Inquiring Winemaker
The Sticky Side of Tannin Management
Top researchers from around the world gathered at an all-day Tannin Symposium held in conjunction with the annual meeting of the American Society for Enology and Viticulture in June. But the biggest tannin buzz came from another presentation given the day before, one that was generating reference after reference in subsequent panels. I’m not sure enologists make much use of Twitter to air what’s on their minds, but if they did, this paper would surely have been trending, with the hashtag #hybridtannin.
The unlikely object of attention was a talk by Lindsay Springer of Cornell University based on work she and her advisor, Gavin Sacks, had done on tannin extraction from French-American hybrid grape varieties. Hybrids might seem an odd focus for examining tannin extraction, since they are notorious for producing virtually tannin-free wines, but that was exactly the point. The refusal of hybrids to yield their tannins to any and all exertions by winemakers turned out to be a great way to shine a spotlight on the dicey, unpredictable movement of tannins from grapes to wine.
Hybrids may be an extreme case, but even for mainstream vinifera, the relationship between tannin levels in the fruit and in the resulting wine is anything but straightforward. Sugar turns into alcohol in a predictable fashion, and acid comes across as acid, but tannin levels in grapes are not a reliable harbinger of tannin levels after fermentation. Some lots of high-tannin Cabernet come out wimpy; some lots of tannin-shy Pinot Noir come out hard as nails.
Extraction is the first link in the chain of tannin management in the cellar, and it’s full of mysteries. The most irritating part is there’s not a lot winemakers can do about it.
Certainties and uncertainties
Researchers know that tannins are famously sticky, always looking for something to glom on to. Depending on the situation that might be protein, polysaccharides, each other, glass, whatever. It’s this same tendency to be sticky that hitches wine tannins to salivary proteins and produces astringency in the mouth.
Not surprisingly, large-chain, higher molecular weight tannins seem to be the stickiest—or at least the stuckest. It’s like Velcro; the more hooks and loops on a molecule, the more opportunities to fasten. Thus, proportionately fewer of the longer tannins make the extraction cut. Shorter seed tannins also are underrepresented, unless they come out during extended maceration. Thus the tannin length profile in wine is different from that in grapes, in addition to the differences in sheer amount.
Tannin binding mostly comes as weak bonds, either hydrogen attachments or hydrophobic hookups, aggregations of molecules seeking to avoid contact with water. These relatively weak bonds are, at least in principle, able to be broken up to some degree by one or another winemaking element—by heat, by ethanol, or various production techniques. Unless, of course, the grapes happen to be hybrids.
Hybrids have the same kinds of tannins as vinifera grapes, and indeed the degree of vinifera parentage correlates well with the concentration of tannins. But they are reluctant to come out, or at least to stay out, under any circumstances, even though aggressive fermentation practices leave the skins utterly disintegrated, far more so than vinifera skins. This different behavior again suggests that we need to understand something about the cell walls, not just about the tannins.
Skins and cell walls
Skins, seeds and grape pulp all contain some tannin—and, of course, some cell walls. But skins are the main focal point of extraction studies; they only comprise 10% of the weight of grape berries, but they contain two-thirds of the cell wall material (excluding seeds), the likely sticking points. Tannins have to come into solution from the “inside” of the grape skins; the outer cuticle is too hard to penetrate.
The most detailed examination of tannin/cell wall dynamics comes from the work of Keren Bindon at the Australian Wine Research Institute. Using mainly model wines and model fermentations, Bindon has produced a long list of intriguing results, and a concern for saving trees prohibits listing them all in this magazine. But I’ve included a few of the more tantalizing findings.
In immature grapes, tannins tend to stick to pectins, which are plentiful at that stage. As grapes ripen, the skins get more porous. You might think that would aid extraction by providing more escape hatches, but in fact it seems to provide more places for tannins to get trapped between layers and layers of stuff.
Whatever the methods of winemaking employed, any given batch of grapes seems to have a plateau maximum level of tannin extraction. Some of the tannins pried loose from their original homes end up stuck back on something else and falling out of solution by the end of fermentation. Tannins may stick onto other pieces of skin, or stray polysaccharides, or yeast/lees, in an endless round of extraction, adsorption and desorption. Extraction thus is not an either/or process—you either extract the tannins or you don’t—but a complex chemical dance.
Bindon’s observation of a plateau effect on tannin extraction from vinifera grapes resonates with a pattern found in hybrids. Extension specialist and researcher Anna Katherine Mansfield and her cohorts at Cornell University studied several processing methods for fermentations of three hybrid reds—cold soak, enzyme additions, tannin additions and hot press (thermovinification)—and found some differences in extraction, particularly for the hot-press method. But alas, after the completion of fermentation, the level of differences in the finished wines was negligible—implying that something about the hybrid grape composition or cell wall structure was putting a very low lid on stable extraction.
Bindon’s trials in Australia also suggest that the presence of anthocyanins somehow helps enable extraction. Anthocyanins, too, are a non-linear bunch when it comes to extraction, having their own ways of sticking to fermentation flotsam and falling out of solution. But in some yet-to-be-explained fashion, anthocyanin levels may matter as much or more than extraction techniques.
Jim Harbertson at Washington State University suggested to me that there may be different kinds of bonds at work in different cases. The “weak” hydrogen and hydrophobic bonds are the most common and most studied, but maybe there are circumstances in which tannins get strongly bonded—covalent bonds—to whatever works for them, rendering them nearly impervious to conventional winemaking methods. Jim Kennedy at California State University, Fresno, (the two Jims co-chaired the ASEV Symposium) says the evidence for this high-test bonding so far is mostly anecdotal, but quite intriguing.
There may be varietal differences in extractability, even among vinifera cultivars, let alone between vinifera and hybrids, but this has not gotten much attention yet. (My amateur speculation: Maybe grapes like Tannat and Sagrantino, sources of infamously tannic wines, are just more willing to let their tannins go?) It is also entirely possible that site and viticultural factors exert an influence that persists all the way through winemaking. Doug Adams at the University of California, Davis, for example, mentioned a field trial done a few years back on own-rooted Pommard clone Pinot Noir in Oregon involving two vineyards near each other, one on a hillside, one on flatlands. Tannin levels in the fruit were comparable; the hillside grapes had three times as much tannin in the finished wines.
Go figure.
Measuring and managing
Right now, there is no handy-dandy instant tannin extractability assay, no nifty attachment to your refractometer that indicates you will get 43.8% of the tannins out of these grapes. What’s more, there is no particular demand for such a test. “I’ve never heard a winemaker say, ‘If I could only get my cell wall interaction right…’” Kennedy observes.
But some kind of rapid test might help growers and winemakers avoid unpleasant surprises of over- or under-extraction that complicate the pursuit of a particular wine style. Bindon insists that the level of tannins in the grapes is the best general predictor of tannin levels in wine, even if it’s not perfect. She thinks relatively primitive tests can easily be done to get a grip on extractability: macerating grapes in an acidified solution of 15% ethanol, or hitting grapes with a 70% acetone solution.
Instead, what vinifera winemakers generally do is get used to their fruit sources and learn to live with them. Careful winemakers track their various lots year after year, and if a particular block is always tannin-deficient, or a particular vineyard is always tannin-heavy, those lots are treated accordingly and set on track for a particular wine tier or blend. This kind of informed experience doesn’t explain all of the tannin extraction anomalies, but it handles them in a practical fashion over time.
Good red hybrid winemakers face much tougher challenges (see “How two winemakers maximize tannins in hybrids” on page 25) and certainly earn their pay in meeting them. But even in their wineries, the mysteries of extraction and non-extraction probably aren’t the most pressing problem at harvest time. Advancing our understanding of tannin dynamics could lead to more informed decisions both in the vineyard and the winery, and someday that will happen. In the meantime, there are plenty of cool riddles wrapped in these mysteries, all part of the great tannin enigma.
Tim Patterson is the author of “Home Winemaking for Dummies.” He writes about wine and makes his own in Berkeley, Calif. Years of experience as a journalist, combined with a contrarian streak, make him interested in getting to the bottom of wine stories, casting a critical eye on conventional wisdom in the process.
The unlikely object of attention was a talk by Lindsay Springer of Cornell University based on work she and her advisor, Gavin Sacks, had done on tannin extraction from French-American hybrid grape varieties. Hybrids might seem an odd focus for examining tannin extraction, since they are notorious for producing virtually tannin-free wines, but that was exactly the point. The refusal of hybrids to yield their tannins to any and all exertions by winemakers turned out to be a great way to shine a spotlight on the dicey, unpredictable movement of tannins from grapes to wine.
Hybrids may be an extreme case, but even for mainstream vinifera, the relationship between tannin levels in the fruit and in the resulting wine is anything but straightforward. Sugar turns into alcohol in a predictable fashion, and acid comes across as acid, but tannin levels in grapes are not a reliable harbinger of tannin levels after fermentation. Some lots of high-tannin Cabernet come out wimpy; some lots of tannin-shy Pinot Noir come out hard as nails.
Extraction is the first link in the chain of tannin management in the cellar, and it’s full of mysteries. The most irritating part is there’s not a lot winemakers can do about it.
Certainties and uncertainties
Some pieces of this puzzle are pretty clear. Some varieties tend to have higher concentrations of tannins than others—Cab more than Pinot, for example. For any variety, the concentration in the berries is higher than in the wine they make; even badly over-extracted wines don’t get everything the grape has to offer. Even though a number of winemaking techniques—macerating enzymes, more punch downs, thermovinification, extended maceration—can enhance extraction, some tannins always remain sequestered. (Much like parts of the Federal budget.) Or at least that’s the pattern with vinifera grapes; hybrids seem more resistant to any winemaking blandishments.
Researchers know that tannins are famously sticky, always looking for something to glom on to. Depending on the situation that might be protein, polysaccharides, each other, glass, whatever. It’s this same tendency to be sticky that hitches wine tannins to salivary proteins and produces astringency in the mouth.
Not surprisingly, large-chain, higher molecular weight tannins seem to be the stickiest—or at least the stuckest. It’s like Velcro; the more hooks and loops on a molecule, the more opportunities to fasten. Thus, proportionately fewer of the longer tannins make the extraction cut. Shorter seed tannins also are underrepresented, unless they come out during extended maceration. Thus the tannin length profile in wine is different from that in grapes, in addition to the differences in sheer amount.
Tannin binding mostly comes as weak bonds, either hydrogen attachments or hydrophobic hookups, aggregations of molecules seeking to avoid contact with water. These relatively weak bonds are, at least in principle, able to be broken up to some degree by one or another winemaking element—by heat, by ethanol, or various production techniques. Unless, of course, the grapes happen to be hybrids.
Hybrids have the same kinds of tannins as vinifera grapes, and indeed the degree of vinifera parentage correlates well with the concentration of tannins. But they are reluctant to come out, or at least to stay out, under any circumstances, even though aggressive fermentation practices leave the skins utterly disintegrated, far more so than vinifera skins. This different behavior again suggests that we need to understand something about the cell walls, not just about the tannins.
Skins and cell walls
Skins, seeds and grape pulp all contain some tannin—and, of course, some cell walls. But skins are the main focal point of extraction studies; they only comprise 10% of the weight of grape berries, but they contain two-thirds of the cell wall material (excluding seeds), the likely sticking points. Tannins have to come into solution from the “inside” of the grape skins; the outer cuticle is too hard to penetrate.
The most detailed examination of tannin/cell wall dynamics comes from the work of Keren Bindon at the Australian Wine Research Institute. Using mainly model wines and model fermentations, Bindon has produced a long list of intriguing results, and a concern for saving trees prohibits listing them all in this magazine. But I’ve included a few of the more tantalizing findings.
In immature grapes, tannins tend to stick to pectins, which are plentiful at that stage. As grapes ripen, the skins get more porous. You might think that would aid extraction by providing more escape hatches, but in fact it seems to provide more places for tannins to get trapped between layers and layers of stuff.
Whatever the methods of winemaking employed, any given batch of grapes seems to have a plateau maximum level of tannin extraction. Some of the tannins pried loose from their original homes end up stuck back on something else and falling out of solution by the end of fermentation. Tannins may stick onto other pieces of skin, or stray polysaccharides, or yeast/lees, in an endless round of extraction, adsorption and desorption. Extraction thus is not an either/or process—you either extract the tannins or you don’t—but a complex chemical dance.
Bindon’s observation of a plateau effect on tannin extraction from vinifera grapes resonates with a pattern found in hybrids. Extension specialist and researcher Anna Katherine Mansfield and her cohorts at Cornell University studied several processing methods for fermentations of three hybrid reds—cold soak, enzyme additions, tannin additions and hot press (thermovinification)—and found some differences in extraction, particularly for the hot-press method. But alas, after the completion of fermentation, the level of differences in the finished wines was negligible—implying that something about the hybrid grape composition or cell wall structure was putting a very low lid on stable extraction.
Bindon’s trials in Australia also suggest that the presence of anthocyanins somehow helps enable extraction. Anthocyanins, too, are a non-linear bunch when it comes to extraction, having their own ways of sticking to fermentation flotsam and falling out of solution. But in some yet-to-be-explained fashion, anthocyanin levels may matter as much or more than extraction techniques.
Jim Harbertson at Washington State University suggested to me that there may be different kinds of bonds at work in different cases. The “weak” hydrogen and hydrophobic bonds are the most common and most studied, but maybe there are circumstances in which tannins get strongly bonded—covalent bonds—to whatever works for them, rendering them nearly impervious to conventional winemaking methods. Jim Kennedy at California State University, Fresno, (the two Jims co-chaired the ASEV Symposium) says the evidence for this high-test bonding so far is mostly anecdotal, but quite intriguing.
There may be varietal differences in extractability, even among vinifera cultivars, let alone between vinifera and hybrids, but this has not gotten much attention yet. (My amateur speculation: Maybe grapes like Tannat and Sagrantino, sources of infamously tannic wines, are just more willing to let their tannins go?) It is also entirely possible that site and viticultural factors exert an influence that persists all the way through winemaking. Doug Adams at the University of California, Davis, for example, mentioned a field trial done a few years back on own-rooted Pommard clone Pinot Noir in Oregon involving two vineyards near each other, one on a hillside, one on flatlands. Tannin levels in the fruit were comparable; the hillside grapes had three times as much tannin in the finished wines.
Go figure.
Measuring and managing
Right now, there is no handy-dandy instant tannin extractability assay, no nifty attachment to your refractometer that indicates you will get 43.8% of the tannins out of these grapes. What’s more, there is no particular demand for such a test. “I’ve never heard a winemaker say, ‘If I could only get my cell wall interaction right…’” Kennedy observes.
But some kind of rapid test might help growers and winemakers avoid unpleasant surprises of over- or under-extraction that complicate the pursuit of a particular wine style. Bindon insists that the level of tannins in the grapes is the best general predictor of tannin levels in wine, even if it’s not perfect. She thinks relatively primitive tests can easily be done to get a grip on extractability: macerating grapes in an acidified solution of 15% ethanol, or hitting grapes with a 70% acetone solution.
Instead, what vinifera winemakers generally do is get used to their fruit sources and learn to live with them. Careful winemakers track their various lots year after year, and if a particular block is always tannin-deficient, or a particular vineyard is always tannin-heavy, those lots are treated accordingly and set on track for a particular wine tier or blend. This kind of informed experience doesn’t explain all of the tannin extraction anomalies, but it handles them in a practical fashion over time.
Good red hybrid winemakers face much tougher challenges (see “How two winemakers maximize tannins in hybrids” on page 25) and certainly earn their pay in meeting them. But even in their wineries, the mysteries of extraction and non-extraction probably aren’t the most pressing problem at harvest time. Advancing our understanding of tannin dynamics could lead to more informed decisions both in the vineyard and the winery, and someday that will happen. In the meantime, there are plenty of cool riddles wrapped in these mysteries, all part of the great tannin enigma.
Tim Patterson is the author of “Home Winemaking for Dummies.” He writes about wine and makes his own in Berkeley, Calif. Years of experience as a journalist, combined with a contrarian streak, make him interested in getting to the bottom of wine stories, casting a critical eye on conventional wisdom in the process.
SHARE »
