Skip to main content
NCBI home page
Food Chemistry: X logoLink to Food Chemistry: X
. 2023 Apr 26;18:100690. doi: 10.1016/j.fochx.2023.100690

Investigation of aroma characteristics of seven Chinese commercial sunflower seed oils using a combination of descriptive Analysis, GC-quadrupole-MS, and GC-Orbitrap-MS

Jiani Liu a,1, Huimin Zhao b,1, Xiaomin Chang a,1, Xiaolong Li b, Yu Zhang a, Baoqing Zhu a,⁎, Xiangyu Wang b,c,d,⁎
PMCID: PMC10172861  PMID: 37179977

Highlights

  • •

    46 and 107 volatiles were quantitatively detected by GC-quadrupole-MS and GC-Orbitrap-MS respectively in 7 Chinese commercial sunflower seed oils.

  • •

    23 volatile compounds were identified firstly in sunflower seed oil.

  • •

    The volatile composition and aroma profiles of these 7 samples varied dramatically.

  • •

    PLSR was used to detect volatiles that positively correlated to each aroma attribute.

Keywords: Sunflower seed oil, Aroma characteristics, GC-Orbitrap-HRMS, PLSR

Abstract

The aroma characteristics of seven commercial Chinese sunflower seed oils were investigated in this study using descriptive analysis, headspace solid-phase microextraction coupled with GC-quadrupole-MS (LRMS, low-resolution mass spectrometry), and GC-Orbitrap-MS (HRMS, high-resolution mass spectrometry). GC-Orbitrap-MS quantified 96 compounds, including 18 alcohols, 12 esters, 7 ketones, 20 terpenoids, 11 pyrazines, 6 aldehydes, 6 furans, 6 benzene ring-containing compounds, 3 sulfides, 2 alkanes, and 5 nitrogen-containing compounds. Moreover, 22 compounds including 5 acids, 1 amide, and 16 aldehydes were quantified using GC-Quadrupole-MS. To our knowledge, 23 volatile compounds were reported for the first time in sunflower seed oil. All the seven samples were found to have a ‘roasted sunflower seeds’ note, ‘sunflower seeds aroma’ note and ‘burnt aroma’ note and only five of them had ‘fried instant noodles’ note, three had ‘sweet’ note and two had ‘puffed food’ note. Partial least squares regression was used to screen the candidate key volatiles that caused the aroma differences among these seven samples. It was observed that ‘roasted sunflower seeds’ note was positively correlated with 1-octen-3-ol, n-heptadehyde and dimethyl sulfone, whereas the ‘fried instant noodles’ and ‘puffed food’ demonstrated a positive correlation with pentanal, 3-methylbutanal, hexanal, (E)-2-hexenal and 2-pentylfuran. Our findings provide information to the producers and developers for quality control and improvement of sunflower seed oil.

Introduction

Sunflower (Helianthus annulus L.) is one of the primary oil crops worldwide and is cultivated in Europe, the Middle East, Asia, America and Africa (Pilorgé, 2020). Sunflower seed oil is typically extracted from raw or roasted sunflower seeds. Sunflower seed oil is renowned among consumers worldwide due to its rich nutrient content and distinct flavor. In 2020, the global annual production of sunflower seed oil was approximately 19,158 million tons, rising to 21,796 million tons in 2021 (USDA, 2021). The aroma of vegetable oil is an essential characteristic of its quality and has a considerable impact on consumers’ purchasing inclinations. More and more attention has been paid to the research on the aroma of sunflower seed oil.

Both solvent-assisted flavor evaporation (SAFE) and headspace solid-phase microextraction (HS-SPME) pretreatments had been used to concentrate volatile compounds in sunflower seed oil (Guillen et al., 2005, Yin et al., 2022). HS-SPME is one of the most widely reported techniques for the extraction of volatile compounds from sunflower seed oil due to its remarkable advantages such as ease, speed, small sample size, solvent-free system, and automation of the entire extraction process (Guillen et al., 2005, Marasca et al., 2016, Nieva-Echevarria et al., 2019, Petersen et al., 2012). Gas chromatography coupled with single quadrupole mass spectrometry (GC–MS) is commonly used to identify volatile compounds in sunflower seed oil (Aydeniz-Guneser and Yilmaz, 2022, Fu et al., 2019, Hu et al., 2014, Petersen et al., 2012, Valdés García et al., 2021, Yin et al., 2022). Notably, a previous study reported that two-dimensional comprehensive chromatography-time-of-flight mass spectrometry could identify additional volatile compounds in sunflower seed oil (Hu et al., 2014), indicating that advanced analytical instruments could contribute to a better understanding of the sunflower seed oil aroma.

Recently, GC-Orbitrap-MS has emerged as a compelling substitute for single quadrupole MS because accurate mass measurements can reveal substantially lower background signals, thereby enhancing the signal-to-noise ratio (S/N) of the target analyte and minimizing background ion interference. In 2005, Alexander Makarov developed the first Orbitrap-based mass spectrometer (Eliuk & Makarov, 2015) and it was applied to GC/MS for the first time in 2010 (Peterson, McAlister, Quarmby, Griep-Raming, & Coon, 2010). GC-Orbitrap-MS could provide high quality resolution (upto 120,000 FWHM) (m/z 200) and high-quality accuracy (<3 ppm) (Belarbi, Vivier, Zaghouani, Sloovere, Agasse-Peulon, & Cardinael, 2021). Several GC-Orbitrap-MS methods have been recently developed for the quantitative analysis of pesticide residues in kinds of foods (Belarbi et al., 2021, Mol et al., 2016), as well as persistent organic pollutants in environmental samples. GC-Orbitrap-MS has been demonstrated as a powerful technique for the qualitative assessment of trace aroma compounds (Dominguez, Arrebola, Martinez Vidal, & Garrido Frenich, 2020). We recently used GC-Orbitrap-MS to analyze aroma compounds in fruit wines (Lin et al., 2022; R. Liu, Liu, Zhu, Kortesniemi, Zhu, & Li, 2022; Y. Liu et al., 2022). Similarly, high-resolution and sensitive instruments were found that could identify more key volatiles in foods such as green tea, fruits and wines (Lim et al., 2020, Qian et al., 2020, Yang et al., 2020). The application of GC-Orbitrap-MS could contribute to the identification of additional aromatic compounds in sunflower seed oils.

Quantitative descriptive analysis (QDA) is commonly used in food aroma evaluation (Da Costa et al., 2020, Xu et al., 2023), and it had already been used in sunflower seed oil. Yin used the QDA method to evaluate one roasted and one cold-pressed sunflower seed oils and their corresponding reconstituted samples, respectively (Yin, Shi, Li, Ma, et al., 2022). This study found that the two samples were very different in aroma sensory and were determined by their volatiles composition.

Roasting is a standard process to enhance and modify the flavor and nutritional value of vegetable oils including sesame, cactus seed, peanut, walnut and other raw materials (Bi et al., 2022, Nounah et al., 2020; K. Yang et al., 2021, Yin et al., 2021; W. Zhang, Cao, & Liu, 2020; Y. Zhang, Li, Lu, Sun, & Wang, 2021). As a processing technique, roasting also has the potential to increase the oil yield and antioxidant activity (Xu, Zhu, Liu, Karrar, Ouyang, & Li, 2022). In China, the demand for heat-treated sunflower seed oil has increased substantially. However, researches on roasted sunflower seed oil were mainly focused on its nutritional properties, safety and physicochemical indexes but not on flavor (Ji et al., 2022, Kiczorowska et al., 2019, Özcan and Köse, 2022; C.-X. Zhang, Xi, Zhao, Ma, & Wang, 2020). Therefore, it is necessary to investigate the sensory attributes and volatile compounds composition of roasted sunflower seed oil available in the Chinese market.

Partial least squares regression (PLSR) is a widely used statistical analysis method that connects two matrices, X and Y, to create linear multivariate models (Gu et al., 2018). It has been reported to be an effective statistical prediction tool for chemometric datasets with several variables and a relatively small sample size (Wu et al., 2022). Currently, PLSR is mostly used to identify key aroma compounds present in vegetable oils, including onion oil (Tian, Zhan, Tian, Wang, Lu, & Zhao, 2020), A. tenuissimum flower oil (C. Zhang, Wang, Ding, Su, & Zhao, 2022), and even cold pressed sunflower seed oils from Italy (Bendini, Barbieri, Moradi, Palagano, Valli, & Toschi, 2014). We believe that, till date, there has been no research on the application of the PLSR model to analyze the aroma of commercial sunflower seed oils that are popular in China.

To above all, seven commercially available sunflower seed oils from China market were analyzed by using both GC-Quadrupole-MS (LRMS, low-resolution mass spectrometry) and GC-Orbitrap-MS (HRMS, high-resolution mass spectrometry) for volatile identification and the QDA method for sensory evolution. We aimed to investigate the aroma characteristics and the potential key aroma compounds of seven commercial sunflower seed oils from Chinese market. Our findings can be useful in quality control and the improvement of sunflower seed oil. We believe that this is the first study to utilize GC-Orbitrap-MS for the identification of volatiles in sunflower seed oils.

Materials and methods

Samples

The samples selected in this study were seven commercially available sunflower seed oils within the shelf life. Most were produced by different manufacturers, and four samples were from the same manufacturer, albeit with differences in the raw materials and processes. The samples were designated HG, JZ, JN, LH, JXP, JXZ and HK. All the products implement GBT10464-2017 standard.

Chemical

All standards were of chromatographic grade, dissolved in ethanol and stored at -20 °C. Except for standard products, the reagents without special instructions were of analytical grade purity, and their details are shown in Table S1. An n-alkanes solution (C6-C24) was obtained from Supelco, Bellefonte, PA, USA.

HS-SPME-GC-Quadrupole-MS detection of volatile compounds in sunflower seed oil

Volatile compounds in sunflower seed oil samples were extracted by HS-SPME. Five grams of sunflower seed oil sample was accurately weighed into a 20 mL vial, tightly capped with a PTFE-silicone septum and 10 µL of 4-methyl-2-pentanol (solution dissolved in ethanol, 0.2 g/L) was added.

The chromatographic procedure for HS-SPME-GC-Quadrupole-MS was optimized based on existing studies (Jia et al., 2021, Ni et al., 2021, Ni et al., 2022, Zhou et al., 2019). The sample was equilibrated at 60 °C for 30 min. A solid-phase microextraction fiber (50/30 μm DVB/CAR/PDMS; Supelco, Bellefonte, PA, USA, with preheat treatment according to the manufacturer’s recommendations before use) was exposed for 30 min (60 °C) in the vial headspace and then desorbed into the GC injection port at 250 °C for 8 min.

The volatile compounds in seven sunflower seed oil samples were analyzed using an Agilent 7890 A GC System (Agilent Technologies, Santa Clara, CA, USA) on a HP-INNOWAX capillary column (60 m × 0.25 µm, 0.25 µm, J & W Scientific, Santa Clara, CA). The temperature increase procedure for chromatography was as follows: 40 °C for 5 min, 5 °C per min up to 240 °C, and held for 5 min. The flow rate of carrier gas (nitrogen with 99.99% purity) was set at 1 mL/min. All mass spectra were acquired in electron ionization mode at 70 eV using full scan with a scan m/z range of 25 to 300. Under the same chromatographic and mass spectrometric conditions, a C7–C30 n-alkane series (500 μg/L, Supelco, Bellefonte, PA, USA) were analyzed to calculate the retention indices (RIs).

The compounds were identified by combining the calculated RI value, the mass spectrum information in NIST13 (S. Liu et al., 2018, Wang et al., 2018) and the retention index. The relative abundances of the compounds in the samples were expressed as peak area ratios (J. Zhang, Li, Gao, Wang, Gao, & Jiang, 2010). The volatile compounds detected were quantified by calculating the peak area ratio using the internal standard method, with 4-methyl-2-pentanol as the internal standard. Three repeated injections were performed for each sample.

HS-SPME-GC-Orbitrap-MS detection of volatile compounds in sunflower seed oil

Sunflower seed oil samples were pretreated by HS-SPME, as described in Section 2.3. A TriPlus RSH autosampler (Thermo Fisher Scientific, Bremen, Germany) was used for automated headspace solid-phase microextraction.

A Thermo Scientific Trace 1300 gas chromatography coupled with a Thermo Scientific Q-Exactive Orbitrap mass spectrometer (GC-Orbitrap-MS, Thermo Scientific, Bremen, Germany) was used to analyze volatile compounds in sunflower seed oil. The chromatographic column is HP-INNOWAX capillary column (60 m × 0.25 µm, 0.25 µm, J & W Scientific, Santa Clara, CA). The temperature increase procedure of chromatography was as follows: 40 °C for 5 min, 150 °C at a rate of 3 °C/min, and then 230 °C at a rate of 5 °C/min. The carrier gas was helium (purity: 99.999%), and the flow rate was 1 mL/min. MS acquisition was performed in profile mode using an m/z range of 33–350. The resolution power was set at 60,000 full widths at half-maximum (FWHM) at m/z 200. Data acquisition and analyses were performed using the Xcalibur version 4.1 software, with the processing setups Quan Browser and Qual Browser (Thermo Fisher Scientific, Les Ulis, France). Each analysis was performed in duplicate (Y. Liu et al., 2022). The retention indices were obtained by injection of the C6–C24 n-alkane series (Supelco, Bellefonte, PA, USA) mixture under the same chromatographic conditions.

The compounds were qualitatively analyzed by combining the calculated RI value, mass spectrum information, and retention index. The relative abundance of the compounds in the samples was expressed the peak area ratio (J. Zhang et al., 2010). Three repeated injections were performed for each sample.

Quantitative descriptive analysis

The sensory characteristics of sunflower seed oil products were evaluated by scoring the intensity of each sensory attribute of sunflower seed oil products. The panelists in our evaluation team all have more than two years of experience in sensory evaluation. The definitions of six sunflower seed oil sensory property descriptors including ‘sunflower seeds’, ‘burnt’, ‘fried instant noodles’, ‘sweet’, ‘puffed food’ and ‘roasted sunflower seeds’ were determined through discussion, and reference samples were prepared for evaluator training. The six sensory descriptors are defined in Table S2. The panel were required to score the intensity of each characteristic sensory attribute, with a score range of 0–10 points (linear scale). After a month of training, they started the evaluating of the samples. The higher the score, the greater the intensity of the representative attribute. The evaluation was repeated twice.

Data analysis

Data analyses were performed using Origin 2021 (MicroCal, Massachusetts, USA) and XLSTAT 19 (Addinsoft, New York, NY). One-way analysis of variance (ANOVA) was used to compare the significant differences between the means using Tukey’s post hoc test. Statistical significance was set at α = 0.05. A bubble chart was constructed to show the peak area ratios of the compounds jointly detected using these two methods. The compounds detected in this study and their peak areas in each sample were analyzed using a cluster analysis model of agglomerative hierarchical clustering (AHC). Principal component analysis (PCA) was performed on the average intensity values of each sensory attribute obtained from the QDA to determine the sensory characteristics of the samples and their relationships with volatile compounds. Partial least squares regression (PLSR) based modeling was performed to explore the relationship between sensory attributes and volatile composition. In addition, only chemical components with variable importance to projection (VIP) value>1 were retained in the final result of the PLSR models. The data were auto-scaled (mean-centered and divided by the standard deviation of each variable).

Results

Identification of volatiles from sunflower seed oil samples

The volatile compounds were analyzed in seven samples based on HS-SPME-GC-Quadrupole-MS (Table 1) and HS-SPME-GC-Orbitrap-MS (Table 2), respectively.

Table 1.

List of volatile compounds identified in the seven sunflower seed oils studied by GC-Quadrupole-MS.

Categories Compound Structure CAS Quantitative ion Actual RI Identa
Pyrazines Pyrazine C4H4N2 290–37-9 80 1357 St, RI, QMS
2-Ethyl-6-methylpyrazine C7H10N2 13925–03-6 121 1389 St, RI, QMS
2,3,5-Trimethylpyrazine C7H10N2 14667–55-1 42 1401 St, RI, QMS
Methylpyrazine C5H6N2 109–08-0 94 1418 St, RI, QMS
3-Ethyl-2,5-dimethylpyrazine C8H12N2 13360–65-1 135 1440 St, RI, QMS
2-Ethyl-5-methyl pyrazine C7H10N2 13360–64-0 121 1441 St, RI, QMS
2,5-Dimethylpyrazine C6H8N2 123–32-0 108 1482 St, RI, QMS
2,6-Dimethylpyrazine C6H8N2 108–50-9 108 1488 St, RI, QMS
Alcohol Methyl Alcohol CH4O 67–56-1 31 968 St, RI, QMS
1-Pentanol C5H12O 71–41-0 55 1253 St, RI, QMS
1-Octen-3-ol C8H16O 3391–86-4 57 1440 St, RI, QMS
2-Furanmethanol C5H6O2 98–00-0 98 1653 St, RI, QMS
Furan 2-Methylfuran C5H6O 534–22-5 82 936 St, RI, QMS
Dihydro-2-methyl-3(2H)-furanone C5H8O2 3188–00-9 43 1271 St, RI, QMS
2-Pentylfuran C9H14O 3777–69-3 81 1376 St, RI, QMS
2(5H)-Furanone C4H4O2 497–23-4 84 1967 St, RI, QMS
Aldehyde 2-Methylpropanal C4H8O 78–84-2 43 814 St, RI, QMS
Butanal C4H8O 123–72-8 44 878 St, RI, QMS
Pentanal C5H10O 110–62-3 44 981 St, RI, QMS
3-Methylbutanal C5H10O 590–86-3 44 982 St, RI, QMS
α-Methylbutanal C5H10O 96–17-3 57 995 St, RI, QMS
Hexanal C6H12O 66–25-1 44 1095 St, RI, QMS
n-Heptaldehyde C7H14O 111–71-7 70 1193 St, RI, QMS
(E)-2-Hexenal C6H10O 6728–26-3 41 1228 St, RI, QMS
(E)-2-Octenal C8H14O 2548–87-0 41 1329 St, RI, QMS
n-nonanal C9H18O 124–19-6 57 1390 St, RI, QMS
Furfural C5H4O2 98–01-1 96 1463 St, RI, QMS
(Z)-2-Heptenal C7H12O 57266–86-1 83 1485 St, RI, QMS
Decyl aldehyde C10H20O 112–31-2 43 1493 St, RI, QMS
Benzaldehyde C7H6O 100–52-7 106 1523 St, RI, QMS
(E,E)-2,4-decadienal C10H16O 25152–84-5 81 1814 St, RI, QMS
Acid Acetic acid C2H4O2 64–19-7 43 1442 St, RI, QMS
Propanoic acid C3H6O2 79–09-4 74 1526 St, RI, QMS
Butanoic acid C4H8O2 107–92-6 60 1616 St, RI, QMS
4-Hydroxybutanoic acid C4H8O3 591–81-1 86 1829 St, RI, QMS
Hexanoic acid C6H12O2 142–62-1 60 1838 St, RI, QMS
Easter Methyl acetate C3H6O2 79–20-9 43 828 St, RI, QMS
Terpene α-Pinene C10H16 80–56-8 93 1022 St, RI, QMS
α-Phellandrene C10H16 99–83-2 93 1123 St, RI, QMS
α-Thujene C10H16 2867–05-2 93 1131 St, RI, QMS
Camphene C10H16 79–92-5 93 1174 St, RI, QMS
α-terpineol C10H18O 98–55-5 93 1249 St, RI, QMS
γ-Terpinene C10H16 99–85-4 93 1249 St, RI, QMS
Amide Acetamide C2H5NO 60–35-5 59 1962 St, RI, QMS
Pyrrole 1-(1H-pyrrol-2-yl)1-ethanone C6H7NO 1072–83-9 94 1978 St, RI, QMS
1H-Pyrrole-2-carboxaldehyde C5H5NO 1003–29-8 95 2035 St, RI, QMS
Open in a new tab

aBasis for identification: St, standard compound; QMS, quadrupole GC–MS; RI: RI agreed with data base of NIST11.

Table 2.

List of volatile compounds identified in the seven sunflower seed oils studied by GC-Orbitrap-MS.

Categories Compound Structure CAS Quantitative ion Actual RI Identa
Pyrazines Pyrazine C4H4N2 290–37-9 80.036942 1200 O MS
Methylpyrazine C5H6N2 109–08-0 94.05249 1249 St, RI, O MS
2,5-dimethylpyrazine C6H8N2 123–32-0 108.06814 1300 St, RI, O MS
Pyrazine, 2,6-dimethyl- C6H8N2 108–50-9 108.06822 1307 O MS
Pyrazine, ethyl- C6H8N2 13925–00-3 107.06043 1312 St, RI, O MS
Pyrazine, 2,3-dimethyl- C6H8N2 5910–89-4 108.06822 1324 O MS
2-Ethyl-5-methyl pyrazine C7H10N2 13360–64-0 121.07604 1364 St, RI, O MS
2-Ethyl-6-methylpyrazine C7H10N2 13925–03-6 121.07604 1370 St, RI, O MS
2,3,5-Trimethylpyrazine C7H10N2 14667–55-1 122.08369 1383 St, RI, O MS
3-Ethyl-2,5-dimethylpyrazine C8H12N2 13360–65-1 135.09175 1426 St, RI, O MS
5-Ethyl-2,3-dimethylpyrazine C8H12N2 15707–34-3 135.09184 1443 St, RI
Alcohol Methyl Alcohol CH4O 67–56-1 33.033463 905 O MS
2-nonanol C9H20O 628–99-9 45.033535 932 St, RI
2-Propanol, 1-methoxy- C4H10O2 107–98-2 45.033535 1127 O MS
1-Propanol, 2-methyl- C4H10O 78–83-1 41.038399 1147 St, RI
1-Pentanol C5H12O 71–41-0 41.03841 1245 St, RI
Acetone alcohol C3H6O2 116–09-6 43.017773 1277 St, RI
3-Cyclohexene-1-methanol, 6-methyl- C8H14O 5259–31-4 93.069931 1317 O MS
1-Pentanol, 4-methyl- C6H14O 626–89-1 41.038399 1342 O MS
1-Butanol, 2-ethyl- C6H14O 97–95-0 43.017803 1367 O MS
1-Octen-3-ol C8H16O 3391–86-4 57.033539 1440 St, RI, O MS
1-Heptanol C7H16O 111–70-6 41.038399 1446 St, RI
2,3-Butanediol C4H10O2 513–85-9 45.033535 1531 St, RI
2,3-Butanediol, [S-(R*,R*)]- C4H10O2 19132–06-0 45.033535 1571 O MS
Propylene Glycol C3H8O2 57–55-6 45.033535 1583 O MS
2-Furanmethanol C5H6O2 98–00-0 97.02845 1654 St, RI
2-Furanmethanol, 5-methyl- C6H8O2 3857–25-8 112.05201 1718 O MS
Benzenemethanol, α,α-dimethyl- C9H12O 617–94-7 121.06496 1754 O MS
Benzyl alcohol C7H8O 100–51-6 79.054298 1870 O MS
Furan Furan, 3-methyl- C5H6O 930–27-8 82.041397 885 O MS
2-Pentylfuran C9H14O 3777–69-3 81.033531 1221 St, RI, O MS
3(2H)-Furanone, dihydro-2-methyl- C5H8O2 3188–00-9 43.017803 1247 O MS
Acetylfuran C6H6O2 1192–62-7 95.012772 1483 St, RI
1-Pentanone,1-(2-furanyl)- C9H12O2 3194–17-0 110.03626 1749 St, RI
Benzenes Styrene C8H8 100–42-5 104.06213 1239 St, RI
Benzene, 1-methyl-2-(2-propenyl)- C10H12 1587–04-8 117.07002 1413 O MS
Naphthalene, 1-methyl- C11H10 90–12-0 141.07001 1829 O MS
Naphthalene, 2-methyl- C11H10 91–57-6 141.07001 1865 O MS
Butylated Hydroxytoluene C15H24O 128–37-0 205.15872 1908 O MS
Phenol C6H6O 108–95-2 94.041405 1991 St, RI
Sulfur Carbon disulfide CS2 75–15-0 75.943642 713 O MS
Dimethyl Sulfoxide C2H6OS 67–68-5 78.013474 1556 O MS
Dimethyl sulfone C2H6O2S 67–71-0 78.984879 1885 O MS
Aldehyde 2-Methylpropanal C4H8O 78–84-2 41.03841 859 St, RI
Hexanal C6H12O 66–25-1 41.03841 1078 St, RI
n-Heptaldehyde C7H14O 111–71-7 55.054306 1176 St, RI
(E)-2-Hexenal C6H10O 6728–26-3 41.03841 1204 St, RI
Octanal C8H16O 124–13-0 41.038399 1271 St, RI
2-Heptenal, (E)- C7H12O 18829–55-5 83.04921 1300 St, RI
n-nonanal C9H18O 124–19-6 41.03841 1372 St, RI
5-Ethylcyclopent-1-enecarboxaldehyde C8H12O 36431–60-4 67.054298 1389 O MS
(E)-2-Octenal C8H14O 2548–87-0 41.03841 1405 St, RI
Furfural C5H4O2 98–01-1 95.012825 1442 St, RI
Trans −2- nonanal C9H16O 18829–56-6 41.03841 1520 St, RI
5-Methyl-2-furancarboxaldehyde C6H6O2 620–02-0 109.02845 1553 St, RI
Benzeneacetaldehyde C8H8O 122–78-1 91.054245 1621 St, RI
Benzaldehyde, 2-hydroxy- C7H6O2 90–02-8 121.02856 1658 O MS
2E,4E-Decadienal C10H16O 25152–84-5 81.033554 1798 St, RI
1H-Pyrrole-2-carboxaldehyde C5H5NO 1003–29-8 94.028748 2001 St, RI, O MS
Esters Methyl acetate C3H6O2 79–20-9 43.017773 867 St, RI
Isoamyl acetate C7H14O2 123–92-2 43.054199 1054 St, RI
Butyl propionate C7H14O2 590–01-2 75.044151 1138 O MS
Butyl acrylate C7H12O2 141–32-2 55.017899 1170 O MS
Ethyl lactate C6H8N2 97–64-3 45.033535 1336 St, RI
Allyl tiglate C8H12O2 7493–71-2 83.04921 1347 O MS
Ethyl (S)-(-)-lactate C5H10O3 687–47-8 45.033535 1385 O MS
trans-3-Hexenyl butyrate C10H18O2 53398–84-8 67.054298 1481 O MS
(-)-Bornyl acetate C12H20O2 5655–61-8 95.08564 1562 O MS
Citronellyl acetate C12H22O2 150–84-5 95.04924 1641 St, RI
Myrtenyl acetate C12H18O2 1079–01-2 91.054306 1643 O MS
Methyl salicylate C8H8O3 119–36-8 120.02071 1757 St, RI
Acid Propanoic acid C3H6O2 79–09-4 74.03624 948 St, RI
Acetic acid C2H4O2 64–19-7 43.017757 1428 St, RI
Hexanoic acid C6H12O2 142–62-1 73.028442 1893 St, RI
Terpene α-Pinene C10H16 80–56-8 91.054245 1013 St, RI
Camphene C10H16 79–92-5 93.069878 1055 St, RI
β-Pinene C10H16 127–91-3 93.069878 1096 St, RI, O MS
Sabinene C10H16 3387–41-5 91.054245 1112 St, RI, O MS
α-Phellandrene C10H16 99–83-2 91.054245 1154 St, RI
γ-Terpinene C10H16 99–85-4 91.054245 1230 St, RI
p-Cymene C10H14 99–87-6 119.08569 1251 St, RI
α-Campholenal C10H16O 4501–58-0 93.069931 1466 O MS
5,7-Octadien-4-one, 2,6-dimethyl-, (Z)- C10H16O 3588–18-9 95.08564 1487 O MS
(±)-Linalool C10H18O 78–70-6 93.069878 1540 St, RI
Cyclosativene C15H24 22469–52-9 161.13246 1541 O MS
Calarene C15H24 17334–55-3 161.13246 1569 O MS
(±)-Terpinen-4-ol C10H18O 562–74-3 93.069878 1588 St, RI
Menthol C10H20O 89–78-1 81.069984 1632 St, RI
Verbenol C10H16O 473–67-6 79.054298 1671 O MS
Bicyclo[3.1.1]hept-3-en-2-one, 4,6,6-trimethyl- C10H14O 80–57-9 107.08566 1686 O MS
α-terpineol C10H18O 98–55-5 121.10121 1694 St, RI
Myrtenol C10H16O 515–00-4 79.054298 1787 St, RI
Ketone 3-Penten-2-one, (E)- C5H8O 3102–33-8 69.033562 1121 O MS
Acetoin C4H8O2 513–86-0 45.033497 1265 St, RI, O MS
4-Cyclopentene-1,3-dione C5H4O2 930–60-9 68.025772 1563 O MS
2(3H)-Furanone, dihydro-4,5-dimethyl- C6H10O2 6971–63-7 42.046307 1602 O MS
2(5H)-Furanone C4H4O2 497–23-4 55.017818 1733 St, RI, O MS
Pantolactone C6H10O3 599–04-2 71.04921 2006 St, RI
Alkanes Cyclohexene, 1-ethyl- C8H14 1453–24-3 81.069984 1481 O MS
Azulene C10H8 275–51-4 128.06216 1717 O MS
Nitrogen-containing heterocycles Pyridine C5H5N 110–86-1 79.04174 1173 O MS
Pyrrole C4H5N 109–97-7 67.041733 1498 O MS
1H-Pyrrole, 3-methyl- C5H7N 616–43-3 80.049561 1558 O MS
1H-Pyrrole-2-carboxaldehyde, 1-methyl- C6H7NO 1192–58-1 108.0445 1599 O MS
1-(1H-pyrrol-2-yl)1-ethanone C6H7NO 1072–83-9 94.028748 1958 St, RI, O MS
Open in a new tab

aBasis for identification: St, standard compound; O MS, Orbitrap-MS; RI: RI agreed with data base of NIST11.

A total of 46 volatile compounds were detected using GC-Quadrupole-MS. The detected compounds included 8 pyrazines, 3 lower alcohols, 4 furans, 1 compound with benzene ring, 15 aldehydes, 5 acids, 1 ester, 6 terpenes, 1 amide and 2 pyrrole compounds. In contrast, 107 volatile compounds were detected by GC-Orbitrap-MS, including 11 pyrazines, 9 lower alcohols, 10 higher alcohols, 5 furans, 6 compounds with benzene rings, 3 sulfur compounds, 14 aldehydes, 13 esters, 3 acids, 12 terpenes, 11 ketones, 3 alkanes, 2 nitrogen-containing compounds, 1 pyridine compound and 4 pyrrole compounds.

The HRMS specifically detected 72 volatile compounds, whereas 13 compounds were identified by LRMS specifically, and 33 compounds were detected simultaneous (Fig. 1 (a)). It could be seen from Fig. 1 (b) that among the compounds specifically detected by HRMS, there were 3 pyrazines, 14 alcohols, 4 furans, 6 compounds containing benzene rings, 3 sulfides, 7 aldehydes, 12 esters, 12 terpenes, 5 ketones, 2 alkanes, 4 nitrogen-containing heterocycles. The compounds specifically detected by LRMS included two furan compounds, seven aldehydes, two acids, one terpene, and a single amide. Overall, HRMS detected a higher number of volatile compounds in sunflower seed oil than LRMS. HRMS were more suitable for qualitative detection of alcohols, esters, terpenoids and ketones in sunflower seed oil in terms of the types of compounds specifically detected, LRMS might be more suitable for qualitative detection of aldehydes and acids in the samples. It is evident from Fig. 1 (a) that HRMS can characterize more compounds than LRMS, which may be due to the lower mass spectral sensitivity of LRMS. Interestingly, HRMS, which has the advantages of high resolution and high sensitivity, still failed to characterize 13 compounds. This may be due to the relatively low mass charge of the ionic fragments of the compound, which is difficult to match during characterization. For instance, acetamide, which has a molecular weight of 59, probably failed to be characterized by HRMS because it generates major fragments with a mass-to-charge ratio below 50 m/z during ionization by the ion source. As far as we know, the present study reports for the first time 23 volatile compounds in sunflower seed oil.

Fig. 1.

Fig. 1

Open in a new tab

Comparison of the qualitative effects of HRMS and LRMS.

Esters are generally considered aroma-rich compounds and it has been demonstrated in a variety of media that they have major synergistic and aroma-presenting effects (De-la-Fuente-Blanco et al., 2020, Jia et al., 2021, Xiao et al., 2019). In contrast, the reports of esters in studies on sunflower seed oil were rare. In this study, 8 of the 13 esters including methyl acetate (E1), isoamyl acetate (E2), butyl acrylate, (E4), ethyl lactate (E5), ethyl (S)-(-)-lactate (E7), citronellyl acetate (E10), myrtenyl acetate (E11), methyl salicylate (E12) were first detected in sunflower seed oil by HRMS. These compounds were firstly detected in sunflower seed oils. The esters were probably produced by the decomposition of the hydroperoxides generated by the oxidation of oils (W. Zhang et al., 2020).

In recent years, small-molecule sulfides have been detected that can provide ‘sulfur’, ‘fatty’ and ‘cabbage’ flavors to vegetable oils (Yini Yang et al., 2022). Among the three sulfides detected in this study using HRMS, dimethyl sulfone (S3) was reported for the first time in sunflower seed oil. This compound has been detected in rapeseed oil (Zhou et al., 2019) and raw cabbage and has been described as ‘sulfurous’ and ‘fatty’ (Marcinkowska, Frank, Steinhaus, & Jeleń, 2021). A rencent research generally suggested that the sulfur compounds in vegetable oils came from the oilseed crop itself or that sulfur-containing precursors react with enzymes during processing (Yu et al., 2022).

To our knowledge, this study reported 23 volatiles in sunflower seed oil for the first time, including the 8 esters and 1 sulfur-containing compound mentioned above in addition to verbenol (T14, HRMS), myrtenol (T17, HRMS), α-phellandrene (T5, LRMS and HRMS), p-cymene (T7, HRMS), menthol (T13, HRMS), azulene (ALK2, HRMS), 5-ethylcyclopent-1-enecarboxaldehyde (AL12, HRMS), 1-ethyl-cyclohexene (ALK1, HRMS), phenol (B6, HRMS), styrene (B1, HRMS), 1-methyl-2-(2-propenyl)benzene (B2, HRMS), 1-methyl-naphthalene (B3, HRMS), 2-methyl-naphthalene (B4, HRMS), and butylated hydroxytoluene (B5, HRMS).

Quantitative comparison of volatiles among sunflower seed oil samples studied

The volatile compounds were quantified using an internal standard method, and the peak area ratios were calculated for each compound in the samples. The quantitative results from these two methods are plotted in a bubble diagram for the 33 substances detected simultaneously using HRMS and LRMS (Fig. S1). It can be observed from Fig. S1 that some compounds may be effectively detected in the samples by LRMS, e.g. Acetic acid, which was detected by LRMS in all seven samples, whereas in HRMS the compound was quantified in only two samples; this observation was similar for compounds including n-heptaldehyde and α terpineol. Conversely, some compounds may be better detected in samples by HRMS, such as 2-Ethyl-6-methylpyrazine, which was detected by HRMS in all seven samples but only quantified in four samples by LRMS. The occurrence of the aforementioned may be a result of variations or inherent qualities of the device; however, we decided to quantify these compounds using a more efficient mass spectrometry method. Among these 33 compounds, some compounds could be quantified in LRMS, but with very small relative concentrations and also without differences between samples, whereas these compounds were well represented in HRMS, such as γ-Terpinene and (E)-2-Hexenal. This is probably because the concentrations of these compounds in the samples were below the minimum detection limit of LRMS and outside the linear range of quantification by LRMS. Therefore, they could only be quantified by HRMS, which has a much lower detection limit. According to the aforementioned observations, among these 33 compounds, 23 were quantified using HRMS and the remaining 10 using LRMS.

Finally, 96 compounds, including 18 alcohols, 12 esters, 7 ketones, 20 terpenoids, 11 pyrazines, 6 aldehydes, 6 furans, 6 benzene ring-containing compounds, 3 sulfides, 2 alkanes, and 5 nitrogen-containing compounds, were quantified using HRMS, whereas 22 compounds including 5 acids, 1 amide, and 16 aldehydes were quantified using LRMS. The mean values were calculated and 108 out of 118 compounds were found to have significantly different contents among the samples based on ANOVA (p ≤ 0.05 considered to indicate significantly different) and the results were shown in Table S3. The relative contents did not differ significantly between samples (p > 0.05) for 2-methylfuran, dihydro-2-methyl-3(2H)-furanone, α-thujene, benzaldehyde, α-phellandrene, trans-2-nonanal, 5-methyl-2-furancarboxaldehyde, benzeneacetaldehyde, benzaldehyde, 2-hydroxy-, and α-campholenal, therefore, they were excluded from the subsequent statistical analysis.

After normalization, a cluster heat map was created to observe the distribution of the 108 compounds among the samples (Fig. 2). These compounds could be divided into two major categories. There were 60 compounds in Group 1 (G1), including 11 terpenoids. In sunflower seed oil, these compounds were usually from sunflower seeds. Terpenoids have also been identified as key aromatic compounds in cold-pressed sunflower seed oils (Yin, Shi, Li, Ma, et al., 2022). It should be noted that terpenoids might also be converted into other components during heat treatment (Yin, Shi, Li, Ma, et al., 2022). Group 2 (G2) consisted of 48 compounds, including 10 pyrazines and all 3 sulfur-containing compounds detected in this study.

Fig. 2.

Fig. 2

Open in a new tab

The cluster Heatmap showing the relative content of compounds in seven sunflower oil samples.

Among the 108 compounds that qualified in JXZ, 37 were considerably higher in the samples than the other six, mainly aldehydes and terpenoids. Among these 37 compounds, 3-ethyl-2,5-dimethylpyrazine (NH10) was 2.8 times higher than the content of this compound in JN, two alcohols (Al-ol4 and Al-ol18) were 10 times higher than the samples with the lowest content on average, while seven other compounds (Al-ol7, Al-ol8, Al-ol10, Al-ol18, T14, T17, and Al-ol20) were 12 times higher than the samples with the lowest content on average, and one furan (F5, with a relative concentration of 0.002) was only detected in this sample. Five compounds with benzene rings (B1, B2, B4, B5, B6) were 14 times higher than the samples with the lowest content on average, the four esters (E8, E9, E11, E12) were 19 times higher than the samples with the lowest content on average, the concentration range of 11 terpenoids (T1-7, T9, T11-13) was 0.003–9.114, and the four other compounds (AL9, AL10, T8, T15) were 6 times higher than the samples with the lowest content on average. One alkane (ALK1, 0.088) was only detected in this sample.

Twenty-nine compounds in LH, mainly pyrazines, were significantly more abundant than those in the other samples. Among them, 10 pyrazines (NH1-9 and NH11) were 32 times higher than the samples with the lowest content, 2 alcohols (Al-ol15 and Al-ol17), 3 furans (F1, F3 and F4), 1 amide (AM1), 2 sulfides (S1, S2), 3 ester compounds (E1, E5, and E7), 4 ketone compounds (K2, K4, K5, and K6), 1 pyridine compound (N1), 3 pyrrole compounds (N2, N4, and N5). 19 compounds in JZ that were significantly more abundant than other samples, including 4 alcohols (Al-ol1, Al-ol3, Al-ol4, Al-ol14), 5 aldehydes (AL1, AL5, AL13, AL14, AL16), 1 acid (A4), 3 compounds with benzene rings (B3-5), 2 esters (E2, E3), 1 ketone (K3), and 2 other compounds (T10, ALK2).

The compounds in HK that were significantly higher than those in other samples were: Al-ol5 (relative concentration is 0.122), which was 6 times the content of this compound in LH, F2 (0.349, relative concentration, the same below) was 7 times the content of this compound in LH, AL4 (0.337) was 5.8 times the content of this compound in JXZ, and AL6 (1.664) was 3.8 times the content of this compound in JXZ. There were 12 compounds in HG that are significantly higher than other samples, including 4 alcohols (Al-ol1, Al-ol2, Al-ol12, Al-ol13), 3 aldehydes (AL7, AL13, AL16), 1 acid (A4), 1 compound with benzene ring (B5), 1 sulfide (S3).

Ten compounds with no significant difference (p > 0.05) in content as mentioned previously were excluded from the subsequent statistical analysis.

Sensory characteristics of sunflower seed oil samples

QDA was conducted to quantitatively describe the aroma characteristics of seven commercial sunflower seed oils by a well-trained panel. One-way ANOVA was performed on the quantitative description of the results, and all the six specific notes were found to be significantly different (α < 0.05) among the samples. HCA was performed, as shown in Fig. S2.

The intensity of ‘roasted sunflower seeds’ aroma ranged from 2.8 to 4.9 in the samples, with the intensity of this attribute scoring 4.0 and above in all samples except JXZ. The intensity of ‘sunflower seeds’ aroma ranged from 0.9 to 5.9, reaching the highest value in JXZ and the lowest value in LH. The intensity distribution of ‘burnt’ aroma in the samples ranged from 2.5 to 5.5. The ‘fried instant noodles’ aroma had an intensity distribution in the samples between 0.3 and 4.3, reaching a minimum of 0.3 and 0.4 in JXP and LH. Only JXZ, JZ and HG were described as a ‘sweet’ aroma with intensities of 1.0, 2.0 and 2.7, respectively. The ‘puffed food’ aroma was only identified in JXP and LH, with intensities of 5.3 and 6.3, respectively. Overall, all samples had ‘sunflower seeds’ aroma, ‘roasted sunflower seeds’ aroma and ‘burnt’ aroma with generally high scores.

‘Fried instant noodles’ aroma, ‘puffed food’ aroma and ‘sweet’ aroma were only present in part of the samples or very weakly in some samples. The intensity of ‘fried instant noodle’ aroma in the samples ranged from 0.3 to 4.3, reaching a maximum in HK and a minimum in the JXP, and in fact LH also exhibited a very weak ‘fried instant noodle’ aroma (0.4). A ‘sweet’ aroma was only found in JXZ, JZ and HG, with distributions between 1.0 and 2.7. In parallel, only JXP and LH were described as ‘puffed food’ aroma, with scores of 5.3 and 6.3, respectively.

Principal component analysis on QDA results

The QDA results of these seven samples were used for the principal component analysis, and the peak areas of the compounds were introduced as supplementary variables. The results are shown in Fig. 3. The first two principal components together contributed 80.80% of the variance, and the seven samples were well-differentiated. JXP and LH were identified as ‘puffed food’. JXZ had a more pronounced ‘sunflower seeds’, while the rest of the samples were closer to ‘fried instant noodles’ and ‘sweet’. The panel described LH as having higher ‘puffed food’ and ‘burnt’ scores. As can be seen from Fig. 3, the seven samples can be classified into three categories according to their aroma characteristics, including JXZ, which is closer to ‘sunflower seeds’ aroma; HG, JZ, HK and JN, which are closer to ‘sweet’ and ‘fried instant noodles’; and JXP and LH, which are closer to ‘puffed food’ and ‘burnt’. Undoubtedly, the aroma of sunflower seed oil is altered by thermal processing, and sunflower seed oil exhibits different aroma characteristics depending on the process and the degree of thermal processing.

Fig. 3.

Fig. 3

Open in a new tab

The principal component analysis model showing the sensory attribute scores of six sunflower oil samples.

The distribution pattern of the complementary variables showed that some of the detected compounds may be closely related to the sensory properties of the sample. For example, the ‘puffed food’ aroma was more closely located to several pyrazines (NH2, NH1, NH3, NH8, NH9, etc.), and ketones such as K2, K4, and K5, in addition to AM1, Al-ol6, Al-ol15, F3, and F4. Moreover, S3 was correlated with ‘roasted sunflower seeds’ due to its close location. ‘Sweet’ was closer to AL16, Al-ol13 and other compounds, which may also be correlated.

PLSR analysis

To further understand the relationship of volatile compounds and the six sensory attributes of sunflower seed oils, PLSR models were constructed and the results were shown in Fig. 4, and all models were optimized to Q2 > 0.4 and R2Y > 0.6. Compounds with VIP > 1 and correlation coefficient > 0 in each model were marked in red in Fig. 4 and listed in Table S4.

Fig. 4.

Fig. 4

Open in a new tab

Partial least squares regression analysis for aroma intensity variables and volatile compounds content variables in six sunflower oils.

As shown in Table S4 and Fig. 4 (a), the ‘roasted sunflower seeds’ aroma of sunflower seed oil was positively correlated with 1-octen-3-ol (mushroom-like, 0.045) (Neugebauer, Schieberle, & Granvogl, 2021), n-heptaldehyde (citrus-like, fatty, 0.031) (Neugebauer, Granvogl, & Schieberle, 2020) and dimethyl sulfone (sulfurous, fatty, 0.045) (Marcinkowska et al., 2021). Among them, 1-octen-3-ol was also sniffed in GC-O-MS by SAFE pre-treatment method only in a study by Yin et al (Yin, Shi, Li, Ma, et al., 2022). In addition, the 23 compounds reported for the first time in this study with dimethyl sulfone correlated with this aroma. Xu et al. also mentioned in their study on flavor improvement of fried foods that n-heptaldehyde may provide flavor associated with frying or heating (Xu, Mei, Wu, Karrar, Jin, & Wang, 2022).

From Table S4 and Fig. 4 (b), it was evident that the number of aroma compounds correlating tightly with the ‘sunflower seeds’ was very high, including phenol (ink-like, phenolic, 0.021) (Zhai & Granvogl, 2020) and three other benzene compounds, p-cymene (terpene-like, 0.021) (Steinhaus & Schieberle, 2005), menthol (mint-like, 0.025) (Zhai & Granvogl, 2020) and nine other terpenes. Cyclohexene, 1-ethyl- (0.026), 1-pentanone,1-(2-furanyl)- (0.026), (E)-2-heptenal (green, fatty, 0.026) (Neugebauer et al., 2020) and four other aldehydes and five other esters, 1-heptanol (flowery, soapy, fruity, 0.025) (Polster & Schieberle, 2015) and six other alcohols. One alkane and one furan were also found to be highly correlated with the ‘sunflower seeds’ aroma of sunflower seed oil, and the detailed list of compounds is shown in Table S4. Among them, 1-pentanol, α-pinene, γ-terpinene, octanal, (±)-linalool, 2-heptenal, (E)- had been sniffed in GC-O. Xu et al. have mentioned in their study on green plum seed oil that alkenal compounds are potentially associated with fatty aroma (Xu, Wang, et al., 2023). In addition to the reconstituted sunflower seed oil solution, two compounds, 3-ethyl-2,5-dimethylpyrazine and α-phellandrene, were added to the reconstituted solution (Yin, Shi, Li, Ma, et al., 2022). Yin et al. reported (±)-terpinen-4-ol in sunflower seed oil for the first time using the pretreatment method of SAFE (Yin, Shi, Li, Ma, et al., 2022). In the present study, this compound was identified using HS-SPME, which is a simpler pretreatment method combined with HRMS. In addition, 9 of the 23 compounds reported for the first time in this study were correlated with this sensory property, and these compounds are marked with yellow color in Fig. 4 (b).

The ‘fried instant noodle’ aroma of sunflower seed oil was greatly associated with aldehydes, including pentanal (fatty, green, 0.065) (Mallia, Escher, Dubois, Schieberle, & Schlichtherle-Cerny, 2009), 3-methylbutanal (malty, 0.070) (Pollner & Schieberle, 2016), hexanal (green, grassy, 0.051) (Poehlmann & Schieberle, 2013) and (E)-2-hexenal (green apple-like, bitter almond-like, 0.061) (Poehlmann & Schieberle, 2013) (Table S4 and Fig. 4 (c)). In addition to these aldehydes, 2-pentylfuran (vegetable-like. 0.061) (Neugebauer et al., 2020) also demonstrated a positive correlation with the aroma of ‘fried instant noodles’ from sunflower seed oil. The positive correlation between the above compounds, which were described as having a fruity or grassy aroma in most studies, and the ‘fried instant noodle’ aroma in the present study might be due to the interaction between the compounds. Among them, four compounds, pentanal, 3-methylbutanal, hexanal and 2-pentylfuran, were all components of the sunflower seed oil reconstituted solution in the study of Yin et al. Among them, pentanal and 2-pentylfuran were also sniffed in this study by SAFE-GC-O (Yin, Shi, Li, Ma, et al., 2022).

From Table S4 and Fig. 4 (d), it can be observed that there were a large number of aroma compounds positively correlated with the ‘burnt’ aroma of sunflower seed oil, mainly including 5-ethyl-2,3-dimethylpyrazine (0.024), 2-ethyl-5-methyl pyrazine (0.023), 2-ethyl-6-methylpyrazine (0.027) and 8 other pyrazines, 1-(1H-pyrrol-2-yl)1-ethanone (0.026) and 3 other pyrroles, pyridine (0.03), acetamide (0.054), pantolactone (0.091) and 5 other aldehydes and ketones and acetic acid, methyl ester (0.068) and 2 esters, etc. A detailed list of compounds is provided in Table S4. The nitrogenous compounds mentioned above are usually considered to be generated from the Maillard reaction during sunflower seed oil processing, and aldehydes and ketones and esters from the decomposition of oils and fats, with only the difference being whether they were oxidized or not. It could be seen that the ‘burnt’ aroma in sunflower seed oil might come from the processing procedure. Among them, the two compounds detected by HRMS in this study (2-furanmethanol and pyrrole) were also first detected in sunflower seed oil by LRMS using the SAFE pretreatment by Yin et al (Yin, Shi, Li, Ma, et al., 2022). In addition, 2-furanmethanol, acetyl furan, 2(5H)-furanone, 5-ethyl-2,3-dimethylpyrazine, methylpyrazine, 2,5-dimethylpyrazine, pyrazine, 2,6- dimethyl-, pyrazine, ethyl-, pyrazine, 2,3-dimethyl-, 2-ethyl-5-methyl pyrazine, 2-ethyl-6-methylpyrazine and 1H-pyrrole-2-carboxaldehyde, 1- methyl- have also been reported to be aroma-contributing compounds in sunflower seed oil because they were either smelt in GC-O or were present above their threshold values (Yin, Shi, Li, Ma, et al., 2022). Two esters, acetic acid, methyl ester and propanoic acid, 2-hydroxy-, ethyl ester, were identified for the first time in this study and were correlated with the above sensory properties.

From Fig. 4 (e), it could be seen that the ‘sweet’ aroma of sunflower seed oil mainly associated alcohols and aldehydes, including 2 aldehydes decyl aldehyde (0.219) and n-nonanal (citrus-like, soapy, 0.120) (Neugebauer et al., 2021), 2,3-butanediol (butter-like, 0.078) (Kubícková & Grosch, 1998) and 2,3-butanediol, [S-(R*,R*)]- (0.141) 2 lower alcohols, one alcohol 2-nonanol (fruity, green, 0.179) (Polster & Schieberle, 2015). Additionally, one alkane and one benzene compound (naphthalene, 1-methyl-) were positively correlated with the sweet aroma of sunflower seed oil, and detailed compound information, VIP values and correlation coefficients were shown in Table S4. Among them, naphthalene, 1-methyl- is the first compound reported in sunflower seed oil in this study.

As seen from Fig. 4 (f), the compound base of the ‘puffed food’ aroma of sunflower seed oil was more similar to that of ‘fried instant noodles’, including four aldehydes and ketones such as 2(5H)-furanone (0.038), acetoin (0.04), four pyrazines such as pyrazine, ethyl- (0.019), 2-ethyl-5- methyl pyrazine (roasty, nutty, 0.019) (Angeloni et al., 2020), four pyrrole and pyridine compounds such as pyridine (0.024), pyrrole (0.021), in addition to acetic acid (vinegar-like, 0.085) (Neugebauer et al., 2021), two alcohols such as acetone alcohol (0.085), acetamide (0.059), two esters such as acetic acid, methyl ester (0.049), and two furans such as 3(2H)-furanone, dihydro-2-methyl- (0.056). Two of these esters were detected in sunflower seed oil for the first time in this study. Most of the above compounds were nitrogenous and are mainly derived from the processing of sunflower seed oil. Among them, nine compounds, methylpyrazine, 1-(1H-pyrrol-2-yl)1-ethanone, acetyl furan, pyrazine, 2,6-dimethyl-,2-ethyl-5-methyl pyrazine, pyrazine, ethyl-,1H-pyrrole-2-carboxaldehyde, 1-methyl-2-furanmethanol, 2(5H)-furanone, were also detected in GC-O or at levels above their thresholds and were reported to be the compounds with aroma contribution in sunflower seed oil (Yin, Shi, Li, Ma, et al., 2022).

In conclusion, PLSR was successfully screened for compounds that contributed to aroma among a variety of compounds, some of which were identified in earlier research and, more importantly, some of which had not. The results above demonstrate that HS-SPME-GC-HRMS was capable of detecting a wide range of significant volatiles. 15 of the 23 compounds discovered by HRMS for the first time in sunflower seed oil appeared to contribute to the five distinctive aroma characteristics of sunflower seed oil—‘roasted sunflower seeds’, ‘sunflower seeds’, ‘burnt’, ‘sweet’ and ‘puffed food’—that could not have been identified without HRMS. However, the GC-LRMS specifically identified 13 chemicals that were important to the PLSR model. In order to explore the important fragrance components in sunflower seed oil, it is therefore required to combine these two GC–MS techniques.

Conclusions

In this study, we characterized 46 and 105 volatile compounds in seven commercial sunflower seed oils in Chinese market by using GC-Quadrupole-MS and GC-Orbitrap-MS respectively. 96 compounds including 18 alcohols, 12 esters, 7 ketones, 20 terpenoids, 11 pyrazines, 6 aldehydes, 6 furans, 6 benzene ring-containing compounds, 3 sulfides, 2 alkanes, and 5 nitrogen-containing compounds were quantified using HRMS, and 22 compounds including 5 acids, 1 amide, and 16 aldehydes were quantified using LRMS. QDA was used to characterized the aroma profile of seven sunflower seed oil samples, and the results revealed that all the studied samples had ‘roasted sunflower seeds,’ ‘sunflower seeds’, and ‘burnt’ aroma with generally high scores. A few samples were observed to contain ‘Fried instant noodles, ‘sweet’, and ‘puffed food’ aroma. Positive correlations were observed between ‘Roasted sunflower seeds’ and 1-octen-3-ol, n-heptaldehyde, and dimethyl sulfone. The primary constituents of ‘Sunflower seeds’ included a variety of aldehydes, ketones, benzenes, terpenoids, and esters. Pentanal, 3-methylbutanal, hexanal, (E)-2-hexenal, and 2-pentylfuran positively correlated with ‘fried instant noodles’ and ‘puffed food’ aroma. Pyrazines, pyrroles, pyridines, amides, aldehydes, ketones and esters might collectively contribute to the burnt flavor. On the other hand, the ‘sweet’ aroma was primarily correlated with alcohols and aldehydes. GC-Orbitrap-MS can detect additional key volatiles that may have been overlooked during GC-Quadrupole-MS detection. This study provided comprehensive information on key aroma components of sunflower seed oils and further supports the quality control and processing modification of sunflower seed oil.

CRediT authorship contribution statement

Jiani Liu: Data curation, Writing – original draft. Huimin Zhao: Formal analysis. Xiaomin Chang: Data curation. Xiaolong Li: Visualization. Yu Zhang: Writing – review & editing. Baoqing Zhu: Conceptualization, Methodology, Resources. Xiangyu Wang: Conceptualization.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2023.100690.

Contributor Information

Jiani Liu, Email: liujn@bjfu.edu.cn.

Huimin Zhao, Email: zhaohuimin@cofco.com.

Xiaomin Chang, Email: 2210677265@qq.com.

Xiaolong Li, Email: li-xiaolong@cofco.com.

Yu Zhang, Email: zhangyu2019@bjfu.edu.cn.

Baoqing Zhu, Email: zhubaoqing@bjfu.edu.cn.

Xiangyu Wang, Email: wang_xiangyu@cofco.com.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary data 1
mmc1.docx (836.3KB, docx)

Data availability

Data will be made available on request.

References

  1. Angeloni S., Scortichini S., Fiorini D., Sagratini G., Vittori S., Neiens S.D.…Caprioli G. Characterization of odor-active compounds, polyphenols, and fatty acids in coffee silverskin. Molecules. 2020;25(13):2993. doi: 10.3390/molecules25132993. https://doi.org/0.3390/molecules25132993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aydeniz-Guneser B., Yilmaz E. Sunflower oil-polyglycerol stearate oleogels: alternative deep-fat frying media for onion rings. Journal of Oleo Science. 2022;71(5):651–662. doi: 10.5650/jos.ess21446. [DOI] [PubMed] [Google Scholar]
  3. Belarbi S., Vivier M., Zaghouani W., Sloovere A., Agasse-Peulon V., Cardinael P. Comparison of new approach of GC-HRMS (Q-Orbitrap) to GC–MS/MS (triple-quadrupole) in analyzing the pesticide residues and contaminants in complex food matrices. Food Chemistry. 2021;359 doi: 10.1016/j.foodchem.2021.129932. [DOI] [PubMed] [Google Scholar]
  4. Bendini A., Barbieri S., Moradi L., Palagano R., Valli E., Toschi T.G. Quality evaluation of sunflower and hazelnut cold-pressed oils by a sensory approach. Rivista Italiana Delle Sostanze Grasse. 2014;91(2):103–115. [Google Scholar]
  5. Bi S., Niu X., Yang F., Xu Y., Dai Y., Liu Y., Zhou Q. Roasting pretreatment of walnut (Juglans regia L.) kernels: Improvement of the oil flavor profile and correlation with the chemical composition. Food & Function. 2022;13(21):10956–10969. doi: 10.1039/D2FO01990F. [DOI] [PubMed] [Google Scholar]
  6. Da Costa J.R.O., Dal Bosco S.M., Ramos R.C.S., Machado I.C.K., Garavaglia J., Villasclaras S.S. Determination of volatile compounds responsible for sensory characteristics from Brazilian extra virgin olive oil using HS-SPME/GC-MS direct method. Journal of Food Science. 2020;85(11):3764–3775. doi: 10.1111/1750-3841.15467. [DOI] [PubMed] [Google Scholar]
  7. De-la-Fuente-Blanco A., Saenz-Navajas M.P., Valentin D., Ferreira V. Fourteen ethyl esters of wine can be replaced by simpler ester vectors without compromising quality but at the expense of increasing aroma concentration. Food Chemistry. 2020;307 doi: 10.1016/j.foodchem.2019.125553. [DOI] [PubMed] [Google Scholar]
  8. Dominguez I., Arrebola F.J., Martinez Vidal J.L., Garrido Frenich A. Assessment of wastewater pollution by gas chromatography and high resolution Orbitrap mass spectrometry. Journal of Chromatography A. 2020;1619 doi: 10.1016/j.chroma.2020.460964. [DOI] [PubMed] [Google Scholar]
  9. Eliuk S., Makarov A. Evolution of Orbitrap Mass Spectrometry Instrumentation. Annual Review of Analytical Chemistry. 2015;8(1):61–80. doi: 10.1146/annurev-anchem-071114-040325. [DOI] [PubMed] [Google Scholar]
  10. Fu M., Shen X., Peng H., Zhou Q., Hou R. Identification of rancidity markers in roasted sunflower seeds produced from raw materials stored for different periods of time. LWT. 2019;118 doi: 10.1016/j.lwt.2019.108721. [DOI] [Google Scholar]
  11. Gu J., Wang Z., Kuen J., Ma L., Shahroudy A., Shuai B.…Chen T. Recent advances in convolutional neural networks. Pattern Recognition. 2018;77:354–377. doi: 10.1016/j.patcog.2017.10.013. [DOI] [Google Scholar]
  12. Guillen M.D., Cabo N., Ibargoitia M.L., Ruiz A. Study of both sunflower oil and its headspace throughout the oxidation process. Occurrence in the headspace of toxic oxygenated aldehydes. Journal of Agricultural and Food Chemistry. 2005;53(4):1093–1101. doi: 10.1021/jf0489062. [DOI] [PubMed] [Google Scholar]
  13. Hu W., Zhang L., Li P., Wang X., Zhang Q., Xu B.…Ding X. Characterization of volatile components in four vegetable oils by headspace two-dimensional comprehensive chromatography time-of-flight mass spectrometry. Talanta. 2014;129:629–635. doi: 10.1016/j.talanta.2014.06.010. [DOI] [PubMed] [Google Scholar]
  14. Ji J., Liu Y., Ma Y. Variations of polycyclic aromatic hydrocarbons in vegetable oils during seed roasting pre-treatment. Polycyclic Aromatic Compounds. 2022;42(5):2447–2460. doi: 10.1080/10406638.2020.1834414. [DOI] [Google Scholar]
  15. Jia X., Deng Q., Yang Y., Xiang X., Zhou X., Tan C.…Huang F. Unraveling of the aroma-active compounds in virgin camellia oil (Camellia oleifera Abel) using gas chromatography–mass spectrometry–olfactometry, aroma recombination, and omission studies. Journal of Agricultural and Food Chemistry. 2021;69(32):9043–9055. doi: 10.1021/acs.jafc.0c07321. [DOI] [PubMed] [Google Scholar]
  16. Kiczorowska B., Samolińska W., Andrejko D., Kiczorowski P., Antoszkiewicz Z., Zając M.…Bąkowski M. Comparative analysis of selected bioactive components (fatty acids, tocopherols, xanthophyll, lycopene, phenols) and basic nutrients in raw and thermally processed camelina, sunflower, and flax seeds (Camelina sativa L. Crantz, Helianthus L., and Linum L.) Journal of Food Science and Technology. 2019;56(9):4296–4310. doi: 10.1007/s13197-019-03899-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kubícková J., Grosch W. Evaluation of flavour compounds of Camembert cheese. International Dairy Journal. 1998;8(1):11–16. doi: 10.1016/S0958-6946(98)00015-6. [DOI] [Google Scholar]
  18. Lim V., Gorji S.G., Daygon V.D., Fitzgerald M. Untargeted and targeted metabolomic profiling of Australian indigenous fruits. Metabolites. 2020;10(3):114. doi: 10.3390/metabo10030114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lin Y., Liu Y., Liu S., Kortesniemi M., Liu J., Zhu B., Laaksonen O. Sensory and chemical characterization of Chinese bog bilberry wines using Check-all-that-apply method and GC-Quadrupole-MS and GC-Orbitrap-MS analyses. Food Research International. 2022;151 doi: 10.1016/j.foodres.2021.110809. [DOI] [PubMed] [Google Scholar]
  20. Liu R., Liu Y., Zhu Y., Kortesniemi M., Zhu B., Li H. Aromatic characteristics of passion fruit wines measured by E-Nose, GC-quadrupole MS. GC-Orbitrap-MS and Sensory Evaluation. Foods. 2022;11(23):3789. doi: 10.3390/foods11233789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Liu S., Laaksonen O., Kortesniemi M., Kalpio M., Yang B. Chemical composition of bilberry wine fermented with non-Saccharomyces yeasts (Torulaspora delbrueckii and Schizosaccharomyces pombe) and Saccharomyces cerevisiae in pure, sequential and mixed fermentations. Food Chemistry. 2018;266:262–274. doi: 10.1016/j.foodchem.2018.06.003. [DOI] [PubMed] [Google Scholar]
  22. Liu Y., Qian X., Xing J., Li N., Li J., Su Q.…Zhu B. Accurate determination of 12 lactones and 11 volatile phenols in nongrape wines through headspace-solid-phase microextraction (HS-SPME) combined with high-resolution gas chromatography-orbitrap mass spectrometry (GC-Orbitrap-MS) Journal of Agricultural and Food Chemistry. 2022;70(6):1971–1983. doi: 10.1021/acs.jafc.1c06981. [DOI] [PubMed] [Google Scholar]
  23. Mallia S., Escher F., Dubois S., Schieberle P., Schlichtherle-Cerny H. Characterization and quantification of odor-active compounds in unsaturated fatty acid/conjugated linoleic acid (UFA/CLA)-enriched butter and in conventional butter during storage and induced oxidation. Journal of Agricultural and Food Chemistry. 2009;57(16):7464–7472. doi: 10.1021/jf9002158. [DOI] [PubMed] [Google Scholar]
  24. Marasca E., Greetham D., Herring S.D., Fisk I.D. Impact of nitrogen flushing and oil choice on the progression of lipid oxidation in unwashed fried sliced potato crisps. Food Chemistry. 2016;199:81–86. doi: 10.1016/j.foodchem.2015.11.136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Marcinkowska M., Frank S., Steinhaus M., Jeleń H.H. Key odorants of raw and cooked green Kohlrabi (Brassica oleracea var. gongylodes L.) Journal of Agricultural and Food Chemistry. 2021;69(41):12270–12277. doi: 10.1021/acs.jafc.1c04339. [DOI] [PubMed] [Google Scholar]
  26. Mol H.G., Tienstra M., Zomer P. Evaluation of gas chromatography – electron ionization – full scan high resolution Orbitrap mass spectrometry for pesticide residue analysis. Analytica Chimica Acta. 2016;161–172 doi: 10.1016/j.aca.2016.06.017. [DOI] [PubMed] [Google Scholar]
  27. Neugebauer A., Granvogl M., Schieberle P. Characterization of the key odorants in high-quality extra virgin olive oils and certified off-flavor oils to elucidate aroma compounds causing a rancid off-flavor. Journal of Agricultural and Food Chemistry. 2020;68(21):5927–5937. doi: 10.1021/acs.jafc.0c01674. [DOI] [PubMed] [Google Scholar]
  28. Neugebauer A., Schieberle P., Granvogl M. Characterization of the key odorants causing the musty and fusty/muddy sediment off-flavors in olive oils. Journal of Agricultural and Food Chemistry. 2021;69(49):14878–14892. doi: 10.1021/acs.jafc.1c02228. [DOI] [PubMed] [Google Scholar]
  29. Ni R., Wang P., Zhan P., Tian H., Li T. Effects of different frying temperatures on the aroma profiles of fried mountain pepper (Litsea cubeba (Lour.) Pers.) oils and characterization of their key odorants. Food Chemistry. 2021;357 doi: 10.1016/j.foodchem.2021.129786. [DOI] [PubMed] [Google Scholar]
  30. Ni R., Yan H., Tian H., Zhan P., Zhang Y. Characterization of key odorants in fried red and green huajiao (Zanthoxylum bungeanum maxim. and Zanthoxylum schinifolium sieb. et Zucc.) oils. Food Chemistry. 2022;377 doi: 10.1016/j.foodchem.2021.131984. [DOI] [PubMed] [Google Scholar]
  31. Nieva-Echevarria B., Goicoechea E., Guillen M.D. Effect of adding alpha-tocopherol on the oxidation advance during in vitro gastrointestinal digestion of sunflower and flaxseed oils. Food Research International. 2019;125 doi: 10.1016/j.foodres.2019.108558. [DOI] [PubMed] [Google Scholar]
  32. Nounah I., Chbani M., Matthäus B., Charrouf Z., Hajib A., Willenberg I. Profile of volatile aroma-active compounds of cactus seed oil (opuntia ficus-indica) from different locations in morocco and their fate during seed roasting. Foods. 2020;9(9):1280. doi: 10.3390/foods9091280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Özcan M.M., Köse N. Monitoring of changes in physico-chemical properties, fatty acids and phenolic compounds of unroasted and roasted sunflower oils obtained by enzyme and ultrasonic extraction systems. Journal of Food Measurement and Characterization. 2022 doi: 10.1007/s11694-022-01626-5. [DOI] [Google Scholar]
  34. Petersen K.D., Kleeberg K.K., Jahreis G., Fritsche J. Assessment of the oxidative stability of conventional and high-oleic sunflower oil by means of solid-phase microextraction-gas chromatography. International Journal of Food Sciences & Nutrition. 2012;63(2):160–169. doi: 10.3109/09637486.2011.609158. [DOI] [PubMed] [Google Scholar]
  35. Peterson A.C., McAlister G.C., Quarmby S.T., Griep-Raming J., Coon J.J. Development and characterization of a GC-enabled QLT-Orbitrap for high-resolution and high-mass accuracy GC/MS. Analytical Chemistry. 2010;82(20):8618. doi: 10.1021/ac101757m. [DOI] [PubMed] [Google Scholar]
  36. Pilorgé E. Sunflower in the global vegetable oil system: Situation, specificities and perspectives. Oilseeds and fats, Crops and Lipids. 2020;27:11. doi: 10.1051/ocl/2020028. [DOI] [Google Scholar]
  37. Poehlmann S., Schieberle P. Characterization of the aroma signature of styrian pumpkin seed oil (Cucurbita pepo subsp. pepo var. Styriaca) by molecular sensory science. Journal of Agricultural and Food Chemistry. 2013;61(12):2933–2942. doi: 10.1021/jf400314j. [DOI] [PubMed] [Google Scholar]
  38. Pollner G, Schieberle P. J Agric Food Chem. Characterization of the key odorants in commercial cold-pressed oils from unpeeled and peeled rapeseeds by the sensomics approac. 2016;64:627–630. doi: 10.1021/acs.jafc.5b05321. [DOI] [PubMed] [Google Scholar]
  39. Polster J., Schieberle P. Structure–odor correlations in Homologous series of alkanethiols and attempts to predict odor thresholds by 3D-QSAR studies. Journal of Agricultural and Food Chemistry. 2015;63(5):1419–1432. doi: 10.1021/jf506135c. [DOI] [PubMed] [Google Scholar]
  40. Qian X., Lan Y., Han S., Liang N., Zhu B., Shi Y., Duan C. Comprehensive investigation of lactones and furanones in icewines and dry wines using gas chromatography-triple quadrupole mass spectrometry. Food Research International. 2020;137 doi: 10.1016/j.foodres.2020.109650. [DOI] [PubMed] [Google Scholar]
  41. Steinhaus M., Schieberle P. Characterization of odorants causing an atypical aroma in white pepper powder (Piper nigrum L.) based on quantitative measurements and orthonasal breakthrough thresholds. Journal of Agricultural and Food chemistry. 2005;53(15):6049–6055. doi: 10.1021/jf0506030. [DOI] [PubMed] [Google Scholar]
  42. Tian P., Zhan P., Tian H., Wang P., Lu C., Zhao Y. Effects of different vegetable oils on the aroma characteristics of deep-fried shallot flavoring evaluated by HS-SPME/GC-MS coupled with PLSR. Journal of Food Processing and Preservation. 2020;44(9):14698. doi: 10.1111/jfpp.14698. [DOI] [Google Scholar]
  43. USDA. (2021). Oil,sunflower seed, world production. https://apps.fas.usda.gov/psdonline/app/index.html#/app/compositeViz.
  44. Valdés García A., Beltrán Sanahuja A., Karabagias I.K., Badeka A., Kontominas M.G., Garrigós M.C. Effect of frying and roasting processes on the oxidative stability of sunflower Seeds (Helianthus annuus) under normal and accelerated storage conditions. Foods. 2021;10(5):944. doi: 10.3390/foods10050944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Wang S., Li Y., Ren J., Gu P., Li T., Wu Y.…Zhu B. Comparison on evolution of volatile compounds and aroma attributes in different pH-adjusted fermented bog bilberry syrup wines during bottle-aging period. Food Bioscience. 2018;22:121–128. doi: 10.1016/j.fbio.2018.01.003. [DOI] [Google Scholar]
  46. Wu X., Gao S., Niu Y., Zhao Z., Ma R., Xu B.…Zhang Y. Quantitative analysis of blended corn-olive oil based on Raman spectroscopy and one-dimensional convolutional neural network. Food Chemistry. 2022;385 doi: 10.1016/j.foodchem.2022.132655. [DOI] [PubMed] [Google Scholar]
  47. Xiao Z., Luo J., Niu Y., Wang P., Wang R., Sun X. Olfactory impact of esters on rose essential oil floral alcohol aroma expression in model solution. Food Research International. 2019;116:211–222. doi: 10.1016/j.foodres.2018.08.015. [DOI] [PubMed] [Google Scholar]
  48. Xu L., Mei X., Wu G., Karrar E., Jin Q., Wang X. Inhibitory effect of antioxidants on key off-odors in French fries and oils and prolong the optimum frying stage. LWT. 2022;162 doi: 10.1016/j.lwt.2022.113417. [DOI] [Google Scholar]
  49. Xu L., Wang S., Tian A., Liu T., Benjakul S., Xiao G.…Ma L. Characteristic volatile compounds, fatty acids and minor bioactive components in oils from green plum seed by HS-GC-IMS, GC–MS and HPLC. Food Chemistry: X. 2023;17 doi: 10.1016/j.fochx.2022.100530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Xu L., Wu G., Huang J., Zhang H., Jin Q., Wang X. Sensory-directed flavor analysis of key odorants compounds development of French fries and oils in the break-in, optimum and degrading frying stage. Food Science and Human Wellness. 2023;12:140–150. doi: 10.1016/j.fshw.2022.07.032. [DOI] [Google Scholar]
  51. Xu L., Zhu C., Liu T., Karrar E., Ouyang Y., Li D. Effect of microwave heating on lipid composition, chemical properties and antioxidant activity of oils from Trichosanthes kirilowii seed. Food Research International. 2022;159 doi: 10.1016/j.foodres.2022.111643. [DOI] [PubMed] [Google Scholar]
  52. Yang K., Chao L., Wu C., Ye Z., Chen H. Headspace solid-phase microextraction analysis of volatile components in peanut oil. Molecules. 2021;26(11):3306. doi: 10.3390/molecules26113306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Yang Y., Yu P., Sun J., Jia Y., Wan C., Zhou Q., Huang F. Investigation of volatile thiol contributions to rapeseed oil by odor active value measurement and perceptual interactions. Food Chemistry. 2022;373 doi: 10.1016/j.foodchem.2021.131607. [DOI] [PubMed] [Google Scholar]
  54. Yang Y., Zhang M., Hua J., Deng Y., Jiang Y., Li J.…Dong C. Quantitation of pyrazines in roasted green tea by infrared-assisted extraction coupled to headspace solid-phase microextraction in combination with GC-QqQ-MS/MS. Food Research International. 2020;134 doi: 10.1016/j.foodres.2020.109167. [DOI] [PubMed] [Google Scholar]
  55. Yin W., Ma X., Li S., Wang X., Liu H., Shi R. Comparison of key aroma-active compounds between roasted and cold-pressed sesame oils. Food Research International. 2021;150 doi: 10.1016/j.foodres.2021.110794. [DOI] [PubMed] [Google Scholar]
  56. Yin W., Shi R., Li K., Wang X., Wang A., Zhao Y., Zhai Z. Effect of microwave pretreatment of sunflower kernels on the aroma-active composition, sensory quality, lipid oxidation, tocopherols, heterocyclic amines and polycyclic aromatic hydrocarbons of sunflower oil. LWT. 2022;170 doi: 10.1016/j.lwt.2022.114077. [DOI] [Google Scholar]
  57. Yin W., Shi R., Li S., Ma X., Wang X., Wang A. Changes in key aroma-active compounds and sensory characteristics of sunflower oils induced by seed roasting. Journal of Food Science. 2022;87(2):699–713. doi: 10.1111/1750-3841.16044. [DOI] [PubMed] [Google Scholar]
  58. Yu P., Yang Y., Sun J., Jia X., Zheng C., Zhou Q., Huang F. Identification of volatile sulfur-containing compounds and the precursor of dimethyl sulfide in cold-pressed rapeseed oil by GC–SCD and UPLC–MS/MS. Food Chemistry. 2022;367 doi: 10.1016/j.foodchem.2021.130741. [DOI] [PubMed] [Google Scholar]
  59. Zhai X., Granvogl M. Elucidation of the impact of different drying methods on the key odorants of Toona sinensis (A. Juss.) Roem. Using the sensomics approach. Journal of Agricultural and Food Chemistry. 2020;68(29):7697–7709. doi: 10.1021/acs.jafc.0c02144. [DOI] [PubMed] [Google Scholar]
  60. Zhang C.-X., Xi J., Zhao T.-P., Ma Y.-X., Wang X.-D. β-carbolines norharman and harman in vegetable oils in China. Food Additives & Contaminants: Part B. 2020;13(3):193–199. doi: 10.1080/19393210.2020.1759701. [DOI] [PubMed] [Google Scholar]
  61. Zhang C., Wang Y., Ding D., Su J., Zhao Z. Volatile profiles of Allium tenuissimum L. flower fried by four different oils, using SPME-GC-MS, and sensory evaluation coupled with partial least squares regression. Journal of Food Composition and Analysis. 2022;109 doi: 10.1016/j.jfca.2022.104461. [DOI] [Google Scholar]
  62. Zhang J., Li L., Gao N., Wang D., Gao Q., Jiang S. Feature extraction and selection from volatile compounds for analytical classification of Chinese red wines from different varieties. Analytica Chimica Acta. 2010;662(2):137–142. doi: 10.1016/j.aca.2009.12.043. [DOI] [PubMed] [Google Scholar]
  63. Zhang W., Cao X., Liu S. Aroma modulation of vegetable oils-A review. Critical Reviews in Food Science and Nutrition. 2020;60(9):1538–1551. doi: 10.1080/10408398.2019.1579703. [DOI] [PubMed] [Google Scholar]
  64. Zhang Y., Li X., Lu X., Sun H., Wang F. Effect of oilseed roasting on the quality, flavor and safety of oil: A comprehensive review. Food Research International. 2021;150 doi: 10.1016/j.foodres.2021.110791. [DOI] [PubMed] [Google Scholar]
  65. Zhou Q., Jia X., Yao Y.-Z., Wang B., Wei C.-Q., Zhang M., Huang F. Characterization of the aroma-active compounds in commercial fragrant rapeseed oils via monolithic material sorptive extraction. Journal of Agricultural and Food Chemistry. 2019;67(41):11454–11463. doi: 10.1021/acs.jafc.9b05691. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary data 1
mmc1.docx (836.3KB, docx)

Data Availability Statement

Data will be made available on request.

Articles from Food Chemistry: X are provided here courtesy of Elsevier

RESOURCES