Chapter 6
Sherry, Port and Madeira
6.1 Introduction
The three classic fortified wines, Sherry, Port and Madeira, are very different in their sensory properties. The thing they have in common is that the grapes are grown in fairly hot regions, possibly giving young wines that are not necessarily desirable to drink as they are. In all three cases, the character of the finished wine is determined greatly by the maturation process. There are certain quality characteristics required for the grapes and there are specifications for the fermentation process for the production of the young wines. However, the wines are typically blended extensively during the maturation period and the finished wines tend to bear little recognizable resemblance to the original young wines, or even grapes. The processes used to make these wines are described in Chapter 1. Some of the chemistry relevant to the flavour and colour of these wines will be discussed in this chapter.
6.1.1 Sherry introduction
Sherry is the name of several related fortified wines made from grapes grown in Jerez de la Frontera, in the province Cadiz in the south of Spain. The hot climate of the region would be expected to produce white grapes that would turn into rather bland white table wines. Indeed the white table wine made from the Palomino grape, the main grape variety, is fairly neutral, lacking in acidity, without distinct varietal character and of little interest as a wine in its own right. However, as a result of a unique and complex method of maturation and blending, Sherry wines have evolved with great individuality and style. The neutral base wine, which forms the starting point of the Sherry manufacture, forms an excellent background for the delicate flavours produced as a result of the maturation and blending procedures.
The three main types of Sherry (Fino, Oloroso and Amontillado) are made from the base wine using different ageing techniques. In short, Sherry can be matured under flor (a layer of particular yeasts growing on top of the wine) to develop into Fino, which is pale yellow, dry and pungent. The wine can be also matured without flor yeasts to develop into Oloroso, which is usually dark brown, full bodied with a strong bouquet. A combination of flor maturation followed by a period of ageing without flor, results in Amontillado. This type of wine-making has also been adopted and modified by other wine-making regions but the discussion below will focus on these typical Sherry styles from Spain. The wines can be sold sweetened or dry and are usually fortified to 18%–20% v/v alcohol. A range of commercial products is manufactured, all of which are essentially based on the three main wine styles.
6.1.2 Port introduction
Port is a fortified wine made from grapes grown in the Douro region in Northern Portugal. It is a naturally sweet wine – part of the grape sugars are kept in the wine by arresting the fermentation process approximately halfway, by rapidly pressing the grapes and adding grape spirit (also referred to as brandy) to between 18% and 20% v/v alcohol. The fortifying spirit is not entirely neutral in character, and is thought to influence the maturation of the Port wine. Both red and White Ports are made, but White Ports only form a small part of the production.
There are two main styles of Port wine, Ruby and Tawny, both made from red grapes. Ruby Ports are red, full-bodied and often still quite fruity in character when the wines are ready to drink. Ruby Ports are aged in old wood, or larger tanks and do not usually have any wood-aged characteristics. Some special Ruby Ports (the so-called Vintage Ports) receive considerable bottle ageing, giving lighter red wines, with often a very fruity character, despite having aged for two decades or more. Tawny Ports are generally amber and have a typical flavour developed during prolonged ageing in old oak casks, which is best described as ‘crisp, nutty, with an oaky note’ and giving an impression of dryness. Within these two main styles the Port manufacturers make many products, each having its own sensory characteristics.
Although White Port is made in the same way as red and Ruby Ports, it is aged similarly to a young Ruby Port. Very little information has been published on White Ports, therefore this chapter is concerned only with red Ports.
6.1.3 Madeira introduction
In addition to Port and Sherry, Madeira is the third ‘classic’ fortified wine. The island of Madeira forms part of Portugal, lies in the Atlantic about 1000 km from the mainland of Portugal and about 750 km off the coast of North Africa. The wines are essentially shaped by their maturation, which involves heating the wine to up to 50°C, commonly referred to as the estufagem. This confers a strong and characteristic flavour on the wine. Despite this unusual method of maturation, leading no doubt to the formation of the caramel-like brown colour and distinctive flavour compounds typical for Madeira, it is only recently that scientific literature on the chemistry of these changes has been published, discussed later in this chapter.
There are both dry and sweet Madeira wines, which are drunk before and after dinner, respectively. The drier and lighter styles are Verdelho and Sercial, while the sweeter ones are Bual and Malmsey, named after the grapes from which the wines should be made. Madeira is thought to be one of the longest-living wines: the baking process, the high alcohol and the high acidity of the wine all contribute to its stability and hence, its keeping quality.
6.1.4 Comparisons between fortified wines
The three fortified wines are made by fermenting the crushed grapes, in many ways similar to the non-fortified table wines (Chapter 1). Various aspects of the general discussion in Chapters 3 and 4 regarding the basic taste and the volatile components of wine are also relevant to Sherry, Port and Madeira. Little specific published data on grape composition is available but the compounds are probably within the range of those reported in other hot climate wine-making regions. Volatile compounds formed during the fermentation are likely to be similar to those formed during the early part of table wine fermentations. However, since the maturation methods of these classic fortified wines change considerably the character of the young wines, the volatile compounds formed during fermentation are unlikely to contribute much to the flavour of the end-product. The flavour contribution of the grape to the wine is difficult to ascertain. There are very specific requirements for the grapes used for making each of the fortified wines and one can only assume that Sherry would be quite different in character had it not been made from the Palomino grape. For Port wines, many different grape varieties/cultivars are used, and there is evidence that the grape variety (or varieties) determines the character of the wine, even after the prolonged ageing process. Conversely, Madeira should be made from one of only four classic grape varieties/cultivars, and specific wine styles evolve from each variety, although a number of other varieties are also used, and in fact these, mainly Tinta Negra, account for about 88% of Madeira wine production (Elliot, 2010).
Many volatile compounds have been identified in wines, including the fortified wines. Often, non-quantified trace amounts of such compounds are reported but the presence of only small quantities of some minor volatile compounds may well characterize the typical flavour of the wine. The very different wine-making and maturation techniques of fortified wines mean that some volatile compounds seem to be characteristic of the fortified wines. Due to the similarity in volatile composition of many wines, there is also the view that quantitative differences among the volatile compounds are more important than qualitative differences.
Table wines and fortified wines differ in three ways, which are very likely to contribute significantly to the perceived flavour properties of these wines. Firstly, the ethyl alcohol level is raised to about 18–20% v/v, usually by fortification of either the fermented wine (Sherry), or by arresting the fermentation by the addition of ethyl alcohol, such as is customary in the production of Port wine. Secondly, many, but not all of the fortified wines are sweet. Port wines are all sweet and aged in the presence of a high sugar content. Sherry is aged dry, and can be sweetened before bottling. Madeira is aged dry, or sweet, depending on the wine characteristics required. Thirdly, fortified wines are all aged in ways that are not customary for table wines, usually involving a considerable amount of blending and in most cases, more oxidation than with table wines. Port wines can be aged for a minimum of three years to a decade or more in old oak casks to develop their character. Sherry undergoes numerous blending steps in the solera system (Chapter 1), while the wine is kept in oak butts, and the age of the wine is difficult to establish. In addition, the fermenting flor in Fino-style Sherry will determine the character of the wine. Madeira derives its unique flavour properties from the estufagem, or baking process.
6.1.5 Ethyl alcohol – sensory effect
The formation of alcohol by yeast fermentation of sugars in the must in fortified wine-making is similar to the fermentation procedure in table wine production and is discussed in Chapters 3 and 7. The higher concentration of ethanol in fortified wine is expected to affect the physical behaviour of the volatile compounds that contribute to the flavour of the wine. As discussed in Chapter 4, many such compounds become more soluble as the alcohol content in the water–alcohol mixture of wine is raised. This effect can be calculated by comparing the partition coefficient determined for the compound of interest at the two alcohol levels and is expected to show a lower partition coefficient at higher alcohol concentration. Thus, theoretically, many volatile compounds become less available for sensory perception when the ethanol content is increased to the level common for fortified wine. However, the higher ethanol content may assist in solubilizing compounds that have a limited solubility at lower ethanol levels, for example high molecular weight compounds. The increased amounts of such compounds dissolved in higher alcohol concentrations could contribute to the overall flavour of the wine. Thus, if we taste two almost identical wines, with the only difference being their ethanol content (for example 12% and 20% v/v), we would expect a difference between the wines, with the higher alcohol wine having less very small volatile molecules in its head-space (the air-space immediately above the liquid sample, where people take their sniff from, see Chapter 5).
However, it is difficult to predict whether sensory differences could be determined between such samples. Little information is available on the sensory effects of ethanol on the perception of flavour from alcoholic beverages (Bakker, 1995). To determine significant sensory differences in flavour, relatively large differences in flavour concentration in the head-space need to be present. In addition, the concentration of volatile compounds in the head-space of a glass is likely to vary when tasting a wine and the compounds are also perceived once they are released in the mouth via the retronasal route (Chapter 5). When tasting a wine, it is first sniffed from the glass, then some wine is taken into the mouth, where it is warmed gently, and many professional and amateur wine tasters will swirl the wine around in the mouth. These processes will assist the volatile compounds to escape from the wine and are thought to enhance our perception of the wine volatile compounds. There may also be a time effect, with the most volatile compounds being perceived first, followed by lesser volatile compounds, for which it may take more time to reach the sensory threshold concentration (Chapter 4).
6.1.6 Ethyl alcohol – chemical effect
The fortifying spirit used for both Port and Sherry contains trace volatile compounds that could affect the quality of the wines. However, the more important influence will be the change in flavour development in the presence of the higher ethanol concentration. All three wines are stored for maturation in oak and there will be an increase in wood extractives because of the higher ethanol concentration. In table wines, many esters in wine derived from the fermentation tend to be present in excess and they hydrolyse until they are in equilibrium with their component acids and alcohols. New esters may be formed chemically, usually involving compounds present in high concentrations in wine. The equilibria of the formation of esters and acetals, compounds typically formed during wine maturation (Chapter 4), are likely to be increased due to the higher concentration of alcohol than in table wines.
6.1.7 Sweetness
Sucrose in solution increases the partition coefficients (Chapter 4) but there is no data regarding the effect of the typical sugar levels in wines. Some of the data suggest that sugar does not enhance the sensory flavour intensity of tested foods (Bakker, 1995). However, cognitive associations can give an increase in flavour intensity scores. There are some direct effects of sugar on the chemical processes, in particular the longer maturation times of sweet wines is expected to give some flavour compounds due to sugar degradation processes. In Port wine and Madeira there is evidence of this occurring (see Sections 6.3.9 and 6.4.7).
6.2 Sherry
There is much less information published on the chemistry of Sherry affecting the sensory properties than on other table wines, even though Sherry production (Chapter 1) is a much more complex process. Although all Sherry is made from one grape variety/cultivar, Palomino, the ageing under flor (layer of yeast on the top of the wine) or under oxidative conditions produces a remarkable range of wines with a definite recognizable character. Volatile compounds’ production is influenced by the particular grape variety and its growing conditions, the primary fermentation process, maturation under flor, maturation without flor, storage in wood and the fractional blending system. Much of the knowledge of volatile compounds in Sherry can be attributed to the research of Webb and co-workers, as reviewed, for example, by Webb & Noble (1976). Some reviews describe detailed aspects of wine-making and focus on only the major changes in composition during maturation, without detailed information regarding the overall volatile composition (Goswell & Kunkee, 1977). A review of the different methods of Sherry production, including those used in America, Australia and South Africa, and the associated chemistry has been reported by Amerine et al. (1980). The current technology of fortified wine-making is discussed in detail by Reader & Dominguez (1994). Goswell (1986) has reviewed the microbiology of Sherry style production and Bakker (1993) has reviewed the composition of Sherry. The production of Sherry and its chemistry has been briefly reviewed recently by Jackson (2008). A general overview of flavour occurrence and formation in wine is given by Nykänen (1986).
6.2.1 Wine producers
Sherry has a history as long as its hometown Jerez de la Frontera, and English merchants have been involved in the Sherry production and trade for several centuries. Many of the Sherry companies were established in the first part of the nineteenth century although some date back to the eighteenth century. The bigger companies all have their own vineyards, but grapes are also grown by farmers. The wines are made and then matured in the bodegas, the large well-ventilated winery buildings in which the many oak casks with maturing wines are kept relatively cool. Prior to Spain joining the EU, larger companies exported their wines in bulk, and much of the preparation of the final blends and the bottling was done elsewhere, for example in England. These wines served the British market but were also exported. As part of the EU regulations, the wines are ‘bottled at source’ and the entire process of wine-making, blending and bottling is carried out in the Sherry region in Spain. There has been a considerable reorganization in the business operations as a result. Popular wine books list the Sherry companies, often including the main styles of wine they produce.
6.2.2 Commercial wine styles
Finished Sherries have very complex sensory characteristics. Fino Sherries are usually sold dry. Some fine, older wines can also be sold dry but most Sherries are sold sweetened. Fino has a pale straw colour, is very dry but without much acidity. It has a delicate, pungent bouquet and the alcoholic strength usually lies between 15.5 and 17% v/v. Manzanilla is a regional variation of the Fino style and is matured in the coastal town Sanlúcar de Barrameda. It is bone dry, with a clean and slightly bitter aftertaste, being slightly less full bodied than Fino. Manzanilla has a pale straw colour, much like a Fino and an alcoholic strength between 15.5 and 16.5% v/v. Fino and Manzanilla should both be drunk young and cold and opened bottles should be consumed within a day or so, since the flavour is not stable once the bottle has been opened. Amontillado is dry and clean, with a pungent aroma reminiscent of Fino but ‘nuttier’ and fuller bodied. It is amber and becomes darker with increasing age. The alcoholic strength is 17–18% v/v. Oloroso has a strong bouquet and is fuller bodied but has a less pungent odour than Fino or Amontillado. Even when Oloroso is dry, it has a slightly sweet aftertaste, possibly as a result of the wood extractives. Oloroso has the darkest colour, best described as dark gold and its colour intensity increases with age. Wines marketed as medium dry, medium, cream or pale cream Sherry usually are based on Oloroso wine, with an addition of Fino to lighten the colour and flavour and a small amount of Amontillado. Gonzalez Gordon (1972) describes the nomenclature used by Sherry tasters in Jerez. The descriptive terms are related to alcoholic strength, total acidity, the time the wine has matured under flor and the concentration of acetaldehyde that has developed. Other terms used are clean, dirty, soft, hard, full, empty, soft and dull.
6.2.3 Wine writers’ comments
Although the wines are made from mostly one grape variety, the complex maturation procedures and the range of producers gives an array of different tastes and smells. Clarke (1996) comments that some of the commercial blends are too much aimed at mass acceptance of the wine, resulting in ‘bland and forgettable wines’. However, he describes older Sherries as having ‘positively painful intensity of flavour, mixing sweet and sour, rich and dry all at once and all Sherry, dry or sweet should have a bite to it’. Broadbent (1979) describes the three main Sherry styles. Fino is described as having a ‘pale lemon straw colour, refined fresh flor aroma, dry light and fresh, with a long crisp finish’. Amontillado is described as ‘deeper in colour, very slightly Fino reminiscent, richer and distinctly nutty, dry to medium dry’. Oloroso is described as a total contrast to Fino, ‘deeper in colour, deep amber to warm amber brown, complete absence of flor tang, softer and sweeter on nose and palate, medium to full bodied’. The comment regarding sweeter on the nose presumably means smells associated with sweetness, as sugar itself does not have a sweet smell when in solution. Robinson (1995) describes Manzanilla and Fino as ‘very pale, delicate, prancing, palate-reviving thoroughbreds’, which are ‘bone dry and tingling with life and zest’. The other two major styles, Amontillado and Oloroso, she describes as ‘dark, nutty’ wines which can ‘thrill the palate, with its subtle shadings of mahogany and nuances that are direct and delicious results of extended ageing in oak’. Details of Sherry tasting can also be found on the web (for example wineanorak.com).
6.2.4 Grapes and must
The required base wine quality suitable for the successful establishment of flor is made from must of the Palomino grape with a specific gravity between 1.085 and 1.095, a phenolic compounds content of 300 to 600 mg L−1 (no method given but presumably expressed as gallic acid) and a total acidity between 3.5 and 4.5 g L−1 expressed as tartaric acid, equivalent to between 2 and 3 g L−1 expressed as a sulfuric acid (Goswell & Kunkee, 1977). However, titratable acidity is not a reliable indicator of acidity in Palomino must and a high pH is often associated with high contents of phenolic compounds (Reader & Dominguez, 1994), which are considered not desirable for a base Sherry to be aged under flor.
The grapes are crushed and pressed. The free run juice, containing less than 200 mg L−1 total phenolic compounds (presumably expressed as gallic acid) is most suitable for ageing under flor and is kept separate from the pressings containing higher concentrations of phenolic compounds. An addition of sulfur dioxide may be made during crushing but this is not always considered desirable and there is a trend trying to restrict its use. Sulfur dioxide inhibits the action of oxidative enzymes, thus reducing the onset of browning, but it increases the extraction of phenolic compounds from the grapes into the wine and prevents oxidation and precipitation of the phenolic compounds onto the grape solids. Its content present during fermentation is also believed to influence the development of wine aroma, since it forms complexes with carbonyls, such as acetaldehyde, ethyl 2-ketoglutarate and ethyl 4-ketobutyrate (Webb & Noble, 1976). When these complexes dissociate during maturation, the carbonyls become available to form volatile aroma compounds. When the acidity of the must is too low, it is adjusted by an addition of tartaric acid to the must.
6.2.5 Base wine
When the fermentation is finished, the dry wines have an alcoholic strength of 11–12% v/v. In November the malo-lactic fermentation, which converts the fairly sharp tasting malic acid into the softer tasting lactic acid and carbon dioxide, is completed by lactic acid bacteria naturally present in the wine. The wines are initially classified into two groups on the basis of their quality and fortified accordingly (Reader & Dominguez, 1994). Fino is made from the lighter dry wines, using mainly free run juice and some light pressings. This wine is pale yellow and has a low total of phenolic compounds (200 mg L−1; the authors did not indicate in which compound the concentration is expressed but it is probably expressed as gallic acid), a good pungent aroma, a volatile acidity between 0.3 and 0.5 g L−1 (expressed as acetic acid), a pH between 3.1 and 3.4, contains less than 100 mg L−1 sulfur dioxide and is free of any bacterial spoilage. These wines usually develop flor spontaneously.
Oloroso is made from slightly darker wines, with higher concentrations of total phenolic compounds (up to 475 mg L−1), which are less likely to develop and support flor. These base wines have a more vinous and full-bodied nose and a higher volatile acidity (0.7–0.9 g L−1). Wines made from the higher pressings and subsequently with higher total phenolic compounds (over 550 mg L−1) are classed as the lowest quality wines, although small quantities may after maturation be used for the Oloroso style. These wines are often exposed to higher temperatures during their maturation, occasionally outdoors in the sun, where they develop into dark Oloroso styles. Fino wines are fortified to 15.5% v/v alcohol, while Oloroso wines are fortified to 18.5% v/v alcohol.
6.2.6 Maturation
Wines are stored in seasoned oak casks (so-called Sherry butts), with a capacity of 500–600 L, in bodegas, which are tall, well-ventilated buildings, designed to stay relatively cool. The wine can either be matured under flor, to develop the typical Fino character, or they can be matured without flor to develop Oloroso wines. There are some major changes in volatile compounds content, either as a result of the biological ageing under flor, or due to the oxidative changes during maturation without flor. Since the concentration changes involved are relatively high, they probably impact on the typical flavour developed in the Sherry styles. These alterations in volatile compound content can directly be related to the process of maturation.
There are also minor changes in the type of volatile compounds during maturation of Sherry. Some compounds have been identified that are thought to contribute greatly to the flavour of Sherry, although the typical flavour is assumed to be due to the combined effect of several aromatic volatile compounds. The Solera system, the fractional blending system typical for Sherry, is described in Chapter 1. Variation in composition and resulting flavour of the wines between the different Sherry butts is very large; hence the fractional blending system seems essential to obtain a reliable supply of wine with similar flavour characteristics.
6.2.7 Maturation changes under flor
The flor micro-organisms, consisting of yeasts, occur in the Sherry bodegas, and provided the base Sherry has the required composition, usually a wrinkly film of yeast on the surface of the wine, the so-called flor, will establish on maturing Fino. The wines are fortified after the fermentation to 15.5% v/v alcohol (Chapter 1) with neutral spirit, which is not thought to influence the flavour of the wine. However, the resulting higher alcohol content inhibits film-forming acetic acid bacteria, which would rapidly spoil the wine. These wines are stored in 600 L butts that are kept about 80% full (content about 500 L) to maintain a high surface-to-volume ratio and stored between 15 and 20°C. Due to the temperature dependence of these organisms, there is a seasonal variation in flor yeast activity; the flor is most active between February and June and then declines until October time, when activity increases again. The dry storage conditions will also cause evaporation of water through the wooden butts, and over a four-year maturation period of Fino wines a 15% reduction of the initial volume has been estimated (Martinez de la Ossa et al., 1987b), resulting in a concentration of the wine. Much smaller evaporative losses of alcohol (0.2–0.3% v/v) have also been quoted (Reader & Dominguez, 1994). Presumably the evaporative losses will depend on the storage temperature and the humidity and vary accordingly. The cooler and more humid climate at the coast, where the Manzanillas are matured, is presumed to influence the yeast metabolism, thus resulting in differences between Manzanillas and Finos. The higher humidity is also thought to reduce the evaporative losses during maturation.
Both the origin and the taxonomy of the yeasts growing in the flor are still being researched (Reader & Dominguez, 1994). Strains from several species have been identified, including yeasts species belonging to the Saccharomyces species. Other authors suggest that Saccharomyces fermentati or S. beticus should now be known as Torulaspora delbrueckii (Kunkee & Bisson, 1993). These flor yeasts are physiologically different from the fermenting yeasts that dominate during the fermentation. S. beticus is the dominant yeast in younger wines, while S. montuliensis is dominant in older wines. There seems to be a strong relationship between wine composition and yeasts. For example, Mesa et al. (2000) used molecular techniques to characterize the yeasts in the flor and showed that small differences in the season and base wine resulted in the selection of different yeast genotypes. Flor yeast strains vary in their synthesis of volatile compounds, such as esters, higher alcohols and terpenes (Esteve-Zarzoso et al., 2001). The strains of S. cerevisiae identified in flor appear to be unique, for example they have a greater resistance to the toxicity of ethanol and acetaldehyde, and are quite different from the strains found during wine fermentation. There is still the question whether there is a succession of yeast strains involved in flor maturation or whether there is one dominant strain.
The flor needs oxygen; the blending system in the solera, requiring regular movements of wine, should ensure an adequate supply of nutrients and oxygen for the flor to thrive. The flor yeasts are in a respiratory mode, requiring the absence of glucose and concentrations of oxygen greater than that found dissolved even in a saturated wine to oxidize ethanol (Kunkee & Bisson, 1993). Hence these flor yeasts only grow as a film on the surface of the wine, unless the oxygen concentration in the wine is increased under pressure (Goswell & Kunkee, 1977). Therefore all wine technology books stress the importance of partly filled butts to ensure an adequate surface-to-liquid ratio to ensure a sufficient supply of oxygen for the flor to thrive. The layer of flor, which actively consumes oxygen, protects the bulk of the Sherry from the uptake of oxygen and prevents browning due to the oxidation of phenolic compounds – hence Fino maintains a pale yellow wine colour.
Essentially, the flor yeasts cause a second fermentation. The energy and carbon sources come from ethanol and other alcohols, while oxygen is taken up from the head-space. Flor results in numerous biochemical changes, affects the chemistry during maturation and determines the character of the wine. There are reductions in alcohol, glycerol and volatile acidity, all used as a carbon sources for flor yeast growth. Martinez de la Ossa et al. (1987a) reported reductions in glycerol from 7 to 0.2 g L−1, which may help to explain the very dry sensory perception of the Fino wines, which lack any hint of sweetness that may have been imparted by the presence of glycerol. They also reported the precipitation of potassium bitartrate during maturation, which reduces the tartrate concentration from 3.30 to 2.20 g L−1, contributing to the reduction of total acidity from 5.3 to 4.1 g L−1 and thereby, resulting in a slight increase in pH. Martinez de la Ossa et al. (1987b) have reported a considerable reduction of volatile acidity due to the flor growth, from 0.39 to 0.24 g L−1 (expressed as tartaric acid), and a reduction of 1.6% v/v in the ethanol content. There was also an increase in the acetaldehyde concentration, from 96 to 286 mg L−1, with the greatest increase occurring during the early part of the maturation. Increases have been reported to normally 260–360 mg L−1 but even higher concentrations can be formed. Acetaldehyde makes an important odour contribution to the typical oxidized, somewhat apple-like nose of Fino flor Sherries, but is also a precursor for compounds such as acetoin and 1,1-diethoxyethane and can react with certain phenolic compounds and alcohols, generating other volatile compounds, some of which are thought to be typical for Fino Sherry (see later in this section).
Possible autolysis (spontaneous rupture of cells that releases their contents) of old yeast cells that form part of the sediment on the bottom of the cask may well contribute to the typical flavour. However, little firm evidence to date is available on these more speculative flavour formation possibilities.
Interestingly, the word flor may not have the same meaning in all scientific textbooks. Evidently it was first used by Pasteur and, for example, Ribéreau-Gayon (2006) uses the term flor for spoilage of wine caused by a strain of yeast referred to as Candida mycoderma, which grows on the surface of a wine and oxidizes ethanol into carbon dioxide and water, leaving the wine flat, watery and turbid. A second more common meaning of flor, also used by Ribéreau-Gayon (2006), is the layer of micro-organisms growing in Fino Sherry but also on the yellow wines from Jura.
6.2.8 Maturation changes without flor
Fortification of young Oloroso wines to 18.5% v/v inactivates any flor yeast and prevents flor from growing or being formed. The casks are kept 95% full and their storage temperature is less critical than for the Fino wines. To produce a good Oloroso the neutral base wine needs to contain sufficient oxidizable phenolic compounds and is therefore not suitable for Fino wines. Storage under oxidative conditions results in the dark golden colour of the Oloroso, attributed to oxidation of phenolic compounds. The higher alcohol concentration, often combined with a higher storage temperature, may lead to increased extraction of phenolic compounds from the wood during maturation of Olorosos, explaining the higher concentrations of phenolic compounds. Typical maturation of Oloroso takes about eight years.
Although the wines are fermented to dryness, glycerol formed during alcoholic fermentation (7–9 g L−1) is thought to give a hint of sweetness on taste. A considerable loss in volume during maturation and usually an increase in alcoholic strength has been reported in Oloroso wines, resulting in a higher concentration of non-volatile compounds (Martinez de la Ossa et al., 1987b); the longer maturation time than with Fino makes the effect more pronounced. These authors estimated a 30–40% reduction of the initial volume during the 12 years maturation of Oloroso wine. Therefore some changes in concentration of compounds were thought to be due to this concentration effect, such as increases in alcohols (methanol, n-propanol, n-butanol, iso-butanol). However, they observed oxidation of acetaldehyde into acetic acid, resulting in increases in volatile acidity (0.21 to 0.74 g L−1 expressed as tartaric acid) and the formation of the ester ethyl acetate (from 170 to 280 mg L−1), in excess of the increase due to the concentration effect. They speculated that other oxidation and esterification processes may also occur during ageing, contributing the aroma of Oloroso wines.
6.2.9 Maturation with and without flor
Amontillado Sherry is made by maturation with flor, followed by maturation without flor. During initial maturation under flor, the wines develop all the characteristics of a Fino. However, when wine is not refreshed with additions of younger wines in the solera system, or when wine has reached a considerable age, the wine may start to lose the flor. Further fortification to about 17.5% v/v alcohol is usual, to protect the wine against spoilage by acetic acid bacteria and prevent any further development of flor. The wine is matured in a second solera system in casks 95% full and the ageing processes change, with oxidation of the wine changing the pale yellow colour to amber and dark gold, together with the development of a nutty, complex flavour typical for Amontillado.
During the maturation in the solera under flor these wines develop like Fino Sherry, with reductions in alcohol, glycerol and volatile acidity. In the second solera system during the oxidative maturation, there are increases in volatile acidity (0.21 to 0.74 g L−1 expressed as tartaric acid) and glycerol (0.14 to 4.37 g L−1), in part because of evaporative losses of water (Martinez de la Ossa et al., 1987b). Changes in the second more oxidative solera system were similar to those observed for Oloroso wines, although there were quantitative differences as a result of the prior biological maturation. Acetaldehyde was lost (from 288 to 186 mg L−1), which is attributed to oxidation to acetic acid, with an accompanying increase in ethyl acetate (from 60 to 190 mg L−1) (Martinez de la Ossa et al., 1987a).
6.2.10 Volatile compounds
Numerous publications report volatile compounds in Sherry and there are a number of compiled lists of qualitative data. Over the last decade more quantitative data has become available but small quantities of some minor volatile compounds may well characterize the typical flavour of the wine. A useful compilation of more than 130 volatile compounds identified in Sherry and Sherry-type wines has been collated by Webb & Noble (1976), and also reported by Montedoro & Bertuccioli (1986). In 1989, Maarse & Visscher listed 307 volatile compounds identified in Sherries. The compounds listed consisted of 28 alcohols, one hydrocarbon, 19 carbonyls, 47 acids, 65 esters, 16 lactones, 67 bases, four sulfur compounds, 13 acetals, one ether, 14 amides, 17 phenols, five furans, three coumarins, four dioxolanes and two dioxanes. Many of these compounds have also been identified in standard table wines also (Chapter 4) and are not especially typical for Sherry. Similarly, some of the changes during maturation are expected to be not unlike those described for table wines, as described for example, by Rapp & Mandery (1986) (see also Chapter 4). The focus in the literature tends to be on flor Sherry, with little specific information being available on Oloroso or Amontillado Sherry. Presumably, Oloroso wines are not very different from white wines aged under oxidative conditions in oak casks and some of the flavour compounds derived from the wood (such as oak lactone, see below) would be expected to be present in Oloroso wines. Maarse & Visscher (1989) do not differentiate between the Sherry styles in their compilation of volatile compounds. During maturation under flor, there tend to be increases in aliphatic and aromatic acids, terpenes and carbonyls.
Biological flor ageing does appear to give rise to numerous compounds during this stage. Brock et al. (1984) have made a qualitative study on the formation of volatile compounds during three months ageing of Palomino base wine, using a submerged flor (a variant American system whereby the flor is grown submerged in the wine). They reported the formation of 36 new compounds, which included 14 acetals, two acids, three alcohols, four carbonyls, four esters, one lactam, four lactones and four nitrogen-containing compounds. The authors suggested that many of these compounds may contribute to the characteristic aroma associated with submerged flor Sherry and may only be formed in flor Sherry.
Another study using Fino wines made from Pedro Ximenez grapes matured using flor in Montilla-Moriles (southern Spain) compared the composition of the wines before and after flor maturation (Moyano et al., 2009). There were no details regarding the differences between flor ageing in this region, compared with Jerez, although the grapes used for the wine are different. These authors reported significant changes in composition, summarized in Table 6.1, showing significant increases in aldehydes, esters, acids and lactones as a result of flor ageing. The main odour active compounds identified by these authors are also listed in the Table 6.1.
As a result of the biological changes in wines matured under flor, further reactions of the relatively large amounts of small molecular weight compounds, such as acetaldehyde and glycerol, lead to the formation of new aromatic volatile compounds. Acetaldehyde is the most abundant aldehyde in flor Sherry. Glycerol can arise from several biochemical pathways and is a natural constituent in wine. During biological maturation, glycerol tends to decrease in concentration, whereas acetaldehyde increases.
The accumulation of acetaldehyde in flor Sherry gives the oxidized odour or aroma, or ‘nose’. Four isomeric acetals formed by glycerol (and other polyols) and acetaldehyde were present in sufficient quantities to be detected and identified in Fino Sherry (Muller et al., 1978). The presence of all four acetals (cis- and trans-5-hydroxy-2-methyl-1,3-dioxane, and cis- and trans-4-hydroxymethyl-2-methyl-1,3-dioxolane) is indicative of chemical equilibration reactions, rather than the enzymatic formation. These compounds are thought to be typically formed in Sherry type wines and contribute a certain pungency. The acetal 1,1-diethoxyethane (Webb & Noble, 1976) is considered to be formed in sufficient quantities to give a green note to the wine (Jackson, 2008), although this compound has also been described as fruity (Arctander, 1967). Moyano et al. (2009) confirmed the importance of this compound in flor Sherries, and contributing to the aroma (see Table 6.1). Aliphatic acetals tend to have green or fruity characteristics, while cyclic acetals tend to be pungent.
Table 6.1 Changes in volatile composition as a result of ageing under flor of fine wines made from Pedro Ximenez grapes.
Data adapted from Moyano et al. (2009).
Lactones and the associated oxo and hydroxy compounds are important compounds, since they are thought to be the most significant volatile compounds contributing to the typical Sherry aroma. They are formed during maturation under flor. The biosynthesis of some lactones has been elucidated (Chapter 4), for example glutamic acid additions to maturing Sherry under flor enhance the concentration of some γ-lactones (Wurz et al., 1988). Solerone (4-hydroxy-5-ketohexanoic acid lactone) has been reported to contribute an aroma of Sherry (Brock et al., 1984), occasionally referred to as the Sherry lactone. Another lactone, sotolon (4,5-dimethyl-3-hydroxy-2(5H)-furanone) has also been isolated from Vin Jaune, a wine-like Sherry, made in the Jura in France (Dubois et al., 1976). The authors proposed the formation of this compound by aldol condensation of pyruvic acid and α-keto-butyric acid. The same compound has been identified in flor Sherries (Martin et al., 1990); it is present in sufficient concentrations (22–72 μg L−1, mean 41) to contribute its toasty, spicy odour to the characteristic aroma of flor Sherry. These authors confirmed the presence of sotolon in Vin Jaune (75–143 μg L−1, mean 115), and could not determine any sotolon in either non-flor Sherry or in red or white table wines. Sotolon has also been identified in wines made from botrytized grapes, in sufficient concentrations to be above the sensory threshold (2.5 ppb) (Rapp & Mandery, 1986). Numerous other lactones have been determined in Sherry, such as 4-hydroxybutanoic acid lactone, 4-hydroxydecanoic γ-lactone, δ-5-hydroxydecanoic acid lactone, γ-butyrolactone and various substituted forms (see Brock et al., 1984), some of these are also found in table wines (Maarse & Visscher, 1989; see also Chapter 4). There is insufficient information regarding the sensory contribution that this array of lactones makes to Sherry. Oak-lactone, sometimes referred to as the whisky lactone (3-methyl-γ-octalactone), has also been found in Sherry (Chapter 4); it is extracted from wood during maturation of wines and spirits. It occurs in cis (E) and the trans (Z) forms, the trans form has a sensory threshold approximately ten times lower (0.067 ppm) than that of the cis form (0.79 ppm) (Maga, 1989; see also Chapter 4, Table 4.21). It contributes an oaky, woody character to the wine, as well as its particular character to many oaked table wines and also to whisky and brandy. Moyano et al. (2009) determined both cis and trans oak lactone in ‘Fino-style wine’ aged under flor, and trans lactones was identified well above threshold concentration (Table 6.1).
Due to the complexity of the Sherry maturation, laboratory soleras have been set up to study the volatile compounds in a controlled environment. For example, Criddle et al. (1983) analysed over 100 compounds in Sherry maturing under flor. An investigation on the formation of volatile compounds under flor in a model laboratory study published both quantitative and qualitative data on the changes in the wine (Begoña Cortes et al., 1999). A number of the changes in volatile compounds reported are shown in Table 6.2, along with the initial content, the content when the flor yeast layer was fully formed after about 20 days and the content of compounds at the end of the experiment. Some data quoted by Moyano et al. (2009) have been added.
Some compounds tended to increase in the early part of maturation, while others increased after the flor was fully formed, see also Table 6.1. As expected, the data show increases in acetaldehyde, 1,1-dioxyethane and acetoin content. Most alcohols increased in concentration. The acetates of the higher alcohols decreased, except propyl acetate. Acetate esters usually decrease on maturation, accompanied by a decrease in the fresh and fruity character that these acetates impart to the wine (Rapp & Mandery, 1986; see also Chapter 4). The most abundant esters, ethyl acetate and ethyl lactate, increased in the early part of the Fino maturation, followed by a decrease in concentration. As expected the ethyl esters of other acids increased in content, except for ethyl pyruvate, presumably because there is a decrease in pyruvic acid. Two of the lactones identified in this study, γ-butyrolactone and pantalone, increased, in particular after the flor had been formed. There was also an increase in linalool, a monoterpene. The terpene alcohols linalool, nerolidol and farnesol are produced by flor yeasts (Fagan et al., 1981) and may contribute floral notes to the wine.
Table 6.2 Contents of volatile compounds in Fino Sherry.
aData extracted from Begoña Cortes et al. (1999). Measurements at the beginning of maturation, after formation of entire flor yeast film and after 250 days of maturation. bData for the threshold values are extracted from tables in Chapter 4. cThe Odour Activity Value (OAV) is the approximate ratio of the content after 250 days divided by the sensory threshold value. dValues quoted by Moyano et al. (2009).
Table 6.3 Concentration ranges in ng L−1 of ethyl 2-, 3- and 4-methylpentanoate, and ethyl cyclohexanoate in various commercial Sherry styles.
Data adapted from Campo et al. (2007).
Four branched esters have been analysed and quantified in a range of wines, with very low thresholds (see Chapter 4) and they are typical in many wines, including fortified wines (Campo et al., 2007). They suggested that ethyl 2-methylpentanoate, ethyl 3-methylpentanoate, ethyl 4-methylpentanoate and ethyl cyclohexanoate are formed by esterification reactions with ethanol and the corresponding acids formed by micro-organisms. Their concentrations are especially high in aged fortified wines, such as Sherry and can reach concentrations well above threshold concentrations (Table 6.3) and their OAV values are well above 1 and very high for ethyl 4-methylpentanoate. These compounds are thought to contribute sweet fruity notes.
A detailed gas chromatography olfactometry study elucidated the compounds contributing to the distinctive aroma of Fino Sherry and Pedro Ximenez wines (Campo et al., 2008). They confirmed that the high levels of acetaldehyde significantly contributed to the Fino aroma, and reported that diacetyl, ethyl esters of branched aliphatic acids with 4, 5 or 6 carbon atoms, 4-ethylguiacol and sotolon all contribute to the distinctive Fino character. The esters are formed by slow esterification of ethanol with acids formed by the flor yeast (see also Chapter 7). 4-Ethylguiacol is formed by yeasts (see Chapter 7). Sotolon was also identified as being present well above threshold (15 μg L−1) concentration, and is formed by aldol condensation between acetaldehyde and 2-ketobutaric acid, a reaction mediated by flor yeasts.
Regarding the Pedro Ximenez wines, Campo et al. (2008) listed 3-methylbutanal, furfural, β-damascenone, sotolon, ethyl cyclohexanoate, phenylacetaldehyde and methional. They suggested that sotolon, with concentration up to 540 μg L−1, was formed as part of sugar degradation, as has also been suggested for the formation of this compound in Madeira and Port wines. Madeira, Port wines and Pedro Ximenez wines are all aged under oxidative conditions in the presence of high sugar levels, so it is not surprising that the authors suggested that flavour formation in these wines is comparable. The authors quoted that the development of phenyl acetaldehyde and methional are dependent on the dissolved oxygen in wines, in Pedro Ximenez wines their concentrations were 68 and 20 μg L−1 respectively, whiles in Port the authors quoted values of 78 and 17 μg L−1. The PX wines also contained high levels of the branched esters and cyclohexanoate, confirming findings from Campo et al. (2007) discussed above. β-Damascenone seems specific for Pedro Ximenez wines, averaging at 10 μg L−1 but a maximum of 21.7 μg L−1 was analysed. It may well contribute to the raisin notes of Pedro Ximenez wines. In lower levels this compound is also present in wines, where it is thought to act as an aroma enhancer (see Chapter 4).
Amontillado wines have been aged biochemically and chemically, although generally they are aged for a shorter time under flor than Fino Sherry. Zea et al. (2008) did a study determining which volatile aroma compounds are typical for Amontilado wines. They concluded that the flor maturation has a strong influence on the aroma of Amontillado, and there are many similarities in aromas of Fino and Oloroso style wines, however, sotolon has the highest impact on Amontillado and Oloroso wines, whilst acetaldehyde, ethyl acetate and eugenol were typical for flor aged Fino wines.
6.2.11 Changes during maturation in phenolic compound content
Concentrations determined in young Palomino Fino wines at the beginning and the end of fermentation showed that procyanindin B1, epicatechin, caftaric acid, cis and trans p-coutaric acid and to a lesser extent caffeic acid increased during fermentation (Benitez et al., 2005). Data are given in Table 6.4. Interestingly, these authors also investigated the effect of de-stemming on the phenolic composition but there were no significant differences in phenolic content.
The phenolic compounds in Sherry can form a substrate for oxidative reactions, influencing the organoleptic properties of the product. Flavonols (quercetin, kaempferol and isorhamnetin) have been identified in maturing Finos but not in maturing Amontillados and Olorosos by Estrella et al. (1987), who postulated that their absence from the latter could be due to polymerization reactions whereby, under the oxidative conditions prevalent during maturation, they may contribute to the formation of coloured compounds. The Fino style wines were selected to be low in phenolic compounds but all wines are kept in wood at higher levels of alcohol than typical table wines, Oloroso wines being kept with the highest alcohol content.
The breakdown of lignin by ethanol gives small phenolic compounds in the wine, some of which may be further modified by yeast metabolism (Chapter 3). Estrella et al. (1986) studied non-flavonoid phenol compounds (cinnamic acids, benzoic acids, phenolic aldehydes and coumarins) during ageing in eight Sherry solera systems. They reported a steady increase in phenolic compounds but there were also considerable differences between soleras for the same Sherry style, attributed to different bodegas in which the studies were carried out. The minimum, maximum and mean concentrations of these four groups of non-flavonoid phenolic compounds calculated from their data are shown in Table 6.5. These data show that Fino contained the lowest concentrations but the differences between Oloroso and Amontillado are small. The higher concentration of alcohol in maturing Oloroso and Amontillado may have contributed to the higher extraction of these phenolic compounds from the oak casks. In addition, water loss through evaporation, fractional blending and, in Oloroso and Amontillado, the oxidative maturation conditions will have contributed to the range of values.
Table 6.4 Phenolic contents in fermenting Palomino must and young wines.
Data adapted from Benitez et al. (2005).
Table 6.5 Contents of non-flavonoid phenols in Sherrya.
aCalculated from Estrella et al. (1986).
6.3 Port wine
The production of Port is described in Chapter 1. There is relatively little scientific information on Port wine. The wine style originates from the Douro region in northern Portugal, but this style of sweet fortified wine is sufficiently popular for New World wine makers to make these wines in Australia, California and South Africa. The extent of its production varies with the local popularity of the drink. The discussion below focuses on the Portuguese Port wines. Most Port wines are red and due to the slightly different composition of the young Port wines compared with table wines, the chemistry underlying the colour changes are typical for these wines. The manufacturing procedures for the Port styles are all very similar. Besides the climate and the geological influences, both the choice of fruit and the maturation parameters determine to a great extent the chemical composition and the sensory attributes of the final product and will be discussed next.
The choice of young wine, which can be made from different varieties/cultivars produced in different parts of the extensive Douro region, coupled with a period of maturation in old oak casks (often with a capacity of 500–600 L, referred to as pipes) ranging from 3- 20 or more years, gives wines with a spectrum of different colours and flavours. Several reviews deal mainly with the technology of wine-making (Goswell & Kunkee, 1977; Goswell, 1986; Reader & Dominguez, 1994) and Bakker (1993) has reviewed the composition of Port. A brief review on the production and chemistry of Port wine is also given by Jackson (2008).
6.3.1 Port wine producers
The grapes are either grown by farmers or by the companies actually making the wines. Since traditionally these companies were also responsible for exporting the wines, they tend to be referred to as ‘shippers’. Grapes grown by farmers can be made into wine in small, often quite elementary, wineries on the farm (usually referred to as quinta), as a rule under the supervision of the shipper who has agreed to buy the wines from the farmer. There seems to be a trend for the smaller farms to sell the grapes rather than the wines to the shipper, who will then make wine at their own winery. However, some of the bigger quintas produce their own wines and some of these wines are nowadays marketed as quinta wines. Some shippers also have vineyards of their own, so for part of their wine production they can be entirely in control of the quality of the wines.
The wines are made in the Douro valley, although as a result of the rugged terrain and many small roads there is still quite some transporting to be done to deliver all the grapes to the shipper’s winery. The wines are usually matured in Port lodges downstream the river Douro, in Vila Nova de Gaia, near the quay-side of the river Douro. Oporto is just on the other side of the river. Most wine writers give lists of Port shippers or Port lodges, some with notes on the company and their wines. Many Port lodges were established in the early nineteenth century, although some date back to the seventeenth century, and quite a few have British origins.
6.3.2 Commercial Port wine styles
All finished Ports are sweet, as a result of arresting the fermentation by the addition of grape spirit, resulting in a residual sugar concentration of between 80 and 120 g L−1 and an alcohol concentration of about 20% v/v. There are three main styles of Port, i.e. Ruby, Tawny and Vintage. Within each basic style are a number of different quality categories. The maturation time of these categories is prescribed by the Portuguese authorities. Most Ruby and Tawny Ports are blends of different grape varieties/cultivars and of wines made in different years, which allow the shipper to maintain consistently its unique style, despite seasonal variation. Any indication of age on the label is a typical average age of the wine. Ruby Ports are drunk after about three to five years maturation in old oak casks and have a fruity character, a fairly deep red colour and plenty of phenolic compounds to give a full body to the wine.
Vintage Port has the year of the harvest date on the label and is wine all from one year. The wine is made in exceptionally good years, about three times in a decade and these wines command high prices at wine auctions. It is not treated before bottling and bottled at about two years after its harvest. It needs a decade or more of bottle-ageing before it is considered ready to drink. It will have thrown a considerable deposit. Hence the wine needs careful handling, ensuring that the sediment is not disturbed when removing the cork and it will need decanting before drinking to take the wine off the deposit in the bottle. Vintage Port is the premium Port wine from selected years in which the climatic conditions were ideally suited to produce excellent wine, usually from top ranking vineyards. The wine is matured in wood for two years, followed by a considerable period (often a number of decades) of bottle-ageing, during which it develops a different character to those wines solely matured in casks. The wines remain fruity and with a red colour, even after some of the colour intensity has been lost.
There are also Ruby Port wines that aim to have some of the Vintage characteristics, made in years of not quite Vintage quality. These wines will have received a combination of wood and bottle ageing, for example ‘late bottled Vintage’. Such Ports are made from wine of one specific year (usually not a Vintage year) and are bottled after four to six years of maturation in wood. Because of their longer wood-ageing period, they can be drunk much younger than Vintage Ports. Another example is wine with Vintage Character, usually a blend of several years. A currently popular style is ‘single quinta Vintage’, which is wine from one farm and one year, although not usually a year ‘declared’ by the shipper as Vintage. It can be the wine that would be declared as Vintage in Vintage years and it tends to be released on the market ready to drink.
Tawny Ports are aged ten years or more, even up to 30 years, in old oak casks and develop a nutty, raisin like, even slightly oaky character. The phenolic compounds tend to soften during the prolonged maturation and the wine gains an orange–brown Tawny colour. These quality Tawnies generally have an indication of age on the bottle and may contain wines from a number of different years. Some younger Tawnies of a more commercial style may have matured considerably shorter, using lighter wines as a base; however, these wines lack the complexity of truly aged Tawny Ports. Some shippers produce blends of real wood-aged Tawny Ports from a single vintage, referred to as ‘Colheita’ Ports.
6.3.3 Wine writers’ comments
There are many descriptions of the wines and because of the different styles and the individuality the many Port shippers impose on the wines, there is no doubt that there are many flavours and tastes in these wines. Clarke (1996) considers that all Ports will have a degree of sweetness and fieriness, the latter designed to be reduced during maturation but all Ports will need some ‘bite’ to balance their sweetness. According to Clarke:
The best Ports have a peppery background to a rich fruit, which is both plummy and raisiny, getting a slight chocolate sweetness as they become older, and managing to mix perfumes as incompatible as fresh mountain flowers, cough mixture and old leather …
Clarke (1996)
whilst he considers that ‘Tawnies have a gently brown sugar softness, with some bite to balance the sweetness’. Broadbent (1979) gives separate definitions of taste for Ruby, Tawny and Vintage Ports. According to him ‘rubies should be full and true Ruby in colour, with a fruity, peppery “nose”, not unlike a young “Vintage Port”, invariably sweet, full and fruity, often strappy and hefty’. Some of these tasting terms are not typical for the description of drinks and terms such as ‘strappy’ and ‘hefty’ are not defined and presumably indicate some mouthfeel or astringency sensation. In contrast ‘Tawnies should have an amber-Tawny hue, soft and nutty, sweet and harmonious, extended flavour and fine aftertaste’. Young Vintage wines are described as ‘deep red, very purple, peppery, alcoholic and unyielding, very sweet, full bodied, fruity and slightly rasping’. Robinson (1995) describes the great majority of Ruby Ports as ‘vigorous, juicy stuff’. She refers to cheap Tawny Ports as wines ‘lighter and browner’ than the Ruby Ports. The sensory credits are given to Tawny Ports that have aged for one or more decades in old oak casks. She describes real aged Tawny as having an ‘alluring light shaded jewel-like Tawny colour’ and continues to comment that ‘my most hedonistic Port drinking experiences have been with 20-year old [Tawny] Ports, which taste as good served chilled in the heat of the Douro summer as they do next to the fireside in a British winter’.
6.3.4 Grapes and must
Historically the wines were made from up to 60 varieties and many of these are still in production (Goswell & Kunkee, 1977). Clarke & Rand (2001) suggest that only five of these varieties/cultivars are currently recommended, Touriga Nacional, Touriga Franca (the new name for Touriga Francesa, renamed in 2001), Tinta Roriz, Tinto Cão and Tinta Barroca, while the other varieties/cultivars are permitted but not used in great quantities.
The grapes are harvested when the specific gravity of the must is between 1.090 and 1.100, with a typical total acidity between 0.39 and 0.60 g per 100 ml (expressed as tartaric acid), a pH between 3.3 and 3.7 (in warmer regions the pH can be as high as 4.0) and a phenolic compounds content between 0.4 and 0.6 g L−1 (Goswell & Kunkee, 1977). When the acidity of the must is too low it is adjusted by an addition of tartaric acid. A study on 95 Port wines made from 16 different grape varieties/cultivars grown at five different sites over a six-year period showed a wide range of grape maturity, with specific gravities from 1.071 to 1.119, dependent on variety and climatic conditions. The pH values of the must ranged from 3.27 to 3.90, resulting in Ports with a pH of 0.29 ± 0.10 units higher than the must pH (Bakker et al., 1986a).
To make wines that can mature for many years, even decades, it is important for the wines to have a sufficient amount of colour, due to the anthocyanins, and other tannins. The grape varieties/cultivars differ in their phenol and anthocyanin content, hence the wine maker will select the grapes he or she uses to make wines destined to mature and develop into either Ruby or Tawny Ports. In most vineyards, a mixture of varieties/cultivars is planted, making it difficult to pick all grapes at optimum maturity, which is one reason to plant new vineyards in blocks of the same variety.
In sixteen Vitis vinifera grape varieties/cultivars used to produce Ports, all the anthocyanins are 3-glucosides (Bakker & Timberlake, 1985a). In all but three varieties/ cultivars the anthocyanins are located in the skins and anthocyanins based on malvidin generally predominate. Malvidin 3-glucoside is the major pigment (33–94%), followed by malvidin 3-p-coumarylglucoside (1–51%) and malvidin 3-acetylglucoside (1–18%). Peonidin 3-glucoside (1–39%) is prominent in four varieties/cultivars, but delphinidin 3-glucoside (1–13%), petunidin 3-glucoside (2–12%) and cyanidin 3-glucoside (trace–6%) are present at low concentrations. These seven anthocyanins formed usually account for 90% or more of the anthocyanin composition. Although the same anthocyanins could be found in all varieties/ cultivars, the authors found that the ratio malvidin 3-acetylglucoside/total malvidin 3-glucosides (the latter being defined as the sum of the percentages malvidin 3-glucoside, malvidin 3-acetylglucoside and malvidin 3-p-coumarylglucoside) was characteristic of variety, independent of site and a useful aid to identify the grape varieties/cultivars. The structural formulae and some of the chemical properties of these compounds are described in Chapter 3.
6.3.5 Fermentation and base Port wine
The grapes are picked, crushed in the presence of sulfur dioxide, normally between 50 and 150 mg L−1 and fermented to approximately half its sugar content before stopping the fermentation by adding fortifying spirit to bring the level of ethanol to 18% v/v. The young fortified Ports may still contain a residual concentration of sulfur dioxide (20–100 mg L−1) used during crushing, which will influence the ageing mechanisms occurring in the young Ports. Young wines are usually full bodied, rather high in phenolic compounds, deep red, fruity and with a hint of grape spirit perceptible on the nose.
The short maceration on the skins means the wine maker has to ensure a sufficient extraction of the coloured anthocyanins in a very short time, unlike in red wine-making where the fermenting must as well as the wine can be left in contact with the skins if so desired. Numerous techniques in Port wine-making are tried and used to maximize extraction (for more discussion, see Reader & Dominguez, 1994). Fermentation can take place in a traditional lagar, which is an open stone trough, of varying sizes, but typically about 3 x 3 m, usually filled with approximately 0.5 m grapes, which can be ‘worked’ by foot treading or a mechanical pumping method. The fermentation can also take place in tanks, which are many times higher than wide; there is an opening at the top of the tank, used for filling, and the bottom can usually be opened to empty the tank. Two methods in current use for Port fermentation, the traditional open lagar fermentation, including treading and fermentation in a tank with pumping-over have been compared (Bakker et al., 1996), using grapes of a single variety, picked in the same vineyard and on the same day. Their results showed that the type of fermentation vessel and extraction method used during Port fermentation had a very little effect on the characteristics monitored during fermentation, such as yeast growth, sugar depletion, alcohol formation and the metabolism of amino acids. The lagar method extracted a little more coloured pigments and phenolic compounds than the tank method but as the wines matured these analytical differences became insignificant. After three years maturation the sensory quality of the finished wines was not dependent on the method of production and the analytical differences between the wines was minimal.
A biochemical method of enhancing extraction has also been experimented with. The use of commercial pectolytic enzymes, Vinozym G and Lafase H.E., to extract pigments during the short processing of crushed grape mash prior to fortification to make Port wine has been tested. The pectolytic enzyme preparations were used to evaluate the effect on colour extraction during the short processing time (Bakker et al., 1999). Results showed that both enzyme preparations enhanced colour extraction during vinification and gave darker wines than the control, without any apparent sensory disadvantages for their suitability as Ruby Ports. The wines underwent similar changes during maturation but significant differences in colour were maintained after 15 months maturation. There was also an indication of enhanced fruity character of wine made with pectolytic enzymes but this observation requires further investigation.
The short fermentation period on the skin makes an efficient extraction of anthocyanins (the coloured phenolic compounds located usually in the skins of the grapes) essential. In young single-variety Ports made over three different years from up to 16 varieties/cultivars the total contents of anthocyanins ranged from 143 to 1080 mg L−1, with an average of 330 mg L−1 (Bakker & Timberlake, 1985b). The varieties/cultivars Touriga Nacional and Touriga Francesa (now known as Touriga Franca) tended to give high contents, while Mourisco gave low contents of anthocyanins in the resulting Ports. The distribution of anthocyanins in the young Port wines differs from those determined in the grape skins. The relative amounts of malvidin 3-glucoside were higher in Ports than in grape skins, indicating some hydrolysis of the acylated malvidin 3-p-coumarylglucoside during fermentation. According to Singleton (1992) extraction of phenolic compounds and anthocyanins into wine during the wine-making process is usually only about 40% of the concentration in the grape, with 60% or more of these compounds being ‘lost’ by incomplete extraction, adsorption, precipitation with solids or proteins, conversion into non-phenolic compounds (for example by oxidation to quinones) or polymerization to an insoluble condition.
The climatic conditions also seem to exert quite an influence on the accumulation of anthocyanins (Bakker & Timberlake, 1985b). These authors reported that the concentrations of total anthocyanins in young Ports made from single varieties/cultivars from the same site over three years varied two-fold. This variation was as great as that in the same variety from different sites in any one year. Varieties/cultivars contained consistently more anthocyanins when grown in the upper Corgo (relatively hotter region) than when grown in the lower Corgo (relatively cooler region).
The effect of seasonal variation on the colour of Port wines was investigated in up to seven harvests, and the effects of five different sites were examined in 95 young Port wines (Bakker et al., 1986a). Immediately after processing up to sixteen single variety Port wines were analysed for colour, pigment (a measure of total colour potential independent of the pH of the Port) and phenol content. The authors reported that seasonal variation over six harvests was two-fold for pigments (in agreement with the two-fold variation in anthocyanin concentration, discussed above) and 1.6-fold for phenolic compounds. The differences between the varieties/cultivars were much greater than the effect of season, there was a 12-fold variation between the pigment contents of the Ports, but only a 3.6-fold variation between the phenol contents. These results show that even in this relatively warm wine-making region the seasonal variation will give wines with greatly different colours, presumably explaining why Vintage years are usually declared only two or three times in a decade.
6.3.6 Port wine compared to red table wine
The are two significant differences between fortified Port wine and table wine, in addition to their higher alcohol content discussed above. Firstly, Port wines are sweet, and during their lengthy maturation time breakdown products of the residual grape sugars may contribute to the volatile aroma compounds of the mature wine. Secondly, young Port wines tend to have higher acetaldehyde compounds than red table wines, influencing both the colour and flavour development of these wines. Especially, this second difference is thought to be relevant to the character of Port, particularly after maturation. Acetaldehyde (ethanal) is an important aroma compound in wine (see Chapter 4) and in Port wines it contributes also to the chemistry of the changes occurring during maturation. Its wide-ranging rôle in wine-making has been reviewed (Liu & Pilone, 2000). Acetaldehyde in Port wines is derived from the fermentation, the fortifying spirit, and formed during maturation.
During the fermentation the fermenting yeasts produce acetaldehyde in the must and the highest amount of acetaldehyde is formed when the yeast action is in its most vigorous phase (Whiting & Coggins, 1960; see Nykänen, 1986). Therefore, by terminating the fermentation at halfway, the acetaldehyde concentration in young Port wine remains higher than when the wine would have been allowed to ferment to dryness. This is thought to be one reason why the acetaldehyde concentration is higher in sweet fortified wines than in dry table wine. The amounts of acetaldehyde formed by yeasts also depend on the type of yeast and unpleasantly high concentrations can be formed in wines (more than 600 mg L−1, see Liu & Pilone, 2000). The acetaldehyde is in part derived from the fortifying spirit itself, which is not neutral in character, and contains appreciable amounts of acetaldehyde. During maturation of Port wines under fairly oxidative conditions, further acetaldehyde may be formed by oxidation of ethanol (Wildenradt & Singleton, 1974) and data on Port maturation suggest that acetaldehyde is formed during maturation (Bakker & Timberlake, 1986).
Acetaldehyde binds reversibly but strongly, in equimolar concentrations to sulfur dioxide, forming the acetaldehyde-bisulfite complex (CH3CH(OH)SO3 −) (see Chapter 3). In table wine the usually low content of acetaldehyde tends to be bound to any sulfur dioxide present, the latter being added as a processing aid during various stages of Port wine and table wine-making. In freshly made Port wines the molar concentration of acetaldehyde is usually in excess over sulfur dioxide, hence there is both ‘free’ and ‘bound’ acetaldehyde in Port, giving by addition ‘total’ acetaldehyde (Bakker & Timberlake, 1986). Only free acetaldehyde participates in the polymerization reactions occurring during the maturation. The presence of free acetaldehyde (thus not bound to sulfur dioxide) in Port wine forms an important contrast to table wine, contributing to the different character of the wines after maturation. However, as the sulfur dioxide concentration decreases due to oxidation reactions, bound acetaldehyde is gradually released as free acetaldehyde. The average total acetaldehyde content of 55 young Port wines is reported to be 127.3 ± 25.6 mg L−1, while the free acetaldehyde content is 65.1 ± 24.2 mg L−1 (Bakker & Timberlake, 1986). Figures for red table wines indicate a range of total acetaldehyde from 34 to 94 mg L−1, although the author stated that the free acetaldehyde would be fairly low due to binding with sulfur dioxide (see Lykänen, 1986).
6.3.7 Maturation
Young Ports are usually sweet, intensely red with a purplish tinge, high in tannins and they taste sweet but are harsh and astringent due to tannins. Their smell is fruity and reminiscent of the fortifying spirit. The colour, aroma and flavour of young Ports are due to compounds from the grape, the fermentation and the spirit. These wines need maturation to develop the complex sensory attributes typical for the various Port styles. During maturation, they become browner, changing from a deep red with a purple edge at the rim of the glass, to a brick red or even amber Tawny colour, depending largely on the length of the maturation. Maturation times of three to five years generally lead to the development of Ruby styles, while a considerably longer maturation period (up to 20 years) in wood is needed to produce some of the Tawny styles. The chemical processes underlying the changes occurring during maturation are complex and by no means all understood. Some changes are more typical for Port wines but many of the reactions occurring during the maturation of red table wines are likely to be relevant during Port maturation also. Accompanying the colour changes are alterations in perceived astringency due to modifications in the phenolic compounds. The rather harsh character of the young wines is lost, and the tannins become ‘softer’.
At the same time, the bouquet of the Port changes from its youthful fruity aroma to that of a typical Port. The final aroma depends on the length of maturation and whether the Port has matured in wood (usual for most Port wine styles) or mostly in-bottle (Vintage Port style). In the data on volatile composition of Port wines collected to date no terpenes have been reported (see discussion below), and the fruity part of the aroma of young Ports is, presumably, due to esters, in addition to compounds formed in wines with high sugar and alcohol content, matured under oxidative conditions (see Section 6.3.9).
6.3.8 Colour changes during maturation
Young wines are usually left in wooden (usually old oak) vats for two or three months after vinification. During this time yeast cells, suspended solids from the grapes and excess tartaric acid not soluble in the wine settle at the bottom of the vessel. The wine then receives its first racking and gets taken off the debris at the bottom of the vessel and put into a clean one. There is usually a certain amount of aeration involved in such racking procedures, giving the wine the oxygen that is thought to be involved in the colour changes occurring during the maturation of wines (Singleton, 1992). Unlike table wines, in which oxidation rapidly leads to detrimental effects on quality, the maturation of Port wine is thought to benefit from a limited regime of oxidation over a prolonged period of time.
The colour of young Port wines is mostly due to the red monomeric anthocyanins, but during maturation these anthocyanins are gradually ‘lost’. Polymerization reactions between anthocyanins and other phenolic compounds lead to the formation of larger so-called oligomeric or polymeric molecules, in which the anthocyanins are used as building blocks. In Port wines two different polymerization reactions are thought to occur simultaneously, although the extent of each reaction depends on the chemical concentrations, in particular of acetaldehyde. These reactions coincide with the observed losses of anthocyanins and the qualitative and quantitative changes in colour during maturation.
During the first months of maturation the Port wine becomes more intensely coloured, a change typical for Port wine, and referred to as ‘closing up’. The increase in colour can be up to 80%, measured as colour density [defined as the sum of the colour intensity measured spectrophoto-metrically at 520 nm (the red region) and 420 nm (the brown region)], depending on the free acetaldehyde concentration in the young Port wines (Bakker & Timberlake, 1986). This initial increase in colour during Port ageing is attributed to the formation of acetaldehyde-containing oligomeric pigments that are more coloured at Port pH than the anthocyanins from which they are derived (Bakker & Timberlake, 1986). Other phenolic compounds (mostly flavan-3-ols) are also involved in these reactions, leading to the formation of acetaldehyde-bridged polymers between anthocyanins and other phenolic compounds. These authors suggested that anthocyanins react strongly with free acetaldehyde but the oligomers formed become less reactive with increasing size. When the wine reached its maximum colour, at ‘closing up’, the larger oligomers are thought to become insoluble and start to precipitate. However, while the colour remains on this plateau, for a few months at most, the formation of new polymers and the loss of older and larger polymers by precipitation are thought to be in equilibrium. Gradually the precipitation process is faster than the formation of new oligomers and the colour of the Port becomes less intense. In addition, the polymers which remain soluble could become less coloured by incorporation of other colourless phenolic compounds. These changes in visible colour, viz. an increase in intensity followed by a plateau before a decrease in intensity, are accompanied by a measurable loss of monomeric anthocyanins.
The rate of acetaldehyde-induced polymerization reactions is governed by the free acetaldehyde concentration. During maturation free acetaldehyde is liberated from the acetaldehyde–bisulfite complex by oxidation of sulfur dioxide (Bakker & Timberlake, 1986). This process ceases to be important when all sulfur dioxide has been oxidized into sulfate. These authors also found that additional acetaldehyde was formed by oxidation of ethanol, confirming a previous observation (Wildenradt & Singleton, 1974). Small amounts of acetaldehyde were produced during maturation, presumably by coupled oxidation of ethanol, even in the presence of low concentrations of sulfur dioxide. The normal procedure of racking Ports during maturation is thought to give sufficient aeration for these oxidative polymerization reactions. Direct condensation of anthocyanins with other phenolic compounds also occurs but this process is slower than the condensation involving acetaldehyde. The extent of both reactions in Port wines is thought to depend on the free acetaldehyde concentration (Bakker & Timberlake, 1986), with the formation of acetaldehyde-bridged polymers being more prominent at higher concentrations.
Studies in model solutions have confirmed the occurrence of these reactions and their accompanying colour changes (Bakker et al., 1993). These authors confirmed the existence of these dimers by determining the mass of these dimers, consisting of malvidin 3-glucoside linked to catechin linked by an acetaldehyde bridge. They used, at the time, the still fairly new technique of fast-atom bombardment mass spectrometry (FABMS), which allowed the accurate mass determination of large non-volatile compounds. Model solutions mimicking Port wines (containing free acetaldehyde as well as anthocyanin and flavan-3-ol) also showed large increases in colour during the early part of storage, concurring with the rapid loss of monomeric anthocyanins. Further studies on model Port wines showed that added sulfur dioxide slowed the loss of anthocyanins and the accompanying colour changes (Picinelli et al., 1994). The mechanism of formation of acetaldehyde-containing oligomeric pigments in wine is probably similar to that described in model systems, with the formation of complexes containing an acetaldehyde link between anthocyanin and flavan-3-ol.
Colour changes during Port wine maturation are accompanied by losses in measurable anthocyanins (Bakker et al., 1986b), which can be determined by HPLC. Just after fermentation polymeric pigments contribute between 22% and 30% of the colour, indicating that the chemical changes in colour composition have already started during the short fermentation (Bakker, 1986). After 16 weeks, more or less coinciding with the ‘closing-up’ colour plateau in Port wines, 63–66% of the colour in normal Ports is due to polymers, while in a Port containing a high concentration of acetaldehyde already 91% of the colour was due to polymers. After 46 weeks of maturation, between 78 and 95% of the colour was due to polymeric pigments, with the extent of polymerization depending on the concentration of free acetaldehyde. The losses of anthocyanins are logarithmic with time and the two acylated-anthocyanins (malvidin 3-acetylglucoside and malvidin 3-coumarylglucoside) are lost faster than malvidin 3-glucoside.
Monomeric anthocyanins are lost, mostly by participation in these polymerization reactions. The colour changes in maturing wines can be measured accurately and even relatively easily seen. However, many other reactions also occur, although the extent of each one in maturing wines remains very difficult to quantify. For example, monomeric anthocyanins are usually bleached by sulfur dioxide but as the wine matures it tends to become more resistant to any bleaching (see also Chapter 3). The polymeric compounds formed due to chemical changes during maturation are more resistant to bleaching by sulfur dioxide. Little is known about the colour properties of the polymerized anthocyanins, regarding both the quantity and quality of colour on a molecular basis, so the relative contribution of polymeric compounds to wine colour is difficult to assess. Interestingly, Bakker et al. (1993) determined in model Port wine solutions that maturation in the presence of high levels of acetaldehyde gave relatively less brown solutions than ageing of model wines in the absence of acetaldehyde, indicating that the different polymerization mechanisms may lead to qualitative colour differences in Port wines.
However, anthocyanins also partake in reactions that lead to the formation of modified anthocyanins. Several new and unusual anthocyanins (called vitisins) have been identified. They were found in red Port wines in trace amounts, and were formed in Port wines during maturation (Bakker & Timberlake, 1997; Bakker et al., 1997). One of these compounds, vitisin A, consists of malvidin 3-glucoside containing a C3H2O2 link between carbon-4 and the 5-hydroxyl of the molecule. A tentative structure is proposed for vitisin B, which is decarboxy-vitisin A or malvidin 3-glucoside with a CH=CH structure linking carbon-4 and the 5-hydroxyl group. Vitisin A is less red than malvidin 3-glucoside and vitisin B is orange. Both vitisin A and B also occur in acylated forms, having the 6-position of the sugar acylated with acetic acid. A series of studies in model wines and Port wines with additions of pyruvic acid showed the formation of these vitisins, based on a reaction between malvidin 3-glucoside and pyruvic acid. These studies also confirmed the orange colour contribution these compounds make to Port wine, and vitisins could be formed even in the presence of acetaldehyde, when their formation is in competition with other polymerization mechanisms occurring in Port wines (Romero & Bakker, 2000a; 2000b; 2000c). Numerous other anthocyanins having undergone such a reaction have now been isolated and identified, belonging to the group of newly identified pyranoanthocyanins and described in more detail in Chapter 3.
These vitisin anthocyanins possess unusual chemical properties and have colour properties quite different from anthocyanins found in wines. Vitisin A is entirely protected from bleaching by sulfur dioxide, while vitisin B shows greater resistance than malvidin 3-glucoside. In wines low in acetaldehyde, some reversible bleaching of the anthocyanins can occur but in young Port wine this is unlikely. A more interesting point is the colour expression of vitisins; they are less sensitive to pH since they express more colour up to pH 7 than malvidin 3-glucoside, the dominant anthocyanin in most young wines. There is evidence of differences in the equilibrium concentrations of the various structures, with the dominant mixture containing the coloured flavylium structure and the quinonoidal base. The colour properties of the vitisins at wine pH are expected to influence the red wine colour, and may explain some of the observations made by many enologists regarding the colour changes in red Port wines during maturation, since at typical wine pH the vitisins are relatively more orange–brown than malvidin 3-glucoside, which has a much redder absorbance maximum than the vitisins. The greater colour intensity and the relatively more orange–brown colour of vitisins than the other monomeric anthocyanins in Port wines may mean that despite the low concentrations of the vitisins in Port wines, their contribution to the colour is relatively large.
All young red wines start as red Ruby-like wines; it is the length of maturation in old oak casks that slowly changes the deep red Ruby wine into a lighter orange–brown Tawny wine. One analytical method published tried to distinguish the two styles analytically and is based on measuring the colour hue of the wines through an increasingly thick layer of wine. Older Tawny style Port wines do get an apparently redder character while Ruby wines get an apparently browner character when the colour is measured through an increasingly thick layer of wine (Bakker & Timberlake, 1985c). The colour of Port wines can accurately be assessed by a trained panel and their qualitative and quantitative colour descriptions matched instrumental colour measurements of Port wines (Bakker & Arnold, 1993).
6.3.9 Volatile changes during maturation
During the long maturation period, the volatile components in Ports undergo considerable changes. Little information is available on the chemistry of these changes, although the data available on red table wines may be relevant (Chapters 3 and 4). The volatile components in Port wines are derived from the grapes, yeast fermentation and the fortifying spirit. However, the end character of the Port wine is to a considerable extent determined by the many processes taking place during the more oxidative maturation of these wines compared with table wines, such as oxidative changes, carbohydrate degradation, formation and hydrolysis of esters, formation of acetals and extraction from wood. As discussed above, the higher ethanol content compared with table wine is expected to influence the equilibria of esters and acetals and increase the extraction of compounds from the oak casks.
Of the more than 200 volatile components thus far detected in Ports, 141 have been wholly or partially identified (Simpson, 1980; Williams et al., 1983; Maarse & Visscher, 1989). These volatile compounds consisted of 14 alcohols, one diol, two phenols, two alkoxy alcohols, two alkoxy phenols, five acids, five carbonyls, three hydroxy carbonyls, 81 esters, two lactones, nine dioxolanes, two oxygen heterocyclics, one sulfur-containing component, four nitrogen-containing components, six hydrocarbons and two halogen compounds.
A number of the identified volatile components are present in high enough contents to contribute to the overall Port aroma. The organoleptic importance of the various groups of volatile compounds did not wholly explain the sensory properties of Ruby or Tawny Ports (Williams et al., 1983). Although there are some exceptions, it is not easy to discern the sensory contribution many of these volatile compounds make, since there is scant information on the typical content of the volatile compounds, nor on their typical threshold values in 20% v/v ethanol. Some of the volatile compounds in Port wines are the same as in red table wines, such as those formed during fermentation, but as a result of the different maturation processes there are qualitative and quantitative differences that are likely to contribute to the typical Port character. Table 6.6 shows some of the compounds formed in Ports and, where possible, their proposed origin is indicated.
Oxidation of alcohols during storage may account for the presence of aldehydes, in particular acetaldehyde is usually present in higher concentrations than in wine (see above). Cullere et al. (2007) determined the presence of oxidation related aldehydes in Port wines, many above their sensory threshold level. The branched aldehydes enhanced the dried fruit aroma and masked the more negative notes from the (e)-2-alkenals. An additive sensory effect was established for these compounds.
Oxidation of fusel and other alcohols in Port wines leads to carbonyls, some of which are thought to have relatively low odour thresholds, and to contribute a slightly rancid character to these fortified wines (Simpson, 1980). Although acetaldehyde will influence the maturation of the non-volatile components in the Ports its rôle as part of the overall Port aroma is not clear.
The maturation in the presence of higher ethanol and acetaldehyde concentrations as a result of the fortification, as well as the other alcohols, glycol and glycerol formed during the short fermentation can result in the formation of acetals. Acetals are generally formed in equilibrium with alcohols and aldehydes and their concentrations are thought to be higher in fortified wines than in table wines due to the higher alcohol concentration. However, they have little aroma themselves and are not thought to contribute significantly to the Port aroma. Aliphatic acetals usually contribute fruity or green aroma characteristics, whereas cyclic acetals have more pungent characteristics. Williams et al. (1983) found two acetals but their structure was not elucidated. Ferreira et al. (2002) determined acetal levels in Port wines stored under oxidative conditions to range from 9.4 to 175.3 mg L−1. They found that the concentrations of 5-hydroxy-2-methyl-1,3-dioxane and 4-hydroxymethyl-2-methyl-1,3-dioxolane increased linearly with age. The flavour threshold determined in wine was 100 mg L−1, hence these acetals are only expected to contribute to the ‘old Port wine’ aroma in wines older than 30 years. Interestingly, the authors demonstrated that the acetal formation reactions were blocked in the presence of sulfur dioxide, since it strongly binds with acetaldehyde (see Section 6.3.6).
Esters form the most abundant group of volatile compounds in table wines (Chapter 4) and also in Port wines, with ethyl esters and lower molecular weight acetates being most abundant in Ports (Williams et al., 1983). Most esters can be formed during the short yeast fermentation but the maturation will cause changes in equilibrium. The relatively large concentrations of succinates (approximately 100 μg g−1) should contribute a wine-like and fruity aroma to the overall Port bouquet (Williams et al., 1983). Their formation is attributed to esterification and trans-esterification reactions during maturation, while diethyl succinate can also be derived from maturation in wood. Esters from 2-phenylethanol have fruity sweet aromas, contributing to the bouquet. Ethyl esters of medium-chain fatty acids have generally low flavour thresholds and are expected to contribute to the fruity aroma of Ports.
Table 6.6 Volatile compounds typically found in Port wines, including their likely origin of formation.
aYF: yeast fermentation, WE: wood extraction, EM: esterification during maturation, CD: carbohydrate degradation, OA: oxidation alcohols. AF: acetal formation. bCarD: carotenoid degradation. (1) Simpson (1980). (2) Williams et al. (1983). (3) Campo et al. (2007). (4) Ferreira et al. (2003). (5) Cullere et al . (2007). (6) Ferreira et al . (2002). (7) Ferreira & Pinho (2004).
Table 6.7 Concentration ranges in ng L−1 of ethyl 2-, 3- and 4-methylpentanoate, and ethyl cyclohexanoate in various commercial Port wines.
Data adapted from Campo et al. (2007).
Recently, four new esters have been analysed and quantified in a range of wines, with low thresholds (Campo et al., 2007). The authors suggested that ethyl 2-methylpentanoate, ethyl 3-methylpentanoate, ethyl 4-methylpentanoate and ethyl cyclohexanoate are formed by esterification reactions with ethanol and the corresponding acids formed by micro-organisms. Their concentrations are especially high in aged fortified wines and data for Port wines are shown in Table 6.7. The concentrations well above threshold level with OAV’s between 3 and 50. These compounds are thought to contribute sweet fruity notes.
Oak lactone (β-methyl-γ-octalactone) is a typical wood extractive and has been reported in many wood-aged alcoholic beverages. It contributes to the nutty, oaky note of the aroma. Vitispirane has been isolated from Ports and other matured wines. Its contribution to flavour is described as camphorous and eucalyptus-like (Williams et al., 1983), although it has also been described as flowery or fruity (see Jackson, 2008). Despite the prolonged wood ageing of, in particular, Tawny Ports, no phenolic wood-derived aldehydes are listed under Port wines (Maarse & Visscher, 1989). They are listed under red wines and are expected to be present in Tawny Ports.
Table 6.8 Concentration ranges (μg L−1) of nor-isoprenoids compounds in Port wines.
Data adapted from Ferreira & Pinho (2004).
Furan derivatives, which are readily derived from carbohydrate degradation, have generally sugary, oxidized aromas and probably contribute to the overall Port aroma. A number of these compounds have been identified (Table 6.6). Older Ports tend to have higher concentrations of volatile components, resulting from carbohydrate degradation and wood extraction, such as furfural and oak lactone. Presumably some changes occurring during bottle ageing of red wines also take place in Port wines, in particular in bottle-aged Vintage Ports.
Sotolon has been identified to give the typical ‘nutty’ and ‘spicy’ character to barrel-aged Port wines (Ferreira et al., 2003). Generally the levels were founds to increase linearly with oxidative ageing, typical for ageing of Port wines in oak vats. The concentrations ranged between 5 and 958 μg L−1 for wines between 1 and 60 years old. The flavour threshold in Port wines was 19 μg L−1, hence the contribution of sotolon to flavour is significant after some ageing. Interestingly, additions of sotolon gave wines an increased sensory score of ‘age’.
Breakdown of carotenoid compounds can be related to the formation of β-ionone and β-damascenone in old bottle aged Vintage style Ports, whilst vitispirane, 2,2,6-trimethylcyclohexanone and 1,1,6-trimethyl-1,2-dihydronaphtalene are more typical for old barrel aged Ports (Ferreira & Pinho, 2004, see Table 6.8). These compounds have low sensory threshold values. Since the fermentation Port wines is stopped by adding brandy approximately halfway, carotenoids from the grapes can still be found in young Port wines (see Ferreira & Pinho, 2004). This combined with considerably longer maturation times for some of these wines than for most table wines makes these compounds more typical for mature Port wines. Bottle-aged wines develop floral and violet notes, possibly related to these compounds. Ferreira & Pinho (2004) also determined some of the factors affecting their formation, such as oxygenation (only typical for barrel aged wines) and sulfur dioxide. Studies using supplementions in Port wines with β-carotene showed increases in the formation of β-ionone and β-cyclocitral, whilst lutein additions increased the formation of β-damascenone (Ferreira et al., 2008), confirming that these compounds are likely precursors.
6.4 Madeira
Madeira is a fortified wine, a wine style evolved on the island Madeira, approximately 600 km west of the Moroccan coast and almost 1000 km from Lisbon. Fortification of the wine was introduced and this helped to stabilize the wines during their voyage to the consumers. The process of Madeira production has been described in Chapter 1. The use of heat in wine maturation is fairly unique to Madeira wines, the so-called estufagem process. It is believed to be an accidental discovery; firstly the wines were fortified to help them survive the sea journey and secondly the high temperatures the wines were subjected to during their tropical sea journeys developed all the interesting flavours now valued in these wines.
The styles of Madeira wine range from very sweet to dry, some made from a single cultivar, others made from blends of grapes or wines and even solera type blending, as described for Sherries, can be used. The estufagem process is thought to be the main distinguishing feature of this fortified wine style. They are wines aged between 3–20 years, some can age even longer.
Until recently virtually no scientific papers were published on Madeira wine and most knowledge on these wines was recorded by wine writers. However, over the last decade or so scientific publications on aspects Madeira wines has started to emerge.
6.4.1 Madeira wine producers
Madeira is a small volcanic island, ideally situated on sea trade routes, which may well have helped to establish a wine industry. Typically, the grapes are grown in small vineyards, and each of the major wine-making companies buys the grapes directly from the growers. It is estimated that most vineyards are only a fraction of a hectare. Obviously, this requires an enormous organization, ensuring the grapes are all harvested at the correct time, in good condition, with varieties correctly identified, etc. to ensure adequate quality control.
Almost a decade ago the traditional bulk wine exports have all been stopped under the EU regulations and all wines exported have to be bottled in Madeira.
6.4.2 Commercial Madeira wine styles
There are a number of main styles, mostly depending on the age on the wine. Wines are generally sold without a vintage date on the label, since a solera system (as described for Sherry) is used to prepare the final blend. Part of the skill in this wine-making process is blending, ensuring that the desired styles of wines are maintained. Reserve wine is a blend where the youngest wine is five years old, Special Reserve wine has a minimum of ten years ageing, Extra Reserve has a minimum of 15 years maturity and Vintage is more than 20 years old. A Vintage Madeira is very rare and need not be announced until decades after the wine was made. Experts claim that Vintage Madeiras can last well over 100 years. A new style emerging is Colheita – a wine made from a single harvest; these wines are thought to be almost Vintage quality but released much earlier than Vintage wines.
The most highly regarded Vitis vinifera grapes are Sercial, Verdelho, Boal and Malvasia, however, most wine books claim that these main varieties only form a small part of the total grape production, up to 12.5% and the main red variety grown is Tinta Negra. Terms such as ‘finest’, ‘selected’ and ‘choice’ indicate wines made from blends of Tinta Negra grapes.
6.4.3 Wine writers’ comments
The styles of wines are quite different, ranging from wines suitable as aperitif to digestif, depending on sweetness and intensity of flavours. The sweetness of the wines seems to be a key part of the description of its character, but as explained in wine-making, the amount of sugar left in the wine depends on when the grape spirit was added and is not a characteristic of the grape variety (see Chapter 1, Madeira). Johnson (1977) describes the wine as follows. Sercial is described as ‘light, fragrant and slightly sharp, more substantial than Fino Sherry, but still a perfect aperitif’. Verdelho gives ‘a peculiarly soft and sippable wine, the faint honey and distinct smoke make it a wine before or after meals’. Boal is ‘lighter and less sweet than Malmsey, but still a dessert wine’. Malvasia is the ‘sweetest of them all, probably the best, dark brown, very fragrant and rich, soft-textured and almost fatty, but with a tang of sharpness, a good after-dinner wine.
6.4.4 Sensory properties
One of the first scientific papers on the sensory properties analysing 52 different Madeira wines described the wines in terms of colour, aroma, taste and ‘global appreciation’, using terms such as depth of colour, aroma quality, taste quality, body, finish, tipicity and sweetness (Nogueira & Nascimento, 1999). Surprisingly, two of the terms were strange aromas and strange tastes; there were no definitions of any of the sensory terms used. A detailed study on the sensory properties of four ten year old Madeira wines made from the four classic grape varieties Malvasia, Boal, Verdelho and Sercial and three younger wines showed that descriptors characterizing the wines were candy, nutty, maderized, toasty, laquer and dried fruit (Campo et al., 2006).
6.4.5 Grapes and must
The main grapes used for Maderia wine production are white varieties of Vitis vinifera, and usually the grapes are vinified separately to produce their typical wine styles. Sercial generally is made into the driest wines, with about 25 g L−1 sugar and has a light golden colour, a firm acidity and this wine generally needs ten years maturation, although it is reported to last for a century. Wine books report that this grape is the same as the German Riesling grape, and hence a late ripener and not a good cropper. Producers on the island do not believe that Sercial is the same as Riesling. Verdelho is the softest of the four main grapes, giving the softest Madeira style wines and is usually medium dry (65 g L−1 sugar). Boal is one of the most common grapes, giving rich, brown wines, with firm acidity, and sugar levels around 90 g L−1. Malvasia, thought to originate from Greece, is also referred to as Malmsey and is often quoted as the original and best grape variety and was first planted in Madeira more than five centuries ago. It gives a dark brown wine, with a caramel richness, with a sugar content of 100 g L−1 and good acidity.
Other grapes used are Terrantez (white), Bastado and Tinta Negra (both black), and Tinta Negra makes up currently the bulk of the commercial wine styles produced.
6.4.6 Base wines maturation
The character of these wines is determined by climate, the volcanic soil, grape varieties and to a considerable extent the unique production techniques used. Fermentation traditionally took place in concrete troughs or nowadays mostly in stainless steel fermentation tanks. Small producers also use wooden casks for fermentation. In order to retain natural sweetness in the base wines, the fermentation is stopped by the addition of neutral grape spirit (95% v/v). Nowadays all Madeira wines are fortified by adding neutral grape spirit (95% v/v ethanol) to 14–18% v/v ethanol. The wines are clarified and ready for the maturation with heat treatment.
The base wine is stored in casks or larger vats in a special lodge, referred to as the ‘estufa’, which is gradually heated to 45–50°C, the temperature is raised over a two week period. Usually the wines are kept at this temperature for at least three months. The temperature is allowed to drop slowly to ambient and the wines are matured in oak casks for a further two years, or more. More heat can be used during maturation by, for example, storing casks above the heating room, or in the sun.
Depending on the desired wine style, various maturation steps take place, the use of heat and plenty of time seem to be the common factors in the process. The wines are usually fined to remove the deposit formed during the heating and further aged as required. To ensure the required commercial blends can be made, special sweetening wines (surdo) are prepared by fortifying must just after fermentation has started, or by fortifying juice (abofado). Colouring wine is made by heat concentrating must to about a third of the volume, giving this colouring wine a dark brown colour and a caramel smell.
One of the first scientific papers on Madeira wines was published by Nogueira & Nascimento (1999), publishing data characterising Madeira wine. Small analytical changes were observed during the estufagem and further maturation over ten years; none of these were very significant or different from the expected changes. Of course, significant changes in volatile composition were observed, with three year old wines having a volatile acidity ranging from 0.34–0.50 g L−1, whereas ten year old wines ranged from 0.88–1.0 g L−1.
6.4.7 Volatile compounds
Despite Madeira wines being one of the classic fortified wines, until recently very little published scientific information was available but over the last decade some scientific data on the volatile compounds has become available. Presumably this can be attributed to the development of better, faster and cheaper analytical techniques for volatile compounds (Chapter 4). During the extensive ageing in wood, changes expected to be typical for these wines are wood derived compounds and sugar breakdown compounds, in addition to the changes in ester composition typical for maturing wines. One of the first papers including aspects of volatile composition by Nogueira & Nascimento (1999) indicated that the average 5-hydroxymethylfurfural content increased both with sweetness and with the age of the wine, as expected in these sweet wines, since this is a break down product of hexoses (Table 6.9).
Methanol increased, attributed to pectine break down and acetaldehyde also increased, attributed to the enzymic decarboxylation of pyruvic acid. Acetaldehyde can also be formed under oxidative conditions in wines by oxidation of ethanol (Wildenradt & Singleton, 1974), and has been reported to occur in Port maturation (Bakker & Timberlake, 1986). Whether these oxidative conditions prevail during maturation of Madeira wines is an interesting question but there is insufficient published information available to answer.
Alves et al. (2005) publishes a characterization of the aroma profile of Madeira wine of five grape varieties using sorptive extraction techniques for sample preparation. The general compositions of volatiles are shown in Table 6.10. The authors suggested that the C13 isoprenoids may affect the Madeira wines and they reported a good correlation between cis-oak lactone and ageing, and regarded it as an important compound to characterize older wines. Both trans-oak lactone (trans-5-butyl-4-methyl-4,5-dihydro-2(3H)-furanone) and cis-oak lactone (5-butyl-4-methyl-4,5-dihydro-2(3H)-funanone were identified as significant aroma compounds derived from wood, in older Madeira wines. The cis-oak lactone has a sensory threshold of 87 μg L−1, with aroma descriptors of woody, coconut, vanilla, berry and dark chocolate (quoted by Alves et al., 2005), although coconut seems to be the main term reported (see Table 4.21). Other trace aroma compounds these authors reported to be relevant to the Madeira wine aroma were 5-hydroxymethyl furfural and 2,3-dihydroxy-6-methyl-4H-pyran-4-one; both compounds are known to be formed by Maillard reactions. The aroma of 5-hydroxymethyl furfural is a combination of cinnamon and dried fruit.
Sotolon, previously identified in other sweet wines (see Table 4.21), has been identified in Madeira wines (Camara et al., 2004; Camara et al., 2006). The concentrations of sotolon linearly increased with the age of the wine, with none detected in wines being a year old, and up 100 μg L−1 in wines six years old to 1000 μg L−1 in 25-year-old wines. It is a strong odorant and using the odour threshold value 19 μg L−1 of Port wines (Ferreira et al., 2003) odour activity values greater than 20 were observed. The concentration averages quoted were 825 μg L−1 (Malvasia), 540 μg L−1 (Boal), 430 μg L−1 (Verdelho) and 258 μg L−1 (Sercial), ranging from sweet to dry wine. Furfural, 5-methylfurfural and 5-hydroxymethylfurfural were also formed in increasing amounts during the ageing of Madeira wines, but the odour thresholds of these compounds are high and hence their contribution to these wines is not thought to be very great. 5-Ethoxymethylfurfural can be formed from 5-hydroxymethylfurfural and gives spicy and curry notes.
Table 6.9 Ranges of some volatile compounds determined in dry to sweet Madeira wines, aged between 1–10 years.
Information adapted from Nogueira & Nascimento (1999).
Table 6.10 Percentages of volatile compounds detected in Madeira wines (Sercial, Verdelho, Boal, Malvasia), total 40 compounds.
| Compounds | percentage | |
| min | max | |
| Esters | 80.7 | 89.7 |
| Carboxylic acids | 1.6 | 4.2 |
| Alcohols | 3.5 | 8.2 |
| Aldehydes | 0.9 | 3.7 |
| Pyrans | 0.2 | 1.7 |
| Lactones | <3 | |
| Monoterpenes | 0.1 | 1.4 |
| Sesquiterpenes | 0.1 | 0.8 |
| C13 Norisoprenoids | 1.7 | 6.5 |
Data adapted from Alves et al. (2005).
Analysis of esters in Madeira wines made from Sercial, Verdelho, Boal, Malvasia and Tinta Negra, ranging from sweet to dry wines were analysed by Alves et al. (2005). The authors reported the dominating esters in order of abundance as ethyl octanoate, ethyl decanoate, ethyl decenoate and diethyl succinate (Table 6.11). Octanoate and decanoate fermentation esters were higher in dry and medium dry samples. C13 Norisoprenoids were also detected, below the reported threshold limits. However, although the quantification techniques have much improved over the last decade, the authors did point out that capacity limitations of the sampling techniques may not pick up very low level concentrations of sensory important aroma compounds accurately.
Table 6.11 Concentrations (μg L-1) of esters found in five Madeira wines from single grape varieties (Sercial, Verdelho, Boal, Malvasia and Tinta Negro Mole) ranging from sweet to dry.
aAlves et al. (2005). bSee information in Table 4.10.
Some concentration differences were observed, depending on age or sweetness, as expected. For example octanoate and decanoate fermentation esters were higher in dryer wines, due to the longer fermentation times these wines had undergone. Many esters decreased in concentration during maturation. From available threshold data only some of the esters occur above the sensory threshold values, and are expected to contribute to the overall aroma of the wines. The aroma composition of Madeira wines made from the four grape varieties (Sercial, Verdelho, Boal and Malvasia) show a decrease in fatty acid esters (C6-C16) and increases in ethyl esters of diprotic acids, such as diethyl succinate (Camara et al., 2006).
Free terpenes in musts of Malvasia, Boal, Sercial and Verdelho were 132 μg L−1, 77 μg L−1, 72 μg L−1 and 63 μg L−1 respectively (Camara et al., 2004). Boal and Verdelho were characterized by neral, linalool, citronellol and β-ionone, Malvasia musts contained mostly neral and linalool, and Sercial musts contained mostly α-terpineol, β-damascenone and geraniol. β-Damascenone and β-ionone were present above threshold levels and give fruity aromas to these musts.
Terpenes and C13 norisoprenoids in Madeira wines have also been analysed (Camara et al., 2007), using a technique the authors previously verified (Camara et al., 2006). Wines were made from the four main grape varieties (Sercial, Verdelho, Boal and Malvasia) in three different vintages, the wines were sampled in all cases after eight months maturation. There were no statistically significant differences for any of the compounds in the wines made in the three years. Multivariate statistical analyses showed that the four varieties could be characterized typically by two or three compounds. Only citronellol in Boal wine is in the range of the sensory threshold level, so this compound may have a significant sensory impact on the wine. The other compound is β-damascenone, which has a very low sensory threshold, although there is some debate regarding the sensory role of β-damascenone in wine (see Section 4.3.3).
Table 6.12 Concentrations in μg L−1 (minumum and maximum) of terpenes and C13 norisoprenoids identified in wines made from Sercial, Verdelho, Boal and Malvasia from three vintages after eight months maturation.
aCamara et al. (2007). bInformation from Table 4.25. cInformation from Table 4.14.
In another study of the 68 aroma compounds analysed in aged Madeira wines made from the four grape varieties (Sercial, Verdelho, Boal and Malvasia), 33 were found to be present in concentrations above their sensory detection level, although not all were formed during maturation (Campo et al., 2006). Many compounds were present in both young and old wines, and since the tipicity of the wines seem to depend on the long and heated maturation period, it seems that the typical aroma volatiles should be formed during this type of maturation. The authors stress that not all aroma volatiles had been identified, so some key Madeira odor compounds may still need to be identified. Compounds present in all four wines above their threshold levels are listed in Table 6.13. Sotolon, phenylacetaldehyde, (Z)-whiskylactone and some volatile phenols seem to be the main compounds identified so far typical for old Madeira. Acetals formed by acetalization reactions between acetaldehyde and glycerol leads to the formation of four hetrocyclic acetals, (cis-dioxane, trans-dioxane, cis-dioxolane and trans-dioolane), which were shown to be indicators of age in Maturing Madeira wines (Camara et al., 2003).
Young Tinta Negra Mole wines have been analysed before any ageing occurred (Perestrelo et al., 2006), as shown in Table 6.14. The most important flavour compounds of these young wines were were ethyl octanoate, phenylethanal, ethylhexanoate, octanoic acid and 2-phenylethyl acetate. It would be interesting to see how these wines changed during maturation.
Table 6.13 Compounds present in four aged Madeira wines (Sercial, Verdelho, Boal and Malvasia), above sensory detection level.
| Compound | Odour description |
| Phenylacetaldehyde | green, honey |
| (Z)-whiskylactone | coconut |
| 2-Methoxy-4-vinylphenol | bitumen |
| Ethyl dihydroxycinnamate | flowery |
| Sotolon | spicy |
| 2-Methoxyphenol | smoky |
| Ethyl Cinnamate | flowery |
Information from Campo et al . (2006).
Table 6.14 Volatile compounds identified in Negra Mole wines, samples from young wines ranging from sweet to dry.
Date from Perestrelo et al. (2006).
Acetals have been found in maturing Madeira wines, at levels higher than typical for wines matured under less oxidative conditions (Camara et al., 2003). Oxidative conditions led to higher concentrations of acetaldehyde and increased formation of heterocyclic acetals (1,3-dioxanes and 1,3-dioxolanes), although there was no measured effects of the heating conditions on their formation. There were linear increases with age for both cis-5-hydroxy-3-methyl-1,3-dioxane and cis-4-hydroxymethyl-2-methyl-1,3-dioxolane. Absolute quantities were not reported, so their contribution to Madeira flavour still need to be determined, however, information on Port wines indicates their significance only in very old Port, in part due to the high threshold levels for these compounds (see Section 6.3.9).
Bibliography
General texts
Bakker, J. (1995) Flavour interactions with the food matrix and their effects on perception. In: Ingredients Interactions, Effects on Food Quality (ed. A.G. Gaonkar), pp. 411–439. Marcel Dekker Inc.
Broadbent, M. (1979) Wine Tasting. Christies’ Wine Publications, London.
Clarke, O. (1996) New Essential Wine Book. Webster/Mitchell Beazley, London.
Clarke, O. & Mead, M. (2001) Grapes and Wines. loc. cit. Ch. 2.
Jackson, R.S. (1994, 2000) Wine Science, Principles, Practice, Perception. 2nd edn. Academic Press, San Diego.
Maarse, H. & Visscher, C.A. (1989) Volatile Compounds in Food, Qualitative and Quantitative Data Volume II, 6th edn. TNO-CIVO Food Analysis Institute, Zeist (The Netherlands).
Ribéreau-Gayon, P., Glories, Y., Maujean, A. & Dubourdieu, D. (2006) Handbook of Enology, Vol I, Wine and Wine Making, 2nd Edn. John Wiley & Sons, Ltd, Chichester.
Ribéreau-Gayon, P., Glories, Y., Maujean, A. & Dubourdieu, D. (2006) Handbook of Enology, Vol II, The Chemistry of Wine, Stabilization and Treatments, 2nd Edn. John Wiley & Sons, Ltd, Chichester.
Robinson, J. (1993, 1995) Jancis Robinson’s Wine Course. BBC Books, London.
References: Sherry
Arctander, S. (1967) Perfume and Flavour Chemicals. loc. cit. Ch. 4.
Amerine, M.A., Berg, H.W., Kunkee, R.F. et al. (1980) The Technology of Wine Making. Avi Publising, Westport, USA.
Bakker, J. (1993) Sherry, Chemical composition and analysis. In: Encyclopaedia of Food Science, Food Technology and Nutrition (eds. R. Macrae, R.K. Robinson & M.Y. Sadler), pp. 4122–4126.
Begoña Cortes, M., Moreno, J.J., Zea, L., Moyano, L. & Medina, M. (1999) Response to the aroma fraction in Sherry wines subjected to accelerated biological ageing. Journal of Agriculture and Food Chemistry, 47, 3297–3302.
Benitez, P., Castro, R., Natera, R. & Garcia-Barosso, C. (2005) Effects of grape destemming on the polyphenolc and volatile content of fine Sherry wine during alcoholic fermentation. Food Science and Technology International, 11, 233–242.
Broadbent, M. (1979) Wine Tasting. Christies’ Wine Publications, London.
Brock, M.L., Kepner, R.E. & Webb, A.D. (1984) Comparison of volatiles in Palomino wine and submerged culture flor Sherry. American Journal of Enology and Viticulture, 35, 151–155.
Campo, E., Cacho, J. & Ferreira, V. (2007) Solid phase extraction, multidimensional gas chromatography determination of four novel aroma powerful ethyl esters. Assessment of their occurrence and importance in wine and other alcoholic beverages. Journal of Chromatography A, 1140, 180–188.
Campo, E., Cacho, J. & Ferreira, V. (2008) The chemical characterisation of the aroma of dessert and sparkling white wine (Pedro Ximenex, Fino, Sauternes and Cava) by gas chromatography-olfactometry and chemical quantitative analysis. Journal of Agricultural and Food Chemistry, 56, 2477–2484.
Clarke, O. (1996) New Essential Wine Book. Webster/Mitchell Beazley, London.
Criddle, W.J., Goswell, R.W. & Williams, M.A. (1983) The chemistry of Sherry maturation II. An investigation of the volatile components present in standard Sherry base wines. American Journal of Enology and Viticulture, 34, 61–71.
Dubois, P., Rigaud, J. & DeKimpe, J. (1976) Identification de la dimethyl-4,5-tetrahydrofuranedione-2,3 [Sotolon] dans le vin jaune du Jura. Lebensmittel – Wissenshaft & Technologie, 9, 366–368.
Esteve-Zarzoso, B., Peris-Toran, M.J., Garcia-Maiques, E., Uruburu, F. & Querol, A. (2001) Yeast population dynamics during the fermentation and biological aging of Sherry wines. Applied and Environmental Microbiology, 67, 2056–2061.
Estrella, M.I., Hernandez, M.T. & Olano, A. (1986) Changes in polyalcohol and phenol compound contents in the aging of Sherry wines. Food Chemistry, 20, 137–152.
Estrella, M.I., Alonso, E. & Revilla, E. (1987) Presence of flavonol aglycones in Sherry wines and changes in their content during ageing. Zeitschrift für Lebensmitteln Forschung, 184, 27–29.
Fagan, G.L., Kepner, R.E. & Webb, A.D. (1981) Production of linalool, cis and trans nerolidol, and trans-trans farnesol by Saccharomyces fermentati growing as a film simulated wine. Vitis, 20, 36–42.
Gonzalez Gordon, M. (1972) The Sherry viniculture. In: Sherry, the Noble Wine, pp. 110–136. Cassell & Co. Ltd., London.
Goswell, R.W. (1986) Microbiology of fortified wines. In: Developments in Food Microbiology 2 (ed. R.K. Robinson), pp. 1–19. Elsevier Applied Science Publishers, London.
Goswell, R.W. & Kunkee, R.E. (1977) Fortified Wines. In: Economic Microbiology volume 1, Alcoholic Beverages (ed. A.H. Rose), pp. 477–535. Academic Press, London.
Kunkee, R.E. & Bisson, L.F. (1993) Wine-making yeasts. In: The Yeasts, Vol. 5 (eds. A.H. Rose & J.S. Harrison), 2nd edn., pp. 69–127. Academic Press, London.
Maga, J.A. (1989) The contribution of wood to the flavour of alcoholic beverages. Food Reviews International, 5, 39–99.
Martin, B., Etievant, P.X. & Henry, R.N. (1990) The chemistry of sotolon: a key parameter for the study of a key component of flor Sherry wines. In: Flavour Science and Technology (eds. Y. Bessiere & A.F. Thomas), pp. 53–56. John Wiley & Sons, Ltd, Chichester.
Martinez de la Ossa, E., Perez, L. & Caro, I. (1987a) Variations of the major volatiles through the aging of Sherry. American Journal of Enology and Viticulture, 38, 293–297.
Martinez de la Ossa, E., Caro, I., Bonat, M., Perez, L. & Domecq, B. (1987b) Dry extract in Sherry and its evolution in the aging process. American Journal of Enology and Viticulture, 38, 321–325.
Mesa, J.J., Infante, J.J., Rebordinos, L., Sanchez, J.A. & Cantoral, J.M. (2000) Influence of the yeast genotypes on the enological characteristics of Sherry wines. American Journal of Enology and Viticulture, 51, 15–21.
Montedoro, G. & Bertuccioli, M. (1986) The flavours of wines, vermouth and fortified wines. loc.cit., Ch. 4.
Moyano, L., Zea, L., Villafuerte, L. & Medina, M. (2009) Comparison of odor-active compounds in Sherry wines processed from ecologically and conventionally grown Pedro Ximenez grapes. Journal of Agricultural and Food Chemistry, 57, 968–973.
Muller, C.J., Kepner, R.E. & Webb, A.D. (1978) 1,3-Dioxanes and 1,3-dioxolanes as constituents of the acetal fraction of Spanish Fino Sherry. American Journal of Enology and Viticulture, 29, 207–212.
Nykänen, L. (1986) Formation and occurrence of flavor compounds in wine and distilled alcoholic beverages. American Journal of Enology and Viticulture, 37, 84–96.
Rapp, A. & Mandery, H. (1986) Wine aroma. Experientia, 42, 873–884.
Reader, H.P. & Dominguez, M. (1994) Fortified wines: Sherry, Port and Madeira. In: Fermented Beverage Production (eds. A.G.H. Lea & J.R. Piggott), pp. 159–207. Blackie, Glasgow.
Webb, A.D. & Noble, A.C. (1976) Aroma of Sherry wines. Biotechnology and Bioengineering, Vol. XVIII, 939–1052.
Wurz, R.E.M., Kepner, R.E. & Webb, A.D. (1988) The biosynthesis of certain gamma-lactones from glutamic acid by film yeast activity on the surface of flor Sherry. American Journal of Enology and Viticulture, 39, 234–238.
Zea., L., Moyano., L. & Medina, M. (2008) Odourant active compounds in Amontillado wines obtained by combination of two consecutive ageing processes. European Food Research and Technology, 227, 1687–1692.
References: Port wine
Bakker, J. (1986) HPLC of anthocyanins in Port wines; determination of ageing rates. Vitis, 25, 203–214.
Bakker, J. (1993) Port, Chemical composition and analysis. In: Encyclopaedia of Food Science, Food Technology and Nutrition (eds. R. Macrae, R.K. Robinson & M.Y. Sadler), pp. 3658–3662. Academic Press, London.
Bakker, J., Bridle, P., Timberlake, C.F. & Arnold, G.M. (1986a) The colour, pigment and phenol content of young Port wines: effects of cultivar, season and site. Vitis, 25, 40–52.
Bakker, J. & Arnold, G.M. (1993) Analysis of sensory and chemical data for a range of red Port wines. American Journal of Enology and Viticulture, 44, 27–34.
Bakker, J., Bellworthy, S., Hogg, T.A., Kirby, R.M., Reader, H.P., Rogerson, F.S.S., Watkins, S.J. & Barnett, J.A. (1996) Two methods of Port vinification: a comparison of changes during fermentation and of characteristics of the wines. American Journal of Enology and Viticulture, 47, 37–41.
Bakker, J., Bellworthy, S., Reader, H.P. & Watkins, S.J. (1999) The effect of enzymes during vinification on colour and sensory properties of Port wines. American Journal of Enology and Viticulture, 50, 271–276.
Bakker, J., Bridle, P., Honda, T., Kuwano, H., Saito, N., Terahara, N. & Timberlake, C.F.(1997) Isolation and identification of a new anthocyanin occurring in some red wines. Phytochemistry, 44, 1375–1382.
Bakker, J. & Timberlake, C.F. (1985a) The distribution of anthocyanins in grape skin extracts of Port wine cultivars as determined by high performance liquid chromatography. Journal of the Science of Food and Agriculture, 36, 1315–1324.
Bakker, J. & Timberlake, C.F. (1985b) The distribution and content of anthocyanins in young Port wines as determined by high performance liquid chromatography. Journal of the Science of Food and Agriculture, 36, 1325–1333.
Bakker, J. & Timberlake, C.F. (1985c) An analytical method for defining a Tawny Port wine. American Journal of Enology and Viticulture, 36, 252–253.
Bakker, J. & Timberlake, C.F. (1986) The mechanism of colour changes in ageing Port wine. American Journal of Enology and Viticulture, 37, 288–292.
Bakker, J., Preston, N.W. & Timberlake, C.F. (1986b) Ageing of anthocyanins in red wines: comparison of HPLC and spectral methods. American Journal of Enology and Viticulture, 37, 121–126.
Bakker, J., Picinelli, A. & Bridle, P. (1993) Model wine solutions: colour and composition changes during ageing. Vitis, 32, 111–118.
Bakker, J. & Timberlake, C.F. (1997) The isolation and characterisation of new color stable anthocyanins occurring in some red wines. Journal of Agricultural and Food Chemistry, 45, 35–43.
Broadbent, M. (1979) Wine Tasting. Christies’ Wine Publications, London.
Clarke, O. (1996) New Essential Wine Book. Webster/Mitchell Beazley, London.
Cullere, L., Cacho, J. & Ferreira, V. (2007) An assessment of the role played by some oxidation related aldehydes in wine aroma. Journal of Agricultural and Food Chemistry, 55, 876–881.
Ferreira, A.C.S., Barbe, J.C. & Bertand, A. (2002) Heterocyclic acetals from glycerol and acetaldehyde in Port wines: Evolution with ageing. Journal of Agricultural and Food Chemistry, 50, 2560–2564.
Ferreira, A.C.S., Barbe, J.C. & Bertand, A.T. (2003) 3-Hydroxy-4,5-dimethyl-2(5H)-furanone: A key odorant of the typical aroma of oxidative aged Port wine. Journal of Agricultural and Food Chemistry, 51, 4356–4363.
Ferreira, A.C.S. & de Pinho, P.G. (2004) Nor-isoprenoids profile during Port wine ageing – influence of some technological parameters. Analytica Chimica, 513, 169–176.
Ferreira, A.C.S., Monteiro, J., Oliveira, C.M. & de Pinho, P.G. (2008) Study of major aromatic compounds in Port wines from carotenoid degradation. Food Chemistry, 110, 83–87.
Goswell, R.W. (1986) Microbiology of fortified wines. In: Developments in Food Microbiology 2 (ed. R.K. Robinson), pp. 1–19. Elsevier Applied Science Publishers, London.
Goswell, R.W. & Kunkee, R.E. (1977) Fortified wines. In: Economic Microbiology Volume 1, Alcoholic Beverages (ed. A.H. Rose), pp. 477–535. Academic Press, London.
Liu, S.Q. & Pilone, G.J. (2000) An overview of formation and roles of acetaldehyde in wine making with emphasis on microbiological implications. International Journal of Food Science and Technology, 35, 49–61.
Nykänen, L. (1986) Formation and occurrence of flavor compounds in wine and distilled alcoholic beverages. American Journal of Enology and Viticulture, 37, 84–96.
Picinelli, A., Bakker, J. & Bridle, P. (1994) Model wine solutions: effect of sulphur dioxide on colour and composition during ageing. Vitis, 33, 31–35.
Romero, C. & Bakker, J. (2000a) Effect of acetaldehyde and several acids on the formation of vitisin A in model wine anthocyanin and colour evolution. International Journal of Food Science and Technology, 35, 129–140.
Romero, C. & Bakker, J. (2000b) Effect of storage temperature and pyruvate on kinetics of anthocyanin degradation, Vitisin A derivative formation, and color characteristics of model solutions. Journal of Agricultural and Food Chemistry, 48, 2135–2141.
Romero, C. & Bakker, J. (2000c) Anthocyanin and colour evolution during maturation of four Port wines: effect of pyruvic acid addition. Journal of the Science of Food and Agriculture, 81, 252–260.
Simpson, R.F. (1980) Volatile aroma components of Australian Port wines. Journal of the Science of Food and Agriculture, 31, 214–222.
Singleton, V.L. (1992) Tannins and the qualities of wines. In: Plant Polyphenols (eds. R.W. Hemingway & P.E. Laks), pp. 859–880. Plenum Press, New York.
Wildenradt, H.L. & Singleton, V.L. (1974) The production of aldehydes as a result of oxidation of polyphenolic compounds and its relation to wine ageing. American Journal of Enology and Viticulture, 25, 119–126.
Williams, A.A., Lewis, M.J. & May, H.V. (1983) The volatile components of commercial Port wines. Journal of the Science of Food and Agriculture, 34, 311–319.
References: Madeira
Alves, R.F., Nascimento, A.M.D. & Nogueira, J.M.F. (2005) Characterization of the aroma profile of Madeira wine by sorptive extraction techniques. Analytica Chemica Acta, 546, 11–21.
Camara, J.S., Herbert, P., Marques, J.C. & Alves, M.A. (2004) Varietal flavour compounds of four grape varieties producing Madeira wines. Analytica Chimica Acta, 513, 203–207.
Camara, J.S., Marques, J.C., Alves, M.A. & Ferreira, A.C.S. (2003) Heterocyclic acetals in Madeira wines. Analytical and Bioanalytical Chemistry, 375, 1221–1224.
Camara, J.S., Marques, J.C., Alves, M.A. & Ferreira, A.C.S. (2004) 3-Hydroxy-4,5-dimethyl-2(5H)-furanone levels in fortified Madeira wines: relationship to sugar content. Journal of Agricultural Food Chemistry, 52, 6765–6769.
Camara, J.S., Alves, M.A. & Marques, J.C. (2006) Review. Development of headspace solid-phase micro-extraction-gas chromatography-mass spectrometry methodology for analysis of terpenoids in Madeira wines. Analytica Chemica Acta, 555, 191–200.
Camara, J.S., Alves, M.A. & Marques, J.C. (2006) Changes in volatile composition of Madeira wines during their oxidative ageing. Analytica Chimica Acta, 563, 188–197.
Campo, E., Fereira, V., Escudero, A., Marques, J.C. & Cacho, J. (2006) Quantitative gas chromatography – olfactometry and chemical quantitative study of the aroma of four Madeira wines. Analytica Chemica Acta, 563, 180–187.
Camara, J.S., Alves, M.A. & Marques, J.C. (2007) Classification of Boal, Malvazia, Sercial, and verdelho wines based on terpenoid patterns. Food Chemistry, 101, 475–484.
Campo, E., Cacho, J. & Ferreira, V. (2007) Solid phase extraction, multidimensional gas chromatography determination of four novel aroma powerful ethyl esters. Assessment of their occurrence and importance in wine and other alcoholic beverages. Journal of Chromatography A, 1140, 180–188.
Elliott, T. (2010) The Wines of Madeira. Trevor Elliott Publishing, Gosport.
Ferreira, A.C.S., Barbe, J.C. & Bertand, A.T. (2003) 3-Hydroxy-4,5-dimethyl-2(5H)-furanone: A key odorant of the typical aroma of oxidative aged Port wine. Journal of Agricultural and Food Chemistry, 51, 4356–4363.
Johnson, H. (1977) The World Atlas of Wine. Mitchell Beazley, London.
Nogueira, J.M.F. & Nascimento, A.M.D. (1999) Analytical characterisation of Madeira wine. Journal of Agricultural Food Chemistry, 47, 566–575.
Perestrelo, R., Fernandes, A., Albuquerque, F.F., Marques, J.C. & Camara, J.S. (2006) Analytical characterisation of the aroma of Tinta Negra Mole red wine: Identification of the main odorant compounds. Analytica Chemica Acta, 563, 154–164.