Chemical Property Prediction and Identification of Unknown Chemicals Using AI Technology and Quantum Technology

 

Chemical Property Prediction and Identification of Unknown Chemicals Using AI Technology and Quantum Technology

 

Date: 20^th July, 2026 (Monday)

Time: 5 - 6 pm (HKT)

 

Venue: Lecture Theatre P4, Chong Yuet Ming Physics Building

 

 

Department of Chemistry

Sogang University

 

Dr. Keunhong Jeong is an Associate Professor of Chemistry at Sogang University and a member of the Scientific Advisory Board of the Organisation for the Prohibition of Chemical Weapons. He earned his degree from University of California, Berkeley. His research focuses on artificial intelligence, cheminformatics, quantum hyperpolarization, and quantum computing algorithms.

The increasing complexity of chemical threats—from terrorism to industrial accidents involving unidentified hazardous substances—demands next-generation computational tools that integrate artificial intelligence with quantum technologies. This presentation highlights our recent advances across three complementary domains: AI-driven chemical property prediction, quantum algorithm-based molecular simulation, and quantum hyperpolarization-enhanced chemical detection. First, we present advanced machine learning and deep learning frameworks for predicting critical physicochemical and toxicological properties, including vapor pressure, and toxicity, enabling rapid hazard assessment and informed emergency response. Second, we introduce quantum computing algorithms including quantum machine learning (QML) approach for AI-augmented toxicity prediction, bridging quantum computing with artificial intelligence as an emerging paradigm of Quantum AI. Third, we describe the application of quantum hyperpolarization techniques, which dramatically enhance NMR sensitivity by several orders of magnitude, enabling the detection of trace-level chemical species that are otherwise undetectable by conventional NMR methods. These hyperpolarization-enhanced spectroscopic capabilities are integrated with the Density Functional Theory and Spectroscopy Integrated Identification Method (D-SIIM), which combines experimental analytical techniques—GC-MS, NMR, and IR spectroscopy—with quantum chemical calculations for rapid identification of unknown substances, including chemical warfare agents (CWAs). The synergistic convergence of AI-based property prediction, quantum computing algorithms, and quantum hyperpolarization-enhanced detection offers a powerful and scalable framework for chemical safety, security, and defense applications.