Autocata: Reaction-network-guided autonomous catalyst screening via capability-centric AI
Abstract
Artificial intelligence (AI) has expanded the accessible space for catalyst discovery, yet executing complex screening campaigns still requires constant human intervention to bridge high-level reaction objectives with underlying computational pipelines. Here, we present Autocata, a knowledge-embedded autonomous agent that scales catalytic expertise by encoding multistep reaction pathways into reusable, language-invocable capabilities. Powered by a Transformer pretrained on two million catalyst-adsorbate structures, the system incorporates a capability registry of 79 adsorbate-specialized generative models spanning complex reaction networks across carbon, nitrogen, oxygen, and hydrogen intermediates. Given a high-level natural-language prompt, Autocata deconstructs reaction networks, queries and configures matching generative models, generates candidate surface configurations, enforces physical and geometric validity filtering, and coordinates machine-learning potentials evaluations to rank promising materials. Crucially, the agent adaptively reconfigures workflows when encountering capability boundaries or task bottlenecks. We demonstrate these capabilities across CH4 activation, nitrogen reduction reaction (N2RR), and CO2-to-methanol pathways. By shifting from static toolkits to an adaptive, language-driven computational infrastructure, Autocata lowers the barrier for expert-level catalyst discovery and offers a scalable paradigm for self-driving catalytic research.
Related
- Hierarchical Multi-agent Large Language Model Reasoning for Autonomous Heterogeneous Catalyst Discovery
- Atmospheric pressure water-gas shift reaction and transfer hydrogenation of ketones using a ruthenium pincer catalyst.
- Challenges in Catalyst Design for the Ammonia Oxidation Reaction in Aqueous Media: The Case of Metal Porphyrins
- Bridging the Literature-to-Laboratory Gap with Language Models and Active Learning for Reaction Feasibility Prediction and Autonomous Exploration
- CO2 Driven Soot Oxidation via Reverse Boudouard Reaction over Reducible Catalyst: A Sustainable Pathway for Catalyst Regeneration
BibTeX
@misc{li2026autocata,
title = {{Autocata: Reaction-network-guided autonomous catalyst screening via capability-centric AI}},
author = {Ruili Li and Rui Qi and Qingqing Mao and Ritankar Das and Beien Zhu and Yi Gao},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.39060},
eprint = {2609.39060},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.39060},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Autocata: Reaction-network-guided autonomous catalyst screening via capability-centric AI
AU - Ruili Li
AU - Rui Qi
AU - Qingqing Mao
AU - Ritankar Das
AU - Beien Zhu
AU - Yi Gao
PY - 2026
DA - 2026/09/30
PB - arXiv
DO - 10.48550/arxiv.2609.39060
UR - https://arxiv.org/abs/2609.39060
AB - Artificial intelligence (AI) has expanded the accessible space for catalyst discovery, yet executing complex screening campaigns still requires constant human intervention to bridge high-level reaction objectives with underlying computational pipelines. Here, we present Autocata, a knowledge-embedded autonomous agent that scales catalytic expertise by encoding multistep reaction pathways into reusable, language-invocable capabilities. Powered by a Transformer pretrained on two million catalyst-adsorbate structures, the system incorporates a capability registry of 79 adsorbate-specialized generative models spanning complex reaction networks across carbon, nitrogen, oxygen, and hydrogen intermediates. Given a high-level natural-language prompt, Autocata deconstructs reaction networks, queries and configures matching generative models, generates candidate surface configurations, enforces physical and geometric validity filtering, and coordinates machine-learning potentials evaluations to rank promising materials. Crucially, the agent adaptively reconfigures workflows when encountering capability boundaries or task bottlenecks. We demonstrate these capabilities across CH4 activation, nitrogen reduction reaction (N2RR), and CO2-to-methanol pathways. By shifting from static toolkits to an adaptive, language-driven computational infrastructure, Autocata lowers the barrier for expert-level catalyst discovery and offers a scalable paradigm for self-driving catalytic research.
ER -