On April 22, 2026, Associate Professor Chuanlai Liu from Shanghai Jiao Tong University (SJTU), together with Dr. Yuwei Zhang, Dr. Siyuan Zhang, and Prof. Gerhard Dehm from the Max Planck Institute for Sustainable Materials, published a research article in Nature, titled “Mechanically driven Li dendrite penetration in garnet solid electrolyte.”
The study provides new insights into the long-standing challenge of lithium dendrite-induced failure in all-solid-state lithium metal batteries and proposes an effective strategy to mitigate short-circuiting through defect engineering.
Background: a key challenge for next-generation batteries
All-solid-state lithium metal batteries are widely regarded as a promising technology for next-generation energy storage systems due to their high energy density and intrinsic safety advantages. They are expected to play a critical role in emerging applications such as electric vehicles, low-altitude aviation systems, and intelligent robotics.
However, during battery operation, lithium dendrites can penetrate stiff ceramic solid electrolytes, leading to internal short circuits and severe safety risks. This “soft lithium penetrating hard ceramics” phenomenon has remained highly controversial, with competing hypotheses attributing it either to mechanical fracture or to electron leakage-induced lithium nucleation.
A major obstacle to resolving this debate lies in the difficulty of characterizing lithium at multiple scales, from nanoscale crack tips to microscale fracture networks, under complex electrochemo-mechanical coupling conditions.
Multiscale characterization reveals dendrite growth behavior
To address this challenge, the research team developed a in-plane cell configuration that enables precise localization of lithium dendrite tips. By integrating advanced characterization techniques, including optical microscopy, cryogenic scanning electron microscopy (cryo-SEM), cryogenic transmission electron microscopy (cryo-TEM), and electron energy loss spectroscopy (EELS), the team systematically investigated dendrite growth across nano-, micro-, and macroscopic scales.
The results show that, while lithium dendrites appear to propagate along nearly straight paths at the macroscopic level, their growth at the microscale involves a complex combination of intergranular and transgranular fracture modes. Statistical analysis indicates that approximately 20% of fractures occur through grains, demonstrating that dendrite growth is unlikely to be governed by isolated lithium nucleation along grain boundaries
Fig. 1: Morphology, microstructure and fracture statistics of LLZTO solid electrolyte during lithium dendrite penetration.
Direct evidence of lithium filling cracks at the nanoscale
Through three-dimensional reconstruction and cryogenic STEM-EELS analysis, the researchers directly observed that lithium fully fills nanoscale crack tips and extends into micrometer-scale cracks. In contrast, no lithium enrichment or isolated lithium nuclei were detected ahead of the dendrite tip within the solid electrolyte.
Despite our evidence supporting a mechanically governed mechanism for dendrite propagation in garnet solid electrolyte, an important question remains: how can soft lithium generate sufficient internal stress to fracture a stiff ceramic electrolyte? To address this, we next examine the stress state and plastic activity of lithium dendrites confined within cracks, using grain orientation mapping through EBSD and micromechanical fracture modelling.
Fig. 2: Fractography and elemental distribution at the lithium dendrite tip.
Mechanical origin: hydrostatic stress drives fracture
To uncover the driving force behind dendrite penetration, the team combined experimental observations with phase-field fracture modeling. The results reveal that lithium confined within cracks generates extremely high hydrostatic stress.
This internal pressure is significantly higher than the von Mises stress associated with plastic deformation and is transferred to the surrounding solid electrolyte as tensile stress, thereby driving crack propagation and dendrite growth.
Importantly, the analysis shows that plastic deformation in lithium is highly localized near interfaces, while most of the dendrite remains in a nearly elastic state. This indicates that hydrostatic stress, rather than plasticity, is the dominant factor governing the “soft-penetrates-hard” phenomenon.
Fig. 3: Microstructure of lithium dendrite in LLZTO and phase-field fracturing modelling of lithium dendrite penetration.
Defect engineering enables controlled dendrite growth
Based on the mechanically driven growth mechanism, the researchers proposed a novel strategy to control dendrite propagation by introducing engineered defects.
By creating arrays of pre-designed cracks using Vickers indentation, the team demonstrated that lithium dendrites can be effectively redirected along desired paths. When encountering these engineered defects, dendrites deviate significantly from their original trajectories, avoiding direct penetration through the electrolyte.
Further simulations confirm that the geometry of defects plays a crucial role: transverse defects can redistribute local stress fields and induce dendrite deflection. This defect-guided approach provides a pathway to suppress short-circuiting and improve the cycling stability of solid-state batteries.
Fig. 4: Tailor lithium dendrite growth through engineered voids.
The work was carried out through a collaboration between Shanghai Jiao Tong University, the Max Planck Institute for Sustainable Materials, and international partners.
Read the full paper: https://doi.org/10.1038/s41586-026-10415-9