Intelligent fingerprinting of semi-volatile and intermediate-volatility organic compounds (S/IVOCs) from cooking emissions using pixel-resolved GCxGC-MS analysis

Abstract

Semi-volatile and intermediate-volatility organic compounds (S/IVOCs) emitted from cooking are important secondary organic aerosol (SOA) precursors, yet their broad volatility range, chemical complexity, and extensive unresolved chromatographic signals hinder comprehensive characterization. Here, we applied a pixel-resolved GC×GC-MS fingerprinting framework to extract both molecular-level and chemical-space information from real-world Sichuan canteen cooking emissions. The approach integrates peak-based identification, PLS-DA marker screening, and pixel-level fuzzy-difference analysis to capture process-dependent and treatment-induced changes across the full two-dimensional chromatographic space. Applied to paired before- and after-treatment samples collected during breakfast and main-meal periods, the method quantified 204 gaseous compounds within the measurable chemical space, with a summed average concentration of 263.65 ± 90.79 μg m⁻³ . IVOCs accounted for 44.7% of the total concentration, underscoring the importance of lower-volatility precursors in real-world mixed cooking emissions. The integrated analysis identified main-meal-specific fingerprints associated with lipid oxidation, sulfur-containing seasonings, fatty acids, and spice-related oxygenated compounds, demonstrating its capacity to refine cooking source profiles beyond bulk concentration metrics. The same framework further revealed chemically selective purification: although the oil-fume treatment system reduced the total quantified concentration by 15.2%, its reduction of SOA formation potential (SOAFP) was limited, indicating incomplete control of high-impact precursors. S/IVOCs contributed 39.4% of the estimated SOAFP. This pixel-resolved GC×GC-MS strategy offers a transferable approach for complex source fingerprinting, treatment-performance diagnosis, and prioritization of SOA-relevant cooking-emission precursors.