Study on the Effect of Different Concentrations of SO2 on the Volatile Aroma Components of ‘Beibinghong’ Ice Wine

SO2 plays an important role in wine fermentation, and its effects on wine aroma are complex and diverse. In order to investigate the effects of different SO2 additions on the fermentation process, quality, and flavor of ‘Beibinghong’ ice wine, we fermented ‘Beibinghong’ picked in 2019. We examined the fermentation rate, basic physicochemical properties, and volatile aroma compound concentrations of ‘Beibinghong’ ice wine under different SO2 additions and constructed a fingerprint of volatile compounds in ice wine. The results showed that 44 typical volatile compounds in ‘Beibinghong’ ice wine were identified and quantified. The OAV and VIP values were calculated using the threshold values of each volatile compound, and t the effect of SO2 on the volatile compounds of ‘Beibinghong’ ice wine might be related to five aroma compounds: ethyl butyrate, ethyl propionate, ethyl 3-methyl butyrate-M, ethyl 3-methyl butyrate-D, and 3-methyl butyraldehyde. Tasting of ‘Beibinghong’ ice wine at different SO2 additions revealed that the overall flavor of ‘Beibinghong’ ice wine was the highest at an SO2 addition level of 30 mg/L. An SO2 addition level of 30 mg/L was the optimal addition level. The results of this study are of great significance for understanding the effect of SO2 on the fermentation of ‘Beibinghong’ ice wine.


Introduction
Ice wine is a sweet wine that is made by fermenting the naturally frozen grape juice from the vines when the temperature drops to −7~−8 • C [1].Canada and Germany are the leading producers of ice wine, while China, Austria, and the United States also produce ice wine in large quantities.In recent years, the annual production of ice wine in China has reached 300 million liters, especially in Huanren County, Liaoning Province, and the Yalu River Basin, Jilin Province, where the ice wine industry is developing rapidly [2].The 'Beibinghong' grape is the world's first wild grape variety (Vitis amurensis Rupr) that can make ice wine and is very popular in northeast China.It has the advantages of high cold tolerance and stable yield.Compared to unfrozen grapes, frozen grapes contain high concentrations of sugar, aroma, and flavor compounds, giving the resulting ice wine a rich, fruity flavor.After alcoholic fermentation, ice wine still contains a rich concentration of residual sugar, which gives it a solid, sweet flavor.The 'Beibinghong' ice wine has a more rounded taste and mellow aroma than the 'Beibinghong' dry wine, and ice wine has received a lot of attention because of its unique flavor.
'Beibinghong' is one of the famous high-quality grape varieties; Vitis amurensis Rupr belong to the East Asian grape family.'Beibinghong' is an interspecific cross between Foods 2024, 13, 1247 3 of 22 wine, and identifying the volatile compounds that play a vital role in the different amounts of SO 2 added to the wine.
This study measured the fermentation start and end times, basic physicochemical properties, and volatile aroma compounds of 'Beibinghong' ice wine harvested in 2019.The fingerprints of volatile compounds of ice wine brewed with different concentrations of SO 2 were established, and its brewed ice wine was tasted.In the experiment, SO 2 was added by adding solid potassium metabisulfite (PMS), and each gram of PMS produced 0.56 g of SO 2 , i.e., 10 mg of SO 2 required 17.86 µg of PMS, which was then added in eight treatments, i.e., 10 mg/L, 20 mg/L, 30 mg/L, 40 mg/L, 50 mg/L, 60 mg/L, 80 mg/L, and 100 mg/L, in the following order.Eight treatments were performed to determine the optimal amount of SO 2 addition at the same temperature and under the same yeast strain and enzyme treatment conditions.

Process Flow of Brewing 'Beibinghong' Ice Wine
The process of ice wine production involves harvesting the fruit, screening, and de-stemming.The harvested fruit is then granulated, pressed, and preserved by adding different concentrations of SO 2 .Three sets of replicate brewing experiments were conducted using three fermenters per treatment to ensure the reproducibility of the experiments.Gum reduction is carried out, followed by a low-temperature maceration at 2-4 • C. The wine is then post-tempered to 15 • C. Subsequently, the post-temperature was adjusted to 15 • C, and controlled fermentation was carried out with the addition of CECO1 yeast at a dosage of 250 mg/Kg.After completion of fermentation, fermentation was stopped, and crude filtration was carried out to obtain the original wine.After controlled aging, the wine is then fine-filtered and sterilized.After passing quality tests, the final product is bottled, sealed, and labeled as ice wine.

Sample Labeling
The amount of SO 2 used according to the quality of treated iced grape juice was as follows: sample No. 1 (10 mg/L), sample No.

Detection Methods of Basic Physical and Chemical Indexes
Soluble solids were determined by handheld refractometer, and titrable acid content of wine was determined by indicator method according to The GB/T 15038-2006 [21].The alcohol content is measured by the alcohol meter method.The total sugar content in wine was determined by anthrone and sulfuric acid colorimetric method, and standard koji was prepared by standard glucose solution.The total anthocyanin content in grape juice was determined by pH difference method, i.e., anthocyanin reacted with potassium chloride buffer (0.025 M, pH = 1) and acetic acid buffer (0.4 M, pH = 4.5), and then the difference of 520 nm and 700 nm was calculated.Total phenol content-Folin-Ciocalteu colorimetric method.Dry extract content: refer to the dry extract test method in the national standard (GB/T 15038-2006).

Quantification of Volatile Compounds in 'Beibinghong' Ice Wine by HS-GC-IMS
Headspace-gas chromatography-ion migration spectrometry (HS-GC-IMS) determined volatile substances in wine.The FlavourSpec ® flavor analyzer was used to take a 1 mL sample, place it into a 20 mL headspace bottle, add 20 ppm 4-methyl-2-amyl alcohol 10 µL, incubate at 60 • C for 15 min, and then inject it into the sample.
Chromatographic conditions: the column was WAX column (15 m × 0.53 mm,1 µm), column temperature was 60 • C, carrier gas was N2, IMS temperature was 45 • C, and chromatographic conditions were shown in Table 1.The conditions of automatic headspace injection were as follows: injection volume of 100 µL, incubation time of 10 min, incubation temperature of 60 • C, injection needle temperature of 65 • C, and incubation speed of 500 rpm; 4-methyl-2-pentanol was used as the internal standard for the analysis, and the concentration of 198 ppb, the signal peak volume of 493.34, and the intensity of each signal peak was about 0.401 ppb.
Quantitative calculation formula: Ci is the calculated mass concentration of any component in µg/L, Cis is the mass concentration of the internal standard used in µg/L, and Ai/AIS is the volume ratio of any signal peak to the signal peak of the internal standard.The NIST database and IMS database are built into the software for the qualitative analysis of the substances.

Odor Activity Value (OAV) Calculation
OAV was used to assess the contribution of volatile compounds to the overall aroma of the wine.The concentration of volatile compounds was divided by the odor threshold (OT) to calculate the OAV value.Volatile compounds with OAV > 1 were considered to be types of aroma-active compounds, and the larger the OAV value, the more significant the contribution of components to the flavor; the OAV value can help to determine the critical aroma substances in food or plants [22][23][24], analyze the causes of flavor differences, and optimize the flavor quality of wine aroma characteristics formation plays an important role.

Sensory Evaluation
Sensory evaluation methodology: the wines were subjected to quantitative descriptive analysis (QDA) by a trained sensory panel of 17 tasters (10 women and 7 men, aged 24 to 52 years, with an average of 33 years).These experts were recruited based on their motivation and availability, having been trained according to the national standards ISO 6658 [25] and ISO 8586 [26] prior to the sensory evaluation.According to the definitions in the published literature, according to the definitions in The national standard GB 15038-2006 [27,28], and based on the discussion results, specify the development of a sensory evaluation form (Table 2).The samples were marked with three numbers and submitted to the tasters randomly.

Statistical Analysis of Data
Excel 2010 was used to organize the experimental data statistically, and an analysis of variance (ANOVA) was performed using SPSS (version 22.0, IBM, Armonk, NY, USA).Statistical).Statistical analyses were performed on the experimental data to check for significant differences in the individual results, and all the data were expressed as Foods 2024, 13, 1247 6 of 22 mean ± standard deviation.Differences between the two groups were considered significant at p < 0.05.Simca 14.1 software was used for OPLS-DA and VIP value analysis; GC-IMS assay was done with Savitzky Golay for smoothing and denoising, and migration time normalization was done by setting the RIP position as 1, i.e., dividing the actual migration time by the peak out time of the RIP to obtain the approximate migration time.The Reporter plug-in was used to directly compare the spectral differences between the samples, and the Gallery Plot plug-in was used to compare the fingerprints visually and quantitatively to compare the differences of VOCs between different samples.Heat map and correlation analysis were performed using the OmicShare tools, a free online platform for data analysis (https://www.omicshare.com/tools(accessed on 27 July 2020)).

Fermentation Process and Basic Indexes of 'Beibinghong' Ice Wine with Different SO 2 Additions
The yeast treatment time in this experiment was 11 December 2019, and the initial solid content of 'Beibinghong' grape juice was 40.3%.It can be seen from Table 3 that the fermentation time was 24 h at 10 mg/L and 48 h at 30, 40, and 50 mg/L.The fermentation time of 10 mg/L, 40 mg/L, and 50 mg/L was 23, 23, and 21 days, but the red wine was suitable for slow fermentation at low temperatures.This is in line with the findings of Sun Hening and others [29] that the addition of SO 2 prior to fermentation inhibits yeast activity in the short term and increases the delay in yeast multiplication, leading to a delay in fermentation; however, during this period, it allows the must to remain static and encourages the precipitation of impurities, colloidal substances, and decomposed tartaric acid, which is highly susceptible to the formation of tartar in wines.Wang et al. [30] studied the effect of SO 2 on the fermentation process and quality of pineapple wine and found that SO 2 had a significant effect on the time of starting fermentation when the concentration of SO 2 was 150 mg/L or less, the time of starting fermentation was around ten h.With the increasing concentration of SO 2 , the time of starting fermentation was delayed significantly, and when the concentration of SO 2 was 250 mg/L, the time of starting the fermentation time was more than 3 d at a concentration of 250 mg/L, which was the same as our results, indicating that the higher concentration of SO 2 affects the fermentation of fruit wines.The fermentation time of fruit wines should not be too long, so it should be considered comprehensively.The total sugar of ice wine fermented with different amounts of SO 2 was found to be above 160.0g/L, which indicates that different amounts of SO 2 do not have much effect on the total sugar, which is the same as the results of the study by Mou Jingxia et al. [10].The total acid content of SO 2 was relatively low at 10 mg/L, 20 mg/L, 30 mg/L, 40 mg/L.The level of dry leachate index is closely related to the raw materials, production process, and storage method of wine, and it is one of the essential symbols of the quality of wine [31].Dry leachate was higher at SO 2 additions of 30 mg/L, 40 mg/L; From the above table, adding 30 mg/L and 40 mg/L SO 2 is more appropriate.Color is one of the most critical indicators affecting the sensory quality of red wine, and anthocyanin, a class of flavonoid compounds with a benzopyran structure, is an essential water-soluble natural pigment in red wine, as well as a crucial color-presenting substance, with a variety of critical physiological functions and biological activities [32,33].During grape growth and development, anthocyanosides are biosynthesised via the phenylpropane-flavonoid pathway.In wine, there is an equilibrium between the various states of anthocyanins, and their color expression is closely related to the structure and morphology of the anthocyanin molecule [34].After human consumption of wine, wine anthocyanosides are mainly metabolized and absorbed by intestinal flora in the colon.The type, state, and content of anthocyanosides are essential in red wines' color characteristics and aging potential.Changes in the content of anthocyanosides in 'Beibinghong' ice wine brewed with different SO 2 additions are shown in Table 3.
As shown in Table 4, the content of anthocyanin was higher when SO 2 was added at 30 mg/L, 60 mg/L, and 80 mg/L, especially at 80 mg/L, which reached the highest value.However, the total acid content was slightly higher because 'Beibinghong' belongs to the Vitis amurensis Rupr variety.From the taste and quality perspective, the SO 2 addition was not too high, so 30 and 60 mg/L were more appropriate.A moderate addition of SO 2 has a protective effect on the anthocyanins in wine [35].SO 2 inhibits the action of oxidative enzymes and prevents oxidation of the raw material, which helps to maintain the stability and color vividness of the anthocyanosides.The addition of SO 2 also helps to select the fermentation microorganisms, clarify the fermentation matrix, and regulate the acidity of the fermentation matrix, which indirectly affects the solubilization and stability of the anthocyanosides [36][37][38][39].J Bakker et al. [40] found that as the level of SO 2 increased during winemaking, more anthocyanins were extracted.During maturation, all wines lost color and increased brownness.Wines without added SO2 browned more severely than wines with added SO 2 .This also shows that SO2 is essential to maintain color stability in wine production.Means with different letters in the same column express significant differences (Duncan's test p < 0.05).
However, when the added SO 2 level is too high, it can adversely affect anthocyanin [41].This is in line with our findings that an addition of sulphur dioxide that is too high affects the content of anthocyanosides.High concentrations of SO 2 may generate sulfites in acidic environments, which are capable of reacting with anthocyanoside molecules, leading to the formation of anthocyanoside sulfites, which are colorless, and therefore reduce the anthocyanoside content of wines, thus affecting the color of the wines.SO 2 has both protective and potentially damaging effects on the anthocyanosides in wine, depending on the amount added and the conditions of use.The amount of SO 2 added during the winemaking process needs to be precisely controlled to maximize the positive effects and minimize the negative effects on the anthocyanins, thus ensuring optimal wine quality and taste.

Fingerprints of Volatile Components of 'Beibinghong' Ice Wine Brewed with Different SO 2 Additives
In order to analyze the variability of volatile substances in ice wine brewed with different concentrations of SO 2 , we constructed a fingerprint of volatile flavor compounds based on all the signal peaks in the two-dimensional HS-GC-IMS spectra (Figure 1).Each sample was measured three times in parallel, and the darker color indicated a greater peak intensity and higher content.The composition and differences of volatile flavor compounds in ice wine brewed with different concentrations of SO 2 were revealed from the fingerprints.
Means with different letters in the same column express significant differences (Duncan's test p < 0.05).

Fingerprints of Volatile Components of 'Beibinghong' Ice Wine Brewed with Different SO2 Additives
In order to analyze the variability of volatile substances in ice wine brewed with different concentrations of SO2, we constructed a fingerprint of volatile flavor compounds based on all the signal peaks in the two-dimensional HS-GC-IMS spectra (Figure 1).Each sample was measured three times in parallel, and the darker color indicated a greater peak intensity and higher content.The composition and differences of volatile flavor compounds in ice wine brewed with different concentrations of SO2 were revealed from the fingerprints.As shown in Figure 2, the volatile compounds in the eight wine samples were well separated, and the aroma fingerprints of the wine samples differed significantly, mainly in the content of volatile compounds.
The 'Beibinghong' ice wine from the first treatment group was used as a reference, and the rest of the spectra were subtracted from the signal peaks in the first treatment to obtain the difference spectra (Figure 3).As shown in Figure 2, the volatile compounds in the eight wine samples were well separated, and the aroma fingerprints of the wine samples differed significantly, mainly in the content of volatile compounds.The 'Beibinghong' ice wine from the first treatment group was used as a reference, and the rest of the spectra were subtracted from the signal peaks in the first treatment to obtain the difference spectra (Figure 3).

Analysis of Volatile Components
Aroma is an essential index for evaluating the quality of wine, and the study of wine aroma mainly focuses on the aromatic substances that positively affect it.The critical role of SO 2 in winemaking also includes its role in the aroma.SO 2 also significantly affects the aroma of wine, with the addition of SO 2 affecting the concentration of volatile compounds in the wine by between 33% and 43% [42].Aromatic substances in wine can be categorized into terpenoids, aliphatic compounds, and aromatic compounds according to their chemical structure, of which aliphatic compounds include alcohols, acids, ketones, and esters.The analytical spectra and data were viewed with VOCal for qualitative and quantitative analysis, and the volatile components in the wine samples were characterized by the builtin NIST and IMS databases of HS-GC-IMS.A total of 44 typical volatile compounds were detected (Table 5), and the most significant number of species was 18 esters, 11 alcohols, 3 acids were detected, 5 ketones, and 11 uncharacterized volatile compounds.The volatile aroma compounds detected in the ice wine samples from the eight treatments were the same type, but the contents were significantly different.Among the eight treatments, the highest total volatile compound content was found in the seventh treatment group with 80 mg/L of added SO 2 , with a total volatile compound content of 81,930.42256µg/L.The following treatments with the highest total volatile compound content in descending order were the treatment group with 50 mg/L of added SO 2 , with a volatile compound concentration of 81,394.73104µg/L, treatment group with 70 mg/L SO 2 addition 81,328.35368µg/L, treatment group with 40 mg/L SO 2 addition 81,182.37288µg/L, treatment group with 60 mg/L SO 2 addition 79,569.26936µg/L, and treatment group with 90 mg/L SO 2 addition 79,215.75989µg/L.79,569.26936µg/L for 60 mg/L of SO 2 , 79,215.75984µg/L for 90 mg/L of SO 2 , 78,921.88528µg/L for 30 mg/L of SO 2 , and 77,818.71496µg/L for 20 mg/L of SO 2 .Table 5 shows that alcohols accounted for the most significant proportion of 62.9-64.73%,followed by esters at 24-25.82%, and alcohols and esters were the main aroma compounds in the wine samples.The esters in wine are mainly produced by acyl-coenzyme A and fatty acids and alcohols in yeast cells under the catalytic action of relevant enzymes during alcoholic fermentation, and they have the aroma of fruits or flowers, which play a vital role in the aroma of wine [43,44].Esters give wines a unique and complex fruity flavor, which is a critical component of their aroma composition, and the description of the aroma of the detected ester compounds also shows that the esters are mainly dominated by fruity aroma.The various ester compounds present in wines have coordinated compositional ratios and have synergistic effects on the formation of aroma [43,45].The esters in grapes are mainly found in grape skins, which are fully macerated during fermentation, releasing the variety's unique fruity and floral aromas.As can be seen in Table 4, esters were the compounds with the highest percentage of content in the assay, and from the esters detected, the esters that provided a higher concentration of aromas were isoamyl acetate, ethyl acetate, methyl formate, ethyl butyrate, and Ethyl propionate, the highest content of esters detected among the eight treatments was in treatment group 7 (SO 2 addition of 80 mg/L) at 20,217.62344µg/L, followed by ester volatile compounds in descending order by treatment group 8 (SO 2 addition of 90 mg/L) at 20,201.48088µg/L; treatment group 1 (SO 2 addition of 20 mg/L) was 20,089.12584µg/L; treatment group 5 (SO 2 addition of 60 mg/L) was 19,966.58664µg/L; treatment group 3 (SO 2 addition of 40 mg/L) was 19,897.53192µg/L; and treatment group 6 (SO 2 addition of 70 mg/L) was 19,734.66768µg/L; the fourth treatment group (SO 2 addition of 50 mg/L) was 19,533.09736µg/L; and the second treatment group (SO 2 addition of 30 mg/L) was 19,438.08048µg/L.A study by Teresa Garde-Cerdán et al. found [46] that when volatile aroma compounds were examined in wines with or without the addition of SO 2 , it was found that the concentration of SO2 did not have a significant effect on the total esters in the wines, which is the same as our findings, and this is probably because the wines are rich in unsaturated fatty acids.Therefore, the concentration of oxygen in the medium did not affect the formation of these compounds.Liu et al. [47] found that the addition of an appropriate concentration of SO 2 can also increase the content of isoamyl acetate, ethyl acetate, and ethyl octanoate in wine to a certain extent, which can bring pleasant floral and fruity aroma to wine and increase the complexity of wine aroma.The high concentration of SO 2 can lead to the increase of ethyl acetate, which can hurt the aroma quality of the wine.Therefore, when choosing the concentration of SO 2 , it is essential to consider not only the concentration of the aroma but also the negative effect of the aroma content on the wine, and the amount added should be manageable [42,48,49].Moderate amounts of SO 2 can help stabilize the aroma components in wine and contribute to forming certain aromas.However, excessive amounts of SO 2 may react with the aroma components of the wine, resulting in a change or loss of aroma.Such changes may be manifested as a diminution or loss of certain aromas in the wine or the production of some unpleasant off-flavors.

Alcohols
Alcohols accounted for the most significant percentage of 62.9-64.73% of the wines, and their aroma characteristics were mainly pungent or grassy.The content of alcohols between different treatment groups in descending order was 52,687.70472µg/L in the fourth treatment group (SO 2 addition of 50 mg/L); 52,419.32192µg/L in the seventh treatment group (SO 2 addition of 80 mg/L); 51,936.64448µg/L in the sixth treatment group (SO 2 addition of 70 mg/L); third treatment group (SO 2 addition of 40 mg/L) at 51,900.99208µg/L; fifth treatment group (SO 2 addition of 60 mg/L) at 50,486.39344µg/L; second treatment group (SO 2 addition of 30 mg/L) at 50,107.1648mg/L; the eighth treatment group (SO 2 addition of 90 mg/L) with 50,067.50616µg/L; and the first treatment group (SO 2 addition of 20 mg/L) with 48,951.784µg/L.The highest ethanol content was found in the treatment group with the highest SO 2 addition of 70 mg/L, and the lowest was found in the treatment group with the lowest SO 2 addition of 20 mg/L.The highest ethanol content was found in the treatment group with the highest SO 2 addition of 70 mg/L and the lowest in the treatment group with the lowest SO 2 addition of 20 mg/L.The highest ethanol content was in the treatment group with 70 mg/L of SO 2 , and the lowest was in the treatment group with 20 mg/L of SO 2 ; the highest methanol content was in the treatment group with 50 mg/L of SO 2 , and the lowest methanol content was in the treatment group with 20 mg/L of SO 2 ; 2.4.3.Others Aldehydes, acids, and ketones accounted for a relatively small percentage of the wine.The percentage of aldehydes in the volatile compounds of ice wine was 2.06-2.17%.Among the treatment groups, the most considerable aldehydes content was in the third treatment group (SO 2 addition of 40 mg/L) with 1758.6688.The lowest content was in the eighth treatment group (SO 2 addition of 90 mg/L) with 1627.96816µg/L.The proportion of ketones in the volatile compounds of ice wine ranged from 1.87% to 2.02%, with the most considerable ketone content in the first treatment group (20 mg/L SO 2 ) at 1575.7028 µg/L and the lowest in the fifth treatment group (60 mg/L SO 2 ) at 1511.65784 µg/L.The proportion of acid compounds in the volatile compounds of ice wine ranged from 1.87% to 2.02%.The percentage of acid compounds in the volatile compounds of ice wine ranged from 4.73% to 5.66%, with the most significant amount of acid compounds in treatment group 6 (70 mg/L of SO 2 addition) at 4604.20464 µg/L and the lowest in treatment group 1 (20 mg/L of SO 2 addition) at 3%.
Among the treatments, regarding the total content of volatile compounds, the SO 2 addition of 30 mg/L had a higher content of total aroma substances.Methanol and other substances harmful to the aroma components of 'Beibinghong' ice wine were fewer.The content of the components that impacted the quality was lower, so the quality of the wine was better than that of the other treatment groups from a comprehensive point of view.

Principal Component Analysis (PCA) of Wine Samples
In order to better present and distinguish the differences between the different treatment ice wine samples, the known volatile compounds identified by GC-IMS were analyzed by PCA (Figure 4).Eight treatment groups of samples were well differentiated according to their aroma characteristics.PCA was performed on the samples to discriminate the magnitude of variability between the samples of the groups of different wines, between subgroups, and between samples within groups.The contribution rate of PC1 was 28.4%, and that of PC2 was 21.1%, and the eight groups of samples showed apparent separation trends on the two-dimensional graph, with no outlier samples.The samples of the same kind of wines were clustered well, with high experimental reproducibility.The PCA results show that the differences in aroma substances among the eight groups of samples are significant and clearly distinguishable from other samples.As shown in the figure, it can be seen that the four treatments of A, B, H, and E clustered together, the treatment groups of C and D clustered together, and the groups of G and F clustered together, which indicated that the volatile compounds were similar between these treatment groups.The concentration of the aroma compounds was also somewhat different between the different treatments.

OAV Analysis of Major Aroma Compounds of Different Wine Samples
Generally, the components with OAV greater than one directly impacts the overall flavor and are the main components providing the flavor [50][51][52].Based on the qualitative and quantitative results of GC-IMS, the threshold values of corresponding aroma compounds in water were found in the literature, and their OAV values were calculated.As shown in Table 6, a total of 21 aroma compounds with OAV values greater than one were detected in different ice wine samples, which were ethyl caprylate, 1-hexanol-M, ethyl caproate, 3-methyl-1-butanol, isoamyl acetate, hexanal, 2-methyl-1-propanol, ethyl butyrate, isobutyl acetate, pentanal, ethyl acrylate, ethyl isobutyrate, ethyl acetate, acetone, acetaldehyde, propionaldehyde, (Z)-3-hexenyl acetate, ethyl 3-methyl butanoate-M, ethyl 3-methyl butanoate-D, butyraldehyde, and 3-methyl butanal; studies have shown that the OAV values are directly proportional to the contribution of aroma [53,54].Among the compounds with OAV values greater than 1, esters accounted for the most significant proportion of ten, and the OAV values of esters were significantly higher than those of other types of compounds, indicating that esters contribute to the prominent aroma of 'Beibinghong' ice wine.This is also in line with the GC-IMS results, where esters contributed the primary aromas in 'Beibinghong' ice wine, dominated by fruity and floral notes [55], followed by aldehydes, with six aldehydes having OAVs greater than 1, indicating that the grassy aroma of aldehydes is also a significant contributor to the aroma of 'Beibinghong' ice wine; there was also a ketone compound of acetone among the compounds with an OAV value of greater than 1; and there were three alcohols, which is also a significant contributor to the iconic wine flavor of the 'Beibinghong' ice wine aroma.

OAV Analysis of Major Aroma Compounds of Different Wine Samples
Generally, the components with OAV greater than one directly impacts the overall flavor and are the main components providing the flavor [50][51][52].Based on the qualitative and quantitative results of GC-IMS, the threshold values of corresponding aroma compounds in water were found in the literature, and their OAV values were calculated.As shown in Table 6, a total of 21 aroma compounds with OAV values greater than one were detected in different ice wine samples, which were ethyl caprylate, 1-hexanol-M, ethyl caproate, 3-methyl-1-butanol, isoamyl acetate, hexanal, 2-methyl-1-propanol, ethyl butyrate, isobutyl acetate, pentanal, ethyl acrylate, ethyl isobutyrate, ethyl acetate, acetone, acetaldehyde, propionaldehyde, (Z) −3 -hexenyl acetate, ethyl 3-methyl butanoate-M, ethyl 3methyl butanoate-D, butyraldehyde, and 3-methyl butanal; studies have shown that the OAV values are directly proportional to the contribution of aroma [53,54].Among the compounds with OAV values greater than 1, esters accounted for the most significant proportion of ten, and the OAV values of esters were significantly higher than those of other types of compounds, indicating that esters contribute to the prominent aroma of 'Beibinghong' ice wine.This is also in line with the GC-IMS results, where esters contributed the primary aromas in 'Beibinghong' ice wine, dominated by fruity and floral notes [55], followed by aldehydes, with six aldehydes having OAVs greater than 1, indicating that the grassy aroma of aldehydes is also a significant contributor to the aroma of 'Beibinghong' ice wine; there was also a ketone compound of acetone among the compounds with an OAV value of greater than 1; and there were three alcohols, which is also a significant contributor to the iconic wine flavor of the 'Beibinghong' ice wine aroma.
Principal component analysis (PCA) is a multivariate statistical analysis technique.Many complex and hard-to-find variables in the original sample are represented by identifying several principal component factors.The regularity and variability between samples are assessed based on the contribution of the principal component factors in different samples [56].The PCA results clearly showed that two principal components were Principal component analysis (PCA) is a multivariate statistical analysis technique.Many complex and hard-to-find variables in the original sample are represented by identifying several principal component factors.The regularity and variability between samples are assessed based on the contribution of the principal component factors in different samples [56].The PCA results clearly showed that two principal components were extracted from the PCA analysis of the concentration of volatile aroma compounds with OVA values greater than 1 in different treatments of 'Beibinghong' ice wine samples in a relatively independent space (Figure 5a), with a contribution of 32.9% for PC1 and 29.3% for PC2.Among the different varietal treatments, D, C, and H were located at the junction of one and four quadrants and were positive on PC2.G and D were located in the first quadrant of the score and were positive on PC1 and PC2.F was located in the second quadrant and was positive on PC1 and negative on PC2.E was located in the third quadrant and was negative on PC1 and PC2, and A and B were located in the junction of the third and fourth quadrants and were negative on PC1.This indicates significant differences in the volatile aroma compounds with OVA values greater than 1 in the 'Beibinghong' ice wine samples from different treatments.However, specific treatment groups also showed similarities in the aroma compounds with OAV values greater than 1. quadrant and was positive on PC1 and negative on PC2.E was located in the third quadrant and was negative on PC1 and PC2, and A and B were located in the junction of the third and fourth quadrants and were negative on PC1.This indicates significant differences in the volatile aroma compounds with OVA values greater than 1 in the 'Beibinghong' ice wine samples from different treatments.However, specific treatment groups also showed similarities in the aroma compounds with OAV values greater than 1.Clustering of the concentration of volatile aroma compounds in nine sample wines with OVA values greater than 1.Based on the sample heat map analysis (Figure 5b), it was seen that the red color indicated that the aroma compound component was highly expressed in the sample, and the blue color indicated that the aroma compound was expressed at a lower level in the sample.The concentration of volatile aroma compounds with an OVA value of greater than 1 varied considerably among the samples of each variety.In general, most volatile aroma compounds with high aroma intensity values also have high OAV [57]; moreover, the two methods can be mutually verified.However, a small number of volatile aroma compounds also have low OAV despite high aroma intensity values, or vice versa.In this experiment, screening volatile aroma compounds between different treatment groups using OAV values can more accurately extract the critical volatile aroma compounds that may affect the aroma of 'Beibinghong' ice wine.

Analysis of Volatile Compounds 0PLS-DA in Wine
OPLS-DA is a statistical method for supervised discriminant analysis [54,58,59].The contribution of each variable to the flavor of the wine was further quantified according to the variable important for the projection (VIP) in the OPLS-DA model [59].OPLS-DA was validated with 200 permutations and found that R2 and Q2 were more significant than the model after Y replacement (Figure 6b).Thus, the model predictions were reliable, and variables with VIP > 1 could be used as potential biomarkers.This experiment used the OVA values of compounds with an OAV value greater than 1 in the composition of wine samples from different varieties as Y variables for OPLS-DA analysis.
Screening of compounds with VIP values > 1 as marker compounds for wine (Table 7).The results revealed that the compounds that may affect the aroma differences at different SO 2 concentration treatments might be related to ethyl butyrate, ethyl propionate, ethyl 3-methyl butyrate-M, ethyl 3-methyl butyrate-D, and 3-methyl butyraldehyde.
the variable important for the projection (VIP) in the OPLS-DA model [59].OPLS-DA was validated with 200 permutations and found that R2 and Q2 were more significant than the model after Y replacement (Figure 6b).Thus, the model predictions were reliable, and variables with VIP > 1 could be used as potential biomarkers.This experiment used the OVA values of compounds with an OAV value greater than 1 in the composition of wine samples from different varieties as Y variables for OPLS-DA analysis.Screening of compounds with VIP values > 1 as marker compounds for wine (Table 7).The results revealed that the compounds that may affect the aroma differences at  Sensory evaluation is a crucial way for consumers to assess the quality of a wine [60].Sensory evaluation of wines influences consumer choice.As seen in the sensory evaluation, it can be found (Table 8) that the color and clarity of wine samples from the eight treatment groups were above 9 points, and their color and clarity scores were the highest in the sixth treatment group, and the aroma and taste scores were the highest in the third treatment group; the typicality of the wines from the third and the sixth treatment groups had the highest scores; the highest total score was obtained from the third treatment group, and the lowest total score was obtained from the first treatment group.In summary, the overall flavor of the 'Beibinghong' ice wines was the highest when SO 2 was added to the ice wines at an added amount of 30 mg/L.Adding too little or too much SO 2 can affect the wine's flavor and lead to certain defects.Pelonnier-Magimel E et al. [30] studied whether the Bordeaux quality wines without added SO 2 have their typicality to assess the organoleptic specificity of the wines without added SO 2 .Finally, it was found that wines without added SO 2 had a much higher frequency of defects than wines with added SO 2 (70 percent and 15 percent, respectively).Therefore, the absence of SO 2 in production can significantly impact the wine's flavor.If the wine is to maintain its typicality in production, as little SO 2 as possible can be added, while ensuring the flavor and bactericidal effect [34,61].

Conclusions
In this study, we investigated the effects of SO 2 addition on the fermentation process and volatile aroma compounds of 'Beibinghong' ice wine.The basic physicochemical properties of ice wine and the types and contents of volatile aroma compounds were analyzed in different SO 2 treatment groups, and the fingerprints of volatile compounds in ice wine brewed with different concentrations of SO 2 were constructed.In this study, it was found that the fermentation time of ice wine was shortest and the total acid content was relatively low at 10 mg/L, but the types and contents of volatile aroma compounds did not increase significantly; the total content of volatile aroma compounds of ice wine was highest at 80 mg/L.OPLS-DA calculated the VIP values, and the joint analysis of OAV and VIP values further identified five compounds: ethyl butyrate, ethyl propionate, ethyl 3-methyl butyrate-M, ethyl 3-methyl butyrate-D, and 3-methyl butyraldehyde, which were significantly different among the groups treated with different concentrations of SO 2 .These compounds might be the critical factors of the effect of SO 2 on the volatile aroma compounds of the 'Beibinghong' ice wine.These compounds may be critical factors in the effect of SO 2 on the volatile aroma compounds of 'Beibinghong' ice wine.Tasting of 'Beibinghong' ice wine with different SO 2 additions revealed that the overall flavor of 'Beibinghong' ice wine was the highest at 30 mg/L.In conclusion, the wine's best accumulation of nutrients and flavor was achieved at 30 mg/L of SO 2 additions.To some extent, this experiment reflects the differences in the quality of wines at different concentrations of SO 2 through micro-winemaking, which provided a reference for the development and promotion of wines, and the results of the study provided a basis for optimizing the fermentation process of the 'Beibinghong' ice wine.As SO 2 is a cheap and effective wine preservative, its complete replacement is not feasible.The optimum level of SO 2 addition in the production of 'Beibinghong' ice wine was investigated in order to minimize the excessive use of SO 2 in wine production.The results of this study provide some theoretical basis for optimizing the fermentation process of 'Beibinghong' ice wine and improving the quality of ice wine.

Figure 1 .
Figure 1.Fingerprints of volatile compounds of 'Beibinghong' ice wine under different treatments (Note: A-H are the amounts of SO 2 added, in the order of 10 mg/L, 20 mg/L, 30 mg/L, 40 mg/L, 50 mg/L, 60 mg/L, 80 mg/L, and 100 mg/L.Same below).

Foods 2024 , 22 Figure 2 .
Figure 2. HS-GC-IMS 2D spectrum (top view).Note: The background of the whole graph is blue, and the red vertical line at horizontal coordinate 1.0 is the RIP peak (reactive ion peak, normalized).The vertical coordinate represents the retention time (s) of the gas chromatogram, and the horizontal coordinate represents the ion migration time (normalized).Each point on both sides of the RIP peak represents a volatile organic compound.

Figure 2 .
Figure 2. HS-GC-IMS 2D spectrum (top view).Note: The background of the whole graph is blue, and the red vertical line at horizontal coordinate 1.0 is the RIP peak (reactive ion peak, normalized).The vertical coordinate represents the retention time (s) of the gas chromatogram, and the horizontal coordinate represents the ion migration time (normalized).Each point on both sides of the RIP peak represents a volatile organic compound.

Foods 2024 , 22 Figure 3 .
Figure 3. HS-GC-IMS difference spectrum of the sample.Note: Using A as a reference, the rest of the spectrum subtracts the signal peaks in A to obtain the difference spectrum between the two.Blue areas indicate less substance in this sample than in A. Red areas indicate more substance in this sample than in A. The darker the color, the more significant the difference.The difference between the samples can be seen from the above graph.

Figure 3 .
Figure 3. HS-GC-IMS difference spectrum of the sample.Note: Using A as a reference, the rest of the spectrum subtracts the signal peaks in A to obtain the difference spectrum between the two.Blue areas indicate less substance in this sample than in A. Red areas indicate more substance in this sample than in A. The darker the color, the more significant the difference.The difference between the samples can be seen from the above graph.

Foods 2024 ,
13,  x FOR PEER REVIEW 16 of 22 are significant and clearly distinguishable from other samples.As shown in the figure, it can be seen that the four treatments of A, B, H, and E clustered together, the treatment groups of C and D clustered together, and the groups of G and F clustered together, which indicated that the volatile compounds were similar between these treatment groups.The concentration of the aroma compounds was also somewhat different between the different treatments.

Figure 4 .
Figure 4. PCA analysis of the sample.

Figure 4 .
Figure 4. PCA analysis of the sample.

Figure 5 .
Figure 5. (a) PCA plots and (b) thermograms of volatile aroma compounds with OVA values greater than 1 in ice samples from different treatments of 'Beibinghong'.

Figure 5 .
Figure 5. (a) PCA plots and (b) thermograms of volatile aroma compounds with OVA values greater than 1 in ice samples from different treatments of 'Beibinghong'.

Figure 6 .
Figure 6.Analysis of different treatments of 'Beibinghong' ice wine samples OPLSA-DA ((a) Score chart of the OPLS-DA model of the sample, (b) Sample replacement test chart).

Figure 6 .
Figure 6.Analysis of different treatments of 'Beibinghong' ice wine samples OPLSA-DA ((a) Score chart of the OPLS-DA model of the sample, (b) Sample replacement test chart).

Table 3 .
Fermentation process and basic physicochemical indexes of 'Beibinghong' ice wine with different SO 2 additions.

Table 4 .
Content of anthocyanin in 'Beibinghong' ice wine with different SO 2 additions.

Table 5 .
Composition of volatile compounds in 'Beibinghong' ice wine treated with different concentrations of SO 2 .

Table 6 .
OAV analysis of major aroma compounds of 'Beibinghong' ice wine treated with different SO 2 additions.

Table 6 .
OAV analysis of major aroma compounds of 'Beibinghong' ice wine treated with different SO2 additions.

Table 7 .
Analysis of VIP values of aroma compounds in different treatments of 'Beibinghong'.Sensory Evaluation of 'Beibinghong' Ice Wine Brewed with Different Concentrations of SO 2