Demonstration of computational domain, concentration profile within electrolyte and phase-map between current and electrolyte modulus highlighting different lithium deposition zones.
The garnet‐type phase Li7La3Zr2O12 (LLZO) attracts significant attention as an oxide solid electrolyte to enable safe and robust solid‐state batteries (SSBs) with potentially high energy density. However, while significant progress has been made in demonstrating compatibility with Li metal, integrating LLZO into composite cathodes remains a challenge. The current perspective focuses on the critical issues that need to be addressed to achieve the ultimate goal of an all‐solid‐state LLZO‐based battery that delivers safety, durability, and pack‐level performance characteristics that are unobtainable with state‐of‐the‐art Li‐ion batteries. This perspective complements existing reviews of solid/solid interfaces with more emphasis on understanding numerous homo‐ and heteroionic interfaces in a pure oxide‐based SSB and the various phenomena that accompany the evolution of the chemical, electrochemical, structural, morphological, and mechanical properties of those interfaces during processing and operation. Finally, the insights gained from a comprehensive literature survey of LLZO–cathode interfaces are used to guide efforts for the development of LLZO‐based SSBs.
31 May 2019 in Journal of the Electrochemical Society
Solid state electrolytes (SSEs) with garnet structures and chemical composition Li7La3Zr2O12 (LLZO) are being actively considered as the electrolyte for next-generation lithium ion batteries. These LLZO-based ceramic SSEs are polycrystalline and possess distinct grain-interior/grain-boundary (GI/GB) microstructure. Lithium dendrite growth through these GI and GB domains is considered to be the largest bottleneck preventing their successful implementation. Dendrite growth has been observed to occur predominantly within the GB regions. While this would be expected if the GB regions were more ionically conductive than the GI regions, there is significant disagreement over the relative conductivities of these two regions. The present work demonstrates that dendrite growth can occur preferentially in the GB regions even when the GI regions have higher ionic conductivity. Electrochemical potential changes induced by mechanical stress within the GI and GB domains reveals that the magnitude of effective stress induced current density within the elastically softer GB is larger than in the mechanically stiffer GI, which leads to enhanced lithium deposition in GB and subsequent dendrite growth. The present work additionally examines the impact of applied current density and grain size on current focusing in the GB regions, and also finds that SSE fractures initiate from the GB regions.
19 March 2019 in The Journal of Physical Chemistry B
The majority of the ceramic solid electrolytes (LLZO, LATP) demonstrate polycrystalline grain/grain-boundary (G/GB) microstructure. Higher lithium (Li) concentration and lower mechanical stiffness result in current focusing at the GBs. Growth of Li dendrites through local inhomogeneities and subsequent short circuit of the cell is a major concern. Recent studies have revealed that bulk Li metal is a viscoplastic material that has low (∼0.3 MPa) and high (∼1.0 MPa) yield strength during deformation at smaller and larger rates of strain, respectively. It has been argued that during deposition at smaller current densities, due to its lower yield strength, Li metal should demonstrate plastic flow against stiff ceramic electrolytes, and Li dendrites will be prevented from penetrating through solid electrolytes. In this manuscript, a multiscale modeling framework has been developed for predicting properties of GBs and the bulk of ceramic electrolytes using atomistic calculations for input to mesoscale models. Using the parameters obtained from the atomistic simulations, the mesoscale model reveals that, given enough time, even at low charge rates, lithium dendrites can grow through the GBs of LLZO. The present multiscale model results also provide information regarding the dendrite growth velocity through LLZO.
The growing interest in solid-state batteries has resulted in increased focus on the issue of delamination at the cathode/solid-state electrolyte (SSE) interface. The extent of delamination, and its impact on impedance rise and capacity fade, is determined by a number of material properties, including cathode molar volume change and state-of-charge dependence, mechanical properties of the cathode and SSE, SSE grain boundary microstructure, fracture threshold of the interface, and interfacial exchange current density. The goal of this paper is to develop multiscale mathematical models to link these material properties to performance and provide guidance on possible approaches to minimize delamination, impedance rise, and capacity fade. We examine two cathodes, namely, LiNi0.8Mn0.1Co0.1O2 with a positive molar volume change and LiCoO2 with a negative molar volume change against a Li7La3Zr2O12 (LLZO) SSE. Density functional theory (DFT) is used to determine fundamental material properties at the interface, while mesoscale models are used to link these properties to macroscopic performance. Model results suggest that SSEs with smaller grains and low Young’s modulus and cathodes with low molar volume changes provide opportunities for minimizing delamination.