LC Agent: Toward AI-Driven Recommendation of Liquid Chromatography Conditions for Chemical Reactions

Liquid Chromatography (LC) is a foundational tool for the identification and monitoring of chemical compounds. However, its application to complex chemical reactions remains challenging, as analytical methods optimized for individual compounds often fail to capture the full scope of a reaction. The difficulty lies in establishing LC conditions that achieve simultaneous detection and chromatographic resolution for multiple reactants and products with diverse physicochemical properties. In this work, we investigate the capability of Large Language Models (LLMs) in agentic LC condition recommendation for comprehensive chemical reaction analysis. We present an LLM-based multiagent system that comprises multiple subagents that are context-engineered to perform specific functional roles that an analytical chemist would perform when determining the LC conditions, emulating the decision-making process of an analytical chemist. Given a chemical reaction and user-defined analytical requirements in natural language, the system autonomously searches relevant literature, reasons over compound properties, and proposes plausible LC conditions that can detect all reaction components within a single analytical run. We demonstrate its effectiveness through a case study on organic electronic materials, confirming that the recommended LC conditions are highly suitable for a diverse set of chemical reactions. The source code is available at https://github.com/seokhokang/lc_agent/.