MOF-Derived Porous Iron Oxide With Optimized Photothermal and Interfacial Properties for High-Fidelity LDI-MS Metabolomics and Digital Traceability of Tobacco Leaves

Engineering inorganic matrices with optimized photon absorption and interfacial energy transfer is pivotal for advancing laser desorption/ionization mass spectrometry (LDI-MS) toward high-fidelity metabolomic analysis. Here, we developed a metal-organic framework (MOF)-derived porous iron oxide (pFO) nanomatrix for LDI-MS-based metabolomics. The pFO exhibits strong UV absorption at 355 nm, a high specific surface area (59.7 m 2 /g), and a uniform mesoporous structure, which collectively afford low background noise, enhanced analyte signal intensity, and high analytical reproducibility (coefficient of variation < 9.2%). Using this platform, metabolic fingerprints were rapidly acquired from 269 tobacco leaf samples. Integrated with Random Forest, the approach achieved accurate discrimination of leaf stalk position (AUC = 0.962) and reliable differentiation between southwestern and southeastern production regions (AUC > 0.944). Furthermore, eight representative m / z features were identified to construct a reduced classification panel, which maintained robust predictive performance in the test set (AUC > 0.913). This study establishes MOF-derived pFO as a high-performance inorganic matrix for LDI-MS metabolomics and establishes a rapid and reproducible analytical strategy for digital classification and quality traceability of agricultural products.