Exploring All-Solid-State Batteries using First-Principles Modelling: Effective Computational Strategies towards Better Batteries

Lead Research Organisation: University of Warwick
Department Name: Chemistry

Abstract

Energy storage has a more central role in our society today than ever before and has become one of the greatest research challenges of our time. The UK's Department of Energy & Climate Change has committed to the green-house gas emission reduction of 80% by 2050 through the Climate Change Act and has recently announced an £246-million investment in energy storage R&D. Such moves are motivated by the necessity for the UK to benefit from what is a global transition to new energy sources and more effective storage. However, solving the limitations in the current battery technologies will be key in order for the UK to develop high-performance, sustainable energy storage with low environmental impact.

Since the 1980s, rechargeable Lithium-ion batteries (LIBs) have pioneered clean and effective energy storage and revolutionised portable electronics. Similarly, LIBs can be the key technology for the development of electric vehicles and grid-scale storage of renewable energy. The upscaling of the LIBs is, however, not straightforward due to safety issues. Organic electrolyte solutions -commonly used in the conventional Li-ion batteries- are volatile, flammable and even explosive, potentially causing catastrophic failures, specifically when used in substantial amounts in multi-cell batteries to power energy-intensive applications. As we near the theoretical limits of conventional Li-ion batteries, there is an ever-growing need for next-generation battery technologies that can meet the stringent energy demand.

By replacing the organic electrolyte solutions with solid equivalents, all solid-state batteries (ASSB) can not only mitigate these safety issues, but also provide superior battery performances due to their higher energy density. This renders ASSBs ideal for challenging applications in various industries, on a small (battery on a chip or sensor), medium (electric vehicles) to large scale (grid-level storage for renewables). Three major setbacks, however, still need to be addressed before ASSBs can be fully commercialised: (1) the limited performance of the current ASSB components compared to traditional battery ones; (2) chemical, electrochemical and mechanical incompatibilities between the solid electrolytes and electrodes; (3) globally limited Li reserves, increasing the battery unit costs whilst demands for Li-ion batteries are growing.

The full potential of ASSBs as next-generation batteries can be unlocked by the discovery of new battery materials with superior features compared to current technology, such as higher energy densities, faster charge rates, safer operation, better component compatibility and lower prices. Based on lab-based trial-and-error, the experimental materials discovery can be both expensive and time consuming: a new material must be synthesised and stabilized in the lab before its efficiency as a battery component can be assessed. Computational modelling tools can help accelerate this trial-and-error process both by predicting novel materials from scratch and by providing computer-based experiments to characterize the novel materials, complementing the physical experiments.

In this framework, the main goal of this project is to improve all-solid-state battery technology using a bottom-up approach by tackling these primary limitations at an atomic level using computational modelling. This goal will be achieved by addressing three objectives:
(1) To discover novel ASSB materials with superior performance, namely new solid-state electrolytes and suitable electrodes for the Li-ion and beyond Li-ion (e.g. sodium and potassium) battery technologies.
(2) To engineer better solid electrolyte-electrode interfaces within ASSBs to augment their mechanical and electrochemical stability.
(3) To rationally design ultrathin film deposition strategies to coat ASSB components to augment their compatibility with each other.

Planned Impact

There is growing demand for superior batteries across different applications, with the estimated market value reaching £5 billion in the UK and £50 billion in Europe by 2025. All-solid-state batteries (ASSBs), which provide the ultimate safety by eliminating the flammable liquid electrolytes, have become the holy grail for industries where battery safety is paramount, specifically the automotive sector. Pertinently, to overcome challenges in electric vehicle applications, the Faraday Institution has announced a £42 Million investment in energy storage research, through four consortia, including the solid-state batteries and battery system modelling, reiterating the key role of computational modelling.

The proposed research has ramifications in the field of energy research and great impact potential in the next 5-10 years through the discovery of novel battery materials and the development of molecular-level resolutions to the fundamental limitations in all-solid-state batteries. This ambitious research project will thus contribute to energy security and efficiency across the UK, which is defined as a resilience priority area set by EPSRC in the 2017-2020 delivery plan. In the longer term, the knowledge gained in this project will not only mitigate against the negative consequences of environmental change but also contribute towards the UK becoming a global leader in the materials for energy of the future.

Besides the academic beneficiaries, the research findings will also appeal to the industrial partners in the space of energy materials and atomic layer deposition (ALD) technology in the UK and abroad, including Johnson Matthey, Jaguar Land Rover, Infineum, Tata, Dyson Appliances, Tesla, Oxford Instruments, IMEC, ASM and ASML. Knowledge transfer will be initially realised through the connections of the project partners, whereas new partnerships will be forged at the workshops to be organised, which will be attended by industry representatives. Significant opportunities for IP generation exist, which will be pursued through patents with the professional help from the technology transfer office, Warwick Ventures.

This fellowship also prioritises the fostering of public engagement with next-generation clean energy technologies, and to increase the general awareness of the key role of fundamental computational research in the development of new technologies. These outreach activities include lab visits by schools and the department's demonstrations at the local schools, participation in the local events, such as Warwick Family Days, Science Gala, It's a Matter of Crystals and the broader-audience events, like the British Science Festival and Family Day and the Big Bang Fair. Where possible, press releases on our innovative scientific advances will be published on the Warwick University website.

The proposed project is expected to have a significant impact for academia, for industy for the economy, for society and for the environment. To warrant the widest possible impact, this fellowship will implement a rigorous agenda aiming at disseminating the scientific outputs to a broad audience.

Highlight items from this agenda are:
1. Publishing in high-impact interdisciplinary peer-reviewed journals (open-access where possible);
2. Disseminating the outputs in international scientific conferences, departmental seminars and professional networking environments (e.g. SUPERGEN, STFC Batteries and alike);
3. Organising two international workshops in addition to departmental seminar series on the fellowship topic;
4. Making the developed software, generated datasets and materials database available to the public;
5. Attending/organising outreach and public engagement activities;
6. Publicising the outputs via diverse media (university newsfeed, social media, YouTube, etc.);
7. Liaising with industrial partners, patenting and commercialisation of the novel battery materials.

Publications

10 25 50

 
Description Atomistic Simulations of Novel Materials for the Next-gen All-Solid-State Battery Technology
Amount £200,000 (GBP)
Funding ID 2594260 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 09/2021 
End 03/2025
 
Title Accelerating the prediction of large carbon clusters via structure search: Evaluation of machine-learning and classical potentials 
Description Dataset containing the chemical structure collections obtained by using different interatomic potentials that were used in the benchmark presented in the paper. 
Type Of Material Computer model/algorithm 
Year Produced 2022 
Provided To Others? Yes  
Impact The dataset can be used as basis for the further/future benchmark including other interatomic potentials or quantum chemical methods. This can also used to train ML-based interaction potential models. 
URL https://github.com/bkarasulu/Carbon-PP-Benchmark-Paper-SI
 
Title Collection of structures generated 
Description Consists of about 11,000 DFT-optimised crystal structures for novel Li-V-Nb-O phases. 
Type Of Material Database/Collection of data 
Year Produced 2024 
Provided To Others? Yes  
Impact New Li-V-Nb-O based electrolyte materials were identified using this dataset 
URL https://github.com/bkarasulu/Li-V-Nb-O_structures
 
Title Exploring the configuration space of elemental carbon with empirical and machine learned interatomic potentials 
Description This dataset contains a vertical slice of the data used to generate the results found in the publication "Exploring the configuration space of elemental carbon with empirical and machine learned interatomic potentials" It contains nested sampling input files and trajectory files for each potential studied, as well as the xml files and training data for the new potential, GAP-20U+gr. 
Type Of Material Database/Collection of data 
Year Produced 2022 
Provided To Others? Yes  
URL https://zenodo.org/record/7463705
 
Description Happy Electron Ltd collaboration 
Organisation Happy Electron
Country United Kingdom 
Sector Private 
PI Contribution Performing the computational work (i.e. DFT calculations) needed for the publication (Carbon, 2022), liaising with the co-authors from Happy Electron Ltd and academia to draft a manuscript and editing it.
Collaborator Contribution Contributing to the drafting of the manuscript and editing it, discussion of the computational results.
Impact B. Karasulu et al., "Accelerating the prediction of large carbon clusters via structure search: Evaluation of machine-learning and classical potentials", Carbon, Volume 191, May 2022, Pages 255-266. Link: https://www.sciencedirect.com/science/article/pii/S0008622322000379?via%3Dihub
Start Year 2020
 
Description Intel-Merck collaboration for a PhD studentship 
Organisation Intel Corporation
Country United States 
Sector Private 
PI Contribution Supervision of the PhD student
Collaborator Contribution Funding for a 3-year PhD studentship
Impact Just started.
Start Year 2024