Abstract: To elucidate the volatile composition and characteristic markers of loquat honey, a comprehensive analytical investigation was conducted on samples sourced from four major producing regions in China: Jiangsu, Anhui, Fujian, and Sichuan. A dual-analytical strategy was adopted, integrating qualitative profiling with subsequent targeted quantitative verification. Initial characterization of volatile organic compounds (VOCs) was accomplished using headspace solid-phase microextraction coupled with gas chromatography-mass spectrometry (HS-SPME-GC-MS)—a highly sensitive analytical platform ideally suited for capturing and identifying trace-level, thermally stable volatile and semi-volatile constituents in complex matrices such as honey. Through comparative chromatographic profiling and mass spectral matching against the National Institute of Standards and Technology (NIST) spectral library and combined with retention index (RI), a total of 50 distinct volatile compounds were reliably identified across all loquat honey samples. Critical scrutiny of the qualitative dataset revealed five compounds uniquely detected in loquat honey, namely: p-methoxybenzaldehyde (PMBALd), p-methoxybenzyl alcohol (PMBOH), p-methoxybenzoic acid (PMBA), methyl-p-methoxybenzoate (MPMA), ethyl-p-methoxybenzoate (EPEA). Furthermore, sensory assessment confirmed that the aromatic profiles of these five compounds were consistent with the characteristic organoleptic properties of genuine loquat honey. Consequently, based on their chemical uniqueness and congruence with the distinctive aroma of loquat honey, these compounds are proposed as characteristic volatile markers for loquat honey. Quantitative analysis of the five characteristic components was performed via liquid-liquid extraction coupled with gas chromatography-mass spectrometry (LLE-GC-MS). Methodological validation demonstrated excellent linearity for all five components over the concentration range of 0.1-5 μg/mL, with correlation coefficients (R²) exceeding 0.99. Limits of detection (LODs) were determined to be 0.015 mg/kg for PMBALd, MPMA, and EPEA, with corresponding limits of quantification (LOQs) of 0.05 mg/kg; for PMBOH and PMBA, LODs were 0.06 mg/kg and LOQs were 0.20 mg/kg. Recovery experiments were conducted using both acacia honey (negative matrix) and loquat honey (positive matrix). At three spiked concentration levels, the average recoveries of the five markers in negative acacia honey samples ranged from 86.2% to 113.2%, with relative standard deviations (RSDs)≤6.6%. In positive loquat honey samples, the average recoveries varied from 81.2% to 109.6%, with RSDs≤8.5%. These results confirmed the satisfactory recovery and precision, meeting the technical requirements for quantitative analysis of honey matrices. The identified characteristic markers, in conjunction with established physicochemical parameters (e.g., sugar composition, moisture content, diastase activity), can be integrated to develop a comprehensive and specific quality evaluation and authenticity verification system for loquat honey. This system will facilitate differentiation from other honey types and accurate detection of adulteration. Furthermore, the dual-analytical framework—utilizing HS-SPME-GC-MS for untargeted volatile profiling and LLE-GC-MS for targeted quantification—coupled with the marker identification rationale based on both chemical exclusivity and sensory relevance, offers a valuable methodological reference for investigating volatile compounds and discovering characteristic markers in other monofloral honey varieties.

Key words: loquat honey, gas chromatography-mass spectrometry (GC-MS), volatile compounds, characteristic markers