Skip to main content
NCBI home page
Food Chemistry: X logoLink to Food Chemistry: X
. 2025 Apr 9;27:102444. doi: 10.1016/j.fochx.2025.102444

Exploring the influence of oak barrel aging on the quality of Cabernet Sauvignon wine with a high ethanol content: Interactions with wood grain and toasting level☆

Yu Chen a,1, Xingmeng Lei a,1, Tianyuan Zhang a, Hulin Chen a, Lisa Tang a, Yi Shang a, Zhen Huang d, Yuyang Song a,b,c, Yi Qin a,b,c, Dongqing Ye e,⁎, Yanlin Liu a,b,c,⁎
PMCID: PMC12013396  PMID: 40264444

Abstract

This study investigated and addressed the harm caused by high ethanol levels (above 16 % vol) to the quality of industrial Cabernet Sauvignon wines through aging in oak barrels with three types of wood grains and three toasting levels. Our findings demonstrated that differences in chemical indexes between the initial and final wines were not entirely due to the oak barrels. Barrels with medium grains and heavy toasting stabilized the wine's colour and minimised the loss of red tones; this effect was associated with phenolic components. Wood-unrelated volatiles were responsible for the considerable variations in volatile profiles; these differences were characterised by a decrease in alcohols, alongside an increase in ethyl esters, enhancing the fruity and floral attributes of the final aged wines. High-ethanol wines aged in medium-grained and heavily toasted barrels exhibited the best characteristics. These findings provide information to address concerns that a high ethanol content compromises wine quality.

Keywords: High-ethanol wine, Oak barrel aging, Wood grain, Toasting level, Oak-unrelated volatile

Highlights

  • •

    Industrial Cabernet Sauvignon wines with high-ethanol content were aged for one year in 225-L oak barrels.

  • •

    Barrels with medium grain and severe toasting level stabilized the redness of high-ethanol wine.

  • •

    Barrels with medium grain and severe toasting level exhibited the improved potential in ‘floral’ and ‘toasted’ flavour of wine.

1. Introduction

The excessive accumulation of sugar resulting from climate warming during the grape ripening is a crucial element restricting the rapid development of the wine industry (Hranilovic, Gambetta, Jeffery, Grbin, & Jiranek, 2020). This hampering effect is mainly due to the high ethanol content resulting from excessive sugar, affecting wine quality. To commence, wines with a high ethanol content were noted to have more hot and bitter tastes (Heymann et al., 2013). Under winemaking conditions, the colour development effect between tannin and malvidin-3-O-glucoside will be affected by high ethanol content, causing the colour of the wine to become unstable (Vignault et al., 2019). Moreover, wines with a high-ethanol content have the ripe phenolic level, but their colour stability has been affected in certain warm wine-producing regions like Ningxia and Xinjiang of China, as well as Australia (Fanzone et al., 2020; Ye et al., 2016). On the other hand, high ethanol content affects the release of most volatile compounds in wines, such as furfurals, guaiacol, and vanillin associated with oak (Pollon et al., 2023). The perception of fresh fruity and floral aromas were also lost as the alcohol concentration increased, resulting in degraded aroma complexity (King, Dunn, & Heymann, 2013). Thus, elucidating how a high ethanol content would avoid damaging the quality of wine remains a challenge.

Aging in oak barrel is a traditional winemaking technique that yields high-quality wine (Feng, Martínez-Lapuente, Ayestarán, & Guadalupe, 2023). The characteristics of barrels affect wine quality and are associated with the oak species (Fernández de Simón, Sanz, Cadahía, Poveda, & Broto, 2006), the oak origin (Cerdán, Mozaz, & Azpilicueta, 2002), the grain size (Bosso et al., 2008), and wood toasting techniques (Acuña, Gonzalez, de la Fuente, & Moya, 2014). Studies over the past two decades have provided information on the associations between aged wine quality and certain characteristics of the oak barrel. Bosso et al. (2008) stated that wines had a richer ellagitannin content and a lower content of wood-related volatiles after aging at the concentration of 14.28 % (v/v) ethanol, when the oak barrel had a larger wood grain and was treated at a lower toasting temperature. A comparative study on botanical origins and toasting levels, found that French barrels were superior to American oak barrels in their ability to extract phenolic compounds during aging for one year aging, approximately with a concentration of 13.35 % (v/v) ethanol (Navarro et al., 2016), while heavier toasting level was propitious for this extraction. On the other hand, untoasted barrels were related to the increase in several odor active compounds (mainly alcohols and terpenes) in Muscat Ottonel wines with 12.49 % (v/v) ethanol (Călugăr et al., 2020). Although these results have provided insights into the influence of oak barrel aging on the quality of several dry red wines, few studies have focused on the quality control and improvement of oak barrel aging on Cabernet Sauvignon wines with a high-ethanol content [above 16 % (v/v)]. Meanwhile, how the interaction among the different characteristics of the oak barrel (particularly toasting and wood grain) affects the evolution of quality-related molecules during wine aging is unclear.

Cabernet Sauvignon, an ancient wine-grape variety, has been often used to produce aged dry red wine, which is growth in Shandong, Ningxia, and Xinjiang of China (Ye et al., 2016). This study aims to estimate the effects of wood grains (wide, medium, fine), toasting levels (light, medium, severe), and their interactions on the non-volatile compounds of high-ethanol wines during one year aging. Furthermore, wine volatile profiles from different aging durations in nine oak barrels were analyzed using headspace solid-phase microextraction/gas chromatography–mass spectrometry (HS-SPME/GC–MS), while the evolution of wood-unrelated volatiles was explored during aging at different scales. In addition, structural equation modeling (SEM) and a correlation matrix were used to elucidate the relationships between the non-volatiles and volatiles of wine through random forest analysis and orthogonal partial least squares discriminant analysis (OPLS-DA). As a result, the direct and indirect effects of phenolic compounds on wine volatiles and chromatic parameters were identified with multiple scales. In this study, the initial wines from industrial winemaking have aged in oak barrels with commercially basic volume capacity, and the results are expected to help improve the production practices and quality of Cabernet Sauvignon wine with a high ethanol content.

2. Materials and methods

2.1. Wine aging and sampling

The unaged Cabernet Sauvignon wine used for this test was produced at an industrial scale and fermented to 16 % (v/v) alcoholic content. All grapes were grown at the experimental station of the Qingtongxia wine region of Ningxia in China, and were of the 2019 vintage. The chemical parameters of the wine before aging are listed in Table S1. Nine groups of 225-L oak barrels were purchased, one of each kind of wood to be tested (i.e., with different combinations of wood grains and toasting levels, Supplementary Table S2). These casks were preconditioned before use by vapor fumigation for 30 min, followed by water storage for one month. Monitoring was performed by sampling 80-mL of wine each three month during one year aging. We used a glass pipette (54 cm) to collect the wine samples. An 80-mL sample was taken as a composite from the top, middle, and bottom sections of each barrel to capture potential variations in the contents across different locations. Three flexcubes with the capacity of 225-L were used as the control and the wine stored in them was sampled after one-year aging. Considering the risk of evolution, the samples were stored at 4 °C until analyses (three days at most) and all of them were performed in duplicate. Young wine without aging was considered the initial sample. To replenish any loss of wine in barrels, we utilized the initial wine sample for the refilling operation. This initial wine was stored in sealed stainless steel containers and was protected with nitrogen gas to prevent excessive oxidation prior to its use in topping up the barrels.

2.2. Chemicals and reagents

All chromatographic solvents and standards were of HPLC grade and purchased from Sigma-Aldrich (St. Louis, Missouri, USA). Sulfuric acid was also purchased from Sigma-Aldrich. All reagents had a purity greater than 98 %. The water used for all analyses was commercially available purified water (Hangzhou Wahaha Group Co. Ltd., China).

2.3. Determinations of oenological reference parameters

Upon sampling, high performance liquid chromatography (HPLC), with a Shimadzu VWD ultraviolet detector for organic acid (PDA, 210 nm) and a refractive index detector (RID-20 A), was used to determine the glucose, fructose, glycerol, ethanol, tartaric acid, succinic acid, malic acid, lactic acid and acetic acid in each 20-μL sample (Chen et al., 2023). Briefly, an LC-2050C 3D (Shimadzu, Japan) instrument fitted with an Aminex HPX-87H column (300 mm × 7.8 mm; BioRad, Hercules, CA, USA) was used. The column was kept at 60 °C for a 35-min run. The 10 mM H2SO4 was used as the eluent with a flow rate of 0.6 mL/min. The external calibration curves (R2 > 0.999) were plotted using the LabSolution software (Shimadzu, Japan, version 5.0) for quantifying analytes.

2.4. Determinations of phenolic and colour parameters

For measuring the phenolic parameters in samples, total phenolic, anthocyanin, and flavanols of wines were determined respectively according to the Folin–Ciocalteu method, the pH differential method, and the p-dimethylamino cinnamaldehyde method (Tu et al., 2022). In addition, quercetin and caffeic acid were respectively used as the standard for measuring total flavonols, and tartrate ester, while the methods are followed as the previous descriptions (Ramos-Pineda, García-Estévez, Dueñas, & Escribano-Bailón, 2018). A chrome meter (CR 300 Minolta Chromameter, Osaka, Japan) was analyzed the lightness (L*), redness (a*), yellowness (b*), chroma (c*) and hue (H*) of the wines. In addition, the total colour difference (ΔE*) values were calculated according to the previous equation (Guler, 2023).

2.5. Analysis of volatile compounds

Volatile composition of wine samples was identified and quantified by solid phase microextraction-gas chromatography–mass spectrometry (SPME-GC–MS) method with some modifications (Lu, Cheng, Lan, Duan, & He, 2024). Briefly, each SPME vial (20 mL) was assembled with the 5-mL wine samples and 1.2 g NaCl. A 10-μL internal standard (4-methyl-2-pentanol) was added at a concentration of 1.030 g/L prior to sample analysis. The analysis was carried out using a Trace 1610 (Thermo Fisher Scientific, USA) gas chromatograph equipped with a TriPlus RSH SMART (Thermo Fisher Scientific, USA) auto-sampler and an ISQ 7610 (Thermo Fisher Scientific, USA) mass spectrometer for peak detection and compound identification. After equilibration at 40 °C for 30 min, the auto-sampler was operated in the SPME mode, utilizing a DVB/CAR/PDMS fiber (50/30 μm; Thermo Fisher Scientific, USA) to extract the volatile compounds by shaking at 250 rpm for 30 min. The compounds were separated on a TG-WAX column (60 m × 0.32 mm × 0.25 μm, Thermo Fisher Scientific, USA). The chromatographic separation was performed as follows: desorption for 8 min, the column was kept at 50 °C for 5 min, then the temperature was increased to 230 °C at a rate of 8 °C/min, and the final temperature was maintained for 10 min. The mass acquisition range was set from 29 to 350 m/z for quantitative analysis of detections. Subsequently, the volatile concentrations were calculated by the established calibration curves obtained using a synthetic wine [16.0 % (v/v) ethanol, 5 g/L of tartaric acid, and pH adjusted to 3.5 with 1 M sodium hydroxide]. The estimate of volatile compounds in wine aroma profiles was determined through odor activity values (OAVs), calculated using the formula established previously (Chen et al., 2022). The thresholds used in the calculation of OAV were referred to study from Welke, Zanus, Lazzarotto, and Alcaraz Zini (2014).

2.6. Sensory description analysis

The Ethics Review Executive Committee of Northwest A&F University granted the ethical approval for the involved human subjects in this study. Wines after one year aging were generated a sensory evaluation experiment. Thirty wine fans, consisted of 15 males and 15 females from Tasting Panel (TP) of College of Enology in Northwest A&F University (China), voluntarily constituted a sensory panel for this study. All panelists were aged from 20 to 50 years, who was trained using a 54-aroma kit (Le Nez du Vin, France) for at least two months and generated the sensory tests with their formal consent in full accordance with the voluntary, free principle, and ethics policy of 1975 Declaration of Helsinki. Prior to the sensory assessment, all participants underwent a screening process to ensure their accuracy and repeatability in evaluating the samples. The tasters were selected based on their sensory experience and training, and only those demonstrating a sufficient level of expertise were included. To assess repeatability, multiple test runs were conducted (n = 3), and only the data from participants who showed consistent responses across these trials were used in the final analysis. Considering the varietal heterogeneity, a portion of industrial Cabernet Sauvignon wines were selected for panel training before the formal sensory evaluation. All participants were asked to score the sensory contributions at a five-point intensity scale (1 = very weak, 2 = weak, 3 = medium, 4 = intense, 5 = very intense). Then, the modified frequency (MF, %) was calculated following as the previous formula, while was characterised the wine sensory profile (Chen et al., 2022).

2.7. Statistical analysis

Data from each triplicate sample were normalized and visualized using the Origin 2022 software. The physicochemical, chromatic, phenolic, and volatile characteristics during wine aging were analyzed using K-means clustering in R 4.2.2 with the ‘NbClust’ package. The correlations were carried out with the ‘corrplot’ package. The volatiles of wine from nine oak barrels after one year of aging were subjected to a two-way analysis of variance (two-way ANOVA) by the software SPSS 26.0 (IBM, USA) with a Tukey's post-hoc comparisons. The main effects included were the wood grain, the toasting level, and their interactions. Principal component analysis (PCA) and hierarchical clustering analysis (HCA) of volatiles with OAV > 1.0 was generated and visualized by SIMCA 14.0, alongside the orthogonal partial least squares discriminant analysis (OPLS-DA). The ‘ade4’ and ‘vegan’ packages in R 4.2.2 were generated the principal co-ordinates analysis (PCoA) and permutational multivariate analysis of variance (PERMANOVA), respectively. Structural equation model (SEM) using a priori approach, was often used to help characterise and comprehend interactions of variant subject within a complex network (Chen, Lei, et al., 2023). Here, SEM was used to quantify the direct and indirect relationships between phenolic parameters and chromatic characteristics, alongside different volatile drivers. Volatile drivers comprised volatile composition for wood grain (WG) and toasting level (TL) with variable importance for predictive components (VIP) values >1.0 by OPLS-DA and OAV > 1.0. Moreover, the top 10 important volatile predictors by the Random Forest regression model using R 4.2.2 software with a ‘randomForest’ package, were also among the volatile drivers. R 4.2.2 software was constructed the SEM, followed to estimate the model effectiveness by the previously calculated indexes (χ2/df, the goodness-of-fit test in maximum likelihood; RMSEA, the root mean square error of approximation; GFI, the goodness of fit index) (Chen, Lei, et al., 2023).

3. Results

3.1. Changes in physicochemical and chromatic characteristics during Cabernet Sauvignon wine aging

The general physicochemical parameters of the young Cabernet Sauvignon wine followed the Chinese wine industry (GB/T 15037–2006) and International Vine and Wine Organization (OIV) standards (Table S1). The general parameters of the samples were affected by aging, especially regarding the fructose, tartaric, succinic, malic, lactic, and acetic acid contents (Fig. 1A). The initial fructose concentration (2.72 g/L) was 9.78-fold that of glucose (0.28 g/L) (Fig. 1A, Table S1). Few studies have focused on the changes of sugar composition during wine aging. However, a great variance was observed after the 1-year aged Cabernet Sauvignon wine, as the glucose content was 3.01-fold higher than that of fructose (0.84 g/L). The fructose content fleetly decreased within the first 6 months of aging and gradually stabilized after nine months, while the glucose concentrations remained stable during aging. Five short chain organic acids were detected and quantified in different oak barrels during aging (Fig. S1). Notably, the tartaric acid concentrations suffered a decrease after three months of barrel aging, except for the X6, X7, and X10 barrels, with a mean gain of around 23.02 % compared to the young Cabernet Sauvignon wine. The succinic acid content fluctuated during the one-year aging process, but returned to its original level. Lactic and acetic acid were the acids that increased its level in all one-year-aged samples with respect to the initial wines.

Fig. 1.

Fig. 1

Open in a new tab

Cabernet Sauvignon wines from different aging duration have differential chemical and chromatic characteristics in nine oak barrels, depending wood grains and toasting levels. Evolutions of generally physicochemical (A), chromatic (B), and phenolic (C) parameters using statistical method. Generally physicochemical parameters included fructose, tartrate acid, succinic acid, malic acid, lactic acid, and acetic acid. Chromatic characteristics based on indexes of L* (lightness), a* (red hue), b* (yellow hue), ΔE (colour difference), c* (chroma), and H* (tone). Phenolic parameters referred to total phenolic, total anyhocyanin, toal flavonol, total flavanol, and total tartrate ester contents. Figure legend demonstrated the features of nine oak barrels across different wood grains and toasting levels. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The CIELab coordinates of all samples (L*, a*, b*, ΔE, c*, and H*), are shown in Fig.1B. Colour differences were observed during wine aging in different oak barrels. Furthermore, all wines presented an increase in lightness (L*) and yellow colors (b*) but a decrease in the red colors (a*) during the first three months. Although the colour parameters showed inconsistent trends during wine aging, after one year aging, they reached the same level regardless of the type of oak barrels used (Fig. 1B, and Fig. S1). For instance, the parameter c* significantly decreased in most wine samples from the 6th to 9th month (ANOVA, p < 0.05), except for the wines in X10 and X15 (Fig.1B). Furthermore, one-year aging significantly decreased (p < 0.05) the lightness (L*), red hue (a*), yellow colors (b*), chroma (c*), and tone (H*), while the colour difference (ΔE) had a distinct increase.

3.2. Variations of wine phenolic parameters by wood grain and toasting level

Fig. 1C shows the concentration variations in phenolic parameters (total phenolic, total anthocyanin, total flavanol, total flavonol, and total tartrate ester) during Cabernet Sauvignon wine aging. The total phenolic content ranged from 274.37 ± 2.93 mg/L to 822.35 ± 29.10 mg/L, depending on the type of oak barrel and wine aging duration. Levels of total phenolic, total anthocyanin, total flavanol, and total flavonol were increase during the first 3-months aging (Fig. 1C), of which the first three parameters were more affected by wood grains, whereas total flavonol was more affected by the toasting levels (Fig. S1). The increase in total anthocyanin and total flavanol continued through the first 6 months. A durative decrease was only observed for total tartrate ester parameter, from 257.38 ± 2.34 mg/L to 132.97 ± 8.80 mg/L.

PCA was used to classify the different duration of Cabernet Sauvignon dry red wine aging based on the general, colour, organic acid, and phenolic parameters (Fig. 2A). The explanation of first two principal components has reached to 97.94 % of the total cumulative variance contribution (left side of Fig. 2A). The first principal component (PC1) accounted for 93.54 % of the total variance and` separated the initial and final wines from other samples, indicating that aging treatments affected the chemical and chromatic properties of Cabernet Sauvignon wine. Furthermore, the initial and final wines were also separated by the second principal component (PC2), accounting for 4.40 % of the total variance. Consequently, a limited influence presented in the difference of chemical and chromatic characteristics between T1 and T5 samples after wine aging. The variations in chromatic (a*, b*, c*, H*, ΔE), phenolic (total phenolic, flavanol, flavonol), and generally chemical parameters (glucose, succinic acid) separated the T1 and T4 samples for the first three months of aging (right side of Fig. 2A). Nevertheless, the T2 and T3 wine samples seemed to be indistinguishable, confirming the slight development of Cabernet Sauvignon wines according to the glycerol, lactic acid, L*, and acetic acid indexes (Fig. 1).

Fig. 2.

Fig. 2

Open in a new tab

Their interactions between chemical parameters and chromatic characteristics. A, principal component analysis of the chemical parameters (physicochemical, chromatic, and phenolic) in Cabernet Sauvignon wines: characteristics of oak barrels and aging duration (left), as well as correlation circle (right). B, heatmap of correlation matrix among the chemical parameters, involving the general, organic acid, chromatic, and phenolic parameters.

3.3. Volatile evolutions in different oak barrels during Cabernet Sauvignon wine aging

Fig. 3 presents the volatile metabolome profiles of Cabernet Sauvignon wines, highlighting that the types of oak barrel resulted in substantial modifications of the wine aromatic characteristics during aging. Notable differences were presented between the initial and aged wine samples, with a greater separation of volatile compositions (Fig. 3A). Besides the integrally principal component analysis (PCA), a slight separation of wine samples aged for different periods was observed along PC1 and accounted for 22.3 % of the total explained variance. A total of 61 quantified compounds were categorized into eight chemical families, including alcohols (higher alcohols, C6 alcohols), volatile fatty acids, esters (ethyl, acetate, and others), terpenes, and other compounds (Table S3, Fig. 3B). Regardless of the oak barrel types (wood grains and toasting levels), a decrease of higher alcohol and C6 alcohol content level was appeared during wine aging, while an increase of ethyl ester was presented. Some other compounds, such as acetate esters, terpenes, volatile fatty acids, etc., fluctuate during the aging process. Wine volatile profiles were clearly separated by PCoA using Bray–Curtis dissimilarity algorithm for the aging duration. As a result, the first two components explained 89.24 % of the total variance for Cabernet Sauvignon wine samples (PERMANOVA, R2 = 0.9976, p = 0.034; Fig. 3C). A separation of samples from the first three months from those of the final three months was seen along PC1 and accounted for 62.69 % of the total explained variance. The differences in volatile compositions of the same batch of wine were most notable in the sixth month of aging because they overlap with samples from other periods. The constructed OPLS-DA model was further used to obtain detailed information on the marked chemicals that contributed to the evolution of volatile profiles during wine aging (Fig. 3D, Table S3). Twenty-two volatiles were identified into the time biomarkers during wine aging (VIP > 1.0), including 2-ethyl-1-hexanol (A32), 1-hexanol (A22), 3-methyl-1-pentanol (A19), (E)-3-hexen-1-ol (A23), 2-methoxy-3-(2-methylpropyl)-pyrazine (A35), terpinen-4-ol (A42), octanoic acid (A56), etc..

Fig. 3.

Fig. 3

Open in a new tab

Volatile profiles of Cabernet Sauvignon wines in different aging durations. A, principal component analysis of volatile compounds in wines from different aging durations. The arrows marked the direction of samples with the aging duration. B, cluster heatmap of different chemical families in concentration level with different aging durations. C, principal coordinate analysis (PCoA) of volatile profiles based on Bray-Curtis distances among wine aging durations across the different oak barrels, with a permutational multivariate analysis of variance (PERMANOVA). D, the orthogonal partial least squares discriminant analysis (OPLS-DA) of volatile profiles demonstrated the important components with VIP (variable importance for predictive components) > 1.0 for aging durations.

The evolution kinetics of the 61 volatile compounds were analyzed using K-means clustering, which resolved eight distinct clusters (Fig. 4). Volatile levels with Cluster 4, 5, and 7 displayed rapid increases in the first three months of aging, followed by the related compounds in Cluster 6, 8, and 1. Among them, Cluster 4, 6, and 7 presented declines at different rates and extents in the following three months of aging. The concentrations of volatiles associated with Cluster 2 and 3 decreased sharply during the first three months of aging and then stabilized or decreased slightly. The obtained information on wine volatiles was utilized to create K-means clustering, unveiling the changes that occur during aging, influenced by wood grains (Fig. S2) as well as toasting levels (Fig. S3) of oak barrels. In terms of wood grain, albeit the presence of multiple evolution clusters, all compounds were effectively categorized into six distinct clusters. Cluster 1, 2, and 3 had the highest concentrations in T2 and T3 wines aged in oak barrels with wide grains; the volatiles comprised six higher alcohols and seven ethyl esters. Here, a significant increase of two higher alcohols (A51, benzyl alcohol; A53, phenylethyl alcohol) was observed in Cluster 1 (i.e., wide and fine grain, Fig. S2A and C) and Cluster 3 (i.e., medium grain, Fig. S2B) during the first three months of aging. Within the toasting levels, the two higher alcohols were categorized into Cluster 1 of the three clustered heatmaps (Fig. S3). These increases were more notable in the samples aged in the barrels with wide and fine grains than in those aged in barrels with medium grains. Isoamyl alcohol (A13) with the most prevalent higher alcohol in all wines from three typical barrels (> 72 % in total higher alcohols). Two C6 alcohols [1-hexanol (A22); (E)-3-hexen-1-ol (A23)] were detected into the Cluster 6 and their content level slightly fluctuated after the initial sharp decline in the first three month of aging (Fig. S2). Notably, the three volatiles were consistently grouped in the same cluster, regardless of the toasting level (Fig. S2). Compared to the other two compounds, the final reduction of (E)-3-hexen-1-ol was more affected by the wood grain than by the toasting level (Fig. 5A). The contents of most of higher alcohols tended to decrease, starting from the first three months or the 3–6 months period of aging, which was consistent with the above results (Fig. 3B).

Fig. 4.

Fig. 4

Open in a new tab

K-means clustering of volatile kinetics in wines resolved eight profiles, with wine aging duration. The red strings show the mean values of K-means clustering profiles without the corresponding volatiles symboled, and ‘n’ was the number of volatiles in each clustering profile. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5.

Fig. 5

Open in a new tab

Multivariate analysis of wine volatile profiles after the twelfth months of aging. A, variation in volatile composition of Cabernet Sauvignon wines. The normalized concentration of each volatiles identified in wines was presented in clustering heatmap (up). Code represented the volatile compounds showed in Table S3. Percentages of variation explained by wood grain, toasting level, their interaction and residual as determined by two-way ANOVA (down). B (the characteristics of oak barrels) and C (the correlation circle), principal component analysis of relevant volatiles (OAV > 1.0) of Cabernet Sauvignon wines from nine oak barrels coopered with different grain/toasting treatments after one year of aging. D, hierarchical clustering of wines aroma compounds with OAV > 1.0, controlled as the initial wine and the flexcube.

Regarding esters, a comparable number of compounds was detected (22/61), of which either ethyl acetate (A1) or ethyl lactate (A21) predominated in the wines (Table S3). The finer grain was conductive to reduce the ethyl acetate level (Fig. S2), while the case was also in wines with the more severe toasting level (Fig. S3). Because of the influences of the finer grains and heavier toasting levels, ethyl acetate showed the highest percentage decrease during early aging, from 163.72 to 102.66 mg/L, although the final concentration recovered somewhat (140.06 mg/L). Nevertheless, since the total effects of wood grain and toasting levels on ethyl acetate was much smaller than their interaction effect (Fig. 5A), it was believed that the inter-individual variation trend of ethyl acetate is more worthy of study. Despite the increase in lactic acid (Fig. 1), barrels with finer grain and heavier toasting favoured the accumulation of ethyl lactate because of the availability of lactic acid as its precursor (Fig. S2 and S3). A dramatic decrease in the concentration of ethyl butyrate (A5) was seen in the first three months of Cabernet Sauvignon wine aging except for the wines aged in barrels with wide grains, while ethyl hexanoate was also clustered together with ethyl butyrate (Fig. S2). The changes in contents of ethyl octanoate (A27) and ethyl decanoate (A43) levels were dissimilar. Additionally, in contrast to ethyl esters, the concentrations of acetate esters depended more on the variance in the toasting level than in the wood grain (Fig. 5A).

In addition, barrels with finer grains promoted the quick increase in wine fatty acids, mainly octanoic acid (A56) and n-decanoic acid (A58), which were present at higher concentrations in wines from the third to the sixth months of aging than in other wines (Fig. S2). For all grain and toasting modalities, terpinen-4-ol (A42) was the terpene compound with the highest loss during aging, while the increases in several monoterpenoids (A18, linalool; A50, β-ionone; A60, trans-farnesol) were observed (Fig. S2 and S3).

After aging for one year, the differences in the volatile profiles were observed among the different containers and the Flexcubes used as a control. The concentrations of total ethyl esters, volatile fatty acids, and terpenes were higher in aged wines than in the initial wines (Fig. S4). However, the opposite was observed for total higher alcohols and C6 alcohols. Besides, total acetate esters were affected by the different combinations between grains and toasting levels. The two-way ANOVA analysis revealed the effects of the wood grain, toasting level, and their interactions (lower part of Fig. 5A). The concentrations of (E)-3-hexen-1-ol (A23), 2-propylheptanol (A47), ethyl benzoate (A48), and ethyl isobutyrate (A3) were significantly affected by wood grain (p < 0.05), while nonanal (A25), nerol (A31), and isobutyl isovalerate (A12) were associated with variations of toasting level. For most compounds, their concentration levels were jointly regulated by these two factors because their interaction explains more than 50 %. Volatile profiles (volatiles with OAV > 1.0) were subjected to PCA. The samples aged in barrels with medium grains were separated from the different toasting levels through PC1, which accounted for 43.4 % of the total explained variance (Fig. 5B). The 4 × 9 samples (medium grain and heavy toasting) were differentiated from other samples on PC2, which explained 16.8 % of the total variance and was associated with the highest contents of 1-pentanol, isobutyric acid, and two ethyl esters (ethyl caprylate and ethyl 3-hydroxybutyrate). Hierarchical cluster analysis (HCA) was used to estimate the similarities between the volatile composition of wine samples, classifying these samples into five main categories (Fig. 5D). The initial wines were more distant to aged samples regarding their volatile compositions, which corroborated the results from the PCA. Moreover, the category that comprised most samples did not follow the linear trend of grains and toasting level in volatile development (from wide to fine, or from light to heavily toasted).

3.4. Wine sensory analysis of wines after one-year aging

Despite a comprehensive analysis of chemical characteristics was generated, sensory analysis was not able to be devoid for evaluating the wine quality. Sensory profiles revealed significant differences in eight attributes after 12 months of aging (PERMANOVA, F = 3.6715, R2 = 0.3146, p = 0.023), with the largest variation detected in ‘red fruity’ and ‘plant’ characteristics (Fig. 6A). The highest ‘red fruity’ attribute was scored in X9 wines followed by X3 and X7 samples. Wines from X3 and X7 barrels had a fruity flavour (mainly ‘black fruity’ attribute), whereas wines aged in X6 barrels presented more floral notes. This result indicates the oak barrels with finer grain benefitted the retention of fruity flavour. Besides, the distinct creamy flavour was correlated to the oak barrels with wider grain. The X8 sample had typical dried fruit and baked flavors, while the botanical and spice attributes are rated in the middle. Different from this, the aroma of plants and spices in the X9 wine sample was more prominent.

Fig. 6.

Fig. 6

Open in a new tab

A, sensory contributions of the experimental Chardonnay wines from wines after one year of aging. B, evaluating the direct and indirect effects of phenolic parameters on wine volatile profiles at multiple levels using structural equation modeling (SEM). Volatile drivers of wood grain (WG) and toasting level (TL) comprised volatile compounds (VIP > 1.0, and OAV > 1.0) and top 10 important microbial predictors based on Random Forest analysis. Numbers on arrows are standardized path coefficients and indicative of the effect size of the relationship. C, correlation analysis between physicochemical, phenolic, and volatiles (OAV > 1.0). Significant levels of each path are <0.001 and < 0.01, respectively expressed as *** and **, and no significance was unmarked.

3.5. Relationships between chemical parameters and volatile profiles

To disentangle the interactions between different chemical components, structural equation modeling (SEM) and Pearson correlation coefficient analysis by were used to assess the relationship effect between the chemical parameters and selected volatile compounds (Fig. S5 by Random Forest analysis, and OAV > 1.0). SEM quantified the direct and indirect effect of phenolic parameters on wine chromatic and volatile profiles (Fig. 6B). The results of the maximum likelihood goodness-of-fit test (χ2/df = 1.768) and the root mean square error of approximation (RMSEA = 0.081) showed a successful fit for the constructed model. Phenolic indexes directly drove the development of wine volatiles by affecting the toasting level volatiles (TL volatiles, path coefficient = 0.452, **), and wood grain volatiles (WG volatiles, path coefficient = 0.180, ***), while the volatile compositions were affected by their interactions. WG volatiles had the highest directly negative effects on the TL wine volatile profiles (path coefficient = −0.858, ***). By contrast, the TL volatile profiles had the lowest directly positive effects on WG volatiles, with no significance (path coefficient = −0.082). The chromatic parameters were affected directly and significantly by phenolic indexes (path coefficient = 0.144). Thus, the volatile profiles are mainly affected by the phenolic parameters during wine aging, driven by the type of wood grain, toasting level, and their interactions, consistent with the previous results of two-way ANOVA.

4. Discussion

In this study, the effect of barrels with different characteristics on the non-volatile and volatile compositions were evaluated during aging (0, 3, 6, 9, and 12 months, respectively) of Cabernet Sauvignon wine with high-ethanol content (> 16 % v/v). The barrel characteristics included two orthometric experimental conditions: wood grains (wide, medium, fine) and toasting levels (light, medium, heavy). The wines were analyzed over a period of twelve months of aging in term of non-volatile and volatile profiles.

4.1. Effects of wood grain or toasting level on wine non-volatile compositions

As shown in the results of physicochemical characteristics, the Maillard reaction drives the reduced hexose contents during wine aging (Kertsch, Wagner, & Henle, 2023), lowering the fructose levels regardless of the oak types. As one of secondary products, acetic acid is a common organic acid that appears during the thermal degradation of oak (García-Moreno et al., 2021), while oak voids provide an environment conducive to oxidation reactions (Acuña et al., 2014), increasing the acetic acid content during wine aging. Previous research has shown that oak can reduce tartaric acid content and increase acetic acid content in wine (Dumitriu et al., 2019), while these changes were limited by the type of oak barrel in our study. For example, in the first three months, the tartaric acid content of wines from X6, X7, and X10 oak barrels increased, while the remaining samples conformed to the previous description (Fig. 1A).

Regardless of the type of barrel, the lightness (L*) and yellow (b*) values increase and the red (a*) value decreases in the first three months of wine aging, which may be correlated with the extraction of hydrolyzable tannins (ellagitannins and gallotannins) as well as several phenolic compounds from oak barrels (Li & Duan, 2019). Similarly to other aged wines, phenolics appeared as brown compounds resulting from condensation (Boido, Alcalde-Eon, Carrau, Dellacassa, & Rivas-Gonzalo, 2006), adjusting the colour parameters. The b* value increases rapidly in the early stages of aging (the first three months), making the wine closer to yellow (Fig. 1B). This is mainly due to the fact that the dissolved oxygen in the wine oxidizes ethanol to produce acetaldehyde, which promotes the formation of yellow polymeric pigments (Dumitriu et al., 2019). Many studies have shown that a decrease in red (a*) value during wine aging is inevitable due to the loss of free anthocyanins (Pérez-Magariño & González-San José, 2006; Yu et al., 2022). In our study, changes of red value during the first three months were not affected by the toasting levels, whereas wider grains resulted in more severe red losses.

For phenolic parameters, ellagitannins derived from galloyl units esterified to a sugar core, were transferred from oak wood into wine and explain why several phenolic parameters (total phenolic, total anthocyanin, total flavanol, and total flavonol) increased during the first three months of aging (Del Fresno, Morata, Loira, Escott, & Suarez Lepe, 2020). Except for them, a durative decrease was only observed for total tartrate parameter because of the hydrolysis of tartrate ester and the formation of tartrate during the aging process (Hixson, Hayasaka, Curtin, Sefton, & Taylor, 2016). These observed shifts in phenolic parameters may be due to oxidation processes and polymerization reactions during wine aging (Andrew & Laurie, 2006), and their intensity is determined by the amount of oxygen that penetrates through the gaps in the oak barrel (Sánchez-Gómez, del Alamo-Sanza, Martínez-Martínez, & Nevares, 2020). Similar evolutions were also observed in Tempranillo and Graciano dry red wines (Chira, Pacella, Jourdes, & Teissedre, 2011; Monagas, Bartolomé, & Gómez-Cordovés, 2005).

Linking composition to chromatic characteristics has been traditionally performed in wine analysis because the direct interaction among chemicals generates predictable visual changes (Li & Duan, 2019; Sánchez-Gómez et al., 2020). Thereby, we elucidated the association between phenolic and physicochemical parameters, along with chromatic characteristics (Fig. 2B). Although this might establish some correlations that are difficult to see, it is still meaningful to mine some other key information from them. Colour parameters, except for ΔE values, were positively correlated with total phenolic, total flavonol, total tartrate ester, fructose, and malic acid at a higher significance, and confirmed the crucial influences of these chemicals in modifying the colour of wine during aging (García-Moreno et al., 2021; Hixson et al., 2016). On the other hand, lactic acid and glucose presented highly negative correlations with the above indexes, while lactic acid was positively correlated with glucose. The high acidity derived from organic acid (mainly malic and tartaric acid) positively contributed to the phenolic and chromatic characteristics of Cabernet Sauvignon dry red wine (Tian et al., 2023). Albeit lactic acid might provide the satisfied mouthfeel (Sun, Chen, & Jin, 2018), which contradicts the previous observation (Tian et al., 2023). These findings emphasise the internal relationships between the sensory properties and chemicals.

4.2. Wood grain and toasting level interactions shift volatile compositions during the high-ethanol wine aging

Together with effects of non-volatiles, the volatile aroma profiles were also the mutation of the important sensory impressions affected by wine aging process for the young wines. In the past ten years of research, people have almost exclusively focused on the changes in oak-derived volatile compounds during aging, but have ignored the changes in volatile compounds unrelated to oak during wine aging. In fact, in addition to oak-derived volatile compounds, oak-unrelated volatile compounds in raw grape wine are also an important profile of flavour differences (Dumitriu et al., 2019; Feng et al., 2023). Therefore, selecting oak barrels with different combinations of factors and exploring how these factors affect the evolution of wood-unrelated volatiles is crucial to obtain high-ethanol wines with excellent flavour and quality.

The aging of wines in oak barrels significantly modifies their volatile compound profiles, with variations that are influenced by the interaction between barrel type (wood grains and toasting levels) and wine chemicals. As previously discussed, the loss of water or ethanol during barrel aging can promote esterification reactions, which are critical in shaping the aromatic characteristics of wine (Feng et al., 2023). These reactions align with the findings of the current study, where ester levels, particularly ethyl esters, increased during the aging process, enhancing the wine's complexity. However, this effect is typically observed after 10–12 months of aging (Cerdán, Goñi, Azpilicueta and n., 2004), although our data suggest these changes began earlier in the aging process. Specifically, certain compounds such as ethyl lactate (A32), pentyl acetate (A22), 2-ethyl-1-hexanol (A19), and hexyl acetate (A23) were detected at higher levels during aging, consistent with findings from prior studies (Cerdán, Goñi, Azpilicueta and n., 2004; Garde-Cerdán & Ancín-Azpilicueta, 2006). Among the most significant changes, the levels of benzyl alcohol and phenylethyl alcohol increased sharply during the first three months of aging (Fig. 5A). This early increase is likely attributed to the interactions between the toasting levels and wood grain characteristics (Bosso et al., 2008), which have been shown to influence the volatile compound evolution in wine.

One of the key observations in this study is the reduction in the overall content of higher alcohols in the aged wines. These compounds, which are predominantly produced during fermentation, play a crucial role in yeast metabolism (Adlard, 2018), but their levels typically decrease over time due to the aging process. As higher alcohols can be detrimental to wine quality when present in excess (De-La-Fuente-Blanco, Sáenz-Navajas and Ferreira, 2016), this decline supports the improvement of wine sensory attributes, particularly in wines aged in oak barrels. Conversely, our study also revealed a decrease in ethyl acetate levels in some wine samples, reaching concentrations as low as 150 mg/L, which is considered detrimental (Sumby, Grbin, & Jiranek, 2010). This finding highlights the delicate balance of volatile compounds during aging and the importance of barrel influence on their final concentrations.

Interestingly, previous studies attributed similar changes in volatile profiles to oak barrels with medium toasting levels, but the specific characteristics of wood grains, especially from American oak, were often not considered (Cerdán, Goñi & Azpilicueta, 2004; Feng et al., 2023). In this study, the increase in acetic acid from the toasting process was found to influence the overall volatile composition, particularly enhancing the levels of compounds such as octanoic acid, which imparts floral notes to the wine (García-Moreno et al., 2021; Sumby et al., 2010). The increased concentration of octanoic acid during aging is indicative of a positive effect on the wine's sensory attributes, as it contributes honey and caramel nuances to the wine's flavour profile (Garde-Cerdán & Ancín-Azpilicueta, 2006).

Notably, aging in oak barrels, particularly in X9, X5, and X10 types, significantly impacted the wine's flavour profile, with the X9 barrel showing the highest concentrations of most volatile compounds, especially ethyl esters. However, wines aged in flexcubes exhibited lower retention of these compounds. The sensory impact of this variation was evident in the loss of black fruity characteristics, commonly associated with Cabernet Sauvignon wines (Chira et al., 2011), in wines aged in X9 barrels. In contrast, red fruity attributes were enhanced, and wines aged in X5 and X9 barrels scored higher in plant-like characteristics (Fig. 6A). These results suggest that oak barrels with heavier toasting levels are particularly beneficial for preserving varietal flavors, as evidenced by the higher concentrations of 2-methoxy-3-isobutyl pyrazine (A35), a compound that imparts typical green pepper notes to Cabernet Sauvignon wines (Feng et al., 2023).

4.3. Correlations between non-volatile and volatile compounds

The Pearson coefficients were calculated characterizing the relationships among non-volatiles and volatiles (Fig. 6C). Several volatiles, including 3-methyl-1-butanol (A13), isobutyric acid (A40), and benzaldehyde (A38), presented higher correlations with hexose levels, attributed to the Maillard reaction in wines (Kertsch et al., 2023; Perestrelo, Silva, & Câmara, 2019). Although the recent study showed a significant correlation between glycerol and volatiles (Petretto, Urgeghe, Cabizza, & Del Caro, 2023), in which the coefficients were inferior to the point where it was seen as statistical significance (Pearson coefficient > 0.6) in the study. Organic acid in wines were correlated with volatile compounds, mainly with 3-methyl-1-butanol (A13), benzaldehyde (A38), (E)-oak lactone (A52), and ethyl lactate (A21), suggesting the influences of organic acid on the wine pH to adjust the volatile levels (Tian et al., 2023). Notably, ethyl lactate was associated with lactic and malic acid at the highest Pearson coefficient; this confirmed the conversion of lactic acid as intermediate product (Sun et al., 2018). In addition to the associations between general characteristics and volatile composition, total phenolic and tartrate ester showed significantly positive correlations with benzaldehyde (A38), with a limited effect on other phenolic parameters.

5. Conclusion

This study investigated the impact of oak barrel characteristics—specifically wood grain and toasting level—on the colour, chemical composition, and sensory profile of high-ethanol Cabernet Sauvignon wines during aging. The basic chemical parameters, such as hexose and organic acid levels, remained unaffected by barrel characteristics after one year of aging. However, the total phenolic content and chromatic parameters varied with wood grain and toasting levels. Notably, total phenolics returned to initial levels after one year, correlating with changes in colour intensity, particularly the red hue, which diminished in wines aged in barrels with finer grain and heavier toasting. Volatile compound profiles also evolved, with a decrease in alcohols (including higher alcohols and C6 alcohols) and an increase in ethyl esters, contributing to the enhancement of fruity and floral notes. Oak-derived compounds, such as 4-ethyl-2-methoxy-phenol and (E)-oak lactone, contributed to spicier aromas. The volatile development did not follow a linear trend in relation to barrel characteristics, highlighting the complex interaction between wood grain, toasting, and oxidation rates. Wines aged in barrels with medium grain and heavy toasting exhibited the most favorable sensory and chemical attributes, suggesting their potential for producing high-quality aged high-ethanol wines. The wines aged in other oak barrels experienced a decline in quality during the one-year aging process, including excessive increases in acetic acid, rapid loss of colour, and insufficient fruit aroma. Overall, the results demonstrate that barrel characteristics, particularly wood grain and toasting level, significantly influence the quality of high-ethanol wines.

CRediT authorship contribution statement

Yu Chen: Writing – review & editing, Writing – original draft, Visualization, Software, Methodology, Formal analysis, Data curation. Xingmeng Lei: Writing – review & editing, Software, Methodology, Formal analysis. Tianyuan Zhang: Methodology, Formal analysis, Data curation. Hulin Chen: Software, Methodology, Formal analysis, Data curation. Lisa Tang: Software, Methodology, Formal analysis, Data curation. Yi Shang: Methodology, Formal analysis, Data curation. Zhen Huang: Investigation, Formal analysis, Data curation. Yuyang Song: Methodology, Formal analysis, Data curation. Yi Qin: Writing – review & editing, Methodology, Formal analysis, Data curation. Dongqing Ye: Writing – review & editing, Methodology, Formal analysis, Data curation. Yanlin Liu: Writing – review & editing, Methodology, Funding acquisition, Formal analysis, Data curation.

Declaration of competing interest

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

Acknowledgements

This project was supported by Ningxia Hui Autonomous Region Key Achievement Transformative Project (2023CJE09001), Ningxia Hui Autonomous Region Key R & D Project (2022BBF01003, 2023BCF01002), National Natural Science Key Foundation of China (U21A20269), and China Agriculture Research System of MOF and MARA (CARS- 29-jg-03).

Footnotes

☆

This article is part of a Special issue entitled: ‘Plant-Based Products’ published in Food Chemistry: X.

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.fochx.2025.102444.

Contributor Information

Dongqing Ye, Email: yedongqing@gxaas.net.

Yanlin Liu, Email: yanlinliu@nwafu.edu.cn.

Appendix A. Supplementary data

Supplementary material 1

Supplementary figures: Fig. S1. Cabernet Sauvignon wines from different aging duration have differential chemical and chromatic characteristics in nine oak barrels, depending wood grains and toasting level. The changing trend of generally physicochemical, chromatic, and phenolic parameters clustering for the factor of wood grains (A) and toasting levels (B) by the standardized data. Fig. S2. Clustering heatmap of wines volatile compounds obtained from different wood grains: wide grain (A), medium grain (B), and fine grain (C). The heatmap square represented the content level of each volatile and the fold lines represent the changing trend of each cluster. Code represented the volatile compounds showed in Table S3. T1, the initial wine; T2, T3, T4, T5, were the wine samples from the third, sixth, ninth, and twelfth months of aging, respectively. Fig. S3. Clustering heatmap of wines volatile compounds obtained from different toasting levels: light (A), medium (B), and heavy (C). The heatmap square represented the content level of each volatile and the fold lines represent the changing trend of each cluster. Code represented the volatile compounds showed in Table S3. T1, the initial wine; T2, T3, T4, T5, were the wine samples from the third, sixth, ninth, and twelfth months of aging, respectively. Fig. S4. The concentration of volatile compounds identified in wines, involving higher alcohol, C6 alcohol, volatile fatty acid, ethyl ester, acetate ester, and terpenes.

mmc1.docx (28.6MB, docx)
Supplementary material 2

Supplementary tables: Table S1. Chemical parameters of the wine before ageing. Table S2. Informations of oak barrels. Table S3. Volatile compounds detected in Cabernet Sauvignon wines. Table S4. The calibration curves for analyzed substances.

mmc2.xlsx (18.2KB, xlsx)

Data availability

No data was used for the research described in the article.

References

  1. Acuña L., Gonzalez D., de la Fuente J., Moya L. Influence of toasting treatment on permeability of six wood species for enological use. Holzforschung. 2014;68:447–454. [Google Scholar]
  2. Adlard E.R. In: Wine flavour chemistry. 2nd ed. Adlard E.R.J.B., Clarke R.J., editors. Vol. 81. Chromatographia; 2018. p. 1241. [Google Scholar]
  3. Andrew L.W., Laurie V.F. Oxidation of wine Phenolics: A critical evaluation and hypotheses. American Journal of Enology and Viticulture. 2006;57:306. [Google Scholar]
  4. Boido E., Alcalde-Eon C., Carrau F., Dellacassa E., Rivas-Gonzalo J.C. Aging effect on the pigment composition and color of Vitis vinifera L. cv. Tannat wines. Contribution of the main pigment families to wine color. Journal of Agricultural and Food Chemistry. 2006;54:6692–6704. doi: 10.1021/jf061240m. [DOI] [PubMed] [Google Scholar]
  5. Bosso A., Petrozziello M., Santini D., Motta S., Guaita M., Marulli C. Effect of grain type and toasting conditions of barrels on the concentration of the volatile substances released by the wood and on the sensory characteristics of Montepulciano D’abruzzo. Journal of Food Science. 2008;73:S373–S382. doi: 10.1111/j.1750-3841.2008.00891.x. [DOI] [PubMed] [Google Scholar]
  6. Călugăr A., Coldea T.E., Pop C.R., Pop T.I., Babeș A.C., Bunea C.I.…Gal E. Evaluation of volatile compounds during ageing with oak chips and oak barrel of Muscat Ottonel wine. Processes. 2020;8:1000. [Google Scholar]
  7. Cerdán T.G., Goñi D.T., Azpilicueta C.A., n. Accumulation of volatile compounds during ageing of two red wines with different composition. Journal of Food Engineering. 2004;65:349–356. [Google Scholar]
  8. Cerdán T.G., Mozaz S.R.G., Azpilicueta C.A. Volatile composition of aged wine in used barrels of French oak and of American oak. Food Research International. 2002;35:603–610. [Google Scholar]
  9. Chen Y., Jiang J., Song Y., Zang X., Wang G., Pei Y.…Liu Y. Yeast diversity during spontaneous fermentations and oenological characterisation of indigenous Saccharomyces cerevisiae for potential as wine starter cultures. Microorganisms. 2022;10:1455. doi: 10.3390/microorganisms10071455. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chen Y., Lei X., Jiang J., Qin Y., Jiang L., Liu Y.-L. Microbial diversity on grape epidermis and wine volatile aroma in spontaneous fermentation comprehensively driven by geography, subregion, and variety. International Journal of Food Microbiology. 2023;404 doi: 10.1016/j.ijfoodmicro.2023.110315. [DOI] [PubMed] [Google Scholar]
  11. Chen Y., Qi J., Yang H., Lei X., Jiang J., Song Y., Qin Y., Liu Y.-L. Fungal dynamic during apricot wine spontaneous fermentation and aromatic characteristics of Pichia kudriavzevii for potential as starter. Food Chemistry: X. 2023;19 doi: 10.1016/j.fochx.2023.100862. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chira K., Pacella N., Jourdes M., Teissedre P.-L. Chemical and sensory evaluation of Bordeaux wines (cabernet-sauvignon and merlot) and correlation with wine age. Food Chemistry. 2011;126:1971–1977. doi: 10.1016/j.foodchem.2010.12.056. [DOI] [PubMed] [Google Scholar]
  13. Del Fresno J.M., Morata A., Loira I., Escott C., Suarez Lepe J.A. Evolution of the phenolic fraction and aromatic profile of red wines aged in oak barrels. ACS Omega. 2020;5:7235–7243. doi: 10.1021/acsomega.9b03854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. De-La-Fuente-Blanco A., Sáenz-Navajas M.-P., Ferreira V. On the effects of higher alcohols on red wine aroma. Food Chemistry. 2016;210:107–114. doi: 10.1016/j.foodchem.2016.04.021. [DOI] [PubMed] [Google Scholar]
  15. Dumitriu G.-D., Teodosiu C., Gabur I., Cotea V.V., Peinado R.A., López de Lerma N. Evaluation of aroma compounds in the process of wine ageing with oak chips. Foods. 2019;8 doi: 10.3390/foods8120662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fanzone M.L., Sari S.E., Mestre M.V., Catania A.A., Catelén M.J., Jofré V.P.…Maturano Y.P. Combination of pre-fermentative and fermentative strategies to produce Malbec wines of lower alcohol and pH, with high chemical and sensory quality. OENO one. 2020;54 [Google Scholar]
  17. Feng Z., Martínez-Lapuente L., Ayestarán B., Guadalupe Z. Volatile and sensory characterization of Tempranillo wines aged in Quercus alba oak barrels of different geographical origins in USA. LWT. 2023;173 [Google Scholar]
  18. Fernández de Simón B., Sanz M., Cadahía E., Poveda P., Broto M. Chemical characterization of oak heartwood from Spanish forests of Quercus pyrenaica (wild.). Ellagitannins, low molecular weight phenolic, and volatile compounds. Journal of Agricultural and Food Chemistry. 2006;54:8314–8321. doi: 10.1021/jf061546t. [DOI] [PubMed] [Google Scholar]
  19. García-Moreno M.V., Sánchez-Guillén M.M., Delgado-González M.J., Durán-Guerrero E., Rodríguez-Dodero M.C., García-Barroso C., Guillén-Sánchez D.A. Chemical content and sensory changes of Oloroso Sherry wine when aged with four different wood types. LWT. 2021;140 [Google Scholar]
  20. Garde-Cerdán T., Ancín-Azpilicueta C. Review of quality factors on wine ageing in oak barrels. Trends in Food Science & Technology. 2006;17:438–447. [Google Scholar]
  21. Guler A.C. Effects of different maceration techniques on the colour, polyphenols and antioxidant capacity of grape juice. Food Chemistry. 2023;404 doi: 10.1016/j.foodchem.2022.134603. [DOI] [PubMed] [Google Scholar]
  22. Heymann H., Li Calzi M., Conversano M., Bauer A., Skogerson K., Matthews M. Effects of extended grape ripening with or without must and wine alcohol manipulations on Cabernet Sauvignon wine sensory characteristics. South African Journal of Enology and Viticulture. 2013;34:86–99. [Google Scholar]
  23. Hixson J.L., Hayasaka Y., Curtin C.D., Sefton M.A., Taylor D.K. Hydroxycinnamoyl glucose and tartrate esters and their role in the formation of Ethylphenols in wine. Journal of Agricultural and Food Chemistry. 2016;64:9401–9411. doi: 10.1021/acs.jafc.6b04074. [DOI] [PubMed] [Google Scholar]
  24. Hranilovic A., Gambetta J.M., Jeffery D.W., Grbin P.R., Jiranek V. Lower-alcohol wines produced by Metschnikowia pulcherrima and Saccharomyces cerevisiae co-fermentations: The effect of sequential inoculation timing. International Journal of Food Microbiology. 2020;329 doi: 10.1016/j.ijfoodmicro.2020.108651. [DOI] [PubMed] [Google Scholar]
  25. Kertsch A.-L., Wagner J., Henle T. Selected Maillard reaction products and their yeast metabolites in commercial wines. Journal of Agricultural and Food Chemistry. 2023;71:12300–12310. doi: 10.1021/acs.jafc.3c04512. [DOI] [PubMed] [Google Scholar]
  26. King E.S., Dunn R.L., Heymann H. The influence of alcohol on the sensory perception of red wines. Food Quality and Preference. 2013;28:235–243. [Google Scholar]
  27. Li S., Duan C.-Q. Astringency, bitterness and color changes in dry red wines before and during oak barrel aging: An updated phenolic perspective review. Critical Reviews in Food Science and Nutrition. 2019;59:1840–1867. doi: 10.1080/10408398.2018.1431762. [DOI] [PubMed] [Google Scholar]
  28. Lu H., Cheng B., Lan Y., Duan C., He F. Modifications in aroma characteristics of ‘merlot’ dry red wines aged in American, French and Slovakian oak barrels with different toasting degrees. Food Science and Human Wellness. 2024;13:381–391. [Google Scholar]
  29. Monagas M., Bartolomé B., Gómez-Cordovés C. Evolution of polyphenols in red wines from Vitis vinifera L. during aging in the bottle. European Food Research and Technology. 2005;220:331–340. [Google Scholar]
  30. Navarro M., Kontoudakis N., Gómez-Alonso S., García-Romero E., Canals J.M., Hermosín-Gutíerrez I., Zamora F. Influence of the botanical origin and toasting level on the ellagitannin content of wines aged in new and used oak barrels. Food Research International. 2016;87:197–203. doi: 10.1016/j.foodres.2016.07.016. [DOI] [PubMed] [Google Scholar]
  31. Perestrelo R., Silva C., Câmara J.S. Madeira wine volatile profile. A platform to establish Madeira wine aroma descriptors. Molecules. 2019;24:3028. doi: 10.3390/molecules24173028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pérez-Magariño S., González-San José M.L. Polyphenols and colour variability of red wines made from grapes harvested at different ripeness grade. Food Chemistry. 2006;96:197–208. doi: 10.1021/jf035099i. [DOI] [PubMed] [Google Scholar]
  33. Petretto G., Urgeghe P.P., Cabizza R., Del Caro A. Evaluation of volatile and chemical profile of sherry-like white wine Vernaccia di Oristano from Sardinia by comprehensive targeted and untargeted approach. European Food Research and Technology. 2023;249:1887–1897. [Google Scholar]
  34. Pollon M., Río Segade S., Giacosa S., Botto R., Montanini C., Rolle L. Volatile compound release from oak chips in model wine media: Combined influence of toasting degree, size, time of contact, and ethanol content. Journal of Agricultural and Food Chemistry. 2023;71:13440–13450. doi: 10.1021/acs.jafc.3c03469. [DOI] [PubMed] [Google Scholar]
  35. Ramos-Pineda A.M., García-Estévez I., Dueñas M., Escribano-Bailón M.T. Effect of the addition of mannoproteins on the interaction between wine flavonols and salivary proteins. Food Chemistry. 2018;264:226–232. doi: 10.1016/j.foodchem.2018.04.119. [DOI] [PubMed] [Google Scholar]
  36. Sánchez-Gómez R., del Alamo-Sanza M., Martínez-Martínez V., Nevares I. Study of the role of oxygen in the evolution of red wine colour under different ageing conditions in barrels and bottles. Food Chemistry. 2020;328 doi: 10.1016/j.foodchem.2020.127040. [DOI] [PubMed] [Google Scholar]
  37. Sumby K.M., Grbin P.R., Jiranek V. Microbial modulation of aromatic esters in wine: Current knowledge and future prospects. Food Chemistry. 2010;121:1–16. [Google Scholar]
  38. Sun S.Y., Chen Z.X., Jin C.W. Combined influence of lactic acid bacteria starter and final pH on the induction of malolactic fermentation and quality of cherry wines. LWT. 2018;89:449–456. [Google Scholar]
  39. Tian M.-B., Hu R.-Q., Liu Z.-L., Shi N., Lu H.-C., Duan C.-Q., Wang J., Sun Y.-F., Kong Q., He F. The pH adjustment of Vitis amurensis dry red wine revealed the evolution of organic acids, volatomics, and sensory quality during winemaking. Food Chemistry. 2023;436 doi: 10.1016/j.foodchem.2023.137730. [DOI] [PubMed] [Google Scholar]
  40. Tu Q., Liu S., Li Y., Zhang L., Wang Z., Yuan C. The effects of regions and the wine aging periods on the condensed tannin profiles and the astringency perceptions of cabernet sauvignon wines. Food Chemistry: X. 2022;15 doi: 10.1016/j.fochx.2022.100409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Vignault A., Gombau J., Pascual O., Jourdes M., Moine V., Canals J.M.…Teissedre P.-L. Copigmentation of Malvidin-3-O-Monoglucoside by oenological tannins: Incidence on wine model color in function of botanical origin, pH and ethanol content. Molecules. 2019;24 doi: 10.3390/molecules24081448. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Welke J.E., Zanus M., Lazzarotto M., Alcaraz Zini C. Quantitative analysis of headspace volatile compounds using comprehensive two-dimensional gas chromatography and their contribution to the aroma of chardonnay wine. Food Research International. 2014;59:85–99. [Google Scholar]
  43. Ye D.-Q., Zheng X.-T., Xu X.-Q., Wang Y.-H., Duan C.-Q., Liu Y.-L. Evolutions of volatile sulfur compounds of cabernet sauvignon wines during aging in different oak barrels. Food Chemistry. 2016;202:236–246. doi: 10.1016/j.foodchem.2016.01.139. [DOI] [PubMed] [Google Scholar]
  44. Yu Y., Li L., Xue R., Wang C., Chen M., Ramos J.…Sun B. Impact of different oak chips’ aging on the volatile compounds and sensory characteristics of Vitis amurensis wines. Foods. 2022;11 doi: 10.3390/foods11081126. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Supplementary material 1

Supplementary figures: Fig. S1. Cabernet Sauvignon wines from different aging duration have differential chemical and chromatic characteristics in nine oak barrels, depending wood grains and toasting level. The changing trend of generally physicochemical, chromatic, and phenolic parameters clustering for the factor of wood grains (A) and toasting levels (B) by the standardized data. Fig. S2. Clustering heatmap of wines volatile compounds obtained from different wood grains: wide grain (A), medium grain (B), and fine grain (C). The heatmap square represented the content level of each volatile and the fold lines represent the changing trend of each cluster. Code represented the volatile compounds showed in Table S3. T1, the initial wine; T2, T3, T4, T5, were the wine samples from the third, sixth, ninth, and twelfth months of aging, respectively. Fig. S3. Clustering heatmap of wines volatile compounds obtained from different toasting levels: light (A), medium (B), and heavy (C). The heatmap square represented the content level of each volatile and the fold lines represent the changing trend of each cluster. Code represented the volatile compounds showed in Table S3. T1, the initial wine; T2, T3, T4, T5, were the wine samples from the third, sixth, ninth, and twelfth months of aging, respectively. Fig. S4. The concentration of volatile compounds identified in wines, involving higher alcohol, C6 alcohol, volatile fatty acid, ethyl ester, acetate ester, and terpenes.

mmc1.docx (28.6MB, docx)
Supplementary material 2

Supplementary tables: Table S1. Chemical parameters of the wine before ageing. Table S2. Informations of oak barrels. Table S3. Volatile compounds detected in Cabernet Sauvignon wines. Table S4. The calibration curves for analyzed substances.

mmc2.xlsx (18.2KB, xlsx)

Data Availability Statement

No data was used for the research described in the article.

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

RESOURCES