of concrete TES. The measurement of HTF inlet and outlet TESS temperature over 279 charge and discharge

cycles (6000 h) show a stable and repetitive performance, which demonstrates that the concrete storage medium stays stable with no sign of degradation. The cylindrical storage element was cut into smaller sections and, following inspection, revealed no degradation, for instance spalling or cracking. Moreover, on the samples no separation between steel pipes and concrete storage material has been revealed [150].

A similar system was developed by Storworks, also consisting of steel pipes within concrete modules, which are factory built and can be assembled on site. The material utilizing waste fly ash is however claimed to withstand higher temperatures with operation up to 600 ◦C. A 10 MWh pilot system is being built on a working power plant site [151].

Specifically developed thermal storage high temperature concrete can also take the form of granules, where heat transfer with gaseous fluid is secured by fluidized bed technology, as developed by Kraftblock [152,164]. The material is at least 85% recycled and should withstand up to 1300 ◦C. The technology is currently tested in several pilot industrial applications with containerized units of 4 to 60 MWhth.

The use of silica sand has also been developed separately as a TES system utilizing air fluidized bed heat transfer by Magaldi Green Energy [153,165]. Maximal temperatures of the system can reach up to 1000 ◦C and the design can be performed as 5 to 100 MWh thermal capacity modules, which can then be connected both in parallel and series. One significant advantage of this system is the experience from the pilot application of a very similar TES system applied to a concentrated solar mirror field.

Ceramic materials are another explored option. Structured ceramic blocks using air or other gases for charging and discharging were developed by Dürr together with Kraftanlagen München [154,166]. The technology has been adopted from more complex high temperature applications, such as regenerators in regenerative thermal oxidation processes. These ceramic blocks can withstand up to 1000 ◦C and 350 bar in dynamic thermal cycling. Experience from long term cycling is available as it was tested at Jülich solar plant [167,168]. The suggested sizing of CB using this technology is up to hundreds MW and MWh to GWh capacity.

Alternatively, direct insertion of heating element into the ceramic blocks was proposed as a high temperature TES system by a consortium led by Carboclean with EEW and University of Darmstadt. Storage material is expected to also withstand temperatures above 1000 ◦C with the storage capacity systems expected up to 1 GWhth and volumetric capacity up to 1 MWh/m3. A laboratory system was developed to test this technology with plans for integration primarily into waste incineration combined heat and power plants [155,169,170].

Ceramic material in the form of a firebrick has been additionally suggested and marketed by Electrified Thermal Solutions under trademark Joule Hive as a high temperature heat storage material up to 1700 ◦C [156]. A specifically developed electrically conductive ceramics has the advantage of generating the heat from electricity directly in the entire volume of the material. The temperatures are also well suited to gas turbine applications. The outlet air temperature in the proposed configuration, however, varies during discharging and its control is suggested by mixing with a cold air [171,172].

Use of natural materials of rock beds is also being developed as a standalone TES system, specifically by BrenmillerEnergy [157]. Maximum declared storage temperatures reach 750 ◦C and in the case of steam production, its temperature then reaches 500 ◦C. The TES system combines the storage material with charging and discharging heat exchanger (possibly usable also as steam generator) as integrated compact blocks that are then connected for larger capacity and power as containerized solution, which can be further combined. Alternatively, for electrical charging, heating elements are also directly incorporated. Sizing of the units is distinguished between industrial and utility scale. Pilot applications include projects with charge by waste heat or flue gas and smoothing profile or shifting the energy for high thermal demand in industrial or heating systems and one solar thermal project for up to 24 h a day electricity production.

Last representative of sensible heat TES system here is Lumenion. In this system, steel rods are used together with inert (nitrogen) gas circulation as heat transfer fluid, heated to temperatures up to 650 ◦C. A complete plug and play system TES is developed with pilot applications including 2.4 MWh system for district heating in Berlin operated by Vattenfal (charge 720 kW, discharge 100 kW) as a follow-up of a smaller 450 kWh system. Intended application scale is 0.2 to 20 MW systems [158] A cogeneration regime is proposed with electric efficiency around 25% and thermal 70%. Storage installation costs is estimated at €25/kWh and plans exist for up to 500 MWhth systems [173].

Molten salt is also considered for standalone TES units. Norwegian company Kyoto has developed a modular system named Heatcube fitting separate salt tanks into a 20 feet containers, providing 4–100 MWh systems with a discharge rate of up to 25 MW and around 90% energy efficiency. Charging can take place from electricity or directly from a heat source. A primary application is for industrial heat but production of up to 525 ◦C steam offers the possibility of CHP steam plant application with electric roundtrip efficiency predicted in the range 15%–25%. Two pilot projects below 1 MWh were completed in 2021 while commercial orders for 10–20 MWh units are present [159].

Molten hydroxides were proposed for the storage of heat at up to 700◦C for periods of up to 14 days by company Hyme [160]. It is a spinoff of a small modular reactor company Seaborg considering the hydroxides as a heat transfer and storage fluid. Major know-how of this solution is in a specific corrosion control technology. Compared to molten salt the hydroxides are expected to be substantially cheaper [174].

Chemical high temperature reactions are representative of another two technologies. Reversible reaction of CaO with water to form Ca (OH)2 is used by SaltX, where the granules of salt use nano-size layer coating to secure robust behavior in cycling and prevent agglomeration. Charging is performed by heat at a temperature of 550◦C and higher while discharging process releases heat up to 450 ◦C. A pilot system of 0.5 MW/10 MWh is operated as a pilot in Berlin for steam district heating, again by Vattenfall. It is a follow-up on a small 20 kW experimental system. Further pilots are planned with power to steam or heat to heat applications. The general technology concept is, however, planned to be changed from the heated screw reactor type to fluidized bed reactors [161]. Note that this process has been proposed for energy storage a long time ago [175] and much of the research is still ongoing, focusing for example on kinetics and stability [176], which the SaltX appears to have solved. The process is also similar to the more scientifically explored Calcium looping, which is also usable for high temperature TES application [177].

Most of the TES systems can also be utilized in the utility scale to improve the flexibility of the thermal fossil fired power plants. The various levels and complexity of TES integration, as well as technologies of storage, were proposed. Often the impact of storage integration on flexibility can, however, be considered as too low in the hundred MW scale plants. The maximum load variation is below 5% in [70], for around 13% in [178] and with more extensive TES integration separately for high temperature reheated steam and lower temperature feed water regeneration up to 50% in [179]. Only in the smaller plants, such as industrial CHP, can the examples of larger load variation up to 100% be found as in the case of thermochemical storage in [180]. These works can serve as a basis for the analysis of the current thermal plants' conversions to full CB as such works are absent in the available literature. A large part of the original analyses of increasing plant flexibility, such as TES integration, can remain, though the target operation regime needs to be more dynamic with frequent shutdowns, as found, on the other hand, rather commonly in concentrated solar plants. The major difference would also be in focusing on longer duration storage and on the method of charging, utilizing a surplus of renewable electricity instead of flue gases and steam.

#### **3. Discussion and Conclusions**

In the increasing need of medium and long duration energy storage, Carnot batteries (CB) offer a potentially cost-effective solution with systems ranging from large grid scale applications down to even dozens of kW. Therefore, the concept has attracted not only academic, but already also considerable industrial, research and development. Among many concepts and systems, this work provides an extensive overview of commercially developed technologies, their classification by the level of maturity and experimental development. These technologies are then put into the perspective of the scientific research, providing either background for the technologies or pointing at research opportunities. The general composition of references with actual dates points out how recent and rapidly developing the topic of CB is.

There is a total 30 CB systems reported, the technologies and states of which are described in detail. They cover a large range of concepts including both direct power to heat (P2H) Joule heating conversion and heat pump based conversion, thermodynamic cycles, such as the Rankine cycle (with steam, organic or CO2 fluid), the Brayton cycle, as well as their combination, Stirling cycle or direct heat to power (H2P) conversion with systems such as thermophotovoltaics. The range of the power output and storage capacity (respectively storage duration in hours) summarized from all collected systems is presented in Figure 5. When exact data were not available, an engineering estimate was made. It shows that the CB covers the range from kW to GW systems. The smallest ones are based on ORC and Stirling engines while the largest are generally based on current thermal power plants' technologies. The storage duration confirms the application range between about 4 and 24 h, the time range of so-called medium duration energy storage systems. With the higher maturity of these systems and the increased storage duration required, it might be feasible to add further storage capacity at a relatively low cost.

**Figure 5.** Storage power output and capacity (**a**) and discharge duration (**b**) for the commercially developed CB systems.

Out of them, the largest CB commissioned so far is a liquid air energy storage (LAES) pilot 5 MW system of Highview, where the company is furthermore constructing a full-scale unit with 50 MW and 5 h of operation. The first-of-a-kind challenge is being overcome as contracts are prepared for other LAES units and thus the technology is on a clear path for commercial applications. The second largest is Siemens Gamesa ETES, using direct power to heat conversion, the storage of heat in a rock packed bed and using a steam cycle for power to heat conversion with a 1.2 MWe and 29 MWhe scale pilot system. The only existing commercial installations are, however, much smaller-scale units of Azelio with a 13 kWe output provided by the Stirling engine, while charging is also performed by direct conversion and thermal storage comprised of an aluminum alloy phase change material. Many other systems using direct electricity to heat conversion are then at least in a state of experimental demonstration.

Systems using the heat pump principle (mostly vapor compression or Brayton cycle) for charging are much more favored in the scientific literature due to their potential of higher efficiency. Except for the Highview LAES system, the commercial development is significantly less advanced with many more systems staying at the conceptual state and yet unrealized design. This might however change in the near future as Stiesdal is planning a construction of an MW scale pilot for 2022 with a reversible Brayton and rock bed storage system. A peculiar approach to commercialization is explored by MAN, where the transcritical Rankine CO2 heat pump used for charging can be operated separately and as such is planned to be built at the 50 MWth scale.

For the reversible Brayton cycle CB, there is however also an example of the bankruptcy of the company Isentropic, stressing the need for real CB systems to have a low unit cost with a simple and robust design and the fact that their feasibility is expected to improve with future, more renewable-based grids. The collected data can be displayed from the point of round trip efficiency, which is shown in Figure 6. A wide spread of the values between about 25% and 80% can be seen, with an exception in the fuel co-fired systems reaching numerically values even over 100%. High roundtrip efficiency is often reported in conceptual systems, which is similar to theoretical results with rather optimistic estimates of system assumptions. Therefore, caution must be taken, especially for the systems existing only in the conceptual or early design states, without the support of experimental data. Even though demonstration and pilot plants exist, the values are still mostly projected for full scales and thus are usually yet to be verified.

**Figure 6.** Overview of round trip efficiency in the commercially CB systems (mostly declared as experimental values are limited). In notation (d) stands for demo, (p) pilot, (c) commercial units (built or under construction), \* for systems with additional fuel firing.

Lastly, a brief timeline of the CB commercial development considering systems from laboratory demonstrators and proof of concept systems to commercial installations and large scale pilots (often operating in commercial regime) is presented in Figure 7. It shows how the development is gaining momentum, mostly in recent years, with the first major grid scale pilot coming online in 2019, full commercialization of 13 kW modular unit in 2021 and multiple pilot and commercial systems scheduled for commissioning in 2022 and 2023. Once the experience from these systems is obtained and renewable installations with energy prices' volatility further increase, we may witness even more rapid development.


**Figure 7.** Timeline of experimental commercial CB development. In notation (d) stands for demo or laboratory prototype, (p) pilot, (c) commercial units.

Furthermore, there is a considerable development of thermal energy storage systems where the CB application is specifically suggested as one of the use cases. As the market appears to not yet be fully ready for CB applications due to the still lower added value compared to the production processes, industrial and process heat are the primary target customers. As the situation might change with the larger adoption of intermittent renewables, CB-ready TES systems are listed here as well. They often include electric charging and would require adding a relatively standard, typically steam, heat to power units.

Finally, one interesting trend in CB systems with high temperature storage is a readoption of technologies previously considered for concentrated solar power plants and industrial process heat. This confirms that CB is not that much of a novel technology, but is rather a novel use case and a combination of existing technologies. As such, perhaps the simplicity causes the electrically heated systems to be mostly absent from the scientific literature although they have the prospect of a low cost and a simple/robust system. Techno-economic studies evaluating actual costs and LCOE might especially be a suitable area for future studies.

**Author Contributions:** Conceptualisation, V.N.; methodology, V.N.; investigation, V.N., V.B., P.S.; resources, V.N., V.B., P.S., J.S.; data curation, V.N., V.B., P.S.; writing—original draft preparation, V.N. V.N., V.B., P.S.; writing—review and editing, V.N., J.S.; visualisation, V.N.; supervision, V.N., J.S.; project administration, V.N., J.S.; funding acquisition, V.N., J.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Grant Agency of the Czech Technical University in Prague, grant No. SGS21/111/OHK2/2T/12 and by the Technology Agency of the Czech Republic, grant number TJ04000326.

**Data Availability Statement:** All data presented in this study are contained within the article or its references.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **Nomenclature**

#### **Symbols**

