A chemometric framework for forensic source apportionment of nerve agents: GC×GC-TOFMS-based signature profiling of ethyltabun (EGA) and VM

Organophosphorus nerve agents (OPNAs), including G-agents, EGA (ethyltabun, phosphonamidic acid, P -cyano- N , N -diethyl-, ethyl ester) and V-agents, VM ( O -ethyl S -(2-diethylaminoethyl) phosphonothiolate), are highly toxic chemical warfare agents (CWAs) with severe risks to human health and environmental security. This study proposes a chemometric-driven framework for forensic tracing of their synthetic pathways using high-resolution GC × GC-TOFMS. By integrating advanced statistical analysis, we identified 160 synthesis-associated chemical attribution signatures (CAS) for EGA and 138 process-specific CAS for VM, with 11 overlapping markers, including ethoxyphosphates and diethylaminoethylamine derivatives. Key distinguishing features include sulfur-containing derivatives (e.g., phosphorothioates) for VM and cyanide byproducts for EGA. These data were visualized through innovative chord diagrams and heatmaps, offering unprecedented insights into compound-sample relationships critical for forensic traceability. Multivariate classification using Hierarchical Cluster Analysis (HCA) and Partial Least Squares Discriminant Analysis (PLS-DA) achieved exceptional accuracy—97.2 % for EGA and 100 % for VM via HCA, and 100 % for both agents via PLS-DA. Robust k-fold cross-validation (k = 5 and 10) further confirmed model reliability, with R 2 values approaching unity. This work establishes a comprehensive CAS database and a scalable analytical framework, advancing forensic capabilities from reactive detection to proactive, intelligence-driven prevention.