GBClaw: An agentic AI framework for atomistic grain boundary research and mechanistic discovery

Yazhuo Liu; Kunqing Ding; Xiaoguo Gong; Zifeng Wang; Yichen Yang; Yifeng Che; Wen Chen; Josh Kacher; Olivier Pierron; Shigenobu Ogata; Ting Zhu

arXiv · 2610.28629

GBClaw: An agentic AI framework for atomistic grain boundary research and mechanistic discovery

Yazhuo Liu·Kunqing Ding·Xiaoguo Gong·Zifeng Wang·Yichen Yang·Yifeng Che·Wen Chen·Josh Kacher·Olivier Pierron·Shigenobu Ogata·Ting Zhu

View original source ↗Read source document ↗Export metadata ↓

Abstract

Agentic artificial intelligence (AI) systems hold immense potential for automating complex workflows in materials discovery and design, but extending them to domain-intensive mechanistic investigations remains challenging. Here, we present GBClaw, an agentic AI framework for investigating the structure and deformation of general grain boundaries (GBs). Although general GBs constitute the vast majority of real-world interfaces in polycrystalline materials, they remain understudied because their structural complexity and interfacial mismatch complicate experimental and modeling research. GBClaw integrates large language models (LLMs) into a modular architecture comprising an orchestrator, specialized agents, domain-specific tools, simulation interfaces, and hierarchical memory. Through dynamic replanning and iterative workflow refinement, the framework effectively connects systematic simulations with targeted mechanistic analysis. Applied to 14 general asymmetric tilt GBs in nickel, GBClaw characterized interfacial defect structures and identified diverse deformation modes across these incommensurate boundaries. Comparing boundary sliding with and without secondary GB misfit dislocations revealed substantial differences in shear strength, underscoring the critical role of defect hierarchies in sliding resistance. Furthermore, researcher-guided follow-up calculations traced an anomalously high shear resistance to metastable boundary structures. This work demonstrates that combining agentic AI with domain expertise bridges high-throughput computational screening with deep physical understanding of complex material interfaces.

Explore related subjects

cond-mat.mtrl-sci cond-mat.mes-hall cs.AI cs.MA

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Yazhuo Liu, Kunqing Ding, Xiaoguo Gong, Zifeng Wang, Yichen Yang, Yifeng Che, Wen Chen, Josh Kacher, Olivier Pierron, Shigenobu Ogata, Ting Zhu. 2026-10-20. GBClaw: An agentic AI framework for atomistic grain boundary research and mechanistic discovery. https://arxiv.org/abs/2610.28629

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Mechanical Analyses using Discrete Harmonic Expansions of Grain-Based Field Data

A methodology for computing expansion basis functions using discrete harmonic modes is presented. The discrete harmonic modes are determined grain-by-grain for virtual polycrystals for which finite element meshes are available. The expansion weights associated with representing field variables over grain domains are determined by exploiting the orthogonality of the harmonic modes. The methodology is demonstrated with the representation of the axial stress distributions during tensile loading of a polycrystalline sample. An open source code, MechMonics, is available to researchers wishing to use the methodology to analyze data.

cond-mat.mtrl-sci↗

arXiv2026

Precipitation strengthening as a collective multi-dislocation phenomenon

Precipitation strengthening is a cornerstone of physical metallurgy, delivering otherwise unattainable combinations of strength and ductility. The approach relies on nanoscale precipitates that impede the motion of dislocations, the primary carriers of metal plasticity. Historically, precipitation strengthening has been rationalized via two idealized, limiting mechanisms: dislocations either cut through or bow around precipitates. However, under many technologically relevant conditions, multiple dislocations can interact with a precipitate at the same time, potentially giving rise to distinct precipitate-dislocation dynamics that in situ experiments are yet to fully resolve. Here, using large-scale atomistic simulations (100 million atoms) that capture these dynamics, we demonstrate that in this regime, the cutting-versus-bowing dichotomy is incomplete: strengthening arises as an emergent phenomenon driven by concurrent multi-dislocation interactions. These interactions simultaneously induce dislocation accumulation at interfaces, storage within precipitates, and precipitate-mediated multiplication inside the matrix. Our findings establish a distinct mechanistic regime of precipitation strengthening beyond traditional models and provide a quantitative foundation for predicting its behavior.

cond-mat.mtrl-sci↗

arXiv2026

A Combined Tight Binding with Machine Learning Potential Model for Magnesium Compounds

We present a model for magnesium-based systems that combines density functional tight binding (DFTB) with MACE, a machine learning interatomic potential (DFTB+MACE). In this model, the conventional repulsive potential, pair potential, is replaced by a many-body MACE potential. The MACE component of the model is trained on the difference between density functional theory (DFT) energies and forces and the corresponding DFTB values, but neglecting the pair potential contribution. Using this model we performed structural relaxation of MgO-CO2 adsorption systems, molecular dynamics calculations of water clusters and phonon spectrum calculations of stable fcc-MgO and metastable bcc-MgO structures. We compare the performance of our model with a pure MACE model and with DFT. We demonstrate that the DFTB+MACE model achieves improved accuracy relative to DFTB with a pair potential, in many cases with only a moderate increase in computational cost. In addition, it can provide electronic structures that most of the machine learning potentials cannot. The training dataset, originally developed for MACE, may not fully represent all regions of the potential surface we may encounter during simulations. Expanding the dataset for a wider potential surface is expected to further enhance predictive accuracy of DFTB+MACE model. Overall, the resulting DFTB+MACE framework enables simulations at length and time scales beyond the reach of first-principles methods while retaining an explicit description of electronic structures, making it particularly attractive for studying charge-transfer in materials.

cond-mat.mtrl-sci↗