Why wine smells the way it does

A grandiose topic, which will not be exhaustively explored but rather briefly overviewed

Welcome to the first of many wine-science-lite blog posts, inspired by the topic of our club's weekly activities.

We so often tend to talk about wine aroma in terms of varietals—“this is a quintessential example of a Cabernet Sauvignon”-- but in fact what starts out in the grape often doesn’t end up in the bottle, nor does everything in the bottle come from the grape. That’s a theme that will come up frequently in my future wine-related outbursts, and it seems that the topic of aroma, which was the focus of the first of the SWS intro tastings, is as good a place to start as any.

“Aroma” refers to something that you pick up using olfactory perception, which begins at your nose. This is distinct from the flavors your tongue picks up, which are the simple salty, sour, sweet, bitter, and umami. One scientifically difficult thing about our odor and flavor receptors is that every person has a different suite of receptors—so there are chemical compounds that your neighbor will smell that you may not be able to, or that you might perceive differently. On the other hand, one scientifically wonderful characteristic of aromas is that each can be pinned down to an individual molecule (odorant) that, when it hits our olfactory receptors, creates a signal that we recognize as a smell. Which is not to say that we have a grasp on all of the odorants found in wine, but we do understand a few of them in depth.

Experts divide wine aromas into three major categories, defined by the processes by which they enter the wine: primary, secondary, and tertiary aromas. Here are just a few examples.

 

1. Primary Aromas – varietal aroma

Primary aromas are those that are characteristic of the grape.

Aromas native to the grape

As I understand it, if you’re a veteran wine drinker but with little experience making wines, if someone were to hand you a glass of fresh grape juice made from wine grapes you’d be hard-pressed to identify what grape it comes from. Much of wine flavor comes during fermentation (see point #2). However, there are some characteristic odorants that are created in the grape and survive the winemaking process, emerge in the glass to remind us of the botanical origins of the thing we are drinking.

One of those is rotundone, an oxygenated sesquiterpene, which is the molecule that makes Shiraz taste of black pepper. It was a molecule that evaded scientists for years and made the news (well, the news of the American Chemical Society) when it was discovered in 2008 by a team of Australian researchers. It was a feat of chemical detection, as the quantities of rotundone required to make a red wine smell peppery (the human sensory threshold) is a super-low 16 ng/L, a level at which the instruments commonly used (GC-MS) have a hard time finding it. To everyone’s pleasant surprise, rotundone is actually found in real black pepper as well!

Winemakers who want to create a peppery Shiraz certainly have an interest in knowing what controls rotundone levels in their wine [at least, researchers want to believe they do—whether Shiraz winemakers really follow rotundone research is another question]. In the grape, rotundone is present only in the skin; its biological function is unknown, as are the precise environmental influences that cause some grapes to have higher levels than others. The rotundone is leached from the skins into the juice during the first five days of fermentation if the juice is kept in contact with the skin.

In reference to my statement above about the unidentifiable nature of fresh grape juices—one exception that I’ve encountered in my own experience is Gewurtztraminer juice—which was at one point commercially available from Navarro vineyards (see references). If you drink a Gewurtztraminer wine and then drink the fresh juice, you’ll understand the meaning of primary aroma.

The molecules that make Gewurtztraminer wine taste like the grape (and that both have in common with lychee, I’ll wager) are monoterpene alcohols, such as geraniol, nerol, and linalool. These are all found in the skin of the Gewurtzraminer grape and also in Muscat and other super-aromatic grapes, as well as in many flowers and and aromatic plants.

 

Aromas arising from volatile precursors

My understanding is that the vast majority of known aroma molecules found in wine—including those that we identify with certain grape varietals—cannot be found in the original grapes themselves but in fact arise from the processing, during fermentation, of other precursor molecules that were always there but just not as aromatic. So the case of volatile precursors falls somewhere confusingly between “primary” and “secondary” aromas, in that these molecules are very specific to, and indicative of, the grapes they come from, but their presence in the final wine is dictated by the circumstances of fermentation. (Though I guess, actually, everything in the final wine is somehow dictated by the circumstances of fermentation.) More to the point, if you smell a wine and the aromas help you recognize what grape it is, you won’t necessarily know whether what you’re smelling is a molecule that was there all along, or one that got produced during fermentation.

The monoterpenols described above are, in fact, a good example. Tropical-floral blockbusters like Gewurtz and Muscat are not the only wines that can have these molecules-- all wine grapes contain precursor molecules to the monoterpenols. Those precursors are mostly glycoconjugates—that is, monoterpenols with sugar groups stuck onto them, which for some reason are odorless until the sugar is removed via the fermentation process. They “make up a reserve of grape flavor, which is generally more abundant than the free one” (Baumes 2009) – that is, Gewurtz and Muscat can become more enchantingly floral, and even Chardonnay has some locked-up potential to do so, but only upon subjection to the activity of microorganisms.

 

2. Secondary aromas – fermentation aroma

Secondary aromas are those arising from fermentation and from aging in oak barrels.

Aromas produced by yeast

There are some aroma compounds that microbes can produce regardless of what grape juice they’re working with. Take 4-vinylphenol, which smells barnyard-y. Other words used to describe the smell include medicinal, mousy, and band-aid-like (Wikipedia). Its precursor, p-coumaric acid, is found in all grapes, and the spoilage yeast Brettanomyces can use the enzyme cinnamate decarboxylase to carry out the transformation into 4-vinylphenol. In fact, adding p-coumaric acid to a microbial culture can be a good method for detecting-- by smell—whether Brettanomyces is there.

Some beer brewers like the funky, alternative flavors produced by this alternative yeast and may even purchase it for intentional addition to their brew [ref x]; most winemakers do everything they can to keep it away from their fermentations. Note that I say most. Everything’s a matter of taste.

 

Aromas imparted by oak barrel aging

As you probably know, a great many wines spend some time aging in oak barrels before they are bottled. The effects of oak aging are complex enough to fill books. Some of the most important effects to keep in mind include:

  • oak barrels are slightly porous, and therefore afford the wine some oxygen exposure, which brings with it complex flavor-changing chemical processes
  • oak wood can adsorb (attract and hold securely) flavor compounds from the wine, lessening their effects
  • barrels, especially aged ones, can harbor fungus and bacteria that can alter wine flavor through their activities (such as Brettanomyces)
  • most obviously, oak wood can contribute odorants and odorant precursors to the wine.

Consider how much energy and attention we pay to all the aspects of wine grape cultivation and juice pressing and their effects on the flavor of the resulting grape juice; the same considerations may be paid to oak tree growth and barrel-making, as everything in a barrel’s history has an impact on the wine that ages in it. Consider all the variables :

  • the species of oak (American and European oak varieties contribute different proportions of flavoring molecules—and the difference in the final wine can be perceptible)
  • where the oak was grown (also affects flavoring molecules)
  • the age and growth rate of the tree (these can affect the density of the wood)
  • the degree to which the wood has been seasoned and toasted before being made into a barrel (toasting produces a whole different suite of odorants)
  • the age of the barrel (older barrels are less porus, but more likely to harbor microbes)

That said, there are some flavor characteristics universal to oaked wines. A few examples of key odorant molecules include cis- and trans-β-methyl-γ-octalactone. Also known as “oak lactones,” these are the most prominent component of the odorants extracted from the barrel into wine, and contribute a typical “oak barrel aged” flavor often associated with the descriptors “sweet,” “woody,” “toasted,” “vanilla,” or “coconut.”

Speaking of vanilla flavors, one of the next-most-common oak odorants is vanillin, so-called because it literally tastes of vanilla—in fact, if you purchase artificial vanilla flavoring, this is likely the important molecule in the bottle. In oak barrel aging, it’s a product of the degradation of lignin (the structural polymer that makes wood woody); artificial vanilla flavor is also derived chemically from wood components.

 

3. Tertiary aromas – aging aroma, or “bouquet”

As you know, wine doesn’t stop changing once it’s put in the bottle. The yeast may be mostly done with their biological activity, but chemical changes continue to occur... including the production of dimethylsulfide (DMS), a compound that has long been associated with “bottled bouquet.”

DMS is a remarkable molecule. If you ask a biological oceanographer, you’ll be told that it’s produced by phytoplankton in the open ocean and may have something to do with influencing local weather. But an oenologist will talk about its smell: DMS is one of the primary odorants of black truffles (and, interestingly, canned corn); when humans find it in wine (where our threshold for perception is a mere 25-27 g/L) we also associate it with the aromas of red or black currants, black olive, and in general premium aged red and late-harvest white wines. So we generally feel positively toward it, except when there’s too much of it.

DMS is sometimes produced during fermentation, but it’s very volatile and disappears before the wine is bottled. Once in the bottle, however, it’s trapped. It arises from the transformation of precursors such as dimethylsulfoxide (DMSO), sulfur-containing amino acids, and S-methylmethionine (SMM), all of which are present in the grape in varying quantities that seem to depend not only on varietal but also on terroir and grape ripeness. As with all flavors, it seems.

 

References and further reading

All the molecule images were taken from Wikipedia.

Baumes, Raymond. 2009. “Wine Aroma Precursors.” In Wine Chemistry and Biochemistry, edited by M. Victoria Moreno-Arribas and M. Carmen Polo, 251–74. Springer New York. http://link.springer.com/chapter/10.1007/978-0-387-74118-5_14.

on rotundone:

Herderich, M. J., T. E. Siebert, M. Parker, D. L. Capone, D. W. Jeffery, P. Osidacz, and I. L. Francis. 2012. “Spice Up Your Life: Analysis of Key Aroma Compounds in Shiraz.” In Flavor Chemistry of Wine and Other Alcoholic Beverages, edited by Michael C. Qian and Thomas H. Shellhammer, 1104:3–13. Washington, DC: American Chemical Society. http://pubs.acs.org/doi/abs/10.1021/bk-2012-1104.ch001.

Wood, Claudia, Tracey E. Siebert, Mango Parker, Dimitra L. Capone, Gordon M. Elsey, Alan P. Pollnitz, Marcus Eggers, et al. 2008. “From Wine to Pepper: Rotundone, an Obscure Sesquiterpene, Is a Potent Spicy Aroma Compound.” Journal of Agricultural and Food Chemistry 56 (10): 3738–44. http://www.rsc.org/chemistryworld/News/2008/May/13050801.asp

on monoterpenes:

“Monoterpene.” 2013. Wikipedia, the Free Encyclopedia. http://en.wikipedia.org/w/index.php?title=Monoterpene&oldid=572892188.

“Navarro Vineyards - Gewürztraminer Grape Juice.” 2014. Accessed January 21. http://www.navarrowine.com/shop/productdetail.php?prodid=488.

on 4-vinylphenol:

“Wyeast Laboratories : Commercial : Breweries : Technical Information : Wild Beer Brewing.” 2014. Accessed January 21. http://www.wyeastlab.com/com-lambic-brewing.cfm.

on oak:

cis-3-Methyl-4-Octanolide.” 2014. Wikipedia, the Free Encyclopedia. http://en.wikipedia.org/w/index.php?title=Cis-3-Methyl-4-octanolide&oldid=521910125.

Pérez-Coello, M. Soledad, and M. Consuelo Díaz-Maroto. 2009. “Volatile Compounds and Wine Aging.” In Wine Chemistry and Biochemistry, edited by M. Victoria Moreno-Arribas and M. Carmen Polo, 295–311. Springer New York. http://link.springer.com/chapter/10.1007/978-0-387-74118-5_16.

“Vanillin.” 2014. Wikipedia, the Free Encyclopedia. http://en.wikipedia.org/w/index.php?title=Vanillin&oldid=589603672.

on DMS:

“Dimethyl Sulfide.” 2014. Wikipedia, the Free Encyclopedia. http://en.wikipedia.org/w/index.php?title=Dimethyl_sulfide&oldid=591014054.

Maurizio Ugliano, Paul A. Henschke, and Elizabeth J. Waters. 2012. “Fermentation and Post-Fermentation Factors Affecting Odor-Active Sulfur Compounds during Wine Bottle Storage.” In Flavor Chemistry of Wine and Other Alcoholic Beverages, 1104:189–200. ACS Symposium Series 1104. American Chemical Society. http://dx.doi.org/10.1021/bk-2012-1104.ch012.

Vallina, Sergio M., and Rafel Simó. 2007. “Strong Relationship Between DMS and the Solar Radiation Dose over the Global Surface Ocean.” Science 315 (5811): 506–8. doi:10.1126/science.1133680.