Dynamic monitoring and quality evaluation method for flavor substances in processing process of braised pork

Abstract

The invention belongs to the technical field of food flavor analysis and quality control, and particularly relates to a dynamic monitoring and quality evaluation method for flavor substances in the processing process of braised meat. According to the method disclosed by the invention, various detection means such as an electronic tongue, HS-SPME-GC-MS and HS-GC-IMS are adopted to detect and identify flavor components in different braising stages of the braised pork, so that the comprehensiveness and accuracy of flavor detection in the braised pork processing process are remarkably improved; besides, by improving the detection key conditions of the HS-GC-IMS, especially by adopting the treatment of balancing for 30 minutes at the temperature of 80 DEG C in the headspace equilibrium stage, the low-boiling-point volatile matters in the meat sample can be fully released, the desorption efficiency of small molecules is improved, the volatilization loss or thermal decomposition cannot be caused, and the detection sensitivity in the early stage is further improved.

Description

Technical Field

[0001] This invention belongs to the field of food flavor analysis and quality control technology, and specifically relates to a method for dynamic monitoring and quality evaluation of flavor substances during the processing of braised pork belly. Background Technology

[0002] Braised pork belly is a representative dish of traditional Shandong cuisine, with a long history and unique flavor. It is made primarily with pork belly, braised with various seasonings, resulting in tender, flavorful meat that is not greasy, possessing a savory and complex aroma that is highly popular among consumers. Despite the strong market demand for braised pork belly, related research is still relatively limited, and systematic studies on the formation and dynamic changes of its flavor compounds are even rarer. This, to some extent, restricts the standardization and industrialization of the quality of braised pork belly.

[0003] Flavor is a key factor determining the quality and consumer acceptance of meat products. Flavor originates not only from the thermal degradation of fat, the Maillard reaction of amino acids, and the migration and transformation of seasonings, but is also significantly influenced by processing techniques (temperature, time, etc.). As a typical braised meat product, braised pork belly undergoes a complex process of generation, transformation, and release of flavor compounds during prolonged stewing, resulting in significant differences in flavor expression at different time points.

[0004] Although existing technologies such as GC-MS, GC-IMS, and electronic tongue have been applied in meat product flavor research, most of them are single-method applications, which are difficult to fully reveal the dynamic changes of flavor and the interaction between flavor and taste. In particular, there are no relevant research reports on the traditional Shandong dish "Braised Pork Belly".

[0005] Therefore, there is an urgent need for a new methodology that can dynamically evaluate the formation and changes of flavor compounds in braised pork in multiple dimensions during processing, and construct a correlation model between flavor and taste, so as to provide a scientific basis for the optimization of traditional braised meat products and their industrialization. Summary of the Invention

[0006] To address the above technical problems, this invention proposes a method for dynamic monitoring and quality evaluation of flavor substances during the processing of braised pork belly.

[0007] This invention samples braised pork belly at different time points during the stewing process, then uses an electronic tongue to acquire aroma and taste data of the pork belly samples. Headspace solid-phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS) and gas chromatography-ion mobility spectrometry (HS-GC-IMS) are then used to detect volatile compounds in the pork belly samples. Principal component analysis (PCA), aroma activity value (OAV) analysis, thermogram analysis, and correlation analysis are then performed based on the combined detection data to evaluate the flavor of braised pork belly during the stewing process. The HS-GC-IMS conditions were as follows: the sample was equilibrated at 80°C for 30 min, the headspace injection volume was 200 μL, the inlet temperature was 80°C, an MXT-5 column was used, the column temperature was 40°C, nitrogen was used as both the carrier gas and the drift gas, the carrier gas flow rate was initially 2 mL / min, held for 2 min, then linearly increased to 15 mL / min over 8 min, further increased to 100 mL / min over the next 10 min, and then increased to 150 mL / min over another 10 min, for a total run of 30 min, the drift tube was at 45°C, and the drift gas flow rate was 150 mL / min.

[0008] Preferably, the conditions for the HS-SPME-GC-MS are as follows: The pork belly sample was mixed with 10 μL of 2 mg / L 2-methyl-3-heptanone and equilibrated in a 20 mL vial at 250°C for 30 min in the headspace. The pretreated solid-phase microextraction head CAR / PDMS / DVB fiber was exposed to the headspace under helium, and then thermal desorption was performed at 240°C at the GC inlet for 2 min. Chromatographic separation was performed using a DB-WAX capillary column (60 m × 250 μm × 0.25 μm, Agilent J&W).

[0009] Preferably, the MS parameters include 70 eV electron ionization, the ion source and the transmission line being maintained at 230°C and 250°C respectively, and signal acquisition using a full scan mode with a scan range of 35~500 m / z.

[0010] The present invention has the following advantages and effects compared with the prior art: This invention employs rapid GC-IMS detection, precise GC-MS qualitative analysis, and OAV evaluation, further combined with electronic tongue analysis of flavor characteristics. Ultimately, it identifies key flavor markers and reveals the optimal preparation process time, significantly improving the comprehensiveness and accuracy of flavor detection during the processing of braised pork. In addition, by improving the key detection conditions of HS-GC-IMS, especially by using a temperature equilibration process of 80°C for 30 minutes during the headspace equilibration stage, the low-boiling-point volatiles in the braised pork sample are fully released, improving the desorption efficiency of small molecules without causing volatilization loss or thermal decomposition, thereby further improving the detection sensitivity in the early stages. Attached Figure Description

[0011] Figure 1 The curves showing the changes in the intensity of sour, umami, salty, and bitter flavors during the stewing process of braised pork belly; Figure 2VOCs classification map (A), PCA score map (B), VOCs relative abundance (C), and odor radar map based on volatile compounds with OAV>1 (D); Figure 3 The heatmaps are clustering (A), correlation (B), PLS-DA scoring (C), and VIP scoring (D) based on 37 key VOCs. Figure 4 Cluster association heatmap for 37 key VOCs and flavor characteristics; Figure 5 The peak number and volume change of volatile organic compounds in the meat during the stewing process are shown in Figure (A), the two-dimensional graph of aromatic compounds in the seven stewing stages is shown in Figure (B), and the fingerprint spectrum is shown in Figure (C). Figure 6 Principal component analysis diagram of braised pork belly under different stewing times. Detailed Implementation

[0012] To enable those skilled in the art to better understand the present invention, the present invention will now be further described in conjunction with specific embodiments.

[0013] Example 1 1.1 Materials and Sample Preparation The pork belly meat is sourced from Jinan Weierkang Industrial Group Co., Ltd., and is transported under 4°C refrigeration conditions. It undergoes rigorous water washing to remove surface blood residue.

[0014] Pork belly meat was cut into rectangular pieces measuring 15 cm × 4 cm × 1 cm (length × width × thickness) and placed in a stainless steel steaming container. Each batch consisted of 500 g pieces. The stewing medium was 2.5 L of deionized water, with the addition of 2.5 g bay leaves, 2.5 g cinnamon, 5 g star anise, 115 g fermented soy sauce, 115 g Shaoxing wine, and 10 g dark soy sauce. The meat was heat-treated using an induction cooker with programmed parameters: stewed at 2200 W for 20 min, then at 400 W for 100 min to prepare braised pork belly. Systematic samples were taken at 0, 20, 40, 60, 80, 100, and 120 min for subsequent analysis.

[0015] 1.2 Electronic Tongue Analysis 30 g of pork belly samples were chopped and dissolved in 250 mL of deionized water for different stewing times. The solutions were homogenized at 10000 rpm for 10 min. The resulting suspensions were centrifuged at 6000 rpm for 10 min at 4°C. The supernatant was collected as the test solution and analyzed using a taste sensor system (SA402B electronic tongue). Data acquisition time was 30 s, sampling interval was 0.5 s, stabilization delay was 20 s, and washing duration was 120 s. Two copies of each sample were prepared, and each measurement was repeated five times. The experimental results of changes in the taste characteristics of pork belly during stewing are shown in Table 1. The changes of some relevant indicators over time are shown in [Table 1]. Figure 1 .

[0016] Table 1. Changes in flavor characteristics of braised pork during stewing. Time / min acidity bitterness astringency Umami 0 <![CDATA[-7.72±0.69 a ]]> <![CDATA[-4.75±0.15 e ]]> <![CDATA[-13.29±0.38 ab ]]> <![CDATA[3.8±0.25 d ]]> 20 <![CDATA[-14.58±0.44 cd ]]> <![CDATA[-4.31±0.21 cd ]]> <![CDATA[-13.71±0.32 b ]]> <![CDATA[8.51±0.43 abc ]]> 40 <![CDATA[-15.24±0.35 d ]]> <![CDATA[-3.79±0.06 a ]]> <![CDATA[-13.41±0.23 ab ]]> <![CDATA[8.86±0.4 ab ]]> 60 <![CDATA[-15.12±0.5 d ]]> <![CDATA[-3.87±0.07 ab ]]> <![CDATA[-13.23±0.36 ab ]]> <![CDATA[9.02±0.34 a ]]> 80 <![CDATA[-14.4±0.45 bcd ]]> <![CDATA[-4.51±0.07 d ]]> <![CDATA[-13.53±0.47 ab ]]> <![CDATA[8.38±0.52 abc ]]> 100 <![CDATA[-14.02±0.61 bc ]]> <![CDATA[-3.97±0.1 ab ]]> <![CDATA[-12.91±0.54 ab ]]> <![CDATA[8.06±0.63 bc ]]> 120 <![CDATA[-13.53±0.66 b ]]> <![CDATA[-4.1±0.14 bc ]]> <![CDATA[-12.84±0.63 a ]]> <![CDATA[7.66±0.74 c ]]> Time / min Richness salinity Aftertaste - B Aftertaste - A 0 <![CDATA[-48.38±2.08 b ]]> <![CDATA[6.95±0.54 c ]]> <![CDATA[-35.73±0.67 a ]]> <![CDATA[-18.42±0.32 b ]]> 20 <![CDATA[-46.87±0.67 ab ]]> <![CDATA[16.23±0.4 a ]]> <![CDATA[-35.43±1.32 a ]]> <![CDATA[-17.79±0.91 ab ]]> 40 <![CDATA[-46.51±2.2 ab ]]> <![CDATA[16.06±0.5 a ]]> <![CDATA[-35.14±0.59 a ]]> <![CDATA[-17.74±0.79 ab ]]> 60 <![CDATA[-47.37±1.57 ab ]]> <![CDATA[15.97±0.84 a ]]> <![CDATA[-35.09±0.52 a ]]> <![CDATA[-17.95±0.93 ab ]]> 80 <![CDATA[-47.15±0.83 ab ]]> <![CDATA[15.19±0.87 ab ]]> <![CDATA[-35.16±0.52 a ]]> <![CDATA[-18.15±0.88 ab ]]> 100 <![CDATA[-46.22±1.31 ab ]]> <![CDATA[14.74±1.07 ab ]]> <![CDATA[-34.87±0.68 a ]]> <![CDATA[-17.63±0.25 ab ]]> 120 <![CDATA[-44.9±2.17 a ]]> <![CDATA[14.07±1.18 b ]]> <![CDATA[-34.82±0.53 a ]]> <![CDATA[-16.97±0.41 a ]]> Note: After processing with the three Sigma rule, the results are expressed as mean ± standard deviation. There are significant differences in lowercase letters in the same column (p<0.05).

[0017] Table 1 and Figure 1 The results showed that the umami value of raw braised pork belly was 3.80, increasing by 123.95% during the 0-20 min stewing period. This was attributed to the synergistic effect of glutamate / aspartic acid released through protein hydrolysis and ATP-derived guanine nucleotide / adenosine / guanylic acid (IMP / AMP / GMP), reaching a peak within 60 min, and then decreasing due to prolonged heating time leading to IMP deamination. Compared with the 60 min samples, the umami value of the 100 min and 120 min samples was significantly reduced, indicating that excessively prolonged stewing time is detrimental to the umami characteristics of braised pork belly. Saltiness showed a similar trend to umami, indicating a synergistic effect between umami and saltiness.

[0018] Furthermore, the acidity change in braised pork belly showed an opposite trend to the umami / salty taste, indicating an antagonistic interaction between these taste patterns. After stewing for 20 minutes, the acidity of the braised pork belly decreased by 88.86% for the first time. Figure 1 The reasons for this may be the thermal decomposition of lactic acid in pork muscles, and the esterification of saturated fatty acids released from hydrolysis of adipose tissue with alcohols under thermal conditions. Additionally, during stewing, the acidity of the pork initially decreased and then increased, reaching its highest levels at 120 minutes of stewing, indicating that over-stewing actually enhances these sensory characteristics. Furthermore, the results show that compared to raw meat, stewed pork exhibits a significantly enhanced bitterness, peaking at 40 minutes with a 20% increase in intensity. This is attributed to hydrophobic peptides produced during protein hydrolysis during thermal processing, particularly those containing leucine, isoleucine, and phenylalanine residues, which have been identified as natural bitter compounds.

[0019] 1.2 HS-SPME-GC-MS Detection The volatile organic compounds (VOCs) in the sample were identified using a headspace-solid phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS) system, as follows: Two g of pork belly sample was mixed with 10 μL of 2 mg / L 2-methyl-3-heptanone (internal standard) and equilibrated in a 20 mL vial at 250°C for 30 min vial. A pretreated solid-phase microextraction head, made of divinylbenzene / carboxylic acid / polydimethylsiloxane (CAR / PDMS / DVB) fiber (50 / 30 μm, Supelco), was exposed to headspace under helium (flow rate 0.8 mL / min) and then thermally desorbed at 240°C for 2 min vial at the GC inlet. Chromatographic separation was performed using a DB-WAX capillary column (60 m × 250 μm × 0.25 μm, Agilent J&W). MS parameters included 70 eV electron ionization, and the ion source and transfer line maintained at 230°C and 250°C, respectively. Signal acquisition was performed in full scan mode, with a scan range of 35–500 m / z. MS results were compared with the National Institute of Standards and Technology (NIST) database for MS identification of compounds. Quantitative analysis was performed using the internal standard method, with the content of flavor compounds determined relative to the internal standard. This calculation was based on the known concentration and peak area of ​​the internal standard 2-methyl-3-heptanone. Five replicates were performed for each sample group, and three were selected for analysis.

[0020] (1) Composition analysis of PCA and VOCs A total of 85 VOCs were detected in all braised pork samples, mainly including hydrocarbons (19), alcohols (17), esters (10), aldehydes (9), terpenes (8), ethers (5), ketones (4), heterocyclic compounds (3), organosulfur compounds (3), and additional compounds (7). The distribution of VOCs in all samples is shown in the figure. Figure 2 As shown in Figure A, hydrocarbons account for 22.09% of the total VOCs, followed by alcohols (19.77%) and esters (11.63%), which together account for 53.49% of the total VOCs.

[0021] like Figure 2As shown in Figure B, the principal component analysis (PCA) model reveals that the three principal components cumulatively account for 78.2% of the variance. Tight clustering was observed among samples with similar processing times. Samples stewed for 20 min and 40 min showed greater spatial proximity in the score plot, while samples stewed for 60 min and 80 min showed even greater spatial proximity. This indicates that the VOCs of the pork belly gradually change during stewing. Samples stewed for 20 min and 40 min, and samples stewed for 60 min and 80 min, have similar volatility characteristics. The 0 min and 120 min samples showed significant separation from other samples in terms of VOCs, indicating that stewing and prolonged stewing significantly affect the VOCs profile of pork.

[0022] Figure 2 Figure C shows the relative abundance of VOCs over time during the stewing period, depicting the changes in the relative abundance of VOC categories during each sampling interval. In terms of VOC composition, the samples at 0 min showed significant differences compared to other samples. Samples at 20 min and 40 min showed close aggregation, while samples at 60 min and 80 min formed another distinct component. Samples at 100 min and 120 min exhibited overlapping distributions. This hierarchical similarity pattern is consistent with the PCA analysis results, strengthening the validation of the consistency of component changes in the analytical method. Figure C shows that VOCs in raw meat are mainly esters (22.14%), organosulfur compounds (20.3%), and alcohols (14.64%), with phenols (5.47%) and hydrocarbons (0.87%) contributing less. Compared to stewed meat samples, the proportions of esters, organosulfur compounds, and phenols in raw meat were significantly higher than in stewed meat (detected only in the raw meat state), confirming that heat treatment reduced these compound categories in stewed meat. Aldehydes, terpenes, and heterocyclic compounds were not detected in raw meat but were found only in stewed meat samples. Therefore, phenols, aldehydes, terpenes, and heterocyclic compounds can serve as diagnostic markers for differentiating between raw and stewed pork belly.

[0023] Furthermore, after stewing for 20 and 40 minutes, the proportions of hydrocarbons, alcohols, aldehydes, and ethers in the braised pork belly increased significantly, with alcohols and hydrocarbons becoming the dominant complex groups. This increase was mainly attributed to lipid oxidation; conversely, the proportions of organosulfur compounds and esters decreased. The reduction in organosulfur compounds may stem from the instability of thiol groups, which makes thiol compounds prone to polymerization or oxidation under prolonged heating. Similarly, esters are easily hydrolyzed at high temperatures, forming alcohols. These compound classes are considered to have high odor thresholds and contribute slightly to aroma enhancement. After stewing for 60 minutes, the aldehyde content in the braised pork belly was the highest (accounting for 30.24% of total VOCs), while the ethanol content decreased to 17.42%. At 80 minutes, aldehydes dominated (26.84%), consistent with previous findings. After stewing for 120 minutes, aldehydes decreased by 6.33%, while ketones and ethers increased, with the latter reaching 12.82%. Furthermore, starting from 20 minutes of stewing, the hydrocarbon content remained within a narrow range (21.62%–25.99%), with minimal changes observed throughout the stewing process. This stability is attributed to the inherent chemical inertness of hydrocarbons, which exhibit low reactivity under prolonged heating conditions.

[0024] (2) Key VOCs Analysis Table 2 records the specific chemical composition, concentration, threshold, and odor of VOCs detected during the stewing process of braised pork belly using HS-SPME-GC-MS.

[0025] (3) Aroma profile analysis Aroma activity value (OAV) is used to assess the contribution of VOCs and is defined as the ratio between the compound concentration and its sensory threshold. Based on established odor impact criteria, VOCs with an OAV > 1 are identified as aroma active components. Comprehensive VOC characterization was achieved by integrating HS-SPME-GC-MS and OAV methods, enabling systematic quantification of VOCs and identification of key VOCs.

[0026] Aldehydes are considered important fatty acid oxidation products, significantly influencing meat aroma due to their fatty odor and low sensory threshold. Nine aldehyde compounds were detected during the stewing process of braised pork belly, with an OAV > 1, primarily formed via fatty acid oxidation and supplemented by an early Maillard reaction. The highest concentrations of butyraldehyde, pentanal, hexanal, heptanal, benzaldehyde, and 4-methylbenzaldehyde were observed at 60 min of stewing, while nonanal concentration increased at 80 min. Five fatty aldehydes (butyraldehyde, pentanal, hexanal, heptanal, and nonanal) were identified as oxidation products of linoleic and oleic acids, contributing fatty, herbaceous, and fruity aromas. These aldehyde compounds are considered major contributors to meat aroma due to their low odor threshold. Hexanal was identified as the major aldehyde, reaching a concentration of 128948.92 μg / kg at 80 min with an OAV of approximately 17193, thus confirming its status as the major aldehyde. Hexanal possesses fatty and herbaceous aroma characteristics and was identified as a major contributor to fatty aroma. Pentanal is considered the second dominant aldehyde, contributing to the nutty and fruity aromatic properties. Both aldehydes are mechanistically linked to the oxidation of n-6 polyunsaturated fatty acids. At 80 min of stewing, the highest concentration of fatty aldehydes was 137,716.83 μg / kg, corresponding to the peak intensity of the fatty flavor. Benzaldehyde was shown to be derived from both fatty acid oxidation and Strecker degradation of phenylalanine / tyrosine, contributing to almond and cherry aromas and enhancing aroma complexity. Aldehyde depletion after 80 min was attributed to the thermal degradation pathway (oxidation / polymerization) under prolonged heating, but an increase in 3-methylbutyraldehyde levels was observed at this stage. This was attributed to the hydrolysis of internal adipose tissue or intramuscular phospholipids induced by prolonged stewing, releasing additional fatty acid precursors oxidized via thermal pathways, thus contributing nutty and cheese aromas to the braised pork.

[0027] Alcohols were identified as the most abundant VOCs during stewing, contributing subtle aromas to braised pork. These compounds primarily originate from lipid oxidation pathways and typically exhibit high odor thresholds, resulting in limited sensory impact recorded in meat aroma studies. For example, 3-methyl-1-butanol (2687.34 μg / kg) had a negligible olfactory effect despite its high concentration due to its extremely high detection threshold (30000 μg / kg). 1-Octen-3-ol, characterized by a mushroom aroma, is considered a double-edged aromatic compound—positively contributing to the aroma of cooked meat while potentially enhancing blood / metallic off-flavors. 1-Hexanol was identified as the major contributor to the grassy aroma. Based on lipid peroxidation kinetics, 1-octen-3-ol and 1-hexanol reached their peak values ​​(OAVs of 916.69 and 86.53, respectively) within the first 20 minutes of thermal degradation. Volatile alcohols such as 1-octen-3-ol, eucalyptol, 1-pentanol, α-terpineol, and phenylethyl alcohol showed a two-stage increase during stewing (initially increasing, then decreasing, and then increasing again). This may be because in the early stage of stewing (0-40 min), the oxidation of unsaturated fatty acids and ester hydrolysis are the main causes of alcohol accumulation; subsequently, alcohol consumption is related to their oxidation into aldehydes. For example, during stewing for 40-60 min, 1-pentanol (6447.78→2842.12 μg / kg, P<0.05) is reduced by oxidation to pentanal (0→7137.2 μg / kg, P<0.05), leading to an increase in its content; after prolonged heating (60-120 min), the oxidation of adipose tissue releases deeper lipid precursors, resulting in a further increase in alcohols.

[0028] Esters are formed through acid-alcohol esterification reactions and possess a sweet and fruity flavor. Ten esters were identified during stewing, most of which (including methyl 2-methylbutyrate, methyl isovalerate, ethyl acetate, methyl acetate, ethyl butyrate, and ethyl propionate) were undetectable at 60 and 80 minutes, although levels increased in the early and late stages. This deficiency is attributed to the thermal hydrolysis of esters to alcohols under prolonged high-temperature conditions. Figure 2 The negative concentration trend between esters and alcohols shown in C confirms this mechanism. In the later stages (100–120 min), subsequent ester regeneration is attributed to secondary esterification between oxidatively derived fatty acids (from deep lipid pools) and alcohols. Among the detected esters, ethyl acetate exhibited the highest odor contribution due to its highest concentration and low sensory threshold (Table 2). This pineapple aromatic compound, derived from soy sauce and rice wine additives used in the stewing process, was identified as a key flavor enhancer.

[0029] Ketones are primarily produced through the Maillard reaction, amino acid degradation, and thermal fat breakdown, contributing to chocolate and fruity aromas. Four ketone compounds were identified from braised pork: 2,5-dimethyl-3-hexanone, 2-heptanone, 2,3-octanedione, and 2,3-butanedione. The dicarbonyl ketones (2,3-octanedione and 2,3-butanedione) are derived from the cleavage of Amadori compounds in the intermediate Maillard stage, subsequently participating in Strecker degradation to form pyrazines and other heterocyclic flavor compounds. 2,3-Butanedione is a major component of pork volatiles, possessing a distinctive buttery and fruity aroma.

[0030] Hydrocarbons primarily originate from the oxidative cracking of saturated fatty acids. From 0 to 40 min, low-boiling-point alkanes (heptane: 12076.66 μg / kg; octane: 31902.01 μg / kg) accumulate rapidly, followed by accelerated volatilization with increasing processing time. Terpenes with an OAV > 1 were identified as important components derived from flavor blends, possessing a unique pungent aroma and complex flavor.

[0031] Furan derivatives are considered hallmark products of the terminal Maillard reaction stage, and their presence can serve as a chemical marker of the progression of higher-order reactions. However, HS-SPME-GC-MS analysis detected only two furan derivatives, 2-pentylfuran and 2-furanmethanol, suggesting that the Maillard reaction may not be the primary pathway in the braised pork stew system. 2-Pentylfuran, formed via the oxidation of linoleic acid (C18:2 / C18:3) and the later Maillard reaction, is considered a key contributor to the aroma of roasted meat. 2-Furanmethanol, characterized by combustion and a slight odor, was undetectable at 80 min but reached 1035.38 μg / kg at 100 min, indicating its exclusive formation within the 80–100 min time interval. The concentration of sulfides in raw meat was significantly higher, including methanethiol, dimethyl trisulfide, and dimethyl disulfide, which may originate from microbial metabolic activity. Methanethiol was shown to undergo an oxidation pathway, producing dimethyl trisulfide and dimethyl disulfide as secondary metabolites. These sulfides are reportedly closely linked to the formation of off-flavors in cooked meat products, masking the desirable aroma characteristics of meat.

[0032] During the stewing process, a total of 86 volatile organic compounds were detected. However, due to the concentration threshold dependence of olfactory perception, not all compounds were considered aroma-producing substances. Based on the quantitative VOC concentrations and thresholds (Table 2), 37 compounds with OAV > 1 were selected. To describe the aroma evolution, a radar chart was constructed by prioritizing the top 20 most abundant odor substances. Figure 2 (D diagram), and perform hierarchical cluster analysis to generate a heatmap ( Figure 3 (Figure A in the middle)

[0033] The radar chart shows that the aroma intensity within 0 minutes is negligible, with only minimal sweetness, fruitiness, fatty notes, and greenness. These subtle sensory characteristics originate from methyl 2-methylbutyrate, dimethyl disulfide, and methyl isovalerate, which exhibit relatively high initial concentrations ( Figure 3 (Figure A). The observed sweetness properties of the fruit are consistent with the dominance of the initial esters (Figure A). Figure 2 (See Figure C), because ester compounds typically have sweet and fruity flavor characteristics.

[0034] After initial stewing, the pork belly exhibited similar aroma characteristics at 20 and 40 minutes, consistent with the PCA analysis results. Compared to 0 minutes, the aroma spectrum expanded to include a strong sweetness, some green and fruity notes, and less camphor, oily, plastic, and minty notes. These aromas were hypothesized to originate from ethyl acetate, ethyl butyrate, β-anenetene, α-pinene, 1-hexanol, 1-octen-3-ol, eucalyptol, and 2-pentylfuran, all of which showed a significant increase in concentration at 20 minutes. The green, sweet, and fruity aromas originated from α-pinene, 1-hexanol, eucalyptol, ethyl butyrate, and 2-pentylfuran, while the minty, camphor, and oily characteristics were associated with β-anenetene, eucalyptol, and 1-octen-3-ol. At this stage, the meat aroma compounds were not yet fully formed.

[0035] As the stewing time increases, the aroma of the braised pork initially matures at 60-80 minutes, exhibiting a rich fruity, green, fresh, fatty, and aldehyde aroma, while camphor, sweetness, mint, oily, and plastic-like notes decrease. These residual aromas are attributed to residual volatile compounds such as limonene, 3-methylbutanal, D-limonene, estradiol, and (E)-cinnamaldehyde. Furthermore, α-terpineol, toluene, (1R)-2,6,6-trimethylbicyclo[3.1.1]hept-2-ene, β-myrcene, and eucalyptol were identified as the main spice and seasoning components. Simultaneously, the fatty, fresh, fruity, green, and aldehyde aromas significantly intensify, reaching their maximum intensity at 80 minutes. Figure 3 As shown in Figure A, these characteristics are mainly attributed to the increased concentration of aldehydes (3-methylbutyraldehyde, butyraldehyde, pentanal, hexanal, heptanal, benzaldehyde, and nonanal), indicating that the aroma balance and quality are optimal at 80 min.

[0036] At 100 minutes, due to the consumption of aldehydes, the previously dominant aromas of the pork belly, including fat, freshness, fruitiness, greenness, and aldehydes, significantly decreased, and none of the aromas became prominent, resulting in a reduced overall flavor complexity. This is likely due to the significant volatile nature of the aromas caused by prolonged heating, hence the lack of distinctive aroma characteristics in the 100-minute sample. Further stewing to 120 minutes revealed that the pork belly was dominated by citrus and herbal flavors, with the overall flavor further diminishing. This is primarily attributed to the further reduction of meat aroma components (such as aldehydes) during prolonged stewing, while the concentration of volatile compounds in spices such as 2-heptanone, methyl butyrate, linalool, methyl propionate, α-terpineol, and eugenol increased. The aromas of these spices further dominated the sample's aroma, a trend consistent with... Figure 2 C Figure 1 To.

[0037] (4) Correlation analysis of VOCs Figure 3 Correlation heatmap analysis in Figure B shows that hexanal, nonanal, and heptanal are correlated in deep red, indicating that they share a lipid peroxidation pathway and synchronous accumulation / consumption kinetics. Aldehydes and alcohols are mainly negatively correlated, reflecting the potential oxidation of aldehydes by alcohols during the stewing of braised pork. For example, 1-hexanol and hexanal are negatively correlated, indicating that 1-hexanol is converted to hexanal. This alcohol consumption, coupled with aldehyde enrichment, is shown to drive the flavor shift from "raw meat flavor" to "cooked flavor". Conversely, esters and alcohols are negatively correlated, a pattern attributed to competitive biochemical pathways involving the esterification of fatty acids with alcohols and the thermal decomposition of esters into alcohols. The negative correlation between esters and aldehydes is due to the competitive utilization of shared alcohol precursors; esters require alcohols for synthesis, while aldehydes depend on alcohol oxidation. 2-Pentylfuran is negatively correlated with various aldehydes (4-methoxybenzaldehyde, pentanal, butyraldehyde, hexabenzaldehyde, and heptanal), a pattern attributed to Maillard reaction kinetics. Hypothetically, specific aldehydes act as intermediates, interacting with amino acids to form Amadori compounds, which are then converted into heterocyclic derivatives (such as furans), thereby decreasing aldehyde concentrations while increasing furan levels. Positive correlations exist between α-pinene, (E)-cinnamaldehyde, toluene, estradiol, γ-terpinene, β-myrcene, D-limonene, and limonene, further confirming that these compounds originate from the volatilization of spices during prolonged stewing.

[0038] (5) PLS-DA analysis Partial least squares discriminant analysis (PLS-DA) was used for multivariate data analysis to distinguish cooking time-dependent metabolic changes, and key biomarkers were identified using VIP scoring. Figure 3 As shown in Figure C, principal components 1 and 2 explain 41.5% and 29.25% of the total variance, respectively (model validity: R²X = 0.982, Q² = 0.991). The results regarding differences and similarities between sample groups are consistent with the findings discussed earlier. Figure 2 (Figure B in the middle)

[0039] VIP scores are used to assess the relative significance of variables in the X and Y datasets. (See heatmap). Figure 3 As shown in Figure D, five compounds with a VIP > 1.5 (1-octen-3-ol, 1-hexanol, hexanal, pentanal, and α-terpineol) were shown to have a significant impact on stewing kinetics and flavor evolution, and can serve as potential characteristic biomarkers for stewing processes. Their maximum cumulative concentration was observed at 80 min. 1-Octen-3-ol originates from the oxidative cleavage of linoleic acid, α-linolenic acid, and oleic acid, with linoleic acid being the major precursor. During heating, linoleic acid (C18:2) reacts with oxygen via a radical-mediated chain reaction to form hydrogen peroxide 13-hydroperoxy-9z,11e octadecadienoic acid (13-HPODE) at the double bond. This unstable hydrogen peroxide rapidly decomposes into small volatile compounds, including hexanal and pentanal, which are further oxidized to 1-hexenal and 1-pentenal. Specifically, linoleic acid (C18:2) is oxidized at the C10 position to form 10-hydroxy-8,12-octadecadienoic acid (10-HPOD), which then cleaves to generate 1-octen-3-ol. Therefore, 1-octen-3-ol, hexanal, and pentanal are all associated with the oxidation of n-6 polyunsaturated fatty acids (PUFAs) such as linoleic acid. Figure 4 A clustering correlation heatmap of 37 key VOCs (OAV>1) and flavor characteristics indicates that the oxidation pathway of linoleic acid is an important metabolic pathway for producing the characteristic aroma of braised pork belly. These compounds are also key biomarkers of fat oxidation in meats rich in n-6 PUFAs. Linoleic acid in braised pork belly mainly originates from the hydrolysis of glycerophospholipids, playing a dual role in contributing both nutrition and flavor. As mentioned earlier, α-terpineol originates from the thermal decomposition of spices, highlighting its crucial role in shaping the flavor of braised pork belly.

[0040] (4) Correlation analysis between taste characteristics and key VOCs To investigate the relationship between flavor components in braised pork belly, a correlation analysis was conducted between taste attributes and 37 key VOCs (OAV>1), such as... Figure 4As shown. Color gradients were observed to reflect the range of correlations: bitterness, saltiness, umami, and sourness showed abundant correlations and statistically significant differences, indicating strong VOCs-taste associations, while abundance, aftertaste-A, aftertaste-B, and astringency showed limited correlations, indicating weaker interactions. Of the 37 volatile organic compounds (VOCs), 7 were significantly negatively correlated with sourness, indicating that their accumulation is related to a reduction in the sourness of braised pork belly. This sourness mainly originates from amino acids, which are converted into VOCs through Strecker degradation. For example, benzaldehyde and 4-methylbenzaldehyde are both associated with a decrease in acidity, with phenylalanine and tyrosine degradation identified as their main pathways. However, not all deacidifying VOCs originate from amino acid degradation; concurrent lipid oxidation is thought to contribute to their formation. (E)-Cinnamaldehyde, β-cymene, D-limonene, (1R)-2,6,6-trimethylbicyclo[3.1.1]hept-2-ene, toluene, eucalyptol, and 2-pentylfuran were strongly negatively correlated with sourness, all of which are attributed to flavor-derived volatiles that may mask acidity. These seven compounds were also shown to enhance saltiness and umami through a strong positive correlation. Conversely, five raw meat-related VOCs (methanethiol, dimethyl disulfide, dimethyl trisulfide, methyl isovalerate, and methyl 2-methylbutyrate) were significantly positively correlated with sourness and significantly negatively correlated with saltiness / umami. Notably, all aldehydes were shown to have a positive moderating effect on saltiness and umami while suppressing sourness, highlighting their dual role in flavor modulation during stewing.

[0041] 1.3 HS-GC-IMS Detection The analysis method of HS-GC-IMS is as follows: 2g of chopped sample was sealed in a 20mL headspace vial and equilibrated at 80°C for 30min. 200μL of headspace gas was automatically injected into the GC inlet (80°C) via a preheated syringe (≥80°C) under high-purity nitrogen carrier gas (≥99.999%). Chromatographic separation was performed on a 40°C column (MXT-5) with a programmed flow rate: initial 2mL / min (2min), gradually increased to 15mL / min after 8min, then sequentially increased to 100mL / min (10min) and 150mL / min (10min), for a total run of 30min. The separated analytes were transferred to a 45°C drift tube (9.7cm long) with a drift gas flow rate of 150mL / min (N2, ≥99.999%) in positive ion mode. Each measurement consisted of 12 spectral average scans, and analytical reproducibility was ensured through controlled thermal gradients and flow kinetics, following best practices in ion mobility spectrometry. All experimental samples should be prepared in at least triplicate to ensure the reproducibility of the method.

[0042] HS-SPME-GC-MS and electronic tongue data were processed according to the three sigma rule. The Duncan test was used to perform multi-interval tests on differences between groups. HS-SPME-GC-MS and electronic tongue analyses were performed using the Majorbio I-Sanger Cloud and Metware Cloud web platforms and Origin 2021 software. The Pearson algorithm was used for correlation analysis, and the Euclidean algorithm was used for metabolite distance analysis in cluster analysis. HS-GC-IMS analysis was performed using the commercial Vocal-0.4.10 software from GAS Gesellschaft füranalytische Sensorsysteme mbH (Germany) and three types of plugins (Reporter 1.4.00, Gallery Plot 1.2.3, and Dynamic PCA 1.3.0).

[0043] HS-GC-IMS can effectively separate volatile organic compounds from cooked pork. Figure 5 Figure A shows a clear change over time. As the stewing time increases, the peak number and volume of volatile organic compounds gradually change, indicating that the composition and abundance of volatile organic compounds undergo dynamic changes. Figure 5 Figure B is a two-dimensional library of aromatic compounds from the seven stewing stages. A vertical red line at x=1.0 on the drift time axis marks the Reactive Ion Peak (RIP), with each point to its right representing a detected VOC. Compound concentrations are color-coded, with blue indicating low concentrations and red indicating high concentrations (intensity proportional to color saturation). Most signals are localized within a retention time of 100–400 s and a drift time of 1.0–1.5 s. Compared to raw meat, stewed meat shows increased point density and size on the 2D plot, which is interpreted as increased VOC diversity and concentration due to heat treatment.

[0044] The Gallery Plot plugin was used to further analyze the two-dimensional spectrum to generate a Gallery Plot. Figure 5 Figure C shows the volatile organic compounds (VOCs) with rows representing samples and columns representing compounds. The suffixes "M" and "D" indicate monomeric and dimer forms of the same VOCs. A total of 52 VOCs (including monomers and dimers) were detected, including 9 aldehydes, 17 alcohols, 5 ketones, 4 hydrocarbons, 3 esters, 1 ether, 1 miscellaneous compound, and 12 undetected compounds. Notably, although HS-GC-IMS analysis also identified alcohols as the main category (accounting for 32.69% of total VOCs), the proportion of alcohols was higher than the 19.77% detected by HS-SPME-GC-MS. Figure 1Figure A shows that HS-GC-IMS has higher sensitivity for ethanol detection. However, compared with HS-GC-IMS, HS-SPME-GC-MS exhibits a wider range of compounds and higher VOCs diversity, consistent with previous methodological comparisons.

[0045] Compared to HS-SPME-GC-MS heatmaps, the HS-GC-IMS library provides higher visual resolution for tracking VOC concentration dynamics during stewing. Overall, the corridor maps of samples heated to 0 min form a clear cluster, while corridor maps from 20–100 min show high similarity in flavor characteristics. Samples heated to 120 min again cluster separately, conforming to the PCA clustering pattern. At 0 min, alcohols such as 2-hexanol (M / D), 1-pentanol (M / D), and 2-pentanol (M / D) were identified as major components, likely due to the thermal oxidation of triglycerides to fatty acids and alcohols during HS-GC-IMS incubation (80°C, 25 min). During the simmering process of 0–40 min, a large number of aldehydes, ketones, and alcohols were observed to be generated, including all aldehydes except pentanal (M), most ketones (e.g., 2-butanone, 2,3-butanedione), and specific alcohols (2-octanol, 1-pentanol M / D, 2-pentanol D, butanol D, 3-methyl-1-butanol M / D). These compounds are mainly attributed to lipid oxidation and early Maillard reactions, producing carbonyl intermediates.

[0046] The concentration of benzaldehyde (M / D) gradually increased from 20 to 40 min, possibly driven by lipid oxidation, and then significantly increased from 40 to 60 min (bright corridor plot), indicating that the Strecker degradation accelerated and transitioned to the intermediate Maillard phase. This phase was further confirmed by the depletion of 2,3-butanedione (the dicarbonyl precursor of the Strecker reaction).

[0047] The total aldehyde content reached its maximum at 60 and 80 min, consistent with the HS-SPME-GC-MS results. Notably, a different time pattern was observed between HS-SPME-GC-MS and HS-GC-IMS: HS-GC-IMS detected a gradual increase in aldehyde concentration, but HS-SPME-GC-MS showed a sudden surge at 60 min. This difference is attributed to the superior sensitivity of HS-GC-IMS for detecting trace levels and small molecules, enabling the identification of aldehydes in the early stages of stewing when concentrations were below the HS-SPME-GC-MS detection threshold.

[0048] During the 80-100 min simmering stage, a decrease in the concentration of most aldehydes (including benzaldehyde (M / D), heptanal (M / D), 2-hexenal, butyral and 2-methyl-2-propanal) was observed. This phenomenon was thought to be caused by two mechanisms: (1) the rate of lipid oxidation slowed down, resulting in less aldehyde production, as evidenced by stable or decreasing ethanol concentrations, which usually change with the lipid oxidation intermediates; and (2) aldehydes were consumed as substrates during the advanced Maillard reaction stage, leading to gradual consumption.

[0049] During the 100–120 min simmering phase, increased concentrations of several compounds were observed, including 3-perylene (D), β-pinene, benzaldehyde (M / D), butyraldehyde, 2-butanone, 2-methylpropionaldehyde, 2-pentanol (M), and isobutanol (M). This resurgence was hypothesized to be a result of higher lipid oxidation pathways, which may promote the deep formation of these volatiles through the oxidative cleavage of lipid-derived precursors.

[0050] PCA was used to normalize the distribution of VOCs throughout the stewing time. Figure 6 PC1 contributed 57.8% to the differentiation of pork samples stewed for 20, 40, 60, 80, 100, and 120 min, while contributing 26.2% to the differentiation of raw meat samples stewed for 0 min. PC3 (6.3%) provided supplementary resolution for all time points. The cumulative variance explained by PC1+PC2+PC3 (>80%) was validated to be sufficient to represent the overall flavor characteristics. A significant spatial separation existed between the 0 min group and the stewed samples, indicating substantial compositional differences. Within the 20, 40, 60, 80, 100, and 120 min groups, samples closer in time showed closer spatial distributions, a pattern consistent with SPME-HS-GC-MS compositional trends.

[0051] Example 2 This embodiment focuses on determining the detection conditions of HS-GC-IMS, and the specific operation is as follows: Samples of braised pork belly from the same batch were taken after stewing, cooled, and sealed for HS-GC-IMS analysis.

[0052] In this embodiment, adjustments were made to the equilibrium temperature, drift tube temperature, and flow rate scheme, with the specific variables as follows: Comparison A (equilibrium temperature): 60°C, 70°C, 80°C, 90°C; Comparison B (IMS detection temperature / drift tube temperature): 35°C, 45°C, 55°C; Comparison C (flow rate scheme): fixed flow rate 15 mL / min, fixed flow rate 30 mL / min, segmented program 2→15→100→150 mL / min; Table 3. Comparison of peak area and signal-to-noise ratio of six key aroma markers under different GC-IMS conditions.

[0053] Note: The detection signal of volatile compounds is expressed as peak area, which is the integral value of the ion mobility spectrum signal intensity and is a relative response value used to characterize the relative change trend of the same compound under different detection conditions. The signal-to-noise ratio (SNR) is the ratio of the ion signal intensity to the baseline noise intensity, a dimensionless parameter used to evaluate the sensitivity and stability of the detection conditions.

[0054] As shown in Table 3, compared with the 60°C, 70°C, and 90°C conditions, the baseline condition has the greatest improvement in peak area and S / N for all key aldehyde / alcohol signals. The signal is weak at low temperatures, and some low-boiling-point components are lost at high temperatures. The best effect is achieved at 80°C. In addition, drift tube temperatures of 35°C and 55°C both lead to a decrease in resolution or attenuation of light component signals, while the 45°C condition balances mobility resolution and signal strength, making it the optimal choice. Meanwhile, the segmented flow rate program is significantly better than the constant flow or simple linear scheme, showing the best performance in terms of time resolution and peak shape separation.

[0055] Cross-validation: Baseline GC-IMS and HS-SPME-GC-MS showed high consistency in the time dynamics of markers such as 1-hexanol, hexanal, and 1-pentanol (R0). 2 >0.9), demonstrating the qualitative / semi-quantitative reliability of the method, and the present invention significantly improves the detection sensitivity and time-series resolution of key flavor compounds in the braised pork stew system.

[0056] This invention comprehensively and dynamically reveals the flavor evolution process of braised pork belly during stewing. Compared with raw meat, stewed pork belly exhibits increased saltiness and umami, and decreased acidity, with prolonged stewing enhancing these sensory characteristics. HS-SPME-GC-MS technology was used to analyze 86 volatile organic compounds (VOCs) during the stewing process of braised pork belly, identifying 37 key VOCs, including 1-octen-3-ol, 1-hexanol, hexanal, pentanal, and α-terpineol. Due to the contribution of alcohols and aldehydes, braised pork belly stewed for 80 minutes exhibits typical aldehyde, fat, fresh, fruity, and green flavors, with the richest aroma, significantly higher than other stewing times. Stewing beyond 80 minutes leads to the loss of characteristic flavors such as fat and fruit aromas, a decrease in flavor richness, and a more prominent spice aroma. 1-Hexanol and 1-pentanol are characteristic biomarkers jointly discovered by GC-MS and GC-IMS that can be used to distinguish different stewing stages (OAV>1 and VIP>1). Linoleic acid oxidation is a major chemical pathway that significantly affects the aroma profile of braised pork belly during the stewing process. Furthermore, a fingerprint spectrum of dynamic changes in flavor compounds during the stewing process of braised pork belly, primarily driven by lipid oxidation and secondarily by Maillard reactions, was established using HS-GC-IMS analysis. The detection method and results provided in this invention contribute to a deeper understanding of the flavor formation process of braised pork belly and offer stronger theoretical support for its further processing.

[0057] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. All equivalent changes and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A method for dynamic monitoring and quality evaluation of flavor substances during the processing of braised pork belly, characterized in that: Sampling of braised pork belly was conducted at different time points during the stewing process. Then, the aroma and taste data of the braised pork belly samples were acquired using an electronic tongue. The volatile compounds in the braised pork belly samples were then detected by headspace solid phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS) and gas chromatography-ion mobility spectrometry (HS-GC-IMS), respectively. The comprehensive detection data were used for principal component analysis (PCA), aroma activity value (OAV) analysis, thermogram analysis, and correlation analysis to evaluate the flavor of braised pork belly during the stewing process. The HS-GC-IMS conditions were as follows: sample equilibration at 80°C for 30 min, headspace injection volume: 200 μL, inlet temperature: 80°C, using an MXT-5 column, column temperature: 40°C, nitrogen as both carrier gas and drift gas, carrier gas fractional flow rate: initial 2 mL / min, held for 2 min, then linearly increased to 15 mL / min over 8 min, further increased to 100 mL / min over the following 10 min, and continued to increase to 150 mL / min over another 10 min, for a total run of 30 min, drift tube temperature: 45°C, drift gas flow rate: 150 mL / min.

2. The method as described in claim 1, characterized in that: The conditions for the HS-SPME-GC-MS are as follows: The pork belly sample was mixed with 10 μL of 2 mg / L 2-methyl-3-heptanone and equilibrated in a 20 mL vial at 250°C for 30 min in the headspace. The pretreated solid-phase microextraction head CAR / PDMS / DVB fiber was exposed to the headspace under helium, and then thermal desorption was performed at 240°C at the GC inlet for 2 min. Chromatographic separation was performed using a DB-WAX capillary column (60 m × 250 μm × 0.25 μm, Agilent J&W).

3. The method as described in claim 2, characterized in that: In the HS-SPME-GC-MS described above, the MS parameters include 70 eV electron ionization, the ion source and the transport line are maintained at 230°C and 250°C respectively, and the signal is acquired in full scan mode with a scan range of 35~500 m / z.

Images