Trends in Reducing the Effects of Global Warming: Applications of Reverse Osmosis to Obtain Sparkling Wines with Moderate Alcohol Concentrations

Abstract

Sparkling wine can undoubtedly conquer the most demanding tastes due to its qualities, uniquely boosted by carbon dioxide. The quality and characteristics of sparkling wines, their stability, and sensory properties largely depend on the physical–chemical composition of the grapes and the base wine, the production technology applied, the environmental conditions, etc. Several techniques can be implemented to produce low-alcohol wines, and reverse osmosis is a procedure that has been successfully used in recent times to reduce the alcohol concentration while having a low negative impact on the composition of the wine under certain conditions. In the current circumstances of changing climatic conditions in areas with a tradition of producing sparkling wines, it is necessary to identify methods of keeping the alcoholic concentration of the base wine at a moderate level to not change the composition of sparkling wines in relation to consumer preferences. This chapter looks at the effects of reverse osmosis and the implications of inoculated yeasts on sparkling wine quality.

1. Introduction

Sparkling wines are considered special due to their high carbon dioxide concentrations, which explains why they are consumed occasionally and for special events. Such wines were obtained in the Champagne region around Reims, an area generally exploited for pastures and cereal cultivation [1]. The Montagne de Reims wine-growing area was favorable to the cultivation of red grapes, while the region of Aÿ-Champagne for white grapes. Dom Pérignon (1638–1715), a Benedictine monk from the Monastery of Hautvillers in Épernay, was the first who obtained, around 1670, an effervescent drink called Champagne, after the name of the wine region of origin. This beverage, initially accidentally produced through the re-fermentation of wine, had a moderate alcohol concentration (about 10% vol. alc.) and a pressure in a bottle of 1.5–2 atmospheres. The sparkling wine was originally produced from red grapes and had a rosé color. Dom Pérignon laid the foundations for what became known as gentle pressing, which allowed minimal contact of the juice with the solids. The varieties cultivated in the Champagne region in those days are not known in detail, but of the 80 varieties, Pinot Noir and Pinot Gris were predominant. With the invasion of phylloxera, most areas cultivated with wines were destroyed, and the existing varieties were subsequently replanted. Although the beverage was originally obtained from seven Vitis vinifera varieties, nowadays, it can only be produced from Pinot Noir, Pinot Meunier, and Chardonnay [1]. Also, the name Champagne can only be used in the eponymous region. While most sparkling wines are made from a blend of grape varieties, a ‘Blanc de Blancs’ comes exclusively from Chardonnay, while a ‘Blanc de Noirs’ comes from black varieties exclusively (usually only Pinot Noir).

Winemakers are seeking ways to lower this crucial parameter for the consumer due to the rise in the alcoholic concentration of wines made from raw materials that grow under conditions of rising annual temperatures and climate change. The techniques employed in this context, including reverse osmosis, usually use membranes, which filter a mixture of compounds with the smallest molecule and then reassemble the obtained solutions to obtain a product with predetermined characteristics. A base wine that has a lower alcoholic concentration is more appropriate for use as, during the secondary fermentation, this characteristic will likely increase. Therefore, the consumer can continue to appreciate an equilibrated sparkling wine [1].

The production of sparkling wines in Romania started with Ion Ionescu de la Brad, a renowned nineteenth-century agronomist who created this beverage for Mihail Sturdza, a prominent Romanian personality of the time [1]. The production and consumption of sparkling wine have seen an upward trend in the last decade, with consumption shifting from mainly special occasions to less traditional contexts [2].

Among the most famous sparkling wines are Cava (produced in Spain), Espumante (Portugal), Sekt (Austria and Germany), Pezsgo (Hungary), and Shampanskoye (Russia). The grape varieties most frequently used to produce sparkling wines globally are Macabeo, Parellada and Xarello (Cava), Pinot Noir, Chardonnay, Pinot Gris, Aligoté, Pinot Blanc, Grolleau, Cabernet Sauvignon, and Chenin Blanc [3].

2. Production of sparkling wines

The fact that sparkling wines are characterized by high concentrations of carbon dioxide earns them the name of special wines. The classification of these beverages is done according to the type and nature of carbon dioxide and the level of pressure in the bottle. Therefore, according to the Organization of Wine and Vine (OIV) regulations, more categories can be found as sparkling, pétillant wines, and pearl (carbonated and lightly carbonated) wines. According to the pressure inside the bottle, there are wines with relatively high pressure (minimum 3 bars) – sparkling and pearl wines; wines with relatively low pressure (between 1 and 1.5 bars): pétillant and lightly carbonated wines. Of these, sparkling wine is obtained through the secondary fermentation of base wine, and carbon dioxide is of endogenous origin only. It is generally sold with an overpressure of 3 bar minimum at 20°C. Compared to sparkling wines, péttilant wines usually have an overpressure of 1–2.5 bar at 20°C. Pearl wines have an overpressure of over 3 bar at 20°C, while the carbon dioxide is of exogenous or partially exogenous origin [1].

2.1 Grape varieties used to obtain sparkling wines

To produce sparkling wines, either white or red base wines can be used. When red grape varieties are used, they are vinified as white wines to obtain the blanc de noirs. Necessarily, the selection of the grape variety takes into consideration the pedo-climatic conditions of the region, the productivity of the grape variety, and the desired distinctive character of the wine. The existing sparkling wines tend to use the following varieties: Champagne – Chardonnay, Pinot Noir and Pinot Meunier; Cava – Macabeo, Xarel lo, Parellada, and Chardonnay; Talento – Chardonnay, Pinot Nero, and Pinot Bianco; Asti – Muscato Bianco; Lambrusco – Lambrusco Bianco and Lambrusco Nero; Pinotage – a cross between Pinot Noir and Cinsault; Sekt-Riesling, Silvaner, Pinot Blanc, Pinot Noir, and Pinot Gris. Various clones can also influence the quality of sparkling wines. For example, the Chardonnay clone, VCR10, is recommended in the production of base wine for Cava because of its high acidity, which is particularly suitable for producing sparkling wines. Since the phylloxera crisis (late nineteenth and early twentieth centuries), new rootstocks have emerged from a cross between French and American strains. They were selected according to their degree of adaptation to pedological conditions and European varieties. For instance, 41B is adapted to cretaceous soils and continues to be the most used in Champagne. SO₄ rootstock is adapted to moderately calcareous soils, and 3309C is the strain of choice for slightly calcareous soils [4, 5, 6].

2.2 Grape processing

In order to obtain high-quality musts, appropriate conditions during the harvesting and processing of the grapes have to be in place. The harvesting process is extremely important when it comes to quality. It is therefore necessary to harvest the grapes manually and collect them in small buckets or crates to avoid crushing. The fruit loads are shipped immediately for fast processing to avoid the onset of fermentation processes. Sorting the bunches affected by gray rot (Botrytis cinerea spp.), which can affect the foaming capacity of the sparkling wine, is also to be considered. The grapes are then promptly but gently pressed, without being previously crushed, at pressures between 1.5 and 2 bar to avoid oxidation. Next, maceration takes place to extract flavor and color compounds, and the process continues with sulfitation of the resulting must (to avoid spontaneous fermentations) in doses ranging from 3 g/hL to 8 g/hL, depending on pressing fraction and crop health. Treatment with pectolytic enzymes and inoculation of Saccharomyces cerevisiae starter cultures follows. Musts with turbidity values of 200 NTU to 400 NTU are clarified by centrifugation or static decantation (18–24 hours at 6–15°C), and finishing agents can be added (bentonite or casein). During static settling, some must proteins, pectin as well as mineral cations and phenolic substances are removed [4, 5, 6]. The must is further subjected to the fermentation process in order to obtain a base wine with low-alcohol content [7, 8, 9, 10].

2.3 Primary fermentation: Obtaining the base wine

The technology for sparkling wines initially requires making a base wine, to be followed by initiation of the second fermentation in the bottle (Prise de mousse). After clarification of the must, alcoholic fermentation occurs as a result of inoculation (10–15 g/hL) with selected strains of Saccharomyces cerevisiae cv bayanus. From a biochemical point of view, alcoholic fermentation consists of the anaerobic transformation of sugars (mainly glucose and fructose) under the action of yeasts into ethyl alcohol, carbon dioxide, and other secondary products. However, grape must be a nonsterile substrate that contains several types of yeasts and bacteria that can grow and consequently affect the composition and final quality of the wine. The presence of different yeasts in the must depends on several factors, such as the grape variety, the degree of fruit ripening, the application of pesticide treatments, the degree of fungi development, the climatic conditions, as well as the vine culture technology [11].

Alcoholic fermentation usually takes place in stainless steel tanks at temperatures ranging between 15°C and 20°C, with the fermentation process sometimes occurring at temperatures lower than 13°C. Charcoal can be used to even out and lighten the color of wines, though it does affect the foaming capacity and sensory properties of the final product [12, 13]. As the primary alcoholic fermentation progresses, high amounts of alcohol are formed. The influence of ethanol on yeast has been studied extensively [14, 15]. In general, ethanol tolerance is associated with a higher degree of fatty acids unsaturation from cell membranes [16, 17, 18]. Further chemical–physical treatments for clarifying, decanting, and filtering the base wine are followed by bottling. Sparkling base wines will generally have an alcoholic strength of 10–11% vol. alc., low residual sugars, and low volatile acidity. On the other hand, the level of organic acids will be high, which results in increased total acidity (between 12 g/L and 18 g/L tartaric acid) [19].

While the acidity level of the base wine is a key factor for the quality of sparkling wine, its values decrease during the production process due to degradation by yeasts and lactic acid bacteria and precipitation of potassium bitartrate [20].

2.4 Malolactic fermentation

Malolactic fermentation of base wines is widely used in the production of sparkling wines to activate an increase in volatile acidity. The conversion of malic acid-dicarboxylic acid into lactic acid-monocarboxylic acid and carbon dioxide, carried out by lactic acid bacteria, lowers the acidity of the wine and increases its pH. This type of fermentation is important for wines produced from grapes grown in cool climates, which tend to have a high content of organic acids (tartaric and malic) and low pH. In addition, malolactic fermentation ensures microbiological stability. Diacetyl and ethyl lactate are formed during the fermentative process, which gives lactic notes in wines. Malolactic fermentation ought to take place before bottling to prevent the subsequent appearance of bacteria that cause unwanted deposits. Malolactic fermentation is carried out by inoculating lactic bacteria (Oenococcus oeni). Some producers do not use this type of fermentation, though, and prefer to preserve the freshness and fruity notes of the wine by applying higher amounts of SO₂ (8 g/hL to 10 g/hL) [21, 22]. Difficulties in inducing malolactic fermentation are usually attributed to the cumulative inhibitory effects of low pH, high alcohol, and SO₂ content of the wines. Nevertheless, such problems could also arise from inadequacy or imbalance of some nutrients (free amino acids) found in sparkling wines [23, 24].

2.5 Conditioning and fining of base wine

Upon completion of malolactic fermentation, wines are subjected to conditioning treatments (gravitational settling and sedimentation or centrifugation, fining with bentonite, gelatin, tannin or silica gel, charcoal, etc.) [24, 25]. Spontaneous/gravitational settling is done by keeping the wine in the fermentation vessels for a longer time, decanting and sedimenting the yeast deposit, and separating the wine from it. This can be aided by centrifugation insofar as it replaces the filtration operation with cellulose plates or diatomite. Conditioning the base wine by applying fining treatments involves the use of substances such as protein products (gelatin, casein) and tannin. Moreover, while adding bentonite does help to prevent protein and copper casse, it can also negatively impact the foam-forming capacity. In addition, fining can be accelerated by applying alginates, potassium ferrocyanide (for wines with excess iron and copper), synthetic polyamides (polyvinylpyrrolidone, polyvinylpolypyrrolidone – to eliminate compounds susceptible to oxidation), coal (for color improvement), silicic acid (tannin substitute), etc. Fining of wine by filtration is often used as a complementary process to conditioning as well as for final wines. The application of sulfur dioxide provides protection against oxidation. For sparkling wines, low sulfitation is seen as inhibiting the onset of fermentation. Stabilization against tartaric precipitation is frequently done by refrigeration (keeping the base wine at a temperature of −4°C) or by applying agents that inhibit the crystallization of potassium tartrate (metatartaric acid or carboxymethylcellulose). Ion exchange treatment and electrodialysis can also be used. These treatments are to be carried out only after the final base wine is obtained [1].

2.6 Base wine blend

The practice of blending wines from different grape varieties, different origins, and different years (reserve wines) is important for preserving the quality of sparkling wines. Reserve wines are kept for 2–3 years (sometimes on yeast) in tanks, at 12–13°C, and protected from oxygen; reserve wines can also be kept for a longer period (up to ten years) [24]. The aim of the blending stage is to obtain a base wine with well-defined physico-chemical properties and well-balanced sensory characteristics. The quality of the mixture is mainly established through sensory evaluation [1].

2.7 Secondary fermentation and aging on lees

Secondary fermentation in the bottle (prise de mousse) can take place by adding the so-called liqueur de tirage, a mixture of sucrose (18–25 g/L depending on the desired CO₂ concentration), and yeasts, diammonium phosphates and adjuvants (bentonite and alginate). The yeasts must allow fermentation at lower temperatures in the presence of the alcohol formed during primary fermentation. The base wine for bottling must have specific characteristics such as light color, fruity aroma, low residual sugars, moderate alcohol content (10–11% vol. alc.), low volatile acidity (acetic acid), and total acidity of 12–18 g/L tartaric acid. Once the wine is in, the bottles are closed using a crown cap with a plastic cylinder, which is meant for yeast collection. The bottles are kept in specially equipped rooms for decanting sparkling wine, at low temperatures and dim lighting. After filling the bottles, they are placed horizontally. Secondary fermentation is slow and takes an average of six to eight weeks in constant low-temperature conditions (11–12°C). The time interval between the addition of the liqueur de tirage and the liqueur de dosage must be at least 15 months. Once the secondary fermentation is complete, the sparkling wine is matured and aged for a while on the yeast deposit accumulated. Depending on the type of sparkling wine and the legislation in the country of origin, aging varies from 9 to 11 months. During the aging on lees period, the sparkling wine acquires specific organoleptic characteristics acquired through the autolytic process of the yeasts and mediated by hydrolytic enzymes; the latter favor the release of polysaccharides, peptides, fatty acids, proteins, and mannoproteins into the sparkling wine. It is a known fact that oxygen may get into the bottle through the cork, and sensory defects may also occur during aging. Exposure of the bottles to light leads to the degradation of methionine and the formation of volatile compounds with sulfur, which in turn brings about unpleasant smells of cauliflower or wet wool [24].

Yeast autolysis is a very slow process. In general, four main phases occur during aging on lees: (I) during secondary fermentation, the level of amino acids and proteins goes down, and peptides are formed; (II) viable and inactivated cells coexist, peptides are degraded, and amino acids and proteins are released; (III) when no viable cells are present, the release of both proteins and peptides predominates; (IV) approximately 9 months after application of the liqueur de tirage, a decrease in amino acid concentration occurs. Proteases favor the hydrolysis of lysosomal and cytoplasmic membranes; they increase the porosity of the yeast cell wall and facilitate the release of degraded constituents into the wine. The slow rate of enzyme activity delays the autolytic process. Consequently, sparkling wines are left in contact with the yeast for several months or years to benefit from the positive autolytic effects [26]. Some studies center on accelerating yeast autolysis, while other authors suggest choosing strains with strong autolytic capacity, combining positive and negative strains with killer factor, and administering liqueur de tirage made with exhausted yeast [9, 27]. Another option would be the combination of strains with different autolytic capacity [28]. In this respect, exploiting non-Saccharomyces yeasts in combination with Saccharomyces strains is worth considering [29]. With a view to improving the complexity of the aroma and diversifying the assortment range, interest in non-Saccharomyces yeasts for the production of sparkling wine has soared in recent years.

There is very little data regarding the effect of non-Saccharomyces on sparkling wine quality [30, 31], and it mostly covers a limited number of non-Saccharomyces specie including Torulaspora delbrueckii, Metschnikowia pulcherrima, Schizosaccharomyces pombe, and Saccharomycodes ludwigii. The available studies focus on the analysis of amino acids, ammonia, volatile compounds, glycerol, and protein content, all of which have a bearing on the sensory characteristics of sparkling wines. The fermentation process is monitored by analyzing the content of reductive sugars and measuring the pressure by means of an aphrometer [29].

2.8 Yeast selection for the secondary fermentation

The yeasts that perform the secondary fermentation must have a number of additional characteristics compared to those used in the first fermentation of the base wine production. Besides high resistance to ethanol, they must possess a high flocculation capacity, which facilitates removal from the bottle. Flocculation is a distinctive characteristic of yeasts that is frequently found in Saccharomyces cerevisiae, unlike Saccharomyces uvarum. As a result of the biochemical processes, several compounds are released which significantly influence the characteristics of the wine and its sensory quality. The selection of yeast strains is important for improving the quality of sparkling wines. Given that mannoproteins are among the major compounds released by yeasts during autolysis, the search for strains that can release large amounts of these compounds is of major interest if one aims to improve sparkling wine quality [32].

2.9 Factors that influence yeast autolysis

The biochemical process of autolysis is influenced by pH level, temperature, the presence of ethanol, and also by the nature of the yeast strain. High temperatures, up to 60°C, have been reported to favor autolysis in a wine model system. Molnar et al. [33] reported that the optimum temperature for proteolysis with the Champenoise method is between 10 and 12°C. Nunez et al. [34] compared the autolytic capacity of different strains and proposed this as a criterion for yeast selection. The autolytic capacity was evaluated by measuring the amino acids released by the yeast at different temperatures ten days after fermentation. Significant differences were observed in the autolytic capacity of the three strains. It is a fact, therefore, that the yeast strain acts on the amount of nitrogen released into the environment, which could potentially be useful for sparkling wine production [35]. Martinez-Rodriguez et al. [36] suggested that a yeast strain with good autolytic capacity would produce better-quality sparkling wine than yeast with low autolytic capacity. Also, autolytic capacity together with foam analysis, should be used to select yeasts for sparkling wine production. Nunez et al. [34] recently confirmed that the autolytic capacity of yeast is important for sparkling wine quality.

2.10 Riddling and disgorging

During riddling, the bottles are rotated daily for about 15 days until they are perpendicular to the floor. Thus, the yeasts are directed toward the neck of the bottle. In this way, the negative effects of oxygen and biological degradation are avoided. Riddling is not a homogeneous process, as it depends on the type of yeasts present, the variable surface area, and the flocculation characteristics of the yeasts. The addition of both bentonite and the liqueur de tirage aims to normalize sedimentation. The concentrated yeast deposit that reaches the neck of the bottle is then removed through the disgorging process. The process involves placing the bottle in low-temperature brine, freezing the neck of the bottle, and removing the frozen yeast deposit. During this process, part of the liquid is lost, only to be refilled with liqueur de dosage consisting of cane or beet sugar (between 6 g/L and 50 g/L sucrose) and antioxidant substances – SO₂, citric or ascorbic acid. During this operation, the pressure in the bottle decreases due to the loss of carbon dioxide. Care must be taken for the dosage not to raise the alcoholic strength of the sparkling wine by more than 0.5% vol. alc. [8, 10].

2.11 Corking

After applying the liqueur de dosage and filling the bottles with sparkling wine, they are closed with corks (made of cork, synthetic materials – polyethylene, metal with thread). In the case of cork closures, it is necessary to fix them with wire hoods in order to withstand overpressure [1].

3. Methods of producing sparkling wines

The production method is one of the key factors that determine the style and quality of white and rosé sparkling wines. The sparkling wine market has expanded significantly over time, mostly driven by increased global consumer demand. Due to its high added value, the economic impact of sparkling wine is very important, even if production is lower compared to still wines [7]. Current trends highlight a rising consumer interest in sparkling wines, which in turn increases demand on the global market. The type of inoculated yeasts is a fundamental parameter for optimizing the production technology as well as improving the quality of the end-product. According to the regulations and guidelines of the International Organization of Vine and Wine [37], sparkling wines belong to the category of special wines. Sparkling wines differ mainly in terms of production technology, of which the following stand out: the Champenoise method, Cremant (France and Luxembourg), the transfer method, the ancestral method/méthode ancestrale (Limoux, Gaillac), Dioise and the Charmat method (in tanks).

The Champenoise method involves obtaining the base wine by completing the alcoholic fermentation process, followed by a secondary fermentation and aging in bottles (Figure 1). This method involves the removal of the sediments formed by the riddling and disgorging processes [38]. It is used in France to obtain Champagne, which uses varieties such as Chardonnay, Pinot Noir and Pinot Meunier, Arbanne, Petit Meslier, and Pinot Blanc [7].

Figure 1.

Champenoise method.

Transfer method isobarometric transfer. After fermentation and maturation in bottles, the base wine is cooled and then transferred to a pressure tank by means of an automatic or semi-automatic installation (Figure 2). At this stage, dosing of the liqueur de dosage can be done directly in the tank or after filtering the sample. Following aging on lees, the wine is filtered and bottled or simply bottled if filtering was carried out previously. All operations are performed in a carbon dioxide atmosphere. This method ensures a more uniform adding of the liqueur de dosage, as well as the likely elimination of riddling and disgorging. However, the transfer method is expensive, energy intensive, and poses an increased risk of wine oxidation [1, 38].

Figure 2.

Transfer method.

The ancestral method was first devised in the vineyards of Limoux and Gaillac (France), but it is generally difficult to control (Figure 3). The base wine is obtained by processing whole grapes (mainly from the Mauzac variety) or applying classic white winemaking technology to an incompletely fermented wine. At various stages of the winemaking process, it is essential to halt the fermentation each time it starts to accelerate. As such, refrigeration (down to 0°C), sulfiting, and depletion of nutrients from the yeast (by decantation, fining, filtration, or centrifugation) are repeated as often as necessary to regulate or stop yeast activity. The wine is then filtered and stored at 0°C for 2–3 months. It is at this stage that the second fermentation takes place (2–3 months) in the bottle (at rigorously controlled temperatures) due to the sugars left over after the first fermentation. Then, there is riddling and disgorging without adding the liqueur de dosage. This method may also skip the disgorging process and keep a low proportion of sediment at the bottom of the bottle. The method is particularly used in France to obtain Cremant, which incorporates such varieties as Chenin Blanc, Chardonnay, Cabernet franc, Pineau D’Aunis, Grolleau, Gamay, Aligoté, Melon, and Sacy [7].

Figure 3.

Ancestral method.

The Dioise method implements the same winemaking principles as ancestral to obtain a semi-fermented wine from the Muscat à petits grains variety, filtering and refrigeration at 0°C, and the second fermentation in the bottle. To improve the extraction of flavor compounds, pectinolytic enzymes are administered during grape processing. In this case, the must is characterized by a high turbidity value (1000 to 1500 NTU) that needs to be settled by means of the flotation technique. The second fermentation is halted by applying low temperatures and then transferring the content of the bottles to a stainless steel tank to be kept at an isobarometric pressure of CO₂ to avoid loss of carbon dioxide (similar to the transfer method). After filtration, the wine is bottled by applying the isobarometric principle. The final alcoholic concentration of these beverages is generally around 7.5% vol. alc., with 40–50 g/L of residual sugars [21].

The Charmat method (fermentation in tanks). Unlike the Champenoise method, with the Charmat method, the second fermentation occurs in stainless steel tanks. This is a simpler technique that is low-cost compared to the others. Yeasts and sugars are added, and the wine is kept at a temperature of 20–25°C. The second fermentation usually lasts 10 days and is stopped by the application of sulfur dioxide and by refrigerating the wine to −2°C. Having been cold-stabilized at −5°C for several days, the wine is filtered at a low temperature and then bottled using the isobarometric principle. A disadvantage of this method is that it does not mature the wine on the yeast deposit (Figure 4) [24].

Figure 4.

Charmat method.

The method is frequently used to obtain sparkling wines in France (from Cabernet Sauvignon, Chenin Blanc, Cabernet Franc, Sauvignon Blanc, Gamay, Chardonnay, Grollo, Pinot D’Aunis, Pinot Noir, Malbec) and Germany (from Pinot Blanc, Silvaner, Riesling, Pinot Gris and Pinot Noir) [7].

Continuous flow fermentation. The continuous flow method, the so-called Russian method, was created in response to the need to manufacture sparkling wines quickly, cheaply, and with minimal effort. The technological steps are the same as for the tank-based fermentation method, except that maturation takes place in a unique bioreactor-style tank, with specific processes ranging from 3 to 5 years. As a result, this method’s application necessitates the existence of suitable facilities outfitted with high-capacity pressure tanks, which is a disadvantage to its application [1].

The production of red or rosé sparkling wines is relatively low due to the choice of the moment for harvesting the grapes. In order to be suitable for base wine, red grapes often do not reach the appropriate phenolic maturity that gives the wine organoleptic complexity [28].

4. Yeast selection in the technology of sparkling wine production

The selection of starter cultures is generally based on their oenological properties, which reflect the degree of adaptation to the environmental characteristics specific to must and sparkling wine. The properties of yeasts for the first and second fermentation are different. Nevertheless, it is essential to identify the starter cultures that tolerate the specific characteristics of the must and the base wine. The Saccharomyces genus belongs to the Ascomycota division and includes seven species, among which Saccharomyces cerevisiae. It adapts well to conditions of low pH, high concentration of sugars (osmotic stress), progressive increase in ethanol, and depletion of the corresponding nutrients. Saccharomyces cerevisiae is considered the most important yeast species in the winemaking process and is used industrially to carry out high-quality fermentation processes. Saccharomyces cerevisiae strains for the secondary fermentation of the base wine usually show additional physiological and technological characteristics when compared to those proposed for the ‘primary’ starter cultures. The objective is to fully convert the sugars added to the base wine, via the liqueur de dosage, into ethanol and carbon dioxide. Thus, it is necessary that the strains can activate in an environment of at least 10–12% vol. alc., low pH (2.9–3.2), tolerate low temperatures (10–15°C), total SO₂ concentrations of 50–80 mg/L, high total acidity (12–18 g/L tartaric acid), low volatile acidity (0.2–0.4 g/L), high pressure (5–6 bar) and glycerol content of 5–20 g/L. On top of that, some authors highlight the importance of the flocculation and autolytic capacities of yeasts during secondary fermentation. Flocculation is a special characteristic of yeasts that allows for the settling of musts during fermentation. This physiological feature can be described as a natural ability of yeast strains to form compact microbial biomass, which settles at the bottom of the fermentation vessel and, following foam development, facilitates the disgorging of the deposit that accumulates at the neck of the bottle. Not least important, yeast flocculation appears to be associated with improved ester production.

Interspecific hybrid strains are also suitable for the production of sparkling wine re-fermented in the bottle; they are obtained through the appropriate selection of flocculant strains of Saccharomyces cerevisiae and nonflocculant strains of Saccharomyces bayanus. Interspecific hybrids are able to ferment at a temperature range between 6°C and 36°C. The autolytic capacity of yeast is inherent to another selective character of yeast: the killer phenotype, which releases toxic proteins. Under oenological conditions, acceleration of the autolytic process is pursued through the appropriate mix of killer and killer-sensitive yeast strains and the use of mutant yeast strains with autolytic characteristics. Mutagenesis of Saccharomyces cerevisiae strains induces accelerated release of proteins, amino acids, and polysaccharides. Sparkling wines obtained from selected mutant strains show better foaming properties than those inoculated with nonmutant strains. The autolytic capacity of yeasts is closely related to the release of volatile compounds in sparkling wine and their impact on the sensory profile. Research analyzes a specific mannoprotein (PAU5) synthesized by certain Saccharomyces cerevisiae strains. This mannoprotein, encoded by the seripauperin gene PAU5, directly diminishes the undesirable phenomenon caused by the agitation of sparkling wines and, as such, has foam-stabilizing properties. Excessive spontaneous foaming triggered by the release of pressure upon opening the bottle causes important economic losses. This can be caused either by contaminated raw material (for example, the presence of the bacteria Botrytis cinerea spp.) or by noncompliance with the production process. Following the secondary fermentation of sparkling wine, Saccharomyces cerevisiae yeasts are able to produce indole even when the viability of the culture is very low. The proportion of CO₂ released by yeasts and dissolved in the liquid phase during the fermentation process is an additional parameter for the selection of yeast cultures, one of fundamental importance for the organoleptic perception of the end-product [10, 28]. The sensory properties of sparkling wine are influenced by various factors, such as the processing technology, the grape varieties, the chemical composition and structure of the base wine, the selected yeast strains, and the aging time spent on lees. However, the sensory profile of sparkling wines depends more on the autolytic release of volatile compounds during the above-mentioned phase since aroma is one of the most relevant indicators of sparkling wine quality. Primary (prefermentative) aromas are characteristic of the grape variety used, secondary (fermentative) aromas are released following the metabolism of yeasts during fermentation, and tertiary (postfermentative) aromas are formed during aging. Cotea et al. [39] studied the influence of four commercial yeasts on the volatile profile of sparkling wines and found that the enrichment of aromatic compounds is a specific characteristic of inoculated yeast strains. Sensory evaluation and chemical analysis of volatile compounds by qualified experts are among the most important techniques for assessing the aroma profile of sparkling wines. At the end of the second fermentation, the long phase of aging on lees begins and with it, the release of intracellular compounds that causes an increase in free amino acids, often deemed as precursors of aromatic compounds in sparkling wine. All along the aging of sparkling wine, the autolysis of yeasts releases a considerable number of volatile molecules (esters, higher alcohols, aldehydes, sulfur compounds, organic acids, etc.), a process closely correlated to the yeast strains that are present. Being an enzymatic hydrolysis of biopolymers, autolysis is a very slow process associated with the death of yeast cells and the release of volatile constituents.

There are studies that focus on analyzing the different compounds released during the autolytic process and exploring additional methods of inducing autolysis in order to influence the aging character in sparkling wine. There are authors who investigate the different compositional changes that occur during the production of sparkling wines and the factors that most influence a range of sensations at the olfactory level. Polysaccharides can influence the viscosity, foaming capacity, and sensory perception of sparkling wines. Mannoproteins are released in sparkling wine during yeast autolysis to prevent protein disorder or crystallization of potassium bitartrate. Alcohols, carbonyl compounds and organic acids are mainly responsible for the burned, plastic, or rancid note. Aldehydes, such as octanal, nonanal, decanal, and some terpenes, enrich sparkling wines with citrus-like aromas. Intense chemical, roasty aromas, floral, vegetable, and fatty notes are discerned in sparkling wines produced with strains of Saccharomyces cerevisiae bio-immobilized with filamentous fungi – Penicillium chrysogenum. It is therefore clear that not only the strain but also the method of immobilization of the yeasts influences the final sensory characteristics of sparkling wines. The presence of indole and other volatile compounds definitely affects the sensory perception of these products because of the unpleasant plastic flavors. Both Saccharomyces and non-Saccharomyces yeasts can yield high concentrations of indole during primary fermentation. Extensive research on the selection of indigenous yeasts for use in grape must fermentation has been carried out, but few authors have explored the study of indigenous yeasts in sparkling wines. In the latter case, the selection of indigenous yeasts is much more complex because the base wine, into which the specific strain of Saccharomyces cerevisiae is inoculated, is considered an unfavorable environment for the growth of microorganisms due to its high alcohol content and low pH. Indigenous strains of Saccharomyces cerevisiae are selected based on several technological criteria (fermentative strength and vigor, SO₂ and ethanol tolerance, and flocculation capacity) and qualitative characteristics (acetic acid content, glycerol, and hydrogen sulfide production) [8, 40]. An interesting property of indigenous strains of Saccharomyces cerevisiae used in sparkling wine production is their ability to modulate the phenolic components of the end-product. The appropriate selection of the Saccharomyces cerevisiae strain and its proven ability to enhance the varietal properties of grapes are important in the production of a quality sparkling wine with a sensory profile that reflects the typicality of the grape variety [41]. Tufariello et al. [42] studied comparatively the sensory profile of sparkling wines obtained with indigenous strains of Saccharomyces cerevisiae (isolated in Salento, Apulia, Italy) and commercial strains DV10® (Lallemand Oenology, Grenaa, Denmark), respectively. For samples obtained with DV10® yeasts, high concentrations of gluconic acid were recorded, which negatively influenced the foaming properties of the end-product; however, this was not the case with the sparkling wines obtained from autochthonous strains of Saccharomyces cerevisiae. In addition, the use of selected indigenous strains also led to high proportions of volatile compounds (roses and fruity aromas), low values of volatile acidity, and high glycerol content [43]. Other studies showed a reduced flocculation capacity and low volatile acidity of indigenous strains isolated both in Apulia and in the Lombardy area (Italy), compared to commercial ones. In particular, indigenous strains of Saccharomyces cerevisiae isolated from grape berries in Apulia (Italy) were tested for tolerance to different stress factors such as pH, ethanol content, and total SO₂ level. By monitoring fermentation at 6°C and 12°C respectively, most native strains produced very low CO₂ values and exhibited low autolytic and flocculant properties, killer activity, resistance to pH 3.5 and tolerance to ethanol concentrations between 6 and 12% vol. alc., as well as at various concentrations of total SO₂ (100, 150 and 200 mg/L). Also, the volatile acidity of these samples was low, while the pressure in the bottle reached about 5 bar. In the study by Vigentini et al. [40], native strains isolated from the Oltrepò Pavese region (Italy) released low concentrations of glycerol and high amounts of hydrogen sulfide. They were characterized by increased resistance to a concentration of 12% vol. alc., reduced ability to tolerate high concentrations (300 mg/L) of SO_2, and low volatile acidity. Alfonzo et al. [3] selected indigenous strains of Saccharomyces cerevisiae for the secondary fermentation of the base wine obtained from the Grillo variety. The four strains exhibited solid resistance to fermentation and sulfur dioxide, good wine re-fermentation capacity at high total acidity and very low pH, with no off-flavors. Studies have highlighted a high level of genomic diversity within the Saccharomyces cerevisiae species.

Saccharomyces yeast strains (Saccharomyces bayanus, Saccharomyces oviformis Osterwalder) were used by Bozdogan et al. [44] in both free and immobilized form, the latter after immobilization in alginate beads to enhance the secondary fermentation of the base wines obtained from the Emir and Drimit varieties. Significant differences in amino acid concentrations were detected as a factor of the maturation period and the yeast inoculated. The beneficial effects of non-Saccharomyces yeasts (e.g., Torulaspora delbrueckii, Pichia kluyveri, Lachancea thermotolerans, and Metschnikowia pulcherrima) on still wines have been showcased extensively in the literature, but few studies research their impact on the quality of sparkling wines. González-Royo et al. [11] tested two strains of Torulospora delbrueckii and Metschnikowia pulcherrima to produce base wine from the Macabeo variety through successive inoculations with Saccharomyces cerevisiae. They found that the sparkling wine obtained using Metschnikowia pulcherrima had a longer persistence of the foam and an interesting aromatic profile, with notes of smoke and flowers. Reports show that Torulaspora delbrueckii decreases volatile acidity but increases the proportion of glycerol, thus improving the foaming properties of wine due to the autolysis of yeasts [45]. Sparkling wines produced with Saccharomycodes ludwigii and Schizosaccharomyces pombe display important changes in color characteristics, acidity, volatile profile, and biogenic amines [31]. The sequential inoculation of Torulaspora delbrueckii and Saccharomyces cerevisiae helps obtain sparkling wines with high protein content and improved foaming properties [46], but also large amounts of ethyl propanoate, isobutyric and butanoic acid, alcohols, and phenols. However, Velázquez et al. [46] propose that only Torulaspora delbrueckii yeasts be inoculated under strict conditions (high pressure and alcohol content). These yeasts do not complete the secondary fermentation of the sparkling wine and leave high amounts of reductive sugars, plus low CO₂ production and, consequently, low pressure. It is important to point out that some non-Saccharomyces yeasts can have a negative effect on sparkling wines. For example, Zygosaccharomyces species overproduce acetic acid.

Following the tendency of consumers to turn to drinks with a lower alcohol concentration, correlated with the preference of the new generation for different organoleptic sensations, Focea [47] evaluated the influence of yeasts (Saccharomyces spp.) used in the second fermentation on the quality of white sparkling wines obtained from the Muscat Ottonel variety. The results show that yeasts can influence the final quality of sparkling wines in various ways. Considering the physico-chemical characteristics, the type of inoculated yeasts was found to have a minor, though important, impact on the physico-chemical parameters. The metal composition of wine during fermentation, maturation, and storage is not stable. The analysis of variance reveals a significant influence of the metal content according to the type of inoculated yeasts. The results identify a higher content in the organic acids analyzed in the case of the base wine compared to the sparkling wines obtained at a later time. No major differences were signaled in the concentration of the main organic acids and pH values. At an early stage, the concentration of organic acids showed minor variations in relation to the type of yeasts inoculated for the second fermentation (p > 0.05). Significant changes were found between the two analysis points (6 and 11 months of storage). The yeasts administered for the second fermentation contributed to a significant increase in the concentration of most volatile compounds (p < 0.05).

The quantified volatile compounds belonged to different chemical classes including esters, acids, alcohols, and terpenes. In the first category, the presence of ethyl octanoate, ethyl decanoate, ethyl laureate, isopropyl myristate, ethyl palmitate, and ethyl oleate was noted. Their concentrations varied according to the inoculated yeasts. The data showed a significant contribution of the selected yeasts to the enrichment of the volatile fraction of the wines. Regarding organoleptic characteristics, important differences were obtained depending on the yeast product used. Ethyl octanoate and ethyl decanoate were well represented in all variants, which explains the fruity (especially banana, and apple) and floral (elder flower) notes of the samples. In our team studies, the principal component analysis describes the variations in the composition of volatile compounds of sparkling wines produced under the influence of different strains of commercial yeasts. The first factor of the data variability was closely correlated with most of the volatile compounds identified (ethyl octanoate and decanoate, 2-phenethyl acetate, ethyl laurate, hexanoic, octanoic, decanoic and 9-decenoic acid, alcohol isoamyl, 4-octanol, phenylethyl alcohol, linalool L and α-terpineol). As such, these components showed a high correlation with most of the volatile compounds identified in the samples analyzed. The first principal component that explained most of the total variability in the data was highly correlated with isoamyl acetate, ethyl decanoate, ethyl laurate, isoamyl alcohol, and linalool. For the second main component, diethyl succinate and isopropyl myristate showed high and positive values. A positive correlation of 1-heptanol with butyric acid (r close to +1) and a negative correlation with isopropyl myristate (r close to −1) was evidenced. In addition, linalool and ethyl decanoate are positively correlated, while linalool and butyric acid present a negative correlation. Ethyl octanoate, ethyl decanoate, and isoamyl alcohol, the most prevalent volatile substances in the samples analyzed, were also positively correlated. On the other hand, they are negatively correlated with butyric acid and 1-heptanol (positioned in the opposite direction). The variables related to factor 1 allow the differentiation of the samples according to the volatile fraction. From a sensory point of view, the sparkling wines analyzed were characterized by intensely fruity (bananas and apples) and floral (elder flowers) notes associated with high levels of esters (e.g., ethyl octanoate and ethyl decanoate). Regarding concentration, compounds such as isoamyl acetate, ethyl palmitate, 4-octanol or 1-heptanol did not significantly contribute to the sensory profile identified. Samples were described as balanced on the palate, with increased persistence, high acidity (conveying freshness), and good texture. A significant influence of the yeasts on the final sensory profile was noted, with the control sample showing the lowest scores for most of the descriptors followed. The variant obtained with IOC 18–2007™ boasted high acidity and effervescence, a good texture (highest note), and increased persistence. The sample labeled IOC DIVINE™ was highly rated for its intensely fruity notes (apples, melon), while the LEVULIA CRISTAL™ variant stood out for its vegetal, yeasty, and also apple character.

5. Chemical composition of sparkling wines

There are many factors involved in the chemical composition of sparkling wines, such as grape variety, cultivation technology, quality of base wine, type of inoculated yeasts for secondary fermentation, etc. However, the second fermentation and the maturation stage of the yeast sediment are key factors for the final quality of sparkling wines [48].

Water is found in the highest proportions of all existing compounds in sparkling wines (60–85%), with a direct participation in the occurrence of chemical reactions [49].

Alcohols represent, on average, 15% of the total constituents, the predominant ones being ethyl alcohol, 1-propanol, 2-phenylethanol, 2-methyl-1-propanol, etc. They can serve as solvents for some chemical constituents, participate in the formation of compounds such as esters, acetates, and also play an essential role in defining the sensory profile [50].

Acids are found in smaller proportions compared to the first two components, although they show key roles in defining the quality of sparkling wines and have a major impact on the organoleptic profile. Of these, minerals (sulfuric acid and phosphoric acid) and organic acids (tartaric acid, citric acid, malic acid, etc.) stand out [51].

Phenolic compounds can be identified in amounts up to 500 mg/L in white sparkling wines and are of particular interest due to their antioxidant capacity. This category includes phenolic acids (gallic acid, vanillic acid, gentisic acid, syringic acid, etc.), tannins (catechins, procyanidins, and proanthocyanidins), pigments (anthocyanin and flavones), which all engage in numerous redox reactions [48, 50]. Despite the obvious influence of phenolic compounds on the sensory quality of wines, there are not many studies aimed at tracking the changes in these compounds during the technological process. Jeandet et al. [24] notice an intensification of the brick-red color of sparkling wines after 15 months of maturation-aging. This process may be due to the oxidation of phenolic compounds released during autolysis of the yeast, as well as to the action of cytoplasmic (also released during autolysis) and hydrolytic enzymes. In addition, this phenomenon was reduced at the beginning of the autolysis period (between the 12th and 18th month), possibly due to the release of phenolic compounds adsorbed by the yeast. Also, a reduction in resveratrol content during aging on yeast was attributed to the adsorption capacity of the yeast for phenolic compounds.

Aroma compounds can originate in the raw material (primary aromas) or can either be formed during fermentation processes (secondary) or during the maturation-aging period (tertiary). The concentration of these compounds depends on the quality and composition characteristics of the raw material and also on the technology of obtaining sparkling wines. The odorous compounds typical of still and sparkling wines belong to the class of terpenes, phenols, esters (ethyl acetate, isoamyl acetate, benzyl acetate), thiols (4-mercapto-4-methyl-2-pentanone, 3-mercapto-1-hexanol), isoprenoids (geraniol, α-terpineol, linalool), etc. [52]. Most of the early studies related to changes in volatile compounds during the aging of sparkling wines are contradictory. Some authors find an increase in these compounds during aging, while others find a decrease in the concentration of ethyl esters and acetates. This may be accounted for by differences in the experimental conditions as well as the simultaneous degradation and synthesis of volatile compounds that occur during the aging of wine with yeast; as a result, either of these processes predominates at any given moment [48]. Pozo-Bayón et al. [49] showed a major impact of second fermentation and yeast maturation time on the proportion of volatile compounds in sparkling wines. They reported that the proportion of volatile compounds such as hexyl acetate, isopentyl acetate, ethyl butyrate, ethyl octanoate and diethyl succinate can provide information concerning the age of sparkling wines. It has also been suggested that the increased level of C13 norisoprenoids in sparkling wines aged for longer periods of time (over 21 months) may be due to the action of enzymes released during yeast autolysis on the pigments (carotenoids) present in the wine. Of these, vitispirane is considered a marker of the maturation-aging process in sparkling wines [53].

Nitrogen substances are present in wines either in mineral or organic form. These compounds are found in low concentrations in wine and usually come from the raw material. Nitrogen substances play an important role in the development of fermentation processes and foaming capacity [54]. Curioni et al. [55] reported the important antioxidant, surfactant, and antimicrobial role of these compounds. Of all nitrogen substances, amino acids (glycine, valine, proline, phenylalanine, histidine, arginine, etc.) add up to 30% of the total figure in the case of white sparkling wines. Amino acids are food for yeasts during fermentation. During fermentation, the proportion of peptides first increases and then decreases toward the end of the process. The decrease has been attributed to the consumption of peptides by the yeasts and the presence of active acid proteases in the wine [54].

The amino acid fraction of the base wine is the main source of nitrogen during fermentation. These acids also serve as precursors to aromatic compounds that contribute to the characteristics of sparkling wines. Amino acids in sparkling wines come from various sources, such as the grapes used to obtain the base wine, which are not metabolized by yeasts during growth, while others are released by yeasts at the end of fermentation or during autolysis. This may be either because more peptides than amino acids are released during autolysis, or the released amino acids (glutamic acid, arginine, and alanine) are converted by decarboxylation and deamination, which has resulted the reduction of the amino acid fraction [54].

Proteins are important in improving the quality of the effervescence and increasing the stability of the foam. It has also been reported that polysaccharides and mannoproteins play a positive role in enhancing taste perception and that some proteins and peptides released by yeasts may contribute to the sweet taste of wine. In addition, some amino acids, peptides, and nucleotides can contribute to the umami taste and, as such, have been labeled flavor enhancers. It has been found that amino acids and lipids are aroma precursors, and consequently, their release from yeast cells can also contribute to the aromatic complexity of sparkling wines [55].

Sulfur compounds are found in low proportions in sparkling wines, both in inorganic (hydrogen sulfide) and organic form (sulfones, thioalcohols, thioesters, thioethers, etc.). These compounds usually have an unpleasant odor (hatched egg, rotten cabbage or rubber) [50].

Polysaccharides are macromolecules that originate in the raw material, yeasts, bacterial, and fungal contamination of grapes (Botrytis cinerea spp.). Arabinose- and galactose-rich polysaccharides such as type II arabinogalactans and arabinans, type I and type II rhamnogalacturonans and homogalacturonans come from grape seeds, while glucans, mannans, and mannoproteins are released from the yeast either during fermentation or by enzymatic action during aging on yeast, by autolysis. Exogenous polysaccharides such as arabic gum and carboxymethyl cellulose can be found in still and sparkling wines and are considered additives. Their action in wine is dependent on their type but also on the concentration in which they are found. Arabinogalactans greatly influence the filtration process, while mannoproteins are more effective in reducing the cloudy appearance of white wines. Rhamnogalacturonans, mannoproteins, and arabinogalactans influence the aggregation of proanthocyanidins differently, which explains their varying effects on wine characteristics. These compounds could well constitute markers for monitoring the autolysis process [56].

Reductive sugars represent the main classification criterion for sparkling wine, as follows: natural brut (maximum 3 g/L), extra brut (maximum 6 g/L), brut (maximum 12 g/L), extra dry (12.01–17 g/L), dry (17.01–32 g/L), demi-dry (32.01 –50 g/L), and sweet (minimum 50.1 g/L). The reductive sugars are either the result of an incomplete or stopped fermentation or the presence of nonfermentable carbohydrates (arabinose and xylose). Alternatively, they can be added during the technological process (application of sucrose, concentrated must, etc.) [51].

In sparkling wines are usually found large amounts of gaseous compounds such as carbon dioxide, sulfur dioxide, and nitrogen [49].

The predominant vitamins in must and wine include the B complex, C, F, and P. They are growth factors for microorganisms (yeasts and bacteria) during the biochemical processes of fermentation. Vitamin content decreases following bentonite application [56].

The presence of mineral substances in sparkling wines depends on the composition of the raw material, the terroir, and the technology. The presence of potassium, iron, sodium, copper, calcium, magnesium, zinc, etc. has been noted [49].

6. Implications of the technology applied for the quality of sparkling wines

Most of the studies on sparkling wines focus on methods for improving their quality. Given that sparkling wine composition is the result of many factors, different approaches to quality management have been considered. A general trend is the diversification of the assortment range, and to this purpose, there are studies that look at the potential of different grape varieties. The varieties specific to each geographical region have been found to lend a varietal imprint that gives uniqueness to the product [57]. Along the same lines, research has targeted indigenous yeasts [58], seen as microorganisms that reflect the biodiversity of a certain area, which further supports the idea that indigenous yeast strains can be associated with the terroir [59]. However, sparkling wine fermentation is a difficult challenge for indigenous yeasts. Other authors study the influence of secondary fermentation conditions, the selected method, and aging of yeast on the quality of sparkling wines [32].

Ruiz-Moreno et al. [60] investigated the influence of prefermentative maceration and aging on the aroma profile (with emphasis on ester content) of some sparkling wines obtained from the Pedro Ximenez variety. The prefermentative maceration process kept the skin in contact with the must at 10°C for 6 hours. The Champenoise method was applied to obtain the sparkling wines, and the samples were monitored at 3, 6, and 9 months of maturation on the lees. Sparkling wines obtained with prefermentative maceration showed higher contents of branched-chain ethyl esters and cinnamic acids. On the other hand, samples obtained without maceration showed higher levels of ethyl esters of fatty acids and acetates of higher alcohols. The results underscored the impact of prefermentative maceration and aging on the aroma profile of sparkling wines. Prefermentative maceration had a significant bearing on the aroma profile of sparkling wines. Higher levels of branched-chain ethyl esters, cinnamates, fatty acid methyl esters, odd-carbon chain ethyl esters of fatty acids, and various compounds were found in sparkling wines obtained by prefermentative maceration. Ethyl heptanoate, ethyl phenylacetate, methyl hexanoate, and ethyl propanoate stood out as potential volatile markers of prefermentative maceration, while ethyl isovalerate, ethyl isobutyrate, and ethyl 2-methylbutyrate were identified as markers for maturation-aging on lees.

Tofalo et al. [61] focused on the effects of inoculation with Saccharomyces cerevisiae (F6789) and Torulaspora delbrueckii (TB1) or Starmerella bacillaris (SB48) on the quality characteristics of sparkling wines obtained by the Champenoise method. The autolytic properties and the sensory profile of the sparkling wines obtained were also assessed. Secondary fermentation was completed by all mixed and single starter cultures except those driven by Starmerella bacillaris. Sparkling wines produced with a yeast mix of Saccharomyces cerevisiae (F6789) and Starmerella bacillaris (SB48) presented the highest amounts of glycerol (6.51 g/L) and the best autolytic potential (81.98 mg leucine /L). The lowest value for the autolytic indicator was observed for samples obtained only with Saccharomyces cerevisiae – F6789 (53.96 mg leucine/L). The sparkling wines exhibited different aromatic and sensory profiles, showing higher concentrations of esters in the variants produced only with Saccharomyces cerevisiae – F6789 (88.09 mg/L) followed by those obtained with a mixture of Saccharomyces cerevisiae + Torulaspora delbrueckii (87.20 mg/L), and finally, the lowest values were recorded for samples inoculated with a mix of Saccharomyces cerevisiae and Starmerella bacillaris (81.93 mg/L). The ester content was seen to decrease with time, which may be related to adsorption on yeast and chemical hydrolysis. The highest concentrations of higher alcohols were found in sparkling wines produced with Saccharomyces cerevisiae + Torulaspora delbrueckii (27.50 mg/L). The sparkling wines obtained with Saccharomyces cerevisiae and Starmerella bacillaris had distinctive spicy notes, aroma of toasted bread, freshness, and floral smell.

Pérez-Magariño et al. [62] studied the volatile composition of some white and rosé base wines made from different native Spanish grape varieties (Verdejo, Viura, Malvasía, Albarín, Godello, Prieto Picudo, and Garnacha), as well as the quality of sparkling wines obtained by the Champenoise method. To that purpose, the number of amino acids and biogenic amines were analyzed, and the results showed that both the base and sparkling wines obtained from the Albarín, Verdejo, Godello, and Prieto Picudo varieties presented the richest volatile profile, boasting high concentrations of ethyl esters and acetates of alcohols which contribute to the definition of the fruity aroma of the wines. During aging of yeast, an increase in branched ethyl esters, ethyl lactate, and γ-butyrolactone was observed, compounded by a decrease in terpenes (mainly citronellol and geraniol). Albarín and Prieto Picudo wines had the highest concentration of amino acids. Overall, the levels of biogenic amines in all the sparkling wines studied were very low, which confirms the quality and safety of the samples.

González-Royo et al. [11] chose to assess the influence of sequential inoculation of some non-Saccharomyces (Torulaspora delbrueckii and Metschnikowia pulcherrima) and Saccharomyces cerevisiae yeasts on the composition and quality of the base wine. The data indicated an increase in the concentration of glycerol in the samples obtained with Torulaspora delbrueckii Biodiva™ (Lallemand Oenolog, France) as well as a reduction in volatile acidity. However, a positive effect on foaming capacity and foam persistence was noted. On the other hand, the strain Metschnikowia pulcherrimaFlavia® MP346 (Murphy & Son Limited, UK) caused both an increase in the persistence of the foam and a significant change in the aromatic profile by intensifying the smoky notes and the floral aroma.

Martí-Raga et al. [63] analyzed the effect of nitrogen content on the progress of the second fermentation. Three different strains were evaluated at two different fermentation temperatures (12°C and 16°C, respectively). The results showed that the nitrogen consumption during the second fermentation is very low, a fact that calls into question the common oenological practice of adding nitrogen to the base wine before fermentation, which only improves the fermentation kinetics in very old wines low in nitrogen (below 30 mg/L). As this particular study dealt primarily with the factors that affect the second fermentation, their effects on the sparkling wine, their impact on autolysis and the organoleptic characteristics of the end-product after the maturation period remain to be established.

Benucci [64] researched the effect of storage conditions after bottling on the color characteristics and sensory profile of some rosé sparkling wines. To that effect, the samples were bottled in Antique Green glass for 9 months following the disgorging stage. Different storage conditions were tested, sampled at different temperatures and different levels of lighting (30°C, in the dark; 5°C, in the dark; 5°C, under UV irradiation). The data indicated considerable variation in color intensity and hue compared to baseline values. In the case of the variants kept under UV irradiation, both an important reduction in color intensity (−22%) and a considerable increase in shade (+ 46%) were noted. Sparkling wines stored at 30°C, in the dark showed a 16% increase in color intensity and a 33% increase in hue. The variants kept at 5°C, in the dark exhibited an increase of only 4% in color intensity and 9% in hue. All samples presented significant differences in sensory profile. Thus, the first category of sparkling wines (30°C, in the dark) was associated with intense notes of burned material, while the last category of samples (5°C, UV irradiation) gave off unpleasant aromas, such as the smell of wet wool.

Caliari et al. [65] compared sparkling wines obtained from the Moscato Giallo variety produced by traditional methods, Charmat and Asti. As such, they proposed to analyze the volatile profile of the final samples. Sparkling wines produced by the traditional method had the highest concentration of volatile compounds (especially 2-phenylethanol, ethyl octanoate, linalool, and α-terpineol), while those produced by the Asti method showed the lowest concentrations. Principal component analysis confirmed the major bearing of production methods on the volatile composition of Moscato Giallo sparkling wines. These results point out that the method used to produce sparkling wines significantly influences the volatile composition of the final product.

6.1 Use of reverse osmosis

Sparkling wines entail the secondary fermentation of still wines. The quality and characteristics of the sparkling wine, the degree of stability, and the sensory properties are largely dependent on the physical–chemical composition of the grapes and the base wine, the production technology applied, the environmental conditions, etc. [39]. Currently, the alcoholic concentration of wines tends to be higher due to numerous variables, especially climate change [66]. At the same time, many consumers prefer low-alcohol products (9–13% vol. alc.) as a consequence of social aspects (traffic restrictions), but also of health problems associated with frequent consumption [67]. In recent years, several techniques have been tried out to produce wines with a low-alcohol concentration, either by using must with lower sugar concentrations, certain selected yeasts, or by the early interruption of alcoholic fermentation [68]. Also, in order to reach a predetermined alcohol concentration in wines, various practices, such as thermal or membrane-based processes, can be used. Thermal treatments often involve the degradation of some volatile compounds of interest [69]. Various membrane-based procedures are known to reduce the alcoholic concentration of wines while also contributing to preserving the initial sensory properties [68]. These procedures (nanofiltration, reverse osmosis) have a number of advantages, such as low energy consumption when working at low to moderate temperatures. Reverse osmosis involves passing the wine through a very fine filter to separate the alcohol. A certain amount of the alcohol is usually removed before combining all the elements (including the color and flavor compounds that have been filtered out) together again. It can also be used to reduce the amount of water in wine or must to concentrate the aroma compounds. The process can be used to correct some defects: reducing Brett flavor, volatile acidity, and smoke notes in a wine [70].

Carried out at low temperatures, reverse osmosis is successfully used to reduce the alcohol concentration since it generates a minor negative impact on the structure and composition of the wine. As such, it ensures the stability of the aromatic compound and the sensory characteristics [70]. As Pham et al. [71] point out, the application of reverse osmosis has a prominent effect on the quality characteristics of some red wines. Thus, besides the decrease in alcohol concentration and the loss of free sulfur dioxide, the reduction of some organic acids and total acidity were also noted, together with an increase in volatile acidity and pH. Moreover, significant effects were recorded concerning the color of the samples and astringency associated with a change in the content of phenolic compounds and anthocyanin, respectively. With regard to the volatile fraction, an important decrease in ethyl esters was obtained with the reduction in the alcohol content.

The impact of the reverse osmosis process is subject to the operating conditions applied. In this respect, Ivić et al. [72] studied the influence of variable pressure levels (2.5–5.5 MPa), with or without cooling, on some red wines. The retention of phenolic compounds depends on several factors such as the type of membrane used, the size of the pores, the polarity of the membrane and the compounds, the various interactions between the compounds and the membrane, etc. The results indicated that applying a higher pressure and cooling the retainer favor the retention of phenolic compounds and also trigger a stronger antioxidant activity. Changing the pressure did not alter the chromatic characteristics of the polymer significantly, while a high temperature generated by the absence of a cooling regime yielded a higher polymer color associated with the degradation of anthocyanin. These results were confirmed by Cotea et al. [39], who used the reverse osmosis process to obtain the base wine for the production of some sparkling wines in correlation with an appropriate cooling regime. A corresponding decrease in the alcoholic concentration of the base wine from 12.5 to 10.5% vol. alc. was recorded. The available data shows this procedure to be an effective alternative for reducing the alcohol concentration on account of its negligible influence on the physical–chemical properties of the base wines.

Research on using reverse osmosis to obtain wines with reduced alcohol concentration is limited. The causes could be the high cost of the devices and the reduced efficiency in terms of the quantity obtained in a certain time interval, correlated with energy consumption. Also, the complexity of this method requires knowledge from various domains for the efficient adjustment of the parameters.

7. Conclusions

Technological conditions directly affect the chemical composition and structure of sparkling wines, which calls for a well-structured work protocol in keeping with the intended purpose. The type of inoculated yeasts for the two fermentations is vital for defining the phenolic profile, the content of organic acids, metal ions, the volatile fraction, and the sensory characteristics of sparkling wines. Given the growing consumer trend favoring low-alcohol drinks, reverse osmosis is a worthwhile alternative technique for beverages with low alcohol content due to its minimal impact on the quality of the final product.