Extraction: Making Red Wines


Machine harvested Merlot (left) and hand-picked Pinot Noir (right)

One of the key issues in red wine making is getting the flavour and colour out of the skins. It’s called extraction, and how you do it is a vitally important factor in the quality of your wines. Here’s a brief illustrated explanation of this important subject.

The pulp of almost all wine grapes is colourless (the exception here being the rare teinturier grapes, such as Alicante Bouschet and Sousão, which have coloured insides). The skins, however, are richly pigmented, containing a range of compounds such as anthocyanins and tannins that are important contributors to the colour and structure of red wines.

There are many different ways of making red wine, but a common theme to all is the goal of extracting these colour and flavour components from the skins without extracting too much: a common analogy used here is making a brew, where the goal is to take out just enough flavour from the tea leaves, but not letting it stew too long. This extraction occurs via a process of maceration, which really is a bit like brewing a cup of tea.  

Pinot Noir grapes entering and leaving a crusher-destemmer

Most red wine making begins with the crushing and destemming of the grapes, which results in a liquid mass, on which floats a raft of skins and pips. One variant on this theme is to include the stems in the fermentation, which can add a lovely spicy, structural element—it’s most commonly used with some Burgundies. But if this is done, the stems need to be ‘ripe’, or else they will make the wine taste green and bitter. Another variant is to leave the bunches uncrushed, and let fermentation begin from inside the grapes: this is called carbonic maceration. But most of the time red wine fermentation will begin with this semi-liquid mass of seeds, skins, pulp and juice.

Left alone, the skins would form a solid cap on the top of the juice. Bacteria would begin to ferment at the cap–air interface, and the result would be a volatile acidity problem—the wine would reek of vinegar, and be spoiled. Winemakers avoid this by keeping the cap moist, either by plunging it regularly, or keeping it submerged by a mechanical device, or by pumping juice over it.

Plunging, also known as pigeage, is the most traditional method, and can be done by machine, or by a special pole, or even by feet. This is typically done with shallow open-top fermenters. Pumping over is potentially more disruptive, because of the forces involved, but is more appropriate for bigger, closed fermentation vessels such as stainless steel tanks. Plunging or pumping over achieve the dual goal of both keeping the cap wet (preventing it going volatile) and extracting colour and tannin.  

Three types of fermenter. On the left we have a lagar (this is a swanky one with a heat exchanger), ideal for shallow fermentation with good contact between the juice and skins. Traditionally, these are foot-trodden, and this one hails from Portugal's Douro. Below we have a row of open-top fermenters, which are quite small and ideal for processing high-quality red wines (again, this is from a winery in Portugal, this time the Alentejo). Then below left there is a rotary fermenter of the type popular in Australia.

A slightly controversial technique is to use rotary fermenters, which have agitators in them that mix up the cap and juice when the whole tank is rotated mechanically. These have been accused of producing wines that have a slight bitterness to them; advocates suggest that this is because not enough oxygen has been provided to the fermenting juice and the bitterness is a problem of ‘reduction’.

Extraction of the good stuff from the skins of red grapes can occur before fermentation, during it, or after it. If the must and skins are kept cool enough, fermentation will be delayed, and maceration will occur in an aqueous medium. Once fermentation begins, alcohol levels gradually rise, and this alcohol aids extraction. After fermentation has stopped, the maceration that occurs is carried out in an alcoholic medium, and may be more severe, taking lots of stuff out of the skins, than that which occurs before fermentation. A key winemaking decision is when to separate wine and skins after fermentation has finished.  

Red grapes removed from fermenters, ready for pressing

This is the point where pressing occurs. After the wine has been run off from the skins, the winemaker is left with a pulp that is a mix of juice, skins and pips. This gets put into a press, which squeezes out the remaining juice from the skins. How this is done – the force that is used and the type of press – determines the quality of the wine that is thus extracted, known as the pressings. These may be finished off separately from the rest of the wine, or blended back into it. Press too hard and you end up extracting bitter compounds from the skins and seeds that can have an adverse effect on quality.

A traditional basket press of the type favoured for red wines

There's a further issue here, which is that of colour. It's a bit complicated, but it is important. The main colour pigments in the skins of red wines are called anthocyanins: they are responsible for the very vivid colour of just-fermented red wines. However, they aren't very stable. To form stable pigments a variety of chemical reactions need to occur resulting in the formation of pigmented polymers. This series of reactions is only just being worked out, but the story emerging is that the presence of oxygen and the use of oak could be important here. 



The term 'tannin' is commonly used in wine circles, but many people aren't really sure exactly what it means. In this detailed article, Jamie Goode unpacks this important subject, and discusses some exciting new data that challenge the conventional wisdom on this topic.

I’m facing the usual dilemma. I’m writing on a highly technical aspect of wine science, for a mixed readership. I want to keep this piece interesting and understandable enough that non-technical types will stay with me, but I also want to include enough in-depth material so that hardcore wine science dudes will still find it compelling—I think it’s an achievable goal, but ultimately you will have to be the judge.

Why is the subject of tannins an important one for the wine trade at large, and not just winemakers and anoraks? I can think of two reasons. First, I suspect that whatever your involvement in the trade, you’ll be familiar with the term ‘tannin’ and it’s a word that you’ll have used frequently, perhaps, may I humbly suggest, without a clear idea of what you are referring to. Second, it’s a field of active current research, and data that are only now just accumulating are pointing towards a very different understanding of the role of tannins in red wines than that traditionally espoused by wine textbooks. In this feature I’ll present a brief overview of the subject and then look at the new picture that is emerging from recent research. As with many wine science topics, there’s a lot still to be learned, so much of this piece will concentrate on framing the key questions that still need answering. You’ll be relieved to hear that I’m going to try to focus on concepts and ideas, rather than spend too long on chemical structures and formulae.

Introducing Tannins

The term ‘tannin’ is an old one, and comes from the practice of using extracts from plants to cure leather (the process referred to as ‘tanning’). This process exploits one of the key properties of tannins: they have a strong tendency to link up with a range of other chemical entities, most particularly proteins. Applied to animal skins, tannins cross-link the proteins, turning something rather soft and floppy into a material that’s tough and inert enough to make shoes, belts and saddles from.

Tannins are therefore defined functionally. They are polyphenolic compounds that bind to and precipitate proteins. It’s a slightly complicated picture: not all polyphenols can act as tannins, and not all phenolics that bind proteins are tannins, but it’s still a useful definition.

Now would be a good time to introduce some of the key players in this story, in an attempt to make frighteningly chemical-sounding names understandable to a broader audience. First, we have polyphenols. These are a group of compounds that are vitally important in wine, and more specifically red wines. The name stems from the basic building block of this class of chemicals, which is the phenol group. This is a specific chemical structure that consists of a benzene ring with various additions, and it’s highly reactive. It likes to stick to other things, and an important property of phenolic compounds is that they associate spontaneously with a wide range of compounds, such as proteins and other phenolics, by means of a range of non-covalent forces (for example, hydrogen bonding and hydrophobic effects).

Before we get to tannins, we need to take a look at the other group of polyphenolic compounds that are also key participants in this story, the anthocyanins. These are the red/blue/black pigments in grapes, which are almost always found in the skins, giving ‘red’ grapes their colour. Five different anthocyanin compounds are found in red wines, the dominant one being malvidin. We’ll come back to these later when we discuss the connection between tannins, wine quality and wine colour.

Tannins themselves are found principally in the bark, leaves and immature fruit of a wide range of plants. They form complexes with proteins and other plant polymers such as polysaccharides. It is thought that the role of tannins in nature is one of plant defence: they have an astringent, aversive taste that is off-putting to wannabe herbivores. As an animal or insect begins to munch on plant tissue, the tannins are released from cellular compartments and bind with the proteins and other cell components, making them taste unpleasant and rather indigestible. Significantly for winemaking, the grape vine exploits tannins in a rather clever way in its fruit. Grapes start life small, green, mean, and extremely bitter and astringent, through a combination of searingly high acidity and green, aggressive tannins. The grapes are also camouflaged green, the same colour as the rest of the plant. This is because the grape berry’s function in life is to act as a carrier for seeds, and it doesn’t want birds to eat them all before they’re ready. The idea is that the palatability and attractiveness of the berry is timed to coincide with the ripeness of the seed: at the right time, the berry changes colour so it stands out, acidity diminishes, sugar increases and the bitter tannins soften, in order to make it attractive. The birds eat the berries and some time later, the seeds are deposited in a new location. The change in colour from green to red (or purple or black) is brought about through the anthocyanin pigments in the skins. 

Chemically, tannins are large molecules made up of linked subunits. Molecules such as this are known as polymers, with the subunits termed monomers. The monomers here are phenolic compounds that are joined together in a bewildering array of combinations, and can be further modified chemically in a myriad of different permutations.

Exploring Tannins Further

While tannins exist in grapes, what we are actually interested in is the tannins that are found in wine. There’s a difference. Wine tannins come from grape skins, stems and seeds, and their extraction is heavily dependent on the particular winemaking process involved. Some tannins also come from barrels, particularly new ones, where these are used to age wine. The complicating factor here is that the chemical make-up of the tannins is actually changed during the winemaking process. Not only does the chain length change, but the different chemical entities that stick to the sticky bits of the phenolic subunits also changes. According to Dr Paul Smith, a chemist at the Australian Wine Research Institute (AWRI) who’s working on tannins, ‘Wine tannins constitute “evolved” grape tannins plus some grape tannins in the same chemical state as they were in the grape.’ Dr Leigh Francis, also of the AWRI, expands on this: ‘Wine tannins are considered more complex than grape tannins due to the various chemical reactions that occur during winemaking and storage’.

There are two major classes of tannins: condensed and hydrolysable. Hydrolysable tannins aren’t as important in wine: if they’re present, they’ll have most likely come from the oak barrels the wine is fermented and/or aged in. The condensed tannins, also known as proanthocyanidins, are the main grape-derived tannins. They are formed by the polymerization of the polyphenolic flavan-3-ol monomers catechin and epicatechin. These subunits form chains of varying length, referred to by the unit ‘dp’ (for degree of polymerisation, i.e. the number of flavan-3-ol subunits). The main variables in the characteristics of these tannins are the length of the polymer chain and the nature of the individual subunits that compose it. In wine, the bonds between tannin polymers are repeatedly breaking and reforming. Thus a picture is emerging of a complex, dynamic process: the various phenolic subunits of tannins are sticking to each other and other chemical components of the wine in a sequential pattern, with these bonds being broken and reformed in a temporal sequence. No wonder it is a hard subject to study, and that only now, with highly sophisticated analytical devices, are scientists beginning to get a handle on tannins in wine.

Sensing Tannins: Mouthfeel and Astringency

One current research direction involves attempts to work out the relationship between tannin structure and ‘mouthfeel’ of red wines. Tannins contribute two characteristics to red wine character, astringency (most significantly) and bitterness—these are sensations that are sometimes confused by tasters. Bitter perception is quite well understood, since it is one of the five primary tastes and is sensed by a specific receptor found in taste buds on the tongue and soft palate. Astringency perception is much less well understood: the common understanding is that it is actually mediated by the sense of touch rather than by taste. Tannins are thought to taste astringent because they bind with salivary proline-rich proteins and precipitate them out. This leads to increased friction between mouth surfaces, and a sense of dryness or roughness. The term ‘mouthfeel’ has been coined to describe the sensation of wine in the mouth, and it is now recognized that this is an important property of red wines.

Researchers at the AWRI are collaborating with INRA Montpellier to attempt to correlate the mouthfeel properties of different tannins with their structure and composition. It’s an important but daunting task, not least because it is hard to isolate sufficient amounts of well-defined tannin fractions to do rigorous experiments with. ‘The tannins we have isolated are still mixtures, with average degrees of polymerization‘, explains Leigh Francis. ‘We also chemically study the samples carefully using chromatography techniques and mass spectrometry, to find the types of units making up the tannins’. Getting the right tannins to test is only the starting point, though. In order to make the experiments realistic enough to provide relevant results, Francis and his colleagues Liz Waters and Veronique Cheynier have devised a ‘model wine’, with 13% alcohol and wine acidity. They have convened a panel of tasters to do sensory analysis, who rate each sample in triplicate under rigorously defined conditions. ‘We’ve been using very specific terms to describe the astringent sensations, such as degree of coarse texture, drying, adhesive, chalky as well as fullness/viscosity, acidity and bitterness,’ Francis explains. So far, they’ve reached several conclusions. First, grape seed tannins are more coarse and astringent that skin tannins of equivalent size. Part of the explanation for this is that seed tannins have added chemical structures known as galloyl esters. Second, in general, the larger the size of tannin, the more astringent. ‘For example, when we tested tannins isolated from grape skin, a dp3 (degree of polymerization 3 units) tannin was less astringent than a dp8, which was in turn less astringent than a dp12, and a synthesized dp5 tannin was rated intermediate in astringency between the dp3 and dp12’, reports Francis. Thirdly, pigments don’t seem to have any mouthfeel properties—even complex pigment structures isolated from wine. Fourthly, they’ve studied the interaction between tannins and other wine components, finding that polysaccharides isolated from wine diminish the astringency level. Francis adds that these studies are using grape tannins tested in a model system—‘We’d love to find a way of selectively removing or adding different wine tannins in a red wine, but at present this isn’t possible.’  

Tannins During Wine Ageing

Emerging research is suggesting that the traditional account of red wine ageing—that over time tannins get bigger, become insoluble, and fall out of solution—may be wrong. This has been the traditional paradigm of red wine ageing. You are probably familiar with the explanation: a young red wine may be big and tannic, and after several years in the bottle the tannins will ‘soften’, by means of them getting bigger and falling out as a deposit. But this concept isn’t based on good scientific data, and what actually takes place in wine ageing is uncertain. ‘This is a huge question, and I doubt generalizations can be made’, says Paul Smith. ‘Basically, there are alterations and recombinations of all the components. The classic example is the breaking apart and recombination of tannins—perhaps this mellows them, perhaps they get bigger, perhaps they get smaller.’ So, while the traditional explanation of tannins getting bigger and falling out of solution may hold, it could well be that tannins are breaking up in the acidic environment of the wine and are getting smaller. It’s worth bearing in mind that some red wines age wonderfully with very little or no bottle deposit. Because wine ageing is such an important part of the appreciation of fine wine, it would be nice to know what is actually taking place.

Tannins and Red Wine Colour

Here’s another story that could do with some revision. Researchers are now beginning to understand the nature of colour in red wines, and the picture emerging is challenging traditional understanding in this area. Colour in red wines actually falls into three categories. First we have the anthocyanins, the primary pool of colour from the grape. Young wine is packed with anthocyanins, which are very reactive: they interact with both sulphur dioxide and oxygen, which bleaches them. Their colour is also influenced by the pH of the must. At lower (more acidic) pH they are redder; at higher (less acidic) pH they are bluer. It turns out that anthocyanins are unstable, and aren’t that important for the long-term colour of red wines. In addition to anthocyanins there are two major fermentation-derived colour groups. The first of these is the pigmented polymers. These are formed by the chemical linkage between tannins and anthocyanins. This is a covalent (strong) linkage and is very important in forming stable colour in wines. The evidence suggests that most of the pigmented polymer formation occurs during fermentation: ‘This is the window to capitalize on pigmented polymer formation, we believe’, says Paul Smith. The third group is called the anthocyanin-derived pigments, which arise from reactions between anthocyanins and other phenolics and aldehydes. This is a massive, complicated class of non-bleachable pigments, and is an area of intense current research, with new members are being added all the time. The anthocyanin-derived pigments are still quite reactive and they can go and form further combinations with tannins to form pigmented polymers. There’s also current interest in the phenomenon known as copigmentation. This is the stable combination of anthocyanins with phenolic ‘copigments’—colourless molecules which combine with the anthocyanins to increase colour intensity. It’s a head-hurtingly complicated phenomenon, still not fully understood, but it is the basis for cofermenting small proportions of white grapes, such as Viognier, which are rich in copigments, with red grapes, most particularly Shiraz. This is becoming increasingly trendy in Australia, for example, although some winemakers miss the point by combining portions of white and red wine after fermentation where the window for copigmentation may have passed.  

Tannin Management: The Influence of Winemaking and Viticulture

It follows from all this that one of the keys to successful red winemaking is effective tannin management. This first occurs in the vineyard. Grapes, seeds and stems (if the stems are going to be used in macerations) can all contribute significant levels of polyphenols to the wine. Viticultural decisions can influence the extent and nature of the polyphenols that find their way into the must, although this is far from an exact science. While grapes were traditionally harvested on the basis of sugar levels, increasingly they are harvested with a view to achieving physiological or ‘phenolic’ maturity. Indeed, good viticulture can be summed up as encouraging a convergence of phenolic and sugar ripeness, with both at optimal levels at the same time. Shading of grapes is known to reduce the net quantity of skin tannins and also their nature. Unripe red grapes make nasty wines, not just because of high, herbaceous-tasting methoxypyrazine levels but also because of unripe or ‘green’ tannins. Seeds contribute a substantial amount of tannin to red wines and, if these are unripe and green, they can negatively affect wine quality. For this reason, one of the goals of current tannin research is to identify suitable markers of ‘phenolic’ maturity, which would give an indication of the best time to harvest. Another research objective is to identify specific grape tannins that can be used as markers of quality in viticulture.

Once the grapes have reached the winery, the way the polyphenolic substances (principally the tannins and anthocyanins) are extracted has a huge impact on the quality and character of the final wine. Winemakers have plenty of decisions to make about how to macerate red grapes so as to achieve the right level of polyphenol extraction. Some of the significant parameters that can be manipulated are the temperature of fermentation, pumping over or punching down the cap, the choice of fermentation vessel (small volume open-top fermenters, versus large tanks, versus rotary fermenters), the use of prefermentation cold maceration, and malolactic in barrel—and this list is far from complete. There are also new methods of extraction that are only just emerging, such as the flash d’etant system (that involves heating) and cross current extractors, but it’s too soon to say what sort of effect these will have and whether they will have wide take-up. The idea behind these techniques is that current extraction methods only pull out a proportion of the total phenolics present in grape skins, and it may be possible to enhance wine quality by removing more without also extracting unwanted polyphenolics from the seeds.

A subject of great current interest is microoxygenation, which has had a remarkably high take-up worldwide over the last decade considering that there are so far few experimental data backing up the claims of manufacturers of microxygenation devices and service providers who offer this technique to winemakers. It’s likely that oxygen applied at the right time and in the right quantities can have a beneficial effect on the mouthfeel and structure of red wines, but as yet there’s no clear evidence as to the sorts of tannin modifications that are taking place. It seems that microoxygenation is an important tool in tannin management, but winemakers currently have to ‘fly blind’, relying on guesswork and frequent tasting to judge when enough is enough for the particular wine they are working with.  

Concluding Remarks

I hope this will serve as a rapid introduction into the complicated world of tannins, which are so vital for red wine character and quality. The fact that relatively little is known about them reflects the difficulty of studying this important but complex group of chemicals. It’s encouraging to see that this is a field of active current research that promises to yield some valuable data that will help guide viticulturalists and winemakers to make better informed decisions, resulting in more complex and interesting red wines at each price point.


This article is modified from one that originally appeared in Harpers Wine and Spirit Weekly

Back to top

Click Here For Return To Link Page.