📄 chemistry

Self-organized Ag-Sb lamellae redirect cracks in all-solid-state Li metal batteries

This paper demonstrates that an in-situ generated, self-organized Ag-Sb lamellar interlayer at the lithium/solid-electrolyte interface suppresses dendrite-induced short circuits by redirecting through-thickness cracks and lithium growth laterally, thereby enabling high-performance all-solid-state batteries without the tradeoffs associated with stack pressure.

Original authors: Hongli Zhu, Cheng Xu, Shiyao Lin, Bharat Pant, Ye Cao, Jiwei Wang, Han Su, Hua Zhou, Luxi Li, Xianhui Zhao, Ercan Cakmak, Longqing Chen, Chunsheng Wang

Published 2026-08-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hongli Zhu, Cheng Xu, Shiyao Lin, Bharat Pant, Ye Cao, Jiwei Wang, Han Su, Hua Zhou, Luxi Li, Xianhui Zhao, Ercan Cakmak, Longqing Chen, Chunsheng Wang

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The quest for batteries that can power electric vehicles for hundreds of miles on a single charge has long focused on lithium metal. This material is the holy grail of energy storage because it can hold far more charge than the graphite used in current phones and cars. However, pairing lithium metal with a solid electrolyte—a hard, ceramic-like material that replaces the flammable liquid in today's batteries—introduces a stubborn mechanical problem. When lithium is charged into the battery, it does not always settle down smoothly. Instead, it can grow needle-like spikes called dendrites. These spikes push against the solid electrolyte, creating tiny cracks. Once a crack forms, the lithium follows it like water down a riverbed, eventually piercing the entire battery and causing a short circuit. For years, scientists have tried to stop this by making the materials harder or by pressing the battery layers together with immense force. Yet, this pressure, while helpful for keeping contact, often makes the cracks run straight through the battery even faster, leading to a frustrating trade-off where fixing one problem worsens another.

A team of researchers has now found a way to break this cycle by changing the direction of the cracks rather than trying to stop them entirely. Working with a solid-state lithium battery, they introduced a thin layer made of silver and antimony between the lithium metal and the solid electrolyte. This layer is not a static barrier; it is a dynamic, self-organizing structure that changes as the battery charges and discharges. When lithium enters this layer, it reacts differently with the silver and the antimony. Because these two metals expand at different rates when they absorb lithium, they create internal tension. Under the pressure of the battery stack, this tension forces the material to rearrange itself into a pattern of alternating, parallel stripes, much like the layers in a geological rock formation.

The researchers discovered that this striped pattern acts as a guide for the lithium. When a crack begins to form and try to move straight through the battery, it hits the boundary between a silver-rich stripe and an antimony-rich stripe. Instead of pushing straight through, the crack is forced to turn and travel sideways along the stripe. This redirection keeps the lithium growing parallel to the battery layers rather than piercing through to the other side. The process is continuous; as the battery cycles, the stripes regenerate and heal themselves, maintaining this protective pattern even after hundreds of charges. In tests, this approach allowed the battery to operate at high speeds and high pressures without shorting out, whereas batteries without this layer failed almost immediately.

To understand how this works, the team used powerful X-ray imaging and computer simulations to watch the battery in action. They observed that the silver and antimony do not mix randomly. Instead, the lithium causes them to separate into distinct domains that align into a lamellar, or layered, architecture. This structure is unique because it is created inside the battery during operation, rather than being manufactured as a pre-made sheet. The simulations confirmed that the difference in how the two materials handle stress is the key. When a crack reaches the interface between the two different phases, the mismatch in their mechanical properties causes the crack to deflect. It is similar to how a river might be forced to flow along a valley rather than cutting straight through a mountain range, but in this case, the "valley" is a microscopic boundary between two solid metals.

The results were striking. In symmetric cells, which are simplified versions of the battery used to test stability, the new design allowed the battery to run at a current density of 9.0 mA cm-2 with a capacity of 9.0 mAh cm-2 without failing. This is a significant improvement over standard setups, which short-circuited at much lower levels. When the researchers built a full battery using a sulfur cathode, the device cycled stably, delivering an areal capacity of 4.91 mAh cm-2. In contrast, control batteries without the silver-antimony layer failed during the very first charging cycle. The team also noted that the battery could operate at different temperatures, from room temperature up to 80 degrees Celsius, maintaining its stability across the board.

What makes this discovery particularly important is that it solves the problem by working with the physics of the battery rather than fighting against it. Traditional methods try to create a perfect, unbreakable shield to stop the lithium, but this approach accepts that cracks will form and instead programs them to go the wrong way. The self-organizing layer continuously repairs itself, ensuring that even if damage occurs, the lithium remains trapped in a safe, sideways path where it can be reused in the next cycle. This strategy of redirecting cracks offers a new path forward for solid-state batteries, suggesting that the key to safety lies not in building impenetrable walls, but in designing smart, adaptive interfaces that guide the flow of energy away from disaster.

Technical Summary: Self-organized Ag-Sb Lamellae Redirect Cracks in All-Solid-State Li Metal Batteries

Problem Statement
The practical deployment of all-solid-state lithium (Li) metal batteries is hindered by interfacial instability and crack-assisted Li dendrite propagation. While applying stack pressure improves interfacial contact between the Li anode and the solid electrolyte (SE), it simultaneously accelerates through-thickness crack propagation, leading to short circuits. Existing strategies to mitigate this, such as tri-layer electrolyte architectures or externally imposed compressive stress via shape-memory alloys, face challenges including the gradual decay of stress-redirection capabilities over cycling or the difficulty of maintaining in-plane biaxial compressive stress in practical cells. There is a critical need for a strategy that can dynamically redirect stress and crack propagation parallel to the electrode interface during prolonged cycling without relying solely on passive mechanical blocking or chemical passivation.

Methodology
The authors propose an in situ self-organized lamellar interlayer strategy using an Ag-Sb alloy precursor (specifically Ag30Sb70) at the Li/SE interface. The methodology integrates experimental synthesis, advanced characterization, and multi-scale modeling:

Material Synthesis: Ag and Sb powders were ball-milled to create a homogeneous two-phase Ag30Sb70 precursor. This precursor was pressed into pellets and assembled into symmetric Li|Ag30Sb70||SE||Ag30Sb70|Li cells and full Li-S cells using sulfide-based solid electrolytes (Li5.4PS4.4Cl1.6 and Li10GeP2S12).
Characterization: The structural evolution was monitored using Synchrotron X-ray fluorescence (XRF) mapping, Scanning Electron Microscopy (SEM), Energy-Dispersive X-ray Spectroscopy (EDS), and X-ray Computed Tomography (XCT). These techniques tracked elemental distribution and crack propagation paths before and after electrochemical cycling.
Modeling: The study employed coupled phase-field and finite-element modeling (FEM) to simulate microstructural evolution, crack propagation, and Li deposition. The models incorporated lattice-mismatch strains, stack pressure, and electrochemical reactions to predict how the Ag-Sb system self-organizes and redirects cracks.

Key Contributions and Mechanism
The core contribution is the demonstration that sequential lithiation of Ag-rich and Sb-rich phases drives the spontaneous formation of a self-organized, alternating lamellar architecture (Li2AgSb-rich and Li3Sb-rich domains) under stack pressure.

Self-Organization Mechanism: The Ag30Sb70 precursor contains orthorhombic ε (Ag3Sb) and rhombohedral (Sb) phases. During cycling, Sb-rich domains lithiate at a higher potential (~~0.8 V) to form Li3Sb, while Ag-containing domains lithiate at a lower potential (~~0.07 V) to form Li2AgSb. This asynchronous alloying creates differential volume expansion and significant lattice-mismatch strain. Under stack pressure, this strain is preferentially released through interfacial sliding and directional phase rearrangement, driving the spontaneous formation of aligned lamellae parallel to the electrode.
Crack Redirection: Unlike conventional interlayers that may degrade, this architecture continuously regenerates during cycling. The mechanical contrast and fracture-toughness mismatch between the Ag-rich and Sb-rich phases (specifically a low interface-to-material fracture toughness ratio of $\Gamma_i/\Gamma_m = 0.0005$ and cohesive strength ratio of $\hat{\sigma}_i/\hat{\sigma}_m = 0.073$ ) satisfy the criteria for crack deflection. Consequently, through-thickness cracks are redirected into lateral cracks along the lamellar interfaces, preventing Li from penetrating the bulk of the solid electrolyte.
Electrochemical Recovery: The electronically conductive framework of the lamellae allows for the electrochemical stripping of trapped or isolated Li, facilitating the reutilization of "dead Li" and maintaining stable cycling.

Results

Symmetric Cell Performance: Li|Ag30Sb70||SE||Ag30Sb70|Li symmetric cells demonstrated significantly enhanced critical current densities (CCD) compared to bare Li cells. They operated stably up to 9.0 mA cm⁻² at 9.0 mAh cm⁻² (at 80 °C) without short-circuiting, whereas control cells without the interlayer failed during the first charging cycle. Stable cycling was also achieved at 6.0 mA cm⁻²/6.0 mAh cm⁻² for over 500 cycles.
Full Cell Performance: All-solid-state Li-Sulfur (ASSLSB) cells incorporating the Ag30Sb70 interlayer delivered high areal capacities (e.g., 4.91 mAh cm⁻² at 0.15 C and 60 °C) and maintained ~70% capacity retention after 100 cycles. In contrast, cells with bare Li anodes short-circuited immediately.
Structural Integrity: Synchrotron XRF and SEM confirmed the persistence of the lamellar architecture after 100 cycles. XCT visualization showed that cracks propagated laterally within the lamellar framework rather than penetrating through the electrolyte.
Thickness Independence: The self-organization process was observed even when the interlayer thickness was reduced by half, suggesting the strategy is robust and not limited to thick model interlayers.

Significance
The paper establishes a "cycling-adaptive" interfacial strategy that stabilizes Li metal anodes by governing crack trajectories rather than attempting to create a permanently crack-free or mechanically impenetrable barrier. By converting lithiation-induced lattice mismatch into a driving force for self-organization, the Ag-Sb interface dynamically programs crack paths to run parallel to the electrode. This work advances the understanding of solid-state battery failure by demonstrating that dendrite short-circuiting, governed by crack initiation and propagation, can be mitigated through in situ interfacial self-organization. It offers a design principle where interfacial strain, traditionally viewed as a source of degradation, is harnessed to create adaptive protection and redirect dendrite growth.

More like this