Next generation oxide electrolytes for solid-state batteries

New generations of lithium batteries will play a crucial role in the energy systems of the future by storing energy from renewable energy sources such as solar and wind power. It will also help mitigate greenhouse gasses from the transport sector and be necessary for all portable electronics. Fireproof and environmentally-friendly batteries, which can quickly and withstand high temperatures, can be achieved by basing all components of batteries on solid materials such as "All Solid-State-Batteries" (ASSB). That is, batteries that are fireproof, safe during use and tolerant of higher temperatures. In the OxiBat project funded by NFR, a series of new electrolyte compositions have been developed to optimize the ionic conductivity. The conductivity of the electrolyte is one of many challenges in the development of ASSB. The best ones have a conductivity of 1E-3 to 1E-4 S/cm which means that the film should be less than 100 µm for practical applications. A new series of electrolytes based on LLZO, with added Al structurally, has been made with a new recipe that gave dense samples with high homogeneity. Measurements on these electrolytes show that we have achieved very good results with low additions of Al compared to the literature, while for those with the most additions the difference is small, the improvement lies mainly in the production of the materials that give phase purity, i.e. stabilized cubic phase. The Pulse Laser Deposition (PLD) equipment at UiO was available only once to make a series of depositions of battery cell elements at RT and 300 C on gold-coated steel alloys with corresponding TEC. Separate elements were deposited as only cathode or only electrolyte, as well as deposition of symmetrical cells (cathode (NMC)-electrolyte (LLZO)-cathode (NMC)) with and without LNO interface layer to limit unwanted interface reactions. Some anode-free cells ie metal-electrolyte and cathode were also produced. Some were tested directly and some after post calcination at 660C to verify that more crystallization and improvement of interfacial contacts gave better results. Measurement of electrolyte resistance gave a lower conductivity than expected in the range 1E-4 to 1-E5 S/cm. The contact electrodes on the cathode were many and small to get some statistics (8 electrodes with 2mm diameter and 3 with 5mm diameter) but also because deposition was not optimized and some cracks were observed. The lower conductivity for low temperature deposition (300 and 25C) is due to presence of amorphous phases, while two of the post-treated symmetrical cells at 660C 12min gave the expected reduction in resistance related to improved crystallinity. The uncertainties in the measurements is high related to non-optimized film and film quality (cracks & pin holes). DFT calculations have been used to study the effect of aliovalent doping on the charge carrier concentration in the Li Garnet structure and how this doping affects the mobility of the Li vacancies. Tetragonal LLZO was tried doped with Al, Nb and Y. Al ions show a preference for four coordinated Li positions, Nb ions show a preference for Zr positions and Y ions show a preference for La positions. Doping with Y will thus not lead to a charge difference that must be compensated and therefore little or no change in the number of Li vacancies. When sintering with an oxygen pressure of 1 atm, the calculations show that doping with Al increases the number of Li vacancies of approx. 40%. Investigations of the mobility of Li vacancies in tetragonal LLZO, using CI-NEB, showed that doping does not lead to significant changes in migration barriers but only pathways since Al occupy Li sites. Defective modelling of the Garnet structure shows that some of the Li spaces in the grid are empty or partially occupied. These defects, Li vacancies, help the Li ions to jump around the structure more efficiently and thus increase the conductivity. Furthermore, defect modelling shows that doping can affect the conductivity by forming more charge-compensating Li vacancies. Here, the type of dopant and valence charge plays an important role; for example, Al dopants occupy Li sites in the lattice compared to Nb dopants, which occupy Zr sites. Previous calculations show that not all Li atoms are equally mobile; some sites are more energy-efficient than others and thus have higher activation energy for transport. An advanced defect model that addresses two different Li sites has been developed for optimizing the conductivity.

The primary objective of OxiBat is to design, manufacture and operate a novel safe oxide-based all solid-state battery with a lifetime of more than 1000 cycles and degradation less than 10% was not achieved due to availability of UiO’s pulsed laser deposition (PLD) instrument which has been down for 2 years during the project period. Secondary objectives are to: • Develop electrodes that have less than 2% expansion during charging and discharging and obtain charging time less than 15 min for 70% state of charge, was not obtained since quality first generation PLD produced components needed further optimisation. • Design a concept where thin films of batteries can be stacked up to high charging voltages - Concept was made planned with ceramic anode LTO and cathode (NMC/LMN) since Li anode could not survive the PLD conditions. Silver was aimed as an interconnect that could be used, but concept not tested. • Produce DFT modelling tools and NMR tools that can speed up development of new ASSBs - DFT was performed on the electrolyte for a master thesis by Kristoffer Eggestad on "Computational Study of Migration Barriers in the Li Ion Solid State Electrolyte Li7La3Zr2O12". - NMR tolls was elaborated to do 6Li NMR and 27Al NMR to study their location • Educate one PhD and disseminate widely OxiBat results - PhD candidate Madeeha Khalid Pedersen at UiO is writing up thesis and 3 publication manuscripts. Submission and PhD defence expected within 2024. • Strengthen collaboration with industry and internationally leading research groups - Lack of PLD results restricted our interaction with industry and other lading research groups, since this reduced the conferences to attain for networking / collaboration. General impact of our publication on "Effects of Different Doping Strategies in Cubic Li7La3Zr2O12 Solid-State Li-ion Battery Electrolytes" where we present our results combined with literature review, have revealed better understanding that can guide further research in this field. The gain in knowledge on the garnet structure is also achieved through publication of defect chemistry of given structure and DFT calculation on defect mobility and effect of dopant on mobility giving insight in effect of doping strategies. An important outcome of the Oxibat project is an m-era.net spin-off project entitled Lasibat, where we are in collaboration with the Norwegian SME Cerpotech and leading research groups (Fraunhofer) and industry from Germany and Spain. The competence and know-how gained from Oxibat, can also be extended to Na-ion batteries and formed a basis for an interdisciplinary NFR research project proposal (Sosoba) in 2024, where machine learning will be used to develop novel Na-ion conductors. The topic solid-state batteries have been discussed in meetings with Norwegian battery industry. Although their short-term focus is on solving their immediate manufacturing challenges, it is very valuable to build competence also on the next-generation

OxiBat aims to bring about a step-change in the development of the next generation of solid-state batteries. Solid-state batteries are inherently safe and fire-proof, and as well may take higher charging rates compared to the state of the art. The need for faster charging of batteries, especially for cars and large household batteries, can only be met by batteries that can tolerate higher temperatures. For conventional Li batteries containing organic polymers, the possibility for higher charging rate is limited by the heat evolved during charging, which can cause melt-down. By replacing the battery components with oxide ceramics, the battery cannot be set on fire and higher temperatures, and thus higher charging rates, can be tolerated. The development of new solid state electrolyte materials with high Li ion mobility will be strengthened by improved fundamental understanding of Li ion transport and the influence of interfaces towards the electrodes. New knowledge will be obtained through advanced modelling (DFT) and characterization tools (solid state 7Li NMR, impedance spectroscopy, chronopotentiometry and other methods). The project will benefit from collaboration with University of Cambridge (UK), which has one of the world's leading groups in this area. The project will develop fabrication routes based on existing equipment and experience at the research partners. Most emphasis will be on fabrication routes that are up-scalable and have the potential for industrial scale mass production. OxiBat will be a multi-disciplinary project with expertise from a range of fields. It will be a collaborative project involving leading research groups in Norway (SINTEF, UiO and NTNU) and internationally (University of Cambridge).