Grid Scale Thermal and Thermo-Chemical Electricity Storage
Lead Research Organisation:
University of Cambridge
Department Name: Engineering
Abstract
Thermo-chemical energy storage (TCES) has the potential to store very large amounts of energy within a small space and at low cost. This is achieved by converting thermal energy ('heat') to chemical energy via a reversible chemical reaction. For example, by heating a granular metallic oxide to the right temperature and at the right pressure, some of the oxygen is driven off (i.e., the substance is 'reduced') and heat is absorbed during the process. The resulting 'reduced oxide' is stable and can be stored for long periods without degradation. Heat can subsequently be recovered, when required, by passing air at elevated pressure over the reduced oxide. Some of the oxygen in the air is then absorbed but the remaining gas is heated by the reaction and (since it is also at high pressure) can be used to drive a gas turbine to generate electricity.
Other gas-solid reactions are also possible, including 'calcination' of limestone (i.e., heating it up to drive off carbon dioxide) and hydration (with steam) of e.g. calcium oxide. Each reaction has its own set of peculiarities which can be exploited to its advantage. For example, the carbon dioxide emerging from the calcination reaction can be compressed and liquefied. This in itself absorbs electrical energy (in order to drive the compressors) and constitutes an additional, surprisingly compact and stable form of energy storage, from which electricity can be recovered by using the high-pressure CO2 to drive a turbo-generator.
Compared with batteries, TCES has the potential to store energy at much lower cost per kilowatt-hour of storage capacity at grid scale, despite having a lower round trip efficiency. This is because TCES systems can be built based on unit operations and power plant technologies which scale up easily, compared to electrochemical systems. The efficiency for a thermo-chemical system is likely to be in the range 40 to 60%, however the 'conservation of energy', means that the remaining energy need not be wasted: it can be exploited for heating buildings, providing hot water or supplying heat for industrial processes. Furthermore, these systems offer the possibility to provide long duration storage without any safety hazards or pressurised storage facilities. How these technologies can contribute to various grid services, the scale needed and how best to locate them within the distribution network needs to be assessed. Many of the components have inertia, which will provide some frequency support, but the thermal response may limit service provision, particularly if waste heat is also being utilised.
In this grant we will develop and test new materials to enable more efficient and cost effective TCES processes. Issues investigated include the cycle stability of the materials, their capacities and rates of conversion. Lab scale testing will demonstrate key aspects of the cycles and provide information needed for design and modelling work to evaluate these processes. We will conduct modelling on the process flowsheet, with detailed component models to allow losses to be identified and the process and material combinations to be optimised. To understand the value of these technologies to society, we will conduct system level dynamic modelling to understand their ability to provide grid services under various scenarios, including those in which there is the provision of thermal energy for district/industrial heating applications. We will analyse and quantify the grid-scale integration potentialities of TCES technology by adopting a whole-system approach, thus its integration with electricity/heating/cooling/gas networks. This will allow us to unlock the opportunities offered by this novel multi-energy storage technology to enhance the flexibility of the energy grid as a whole, and thus enable a future energy system with a high penetration of renewables
Other gas-solid reactions are also possible, including 'calcination' of limestone (i.e., heating it up to drive off carbon dioxide) and hydration (with steam) of e.g. calcium oxide. Each reaction has its own set of peculiarities which can be exploited to its advantage. For example, the carbon dioxide emerging from the calcination reaction can be compressed and liquefied. This in itself absorbs electrical energy (in order to drive the compressors) and constitutes an additional, surprisingly compact and stable form of energy storage, from which electricity can be recovered by using the high-pressure CO2 to drive a turbo-generator.
Compared with batteries, TCES has the potential to store energy at much lower cost per kilowatt-hour of storage capacity at grid scale, despite having a lower round trip efficiency. This is because TCES systems can be built based on unit operations and power plant technologies which scale up easily, compared to electrochemical systems. The efficiency for a thermo-chemical system is likely to be in the range 40 to 60%, however the 'conservation of energy', means that the remaining energy need not be wasted: it can be exploited for heating buildings, providing hot water or supplying heat for industrial processes. Furthermore, these systems offer the possibility to provide long duration storage without any safety hazards or pressurised storage facilities. How these technologies can contribute to various grid services, the scale needed and how best to locate them within the distribution network needs to be assessed. Many of the components have inertia, which will provide some frequency support, but the thermal response may limit service provision, particularly if waste heat is also being utilised.
In this grant we will develop and test new materials to enable more efficient and cost effective TCES processes. Issues investigated include the cycle stability of the materials, their capacities and rates of conversion. Lab scale testing will demonstrate key aspects of the cycles and provide information needed for design and modelling work to evaluate these processes. We will conduct modelling on the process flowsheet, with detailed component models to allow losses to be identified and the process and material combinations to be optimised. To understand the value of these technologies to society, we will conduct system level dynamic modelling to understand their ability to provide grid services under various scenarios, including those in which there is the provision of thermal energy for district/industrial heating applications. We will analyse and quantify the grid-scale integration potentialities of TCES technology by adopting a whole-system approach, thus its integration with electricity/heating/cooling/gas networks. This will allow us to unlock the opportunities offered by this novel multi-energy storage technology to enhance the flexibility of the energy grid as a whole, and thus enable a future energy system with a high penetration of renewables
Organisations
Publications
Albay A
(2025)
Optimization-based state-of-charge management strategies for supercritical CO2 Brayton cycle pumped thermal energy storage systems
in Journal of Energy Storage
Harrison ARP
(2023)
Kinetic and Thermodynamic Enhancement of Low-Temperature Oxygen Release from Strontium Ferrite Perovskites Modified with Ag and CeO2.
in Energy & fuels : an American Chemical Society journal
Ooi B
(2024)
Kinetics of CO 2 Capture with Calcium Oxide during Direct Air Capture in a Fluidized Bed
in Energy & Fuels
Wang X
(2025)
Distributed optimization of virtual storage plant aggregating large-scale storage for grid reserve provision: Experimental validations
in Control Engineering Practice
Zhang T
(2025)
A corrective control framework for mitigating voltage fluctuations and congestion in distribution networks with high renewable energy penetration
in International Journal of Electrical Power & Energy Systems
Zhang T
(2025)
Distributed control of flexible assets in distribution networks considering personal usage plans
in Sustainable Energy, Grids and Networks
| Description | In the thermochemical system, charging makes use of an endothermic decomposition reaction, charging the reverse. Possible reversible reactions include oxidation/reduction, carbonation/calcination and hydration/dehydration. Hydration/Dehydration of materials does not require large quantities of working fluid to be stored since water is easily stored and steam/wet air are effective working fluids. Both Sr(OH)2/SrO and Ca(OH)2/CaO performed acceptably during dehydration (i.e. charging) with charge efficiencies of ~ 63%. When the entire material space was examined, these simple and readily available materials were close to the "optimal" materials. There was, however, a notable (and inherent) mismatch between materials that are optimal for charge and for discharge, which limits the overall round trip efficiency. Improved materials will not overcome this limitation. The overall round trip (electricity to electricity) efficiencies were 30 % for Sr(OH)2 and 25% for Ca(OH)2. The strontium system's higher equilibrium temperature results in lower exergy destruction. Discharge requires humidification and was one of the largest sources of irreversibility. Wet compression was able to reduce the irreversibility and raise the cycle efficiency to ~ 34% for Sr(OH)2 . Reaching higher round trip efficiencies (electricity to electricity), >40%, with the hydroxide systems requires much more efficient heat exchangers and turbomachinery. Energy densities were promising (up to 200 kWh/m3), and despite the low round trip efficiency, these cycles are still of interest as they can use cheap materials (CaO and SrO are already close to optimum), can use air as the working fluid and don't require the storage of pressurised gas. Making use of carbonation/decarbonation brings with it the issue of storing the gas. Storing the working fluid at low pressure after the discharge is problematic owing to the large volume. We investigated several ways to overcome this issue. For high temperature carbonation/decarbonation materials, the feasibility of integrating with carbon capture was investigated. Models showed that by varying the keeping capture rate constant (e.g. from an industrial source), whilst varying the release of CO2, energy could be stored. The use of lower temperature CO2 sorption materials was also explored, with new configurations of storage systems investigated. In actual operation, most systems will work at part load for a significant part of their lifetime. Dynamic modelling of the part load operation showed a wide range of round trip efficiencies, between 60 and 33% for pumped thermal storage, averaging at 42% for storage integrated with a wind farm. Failure to consider part load operation, which takes into account the performance of the turbomachinery, will lead to overoptimistic estimates of real-world storage efficiencies. This modelling work also began exploring co-generation (i.e., being able to supply heat and electricity) via the energy storage system. The key to making co-generation effective is that very few new components are needed. New algorithms were developed to manage the optimal state of charge of the system when both heat and electricity are demanded. Via dynamic models, new, predictive control strategies were shown to be effective for managing the charge and discharge of energy storage systems. Both model-predictive and model-free control strategies were investigated. The former relied on a linearised response model of the energy storage system, and could effectively control for both peak shaving and voltage support. In the model-free case, the model was replaced with a data-driven system representation, enabling decision-making in a model-free framework. This proved to be effective, for example, by reducing the amount of curtailment needed by more effectively managing the state of charge of the energy stores. |
| Exploitation Route | The models and materials produced in this work provide new insights for workers in the energy storage community. |
| Sectors | Energy Environment |
| Description | Some of the IP from this work is being used by the spinout remedium energy: https://www.remediumenergy.com |
| Title | Research data supporting "Kinetic and Thermodynamic Enhancement of Low-Temperature Oxygen Release from Strontium Ferrite Perovskites Modified with Ag and CeO2" |
| Description | Zip folder containing datasets for all measurements in manuscript and supplementary information, organised by figure number. Data for TGA measurements (Figs. 3 and 4, Figs. S9-12) are arranged in tabular text files, with columns corresponding to time, sample temperature, TGA cell temperature, and sample mass in milligrams. A separate table below the main data includes DSC heatflow measurements, but these are not reported in the main manuscript or SI. Data for Fig. 5 consist of processed TGA measurements, with equilibrium sample composition at various temperatures and partial pressures arranged in a table (.csv file format). Data for Fig. 6 and Fig. 7, and Figs. S18-19 consist of a table of pO2 sensor output (units: atm) as a function of time for 7 measured temperatures, for each sample (.csv file format). Data for Figs. 8 and 9., and Fig. S17 are MATLAB .mat data files, containing the variables 'fit_mat' (corresponding to individual 1/T and k points), 'SFO_k_int_vals', corresponding to the average value of k at each inverse temperature point, 'f_SFOint' and 'gof_SFOint' corresponding to the fitted curves and goodness of fit parameters, and 'E_a' and 'lnA' corresponding to the fitted activation energies and pre-exponential factors, with 95% confidence bands in each fitted parameter. Data for Figs. S1-3 are Excel workbooks corresponding to XRD measurements and fitted reference phases, arranged by columns as 2theta angle, measured intensity, total fitted intensity, difference between measured and fitted intensity, fitted background, and individual fitted phases as reported in the paper SI Data for Fig. S13 consist of reported values in literature for SrFeO3 composition as a function of temperature and pO2, with corresponding references given (.xlsx format). Data for Fig. S14 are composed of estimated values of Thiele modulus for measurements of oxygen release and uptake from each sample at each temperature investigated, arranged in tabular format (.csv file format). Data for Fig. S15 are estimated values of the dimensionless group kRTL/v for each set of measurements of oxygen release, arranged in tabular format (.xslx file format). Data for Fig. S16 are values of maximum rate of reaction for each sample during reduction or oxidation in the packed bed reactor (umol/(g s)), with associated standard deviation over repeated cycles. |
| Type Of Material | Database/Collection of data |
| Year Produced | 2023 |
| Provided To Others? | Yes |
| URL | https://www.repository.cam.ac.uk/handle/1810/357119 |
| Company Name | Remedium |
| Description | Remedium develops technology which aims to capture carbon and create a battery storing energy for the grid. |
| Year Established | 2023 |
| Impact | Part of TechX clean accelerator for the net zero technology centre. |
| Website | https://www.remediumenergy.com/ |
| Description | Conference presentation on TCES |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | A presentation entitles "Thermo-chemical storage of electricity" was delivered by Dr Alex White at the 2023 Off-shore Energy and Storage (OSES) Conference in Malta, 12-14 July 2023. The authors of the presentation were M. Saghafifar, S. Scott and A.J White |
| Year(s) Of Engagement Activity | 2023 |
| URL | https://www.osessociety.com/oses2023 |
| Description | Engaged with Process industries board of IMechE - energy storage working group. |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Industry/Business |
| Results and Impact | Joined working group, orgnaised workshop on energy storage. |
| Year(s) Of Engagement Activity | 2024 |
| Description | ISCRE27 conference |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | ICSRE27 took place in Canada. This is one of the largest Chemical Engineering conferences globally. About 50 people attended our presentation and participated in the discussion afterwards. About 300 people were present at the conference and received our abstract. We were invited to submit a paper to the Chemical Engineering Journal |
| Year(s) Of Engagement Activity | 2023 |
| Description | Keynote at International Conference on Chemical Reaction Engineering. |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | Keynote talk at the 27th International Symposium for Chemical Reaction Engineering (ISCRE 27) on chemical looping technology |
| Year(s) Of Engagement Activity | 2023 |
| Description | Lecture for undergraduate students - UCL |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | Regional |
| Primary Audience | Undergraduate students |
| Results and Impact | I described the project to a cohort of ~120 1st year undergraduate students in mechanical engineering at UCL. The presentation was followed by a very active discussion with about 20 students |
| Year(s) Of Engagement Activity | 2023 |
| Description | Talk given at the 9th IEAHGH high temperature looping meeting |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | Talk on energy storage using chemical looping. |
| Year(s) Of Engagement Activity | 2023 |
