Objective As a premium bio-based material widely used in furniture and automotive interiors, the odor quality of leather directly affects the sensory experience of consumers. However, the odor composition of leather is highly complex, which is generally characterized by a large number of components, significant concentration variations, and intertwined sources. These issues render the traceability of unpleasant odors particularly challenging. Meanwhile, existing detection methods are generally limited by substantial volatile substance loss during pretreatment, insufficient sensitivity, and difficulties in achieving comprehensive analysis. Therefore, in order to enable the systematic characterization of leather odor composition and achieve precise traceability of the sources of odor, the establishment of a detection method for full-components of odor substances in leather is urgently required.

Methods Cowhide sofa leather was selected as the research subject. The leather samples were pretreated using static headspace (HS) extraction and adsorption thermal desorption (TD) to achieve efficient capture of volatile organic compounds (VOCs) and minimize their loss during sample preparation. VOCs were then analyzed by gas chromatography-mass spectrometry (GC-MS). Key parameters, including leather sample mass, incubation time, and multi-stage temperature program, were systematically optimized. During the data analysis phase, qualitative identification of odorants was achieved based on chromatographic retention time, mass spectral data, and standard library. Quantitative analysis was performed using internal standard method and peak area measurement. The detected odorants were described and categorized by odor database retrieval and human olfactory evaluation for further elucidation of odor origins. Based on the physicochemical properties of the identified odorants and the leather manufacturing processes (such as tanning, retanning, fatliquoring, and finishing), a systematic inference regarding the potential sources of odorants was performed. This provides a theoretical basis and data support for leather odor traceability and process control.

Results The optimal pretreatment conditions for the leather samples were established, as a sample weight of 1.0 g, incubation time of 2.0 h, and the addition of 10 μL internal standard. The temperature program for GC was established, i.e., the initial temperature was set at 40 ℃ and held for 6 min, then increased to 100 ℃ at a heating rate of 3 ℃/min, subsequently raised to 150 ℃ at a heating rate of 4 ℃/min and held for 10 min, and finally increased to 230 ℃ at a heating rate of 5 ℃/min and held for 5 min. In this study, the detection method based on HS-TD-GC/MS was successfully developed for the analysis of odorant compounds in leather. This method enabled efficient separation and comprehensive analysis of VOCs in leather. Suspected odorants such as propionic acid, hexanal and nonanal were screened out, which correspond to pungent and oily odors characteristics described in the sensory evaluation of the leather samples. These compounds were traced to potential sources in the leather-making process, such as preservatives, fatliquors, and finishing materials.

Conclusion This study innovatively established an odorant detection method suitable for the complex leather matrix, which not only enhances the capability for detecting volatile compounds in leather and enables the precise analysis of the complex odor profile, but also provides data support for guiding key process optimization, materials selection, and quality control in the production of low-odor and high-grade leather, thereby contributing to the improvement of the sensory quality of leather products.