Inﬂuence of Oak Chips and Oak Barrel Ageing on Volatile Proﬁle in Chardonnay Wine of Romania

The inﬂuence of the addition of oak chips and barrel ageing on basic wine parameters and volatile compounds of Chardonnay wines has been studied. Chardonnay wines were obtained by the traditional wine-making process. Oak chips (4 g/L—non-toasted and light toasted) were added at the ﬁnal stage of the winemaking process for ageing 1, 2 and 3 months, respectively. Also, the control wine was aged in non-toasted barrels for the same period of time. Following Liquid-liquid extraction-gas chromatography-mass spectrometry analysis, alcohols, esters, fatty acids, lactones, and phenolic compounds were identiﬁed and quantiﬁed. The light toasted wine was clearly separated by phenolic compounds (vanillin, p -vinyl guaiacol and acetovanillone). The ﬂoral aroma supplied by 2-phenylethanol was slowly increased by ageing with odor activity values (OAV) higher in aged samples than control wine (1.07). The vanilla scent could be easily perceived in all aged samples, mainly for light toasted chip-treated samples with OAV values between 2.30 and 2.37. After 3 months, the volatile compounds of wine from non-toasted medium (chips and barrels) were almost similar from the volatile proﬁle point of view. This could have economic and viniﬁcation management implications since oak barrels are expensive and the wine oak barrel aging is a long process. All wines studied in this research can provide a viable alternative to young varietal wines.


Introduction
The quality of a wine depends mainly on its chemical composition and also on the expectation of the consumer. Each type of consumer will define wine quality differently depending on their expectations and needs [1]. Expectations of the consumer vary depending on the type of wine: white, red, rose, sweet, dry; young or aged wine [2]. Chardonnay is one of the flexible varieties which could adapt to different growth regions with varied weather and soil characteristics [3]. Among white wines, Chardonnay lends itself to a wide variety of production styles, which can be tailored to the target market and that make this grape variety unique. Presently, two tendencies exist for Chardonnay aroma profiles: fruity and light styles and flavored and complex styles [3]. These styles are targeted by current market trends and popularity of wines. Chardonnay allows winemakers different stylistic choices compared to other white wine varieties. Wines can be aged in oak (barrel or chips) and stored after bottling for different periods of time, depending on the desired style [4]. Oak (Quercus sp.) wood is used in the manufacture of barrels due to its chemical and physical properties. Among the other tree species used in winemaking, the oak is unique due to the size and shape of the annual circles that give it durability when transformed into barrel staves [5]. The technological processes used in carpentry, such as seasoning and toasting the wood staves, give the final characteristics of the barrels [6]. Maturing wines in oak barrels is a traditional practice used to improve sensory and quality characteristics. Although fermentation and maturation in oak barrels is desirable for many wine centers, this is not always possible due to high costs [7,8]. However, an alternative of obtaining wines with significant aromas is the use of oak, non-toasted or toasted (different levels) chips that can be added during the fermentation process of the wine or at the end of alcoholic fermentation [7,9,10]. Some studies examine the addition of wood chips in wine during or after alcoholic fermentation [7,10]. Also, the presence of lees during ageing can improve the organoleptic characteristics of some wines [11]. Recently, the use of alternative vessels to oak barrels during winemaking has become increasingly popular, but little is known about their impact on the chemical composition of the resulting wines [12].
The volatile compounds extracted from the wood depend on a number of factors such as the degree of toasting, the contact duration with wine [13,14], but also on their geographical origin [5,15]. The addition of oak chips to finished wines and their effect on the chemical composition of the wine and its sensory characteristics has been widely studied [3,7,9]. The effects of the chips addition in the alcoholic fermentation of wine led to contradictory results regarding the stabilization of the wine color, but all studies reported changes in the content of volatile and phenolic compounds [5,10,13,15]. Red wine maturation in barrels has been a common practice for centuries. In recent years, the maturation of white wines encountered an increased interest in obtaining wines with improved sensory and quality properties. In Europe, the use of oak wood pieces in winemaking was approved in 2006 (Council Regulation EC No.2165 [16] and since then the addition of oak chips has been generalized. The new EU Delegated Regulation 2019/934 currently regulates the use of oak wood chips in wine practices, which are used for several technological purposes. The main objective is the release of desirable compounds, such as aromas and polyphenolics, from the wood fragments into the wine during aging [16]. The volatile compositions of Chardonnay wines aged in barrels or with chips is studied extensively in Spain [17,18], Greece [19], Italy [20], Australia [21]. Our research is the first attempt to study the volatile composition of Romanian Chardonnay wines, aged through different methods. Chardonnay is one of the grapevine varieties most distributed across the world [17], but in Romania, Chardonnay variety is planted on 2511 hectares representing about 3% of total vineyards [22]. The purpose of the present study was to examine the effects of two types of French oak chips (non-toasted and light toasted) and non-toasted Romanian oak barrel, during short ageing periods on Chardonnay white wine composition by using gas chromatography coupled to mass spectrometry (GC-MS). Unlike most other white wine varieties which are not aged, Chardonnay wines commonly spend between several weeks to a year for ageing in oak barrels. We focused on identifying the chemical compounds according to contact time and wood type. This study provides important scientific content to winemakers.

Oak Wood Origin
The barrels used were made up of Romanian oak from Quercus robur, from the forest located in the Western Romania (Ineu area; 46 • 26 N 21 • 50 E). The raw staves (100 cm × 11 cm × 2.2 cm) were naturally seasoned for 24 months at the Transylvania Bois ® Tonnellerie (Sighetu Marmat , iei, Romania). Once assembled, the barrels (225 L) were not submitted to toasting procedures. For the purpose of the study, non-toasted barrels were provided to the wine cellar. The oak chips (Quercus robur) were provided by Sodinal ® (Wine and Beverage Division of AVeX Group, Paris, France), with dimensions of 0.6 cm × 1.2 cm × 0.15 cm (width × length × thickness). The oak chips used were non-toasted and light toasted, respectively.

White Wine Vinification and Sample Collection
Chardonnay grapes (Vitis vinifera L.) were manually harvested at maturity in Teaca winery (Lechinta, Bistrita Nasaud County, Romania) during the 2017 vintage (225 g/L sugar in grape juice, 8.95 g/L must total acidity expressed as tartaric acid and 104 g weight of 100 berries). The Lechinta grapevine growth area is known for the high acidity of wines, due to climatic conditions (oenoclimatic aptitude index of 4221) [23]. Grapes were crushed and destemmed (Enoitalia ® WE223S, Cerreto, Italy) on the harvest day. A pneumatic press (Vaslin-Bucher ® RPS 50, France) was used, filled at 75-80% of its capacity. Potassium metabisulphite (4.5 ± 0.5 g/hL) was added during the transfer of the must to the stainless steel tank (5000 L capacity). Saccharomyces cerevisiae (Zymaflore VL1, Laffort ® , Bordeaux, France), a commercial active dry yeast was included at a rate of 20 g/hL to perform alcoholic fermentation at 16 • C. Malolactic fermentation (MLF) could take place spontaneously, no bacterial culture was added to performed MLF, and no supplementary sulfitation was made to inhibit the MLF. Three sets of experiments were performed depending on the container in which maturation took place (Table 1). When alcoholic fermentation ended, the control wine was transferred to new non-toasted oak barrels at full capacity (first experiment). The control wine was transferred to demijohns (glass container of 50 L) with 4 g/L non-toasted oak chips (second experiment) and 4 g/L light-toasted oak chips (third experiment). Wines were kept in oak barrels and glass containers with chips for ageing during 3 months at a controlled temperature of 14-16 • C. Sampling was performed at 1, 2 and 3 months of wood contact. Wine samples were stored at 4 • C until the moment of analysis. Each experiment had three sets of containers, meaning three non-toasted barrels, three glass containers with light toasted chips, and three glass containers with non-toasted chips.

Oenological Parameters in Wines
Basic oenological parameters of wines such as pH, alcoholic strength (% v/v), titratable and volatile acidity (g/L tartaric acid, and g/L acetic acid, respectively), were analyzed in accordance with official International Organisation of Vine and Wine methods [24] and Bora et al. [25]. All parameters were determined in triplicate.

Reagents and Standards
The reference standards, 2-phenyl ethanol, ethyl decanoate, butyrolactone, isoamyl alcohol and 1-octanol (used as internal standard) were provided by Fluka ® (Buch, Switzerland). Dichloromethane used for the extraction of volatile compounds was purchased from Merck®(Darmstadt, Germany) Standard stock solutions were prepared by dissolving 10 mg of each reference compound in 10 mL of dichloromethane. The calibration curve of 2-phneyl-ethanol was used for the volatile concentration calculation. The range of standard concentration was 64-12,900 µg/L.

Liquid-Liquid Extraction of Volatile Compounds
The extraction method elaborated by Andujar-Ortiz et al. [26] was adapted and used for the isolation of volatile compounds from wine samples. Into an Erlenmayer flask were introduced 50 mL of wine spiked with 1-octanol (concentration 614 µg/L) as the internal standard and 20 mL of dichloromethane, and it was equipped with a cork. The extraction was performed under continuous stirring in an ice bath for 30 min. The mixture was then kept for 15 min in an Elmasonic S ultrasonic bath (Elma, Germany) at the same temperature to avoid the formation of an emulsion. After separation, the organic layer was dried over Na 2 SO 4 , evaporated under a stream of nitrogen to about 200 µL volume of the extract. From this solution, 1 µL was injected into the GC/MS system. All extractions were performed in triplicate.

Determination of Volatile Compounds
Analysis of volatile organic compounds [27] from wine samples was performed using a Shimadzu QP 2010 PLUS Mass Spectrometer coupled with Gas Chromatograph (Shimadzu, Japan) equipped with a Carbowax type column from Agilent (30 m × 0.32 mm inner diameter and 0.50 µm film thicknesses). Helium (6.0) was used as carrier gas with a flow rate 1.7 mL/min. The injector, the detector and the interface temperature were set at 220 • C. The column temperature program was conducted as follows: 40 • C was the initial temperature for 5 min, increasing at a rate of 4 • C/min to 220 • C, and holding 220 • C for 15 min. Quadrupole mass detector acquisition was carried out using the positive electronic ionization-mode at the 70 eV, with continuous scanning from 40 to 500 amu. Standard compounds in wines were identified by comparing their relative retention times and mass fragmentation with those of computer matching against a commercial library (National Institute of Standards and Technology and Willey).

Odor Activity Values (OAVs)
The contribution of each volatile compound to wine aroma was evaluated qualitatively by its associated descriptor and quantitatively by its odor activity values (OAVs). OAVs were calculated using the ratio between total concentration (in µg/L) of each compound in the wine samples, and the odor threshold value (in µg/L) of the compound in water/ethanol solution [28]; threshold values were obtained from information available in the literature [28].

Statistical Analysis
Univariate analysis was performed using analysis of variance (ANOVA), applying the Tukey multiple range test. Partial least squares regression (PLSR) analysis was also carried out, using the XLSTAT Addinsoft 2014.5.03 version (Addinsoft Inc., NY, USA).

Basic Oenological Parameters
Analysis of basic oenological parameters is presented in Table 2. Alcohol content is high in all samples (above 12.5% v/v), with small but significant differences among variants, with a lower value after ageing in barrel. Samples BAR1M-BAR3M aged in non-toasted oak barrels have a higher level of total acidity than samples N1M-N3M and L1M-L3M, probably due to higher oxygen diffusion through barrel wood pores [29]. All tested wine samples reached dryness (less than 4.0 g/L residual sugar-data not shown). Nevertheless, there is a small and not significant difference in the dry-extract and non-reducing dry-extract level between the control wine and aged variants ( Table 2). The concentration of extract without sugar is very similar in all samples regardless of the ageing method, as previously showed by other studies [11,30]. Concentration of free and total SO 2 in samples is acceptable for all variants.

Volatile Profile of White Wines Aged with Non-Toasted and Light-Toasted Oak Chips
Liquid-liquid extraction-gas chromatography-mass spectrometry analysis of Chardonnay white wines after 1, 2 and 3 months of ageing with non-toasted oak chips, light-toasted oak chips and non-toasted oak barrels, identified five chemical groups: higher alcohols, esters, volatile fatty acids, lactones and phenolic compounds. As noted in Table 3, volatile compounds have significant differences depending on the experimental variant.

Alcohols
In our study, the concentration of volatile alcohols varied both depending on the time and type of ageing. Three major alcohols (iso-butanol, isopentyl alcohol and 2phenylethanol) were quantified in wine samples. Those alcohols showed different trends during ageing. As can be observed in Table 3, the iso-butanol significantly decreased during 3 months of ageing, regardless of the methods, while isopentyl alcohol and 2phenylethanol have increased concentration. After 3 months of ageing, the major alcohol concentrations were higher, regardless of the method (N3M-142 mg/L; L3M-144 mg/L and BAR3M-143 mg/L). By using LLE-GC-MS analysis were detected and quantified 9 minor alcohols (hexanol, 2,3-butanediol, 3-methylthio-1-propanol, benzyl-alcohol). In the minor group, hexanol was the highest alcohol, and its concentration was dependent on ageing time and significantly dependent on ageing method. As can be observed from Table 3, the non-toasted barrel ageing method favored the concentration of hexanol, which supplied a leafy and grassy note to the wine samples [31]. Also, the minor alcohols had different trends during ageing. Most of them (4 methyl-1-pentanol, 2-nonalol, 1-heptanol, 3 methylthio-1-propanol) decreased after 3 months of ageing. The highest increase was detected for 2,3-butanediol in BAR1M-BAR3M variants (1686-4973 µg/L).

Esters
Individual fruity volatile concentrations obtained in the present research were higher than the bibliographic ranges reported by Gambetta et al. [3] in Italy, or González-Centeno et al. [19], in Greece, but in agreement with those obtained by Cheng et al. [31] in Xianjiang (China). Most of them exhibited concentrations above their olfactory perception threshold, providing the wine with pear, apple, pineapple and/or floral notes [30,31]. Among the ethyl esters of straight-chain fatty acids, ethyl caproate (3120-3240 µg/L) and ethyl octanoate (2122-2362 µg/L) were the predominant ones. Isoamyl acetate, characterized by banana flavor, was the main component among the higher alcohol acetates, with values ranging from 2433 to 2538 µg/L, depending on time and method of ageing. Within this family of esters, it was the only volatile present at above threshold level (30 µg/L). Table 2. Oenological analyses of white wines aged with oak chips and barrel for 1, 2 and 3 months; analysis of variance (ANOVA) taking as factors time and ageing method. All values are expressed as means ± SD. Different letters in each row of the same variant are significantly different at the 0.05 level according to ANOVA by Tukey's test. *: p < 0.05; **: p < 0.01; ***: p < 0.001; ns: not significant. CW, N1M, N2M, N3M, L1M, L2M, L3M, BAR1M, BAR2M and BAR3M are the abbreviations for 10 variants of Chardonnay (CW-control wine, N-non-toasted chips, L-light toasted chips, BAR-non-toasted barrel; 1M, 2M and 3M-number of months of ageing).

Fatty Acids
Nine fatty acids were identified in all wine samples. In the present study, there were significant differences in the fatty acid content between control wine and the wine samples aged with different methods. Furthermore, the concentrations of all kinds of fatty acid detected in the variant N1M-N3M (4728-5071 µg/L) and L1M-L3M (4336-4977 µg/L) were lower compared to variant BAR1M-BAR3M (5035-6540 µg/L).

Lactones
Among lactones, butyrolactone showed the highest concentration and depending significantly on the studied factors, as in the case of pantolactone and 3,4-dimethyl-2(5)furanone. Thus, for control sample, an amount of 478 µg/L of total lactones was quantified. Regardless of the ageing method, wines in which non-toasted oak chips were added (N1M-N3M) and in non-toasted oak barrel (BAR1M-BAR3M) presented greater butyrolactone concentrations than the corresponding light-toasted chips (L1M-L3M) wines.

Volatile Phenols
The total content of phenolic volatiles in Chardonnay wines, calculated by adding up the individual concentration of each compound, ranged from 365 to 463 µg/L for N1M-N3M wines, from 393 to 505 µg/L for L1M-L3M wines and from 350 to 468 µg/L for BAR1M-BAR3M. Among methods of ageing, light toasting led to the highest values, and non-toasted to the lowest ones. Volatile phenols that resulted from wood toasting, have significantly higher values in the wines of L1M-L3M variants (p-vinyl guaiacol-58-65 µg/L, acetovanillone-143-171 µg/L and vanilla 133-139 µg/L), compared to the variants from non-toasted wood.

Odor Active Odorants
In order to assess the influence of each single volatile compound, the odor activity value (OAV) was calculated. The OAV represents the ratio between the concentration of the compound and its odor threshold. Results in Table 4 show that 17 out of 42 quantified volatile compounds reached a concentration above the odor threshold in at least one variant (OAV >1). For major alcohols, all wine samples presented an OAV >1, with the highest values for isopentyl alcohol, mainly for L1M-L3M sample. The floral aroma supplied by 2-phenylethanol was slowly increased by ageing with OAV higher in aged samples than control wine (1.07). Among minor alcohols, only 2-nonanol and 3-methylthio-1-propanol reached OAV >1 in at least one sample. An important ester present above its threshold in all samples was ethyl hexanoate, which can supply anise and strawberry aromas to a wine [19]. N-amyl acetate was found above its threshold in all samples, which indicates that this compound is a general contributor to the fruity aroma of wines. Interestingly for N-amyl acetate was that it increased the OAV values for L1M-L3M (83.80-86.76) samples, so a fruity character could be highlighted by ageing with light-toasted chips. Among terpenes, only linalool showed OAV >1. Some authors [32] concluded that the high level for terpenes in wines, could explain the floral descriptors. Three of nine fatty acids showed OAV >1 in all variants. The fatty and rancid aroma supplied by isovaleric acid could be easily perceived in non-toasted oak barrel aged sample, due to the low odor threshold of this compound. Butyrolactone and 3,4-dimethyl-2(5)-furanone were lactones with OAV >1. Butyrolactone extracted from non-toasted wood presented higher values of OAV for N1M-N3M      Figure 1 shows the chemical analysis results of the Chardonnay wines overlaid over the studied variants, with the wines projected on to that space. According to PLSR analysis, the distance between the variable and the center of the circle shows the interpretive degree of the principal components to the variable. The PLSR was established to determine the influence of methods and duration of ageing on volatile compounds. In Figure 1, the volatile compounds of the control wine can be used to distinguish the aroma compounds of the aged wines. Control wine (unaged) is well related to alcohols such as iso-butanol (MA1), 2-nonanol (mA5), and to the terpenes: linalool (T1), terpineol (T2) and trans-geraniol (T3). Esters like hexyl-acetate (ES2), fatty acids like malic acids (FA6) and lactones like 2,4-dimethyl-2(5)-furanone were also related to the control Chardonnay wine. Those compounds are specific for fresh wines. The wines that pass through the maturation process are related to the esters, fatty acids, lactones and volatile phenols [36,37]. As can be seen from Figure 1, the wines aged in non-toasted medium (chips and barrel) appear as grouped against the wines aged with light toasted chips. This finding is similar with other research by Calugar et al. [13], for Muscat Ottonel wines. The non-toasted barrel ageing is correlated with lactones that came from untreated wood-butyrolactone (L1) and pantolactone (L2). The non-toasted medium is also related to esters such as ethyl lactate (ES4), ethyl caproate (ES3), and carboxylic acid, as lactic acid (FA9), octanoic acid (FA4) and isovaleric acid (FA3). The majority of volatile phenols quantified in this research, came from the thermal treatment of wood and are correlated with wines aged with light-toasted chips. In a study, Farrel et al. [38], stated that vanillin (VP5), p-vinyl guaiacol (VP1) and acetovanillone (VP3) are volatile phenols which come from the thermal degradation of wood.

Discussion
The diversity of oak products allows winemakers to achieve different results according to the expected outcome and consumer preferences. Exposure to oak wood may cause unbalanced flavor profiles with strong oak aromas if the maturation period is longer [30].

Basic Parameters of Chardonnay Aged Wines
During wood contact, basic parameters of wines were affected. The decrease of alcohol content could be the result of the evaporative loss of ethanol and its diffusion through the staves, which are higher than that of water during ageing [29]. The increase in titratable acidity is due to the rise in volatile acidity; during the first months this rise could indicate the extraction of carboxylic, phenolic and volatile acids from the wood as suggested by Aiken and Noble, [39]. According to Towey and Waterhouse [40], the pH and ethanol of wines also influenced the amount of volatiles extracted from oak products. It has been stated that the characteristics of the wine itself may also be very important during the ageing process, since pH, ethanol and acidity may have a direct influence on ethanolysis of wood components. Some contradictory results have been found. Garde-Cerdán et al. [41,42] stated that the extraction of furanic aldehydes, oak lactones, phenolic aldehydes and alcohols was influenced by the wine alcohol content, the wine pH being less significant, whereas Ortega-Heras et al. [43] did not find significant correlations between wine ethanol content and the levels of any of the volatile compounds studied.

Volatile Compounds in Chardonnay Aged Wine
The method of ageing influences the amount of chemical compounds at different times of the process. During ageing, specific compounds are transferred to wines as a result of prolonged contact with oak chips [30]. When employing the oak alternatives, the main factors to be considered are the amount and surface area (shape and size) of the materials used. In this respect, chip and barrel-aged wines differ in their content of a range of volatiles, such as vanillin, guaiacol or furfurals [3].
Louw et al. [44] states that young unoaked Chardonnay wines can be discriminated (with 74% correctly classified) by using the following eight volatile compounds: 2-phenylethanol, decanoic acid, diethyl succinate, ethyl hexanoate, ethyl decanoate, ethyl octanoate, hexyl acetate and 1-propanol. In general, alcohols can be the product of sugar fermentation and amino acid metabolism through yeasts. Higher alcohols have a significant impact on the fruity aroma of wine [45]. The major alcohols (iso-butanol, isopentyl alcohol and 2-phenylethanol) varied both depending on the time and type of ageing with different trends. However, at the end of the maturation period, the major alcohol concentrations were higher, regardless of the method. Also, the concentrations of the minor alcohols group were dependent on ageing time and method. The non-toasted barrel ageing method favored the concentration of hexanol, which supplied a leafy and grassy note to wine samples [31]. Generally, 2,3-butanediol is not expected to affect the sensory qualities of wine appreciably, but some authors [46] noted a bitter taste in variants with a concentration over 700 mg/L of this compound in wines samples. In our research, the higher concentration of 2,3 butanediol in the BAR variant wine sample could be explained by enzymatic reduction of diacetyl and acetoin in this compound under malolactic fermentation which could occur in barrels [47].
Overall, the time and method of ageing, significantly influenced the concentration of major and minor alcohols in wine samples, with higher amount for the variant aged in the non-toasted barrel. Increased amounts of higher alcohols during ageing can be caused by acid-catalyzed ester hydrolysis [48].
The esters chemical family contribute to the fruity character of wines and are one of the predominant chemical groups [49]. In this research, the impact of ageing duration and method on the Chardonnay wines concentration of esters was investigated. Most of them exhibited concentrations above their olfactory perception threshold, providing the wine with pear, apple, pineapple and/or floral notes [30,31]. During ageing, the esters had interesting trends; most of them have increased in concentration compared with the control wine and only hexyl acetate and trimethylene acetate decreased after ageing. Ethyl fatty acid esters are considered to be very important for the aroma of wines. The concentration of ethyl lactate progressively increases during malolactic fermentation [46]. This result is beneficial for the wine bouquet due to its fruity, buttery or creamy aromas, also contributing to the sensations of roundness in the mouth [46,50]. Overall, the ageing process, regardless of the method, slightly increased the amount of esters. Even more, after 1, 2 and 3 months of ageing, the concentrations of esters were nearly similar for the same time period of ageing among methods of ageing.
The fatty acids chemical group of volatile compounds is produced by yeast and bacteria during fatty acid metabolism. Volatile fatty acids can contribute to the complexity of the wine bouquet even if they are present under threshold levels. They have a negative effect on wine aroma when above their thresholds [51]. Some acids, such as iso-butyric acid and decanoic acid are not associated with wine quality, but they play an important role in the complexity of the aroma [52]. In the present study, there were significant differences in the fatty acid content between control wine and the wine samples aged by different methods. Decreased acid concentrations could be caused by the loss of volatile acids and esterification reactions between alcohols and acids during the ageing process [53]. Also, Vivas et al. [54] explained that the increase in volatile acidity in wine in contact with oak wood is mainly due to two reasons: firstly, the acidity may increase or decrease as a result of the chip roasting process, and secondly, may be due to acetic metabolism (i.e., acidic bacteria in wine).
As expected, the lactones showed the highest concentration in wine samples treated with non-toasted wood (non-toasted chips and barrel). This could be explained by the fact that some lactones are found in non-toasted wood, and thermal treatment degraded those compounds. Lactones add caramel and sweet flavor to wines [38].
Unlike the expected behavior, for ageing methods in/with non-toasted wood, total phenol volatiles content was similar for both methods, for the three ageing durations. It could be stated that, at least the first three months of ageing in non-toasted barrels are similar with ageing with non-toasted chips, regarding this chemical family. Regardless of the ageing method, wines which underwent the wood ageing presented greater vanillin concentrations, than the corresponding control wine. In our experiments, no control of MLF was conducted (no culture of bacteria and no inhibitors were added), so the increasing of vanilla volatile compounds in wood aged wine could under gone as the capacity of lactic bacteria (spontaneous MLF) to convert vanillin glycoside pre-cursors present in oak wood [55,56]. Spillman et al. [21] states that p-vinylguaiacol is a fermentation product rather than a compound derived from oak contact, formed through decarboxylation of natural grape components ferulic and p-coumaric acids by S. cerevisiae enzymatic activity. The concentration of 4-vinyl-guaiacol in our Chardonnay wine samples could indicate that this compound formed mainly during the fermentation period, but was also extracted from toasted wood products.
Overall woody aroma of barrel-aged wines results from different chemical compounds naturally present in oak wood, originated in the toasting process and/or formed as a consequence of the wood-wine interaction. Gambetta et al. [3] described furfurals, 5methylfurfural, cis-and trans-whiskey lactones, guaiacol, eugenol and vanillin as odorants derived from oak wood contact, or formed during ageing. The statistical test of the experimental data disclosed that the time and the ageing method significantly impact the oak wood volatiles concentration of the studied wines.

Odor Activity Values
Among all the compounds present in Table 3, not all of them present an odor activity value OAV >1 (Table 4), which is normally taken as the value required for a compound to be a likely contributor to the specific aroma. The OAVs based on threshold values in a specific matrix (e.g., neutral white wine or 10% v/v aqueous ethanol) act merely as a guide when determining the importance of a compound to wine aroma, due to a significant effect of the matrix [3]. The highest OAV was exhibited, among all samples, by ethyl octanoate characterized by pineapple and apple notes, specific for Chardonnay wines [19]. Ethyl hexanoate and N-amyl acetate are fruity aroma contributors with concentrations above their threshold in all samples, mainly in wine samples treated with light-toasted chips. The method of wine ageing using non-toasted wood products, could preserve the floral character of Chardonnay wines. These results were in accordance with those presented previously by Calugar et al. [11], on Muscat Ottonel wines. Regarding the fatty acids, the rancid aroma supplied by isovaleric acid could be easily perceived in the non-toasted oak barrel aged sample, due to the low odor threshold of this compound. The spicy and condiment aroma could be highlighted by L1M-L3M samples, where light toast of chips increased the amount of p-vinyl guaiacol extracted by wine. The compounds that give wines a vanilla and roasted note (vanillin-VP5, p-vinyl guaiacol-VP1 and acetovanillone-VP3) are related to the light toasted chips in our study. For short periods of maturation, Chardonnay wines can obtain almost the same qualities if aged with nontoasted chips and non-toasted barrel, from the volatile point of view, but with more efficient cost, from the economical point of view. The PLRS analysis shows that wine samples treated with the French non-toasted oak chips aged for 1 or 2 months have similar wine volatile profiles with those from Romanian non-toasted oak barrels. Some differences between non-toasted variants were observed when the wines were aged for 3 months. Chardonnay is not dominated by a distinct varietal compound as in other varieties like Sauvignon blanc or Muscats. The typicity concept for Chardonnay wines needs more comprehension and requires deeper study, because this is a variety that can express itself in a multitude of profiles. Understanding Chardonnay typicity requires multidisciplinary studies that analyze a number of variables responsible for the final product, rather than a one-directional focus [3]. Nevertheless, the use of oak barrel alternatives, in particular, offers an economic means of imprinting the desirable aroma compounds from oak without the time or expenses associated with ageing wine in oak barrels [57].
It is not easy to distinguish products obtained by aging with chips from those obtained traditionally (barrel aged) from a sensory point of view. The distinction of the two different products is possible through laboratory investigation techniques: analytical methods allow expanding the compositional differences between the two wine products. Among the analytical techniques used, the spectroscopic ones are certainly those that, combined with multivariate statistics techniques, allow to obtain the best results in this field, for commercial purposes and, if necessary, for the control of fraudulent activities [58].

Conclusions
This work is the first study on volatile compounds of young and short matured Chardonnay wines from Romania. The method and time of ageing have a significant influence on basic wine parameters and on the volatile composition. The non-toasted medium (chips and barrel) ageing is correlated with lactones, esters (ethyl lactate and ethyl caproate) and fatty acids (isovaleric and lactic acids). The majority of volatile phenols were quantified in wines aged with light-toasted chips. The oak chip method of ageing the Chardonnay wines could be a useful tool to obtain wines as a viable alternative to the traditionally made Chardonnay wines. The amount of chips used in this study (4 g/L) was selected in order to avoid an excessive impact of the wood character in wines that could produce a negative effect on tasting. For short periods of maturation, Chardonnay wines can obtain almost the same wine aged with non-toasted chips and non-toasted barrels, from the chemical point of view, but with more efficient cost, from the economic point of view. For white wines, the use of oak chips could avoid the oxidation of aromatic volatile compounds that could be produced during barrel ageing, and impart oak notes to wines without decreasing the fresh and fruity characteristics. This approach would enable diversification on the market and increase the range of products on offer to the consumer.