Volatile compound fingerprinting for traditional sesame oil authentication using GC-MS with GC×GC-MS-assisted identification

Sesame is cultivated worldwide for its nutritional value and flavor, especially in India, China, Myanmar, and Tanzania (Yin et al., 2021). Sesame oil (SO) is an edible oil made from sesame seeds roasted at high temperatures to produce the distinct aroma and then processed using pressing, leaching, or other methods (Huang et al., 2024, Ma et al., 2022). This traditional Chinese characteristic edible oil is extremely popular among consumers because of its high nutritional value and unique flavor. The unique aroma is an important indicator of quality and the main feature that distinguishes SO from other vegetable oils. SO contains fatty acids, vitamin E, and phytosterols that support cardiovascular health and anticarcinogenic processes (Xu et al., 2025), along with sesamin and other lignans that impart strong antioxidant and anti-corticosteroid properties (Yin et al., 2024). Owing to its nutritional and sensory values, the high retail price of SO has motivated adulteration with cheaper edible oils or SO essences (SOE). Therefore, the development of SO adulteration detection technologies such as fingerprinting is essential.

Fingerprinting technology typically involves using analytical detection and data processing methods to generate chromatograms/spectra of the separate sample components simultaneously. In China, fingerprinting is primarily used in the fields of traditional Chinese medicine, biology, and food production. To ensure the quality of traditional Chinese medicines, the Chinese Pharmacopoeia Commission developed the “Chinese Medicine Chromatographic Fingerprint Spectra Similarity Evaluation System” (Fingerprint Similarity System). After importing sample chromatograms, the software generates a reference fingerprint spectrum based on the retention times and peak heights, from which the similarity between the individual samples and reference spectrum is calculated. Samples with similarities > 0.900 are considered highly compatible with the reference fingerprint (National Medical Products Administration, 2000). Chromatographic fingerprinting is particularly valuable for traditional SOs and reveals how variations in raw material origin or production processes influence the overall similarity of the volatile component profiles across samples. This enables more effective adulteration detection, origin discrimination, varietal differentiation, and characteristic substance analysis (Fu et al., 2023, Shi et al., 2025; Sun et al., 2023).

Volatile compound fingerprinting has been employed to characterize the volatile profiles of food products, enabling product discrimination and quality assessment (including identification and grading). Sun et al. (2020) analyzed volatile flavor compounds in SO and SO adulterated with soybean oil and identified characteristic compounds that can serve as indicators for adulteration. Wang et al. (2024a) used UV spectral fingerprints in combination with chemometric analytical methods to assess the quality of pressed SO with different levels of adulteration with refined SO. Aghili et al. (2023) analyzed the fatty acid compositions of diluted SO samples using an electronic nose (e-nose) and gas chromatography-mass spectrometry (GC-MS), revealing differences in the volatile profiles between the original and adulterated SO samples, which could aid in distinguishing adulterations. Xing et al. (2019) established fatty acid fingerprints capable of identifying sesame oil adulterated with rapeseed oil at levels as low as 5%. However, the sensitivity decreased for adulteration with corn oil, where only levels above 10% were detectable. Although Zhang et al. (2016) established a simple and rapid ion mobility spectrometry method to identify sesame oil adulterated with flavoring substances, its detection capability is limited to adulteration levels above 10%. However, few studies have assessed the performance of GC-MS-based fingerprinting in the concurrent detection of SO dilution and essence adulteration from volatile profiles.

Although the chromatographic fingerprint similarity evaluation system was originally developed for traditional Chinese medicines, it has recently been successfully applied to edible oils. For example, Li et al. (2025) established volatile fingerprints for fried pepper oils using the "Chinese Medicine Chromatographic Fingerprint Similarity Evaluation System" and adopted a similarity threshold of 0.900 for quality discrimination. Therefore, we characterized the volatile fingerprint of authentic traditional SO using headspace solid-phase microextraction (HS-SPME) coupled with two-dimensional GC (GC×GC)-MS; co-elution issues were resolved by parallel GC×GC-MS analysis, while fingerprint similarity was evaluated using the "Chinese Medicine Chromatographic Fingerprint Similarity Evaluation System" based on GC-MS data. By comparing this fingerprint with those of diluted SO by maize oil (MO) and sesame essence-adulterated oil, we investigated the accuracy of the fingerprint model. This study provides a scientific basis for quality control and authentication of traditional SO.

2. Materials and methods

2.4. GC×GC-MS

GC-MS (8890–5977B, Agilent Technologies, Santa Clara, CA, USA) with a solid-state modulator (SSM) 1800 (J&X Technologies, Shanghai, China) was used to qualitatively analyze the volatile compounds of SO (B. Wang et al., 2023).

Volatile compounds were detected as previously described (Duan et al., 2025), with some modifications. Mass spectra were acquired over the range of 40–450 m/z. Ultra-purity helium (99.999% purity) served as the carrier gas, flowing at a rate of 1.2 mL/min. A splitless injection mode was used. The oven temperature program was set to 40 °C and held this temperature for 2 min, slowly increased to 230 °C at 4 °C per min, and held at 230 °C for 5 min. The detector temperature was set to 250 °C. The quadrupole temperature and transfer line temperature were 150 °C and 250 °C, respectively. An electron impact ion source with an electron energy of 70 eV was used. The ion source temperature was set to 230 °C. A full scan mode was applied for acquisition. Two columns were used inside the GC oven: the first was a polar column, DB-wax (30 m × 0.25 mm × 0.25 μm; Agilent Technologies); the second was a mid-polar column, DB−17 (mid-polar, 1.85 m × 0.18 mm × 0.18 μm; J&X Technologies). SSM1800 was placed between them for the heating and cooling stages. The SSM modulation period was 4 s. GC×GC-MS was used solely for qualitative confirmation of compounds that co-eluted in GC-MS.

2.5. GC-MS

Volatile compounds were analyzed semi-quantitatively by GC-MS (7890A−7000B; Agilent Technologies), and 2-methyl−3-heptanone was used as an internal standard. GC-MS analysis was performed using a DB-wax column (polar, 30 m × 0.25 mm, 0.25 μm; Agilent Technologies, Folsom, CA, USA) to distinguish the volatile compounds. The carrier gas (ultra-pure helium, 99.999%) was delivered at a flow rate of 1.2 mL/min. The method, heating, and mass spectrometry conditions are consistent with those described in Section 2.4. All quantitative and fingerprint similarity analyses were based on the GC-MS data obtained from this section.

3. Results and discussion

3.2. Analysis of volatile compounds

The GC×GC-MS result of JH and the GC-MS results of JH, MSO 5 (50% MO), and MSE 3 (0.6% SOE) are shown in Fig. 1. A total of 85 volatile compounds were identified in the SO, MSO, and MSE samples, and 33 common peaks were identified, including ketone (1), acid (1), nitrogen-containing compounds (16), sulfur-containing compounds (7), furans (3), aldehyde (1), and phenolic compounds (4, Table 3).

Journal of Food Composition and Analysis, 2026, 155:109277: Fig. 1. (A) and (B) respectively represent the three-dimensional chromatogram and two-dimensional chromatogram of volatile compounds within the JH−1 sesame oil sample; (C)-(E) respectively represent the GC-MS results of volatile compounds within the JH−1, MSO 6 (50% MO), and MSE 3 (0.6% MSE).

3.3. Establishment of GC-MS fingerprints and similarity analysis of volatile substances

The GC-MS volatile compound data from the test SO samples were imported into the Fingerprint Similarity System to generate chromatographic fingerprints. The reference fingerprint profile was constructed as the arithmetic mean of all profiles (Fig. 3). Similarity values between the sample fingerprints and reference were calculated to evaluate compositional consistency. The SO samples showed a reference fingerprint similarity of 0.910–0.993 (Table 4), surpassing the 0.900 benchmark (Ma et al., 2024). This demonstrates a consistent composition of key volatile components across samples with minimal quality variation, satisfying the fundamental fingerprinting and quality assessment requirements for traditional SO.

Journal of Food Composition and Analysis, 2026, 155:109277: Fig. 3. Reference fingerprint of volatile compounds in traditional SO established by GC-MS. Different numbers are represented the different compounds in Table 2.

4. Conclusion

This study established a volatile compound fingerprinting method using HS‑SPME‑GC‑MS with GC×GC‑MS‑assisted identification for the authentication of traditional sesame oil. A total of 33 common volatile peaks were identified, with pyrazines and guaiacol as major contributors. OPLS‑DA models (R²Y > 0.98, Q² > 0.97) demonstrated robust discrimination between authentic and adulterated oils, as well as between the two adulteration types. Compounds with VIP > 1 and p < 0.05 were identified as potential key markers. Selected key markers were then subjected to external standard quantification, and the obtained absolute concentrations were used to perform PCA. All calibration curves showed excellent linearity (R² > 0.997), and the external standard PCA clearly separated sesame oil essence from all other samples. Fingerprint similarity analysis, validated by ROC (AUC = 1.000, optimal cutoff 0.8935), showed that adulterated samples had similarity values below 0.9, clearly distinguishing them from authentic oils. The method successfully detected maize oil dilution adulteration as low as 5% and sesame oil essence adulteration as low as 0.2%.

Compared with previous methods (Aghili et al., 2023; Wang et al., 2024a; Xing et al., 2019; Zhang et al., 2016), our approach uniquely detects both dilution and essence adulteration in a single workflow, achieves lower detection limits (5% for MO, 0.2% for SOE), and provides an intuitive 0.900 similarity threshold without complex chemometric modeling. While non‑destructive methods (e‑nose, IMS) are more suitable for rapid screening, our approach offers higher chemical specificity, the ability to distinguish between dilution and essence adulteration, and an intuitive similarity threshold. Current limitations include testing only maize oil and one essence type, and validation on laboratory‑prepared samples. Future work will expand to other adulterants (soybean, rapeseed oil) and blind commercial samples. Overall, this method provides a reproducible, empirically validated tool for traditional sesame oil authentication. Furthermore, external standard quantification in the present study was performed only for a selected set of key volatile markers (VIP > 1 and p < 0.05). The same approach can be extended to other compounds in future studies, enabling odor activity value (OAV) calculations for deeper insights into sensory contributions, more precise adulteration level prediction, and improved model robustness across different production batches.