I did a column a couple months back ("Winemakers Play With Fire--and Win," September 2008) about intentionally edgy winemaking--slowing fermentations down almost to the sticking point, exposing wine to overdoses of oxygen, and so on. Maybe it's because I'm neck deep in my own garage fermentations as I write this, but I'm still preoccupied with that narrow band of winemaking styles between the classically clean and the completely out-of-control.

One of the things keeping me fixated happened up in Portland, near the end of an all-day sensory session at the ASEV meetings in June. Signe Zoller, the former winemaker at Meridian and other places who now custom-crafts designer wines at Zoller Wine Styling in Paso Robles, asked those assembled about the mysterious properties of dimethyl sulfide, a compound she thought was behind the puzzling moment when she and a co-winemaker differed on whether a particular fermentation smelled like gorgeous fruit or sewer backup. Her question never was answered, but I figured that if a past president of the ASEV is confused about this stuff, I had a column idea.

After talking and e-mailing with Bruce Zoecklein at Virginia Polytechnic Institute, Sue Ebeler at the University of California, Davis, and Ken Fugelsang at California State University, Fresno, I've gotten this far. It turns out that dimethyl sulfide (DMS) is just one in a long list of sulfur-containing compounds with split personalities, capable of making your wine more fun or making it downright undrinkable. Nearly all of these are bound to be present, in some concentration or other, in most normal fermentations. So aside from the details, the philosophical question is: How possible is it to manage these critters and encourage the "good" sulfur without egging on the "bad?"

Ambivalent valences

When it comes to sulfur and wine, two compounds get 99% of the attention: SO₂--sulfur dioxide, the protagonist behind the "contains sulfites" warning on most every label and a critical wine preservative, but a definite negative in concentrations strong enough to have a burning, stinging sensory impact; and H₂S--hydrogen sulfide, the flatulence produced by yeast in a stuck, nutrient-deficient fermentation. It's a useful addition to natural gas in home appliances, but the last thing you want in a bottle of Merlot. These two bad actors are enough to give sulfur a terrible reputation and place it high on the wine aroma avoidance list, right up there with TCA.

Those two spotlight-hoggers, however, are just the start of the sulfur-related story, which goes on to include bunches of other compounds. Some of the bit players are never much fun, but quite a number can be useful character actors in the overall vinous production. Dimethyl sulfide, the substance that got me on this track, can add notes of quince or truffle, and it seems to be a prime ingredient in the "cooked corn" aromas fans associate with well-aged wines; 3-mercaptohexylacetate (3-MHA) can put passion fruit in your Sauvignon Blanc; dimethyl disulfide (DMDS) can contribute asparagus (and some of us like that); carbon disulfide can smell downright sweet.

Makes you salivate just hearing those descriptors, right? Except that these nifty effects only happen at thresholds in the single-digit parts per billion; once you get into higher concentrations, they start turning into burnt rubber, rotten garlic, yesterday's cabbage, and, well, sulfide-y. As elsewhere in wine chemistry, too much of a good thing is not a good thing at all. The teensy thresholds get more problematic when you factor in the different sensitivities of different people--perhaps like Signe Zoller and her co-sniffer.

It's important to note that these chameleon compounds are not byproducts of sulfur dioxide additions, nor are they for the most part related to the presence or absence of hydrogen sulfide. All of them--and more--are just among the things that happen in a fermentation, and sometimes afterwards. The quantities they show up in have to do with a multiplicity of factors: vineyard conditions, grape composition, fermentation kinetics, yeast strain, other treatments going on at the same time.

One final complication: These ingredients can be present either as volatile sulfur compounds (VSCs) or in non-volatile forms, and they can move back and forth between those states. The context here is what's known as the redox potential--the balance of tendencies toward reduction and oxidation. Molecules with a high redox potential (like oxygen) are always in search of electrons to hook up with; molecules with a negative redox potential (like SO₂) are looking to shed electrons. This dance of the valences goes on and on, and in the process, aromas get released or masked and even eliminated.

Sulfur Odor Management 101

The methods for combating the major groups of sulfur off-odors are well known as part of the basic winemaking principles drummed into even home winemakers: stop using elemental sulfur in the vineyard well before harvest; make sure your yeast has enough nutrients to get the job done, thus preventing excess formation of H₂S, mercaptans, and the rest of that unpleasant family. If your fermentation stick s anyway, clean the wine up with some combination of aeration, copper sulfate and ascorbic acid; and calibrate SO₂ additions carefully according to the wine's pH to avoid smelling like a burnt match.

In reality, of course, these standard methods can backfire, or carry a cost. In the early phase of fermentation, when the yeast population is still rapidly expanding, nitrogen deficiency can be a problem, and nutrient additions can help prevent H₂S. Near the end, however, excessive nitrogen--for example, from an unnecessary addition--can have the opposite result, encouraging H₂S. Using copper sulfate may succeed in removing the unwanted hydrogen sulfide, but it can take out that appealing passion fruit 3-MHA at the same time.

Knowledge of how to promote the good VSCs is less advanced than knowledge of how to prevent or eradicate the bad ones. Most attention has focused on yeast strains and strategies. A number of commercial yeasts are advertised as encouraging (among other things) this or that sulfur-related aromatic compound; natural/feral yeast fermentations, with their slow pace and mix of unorthodox microbial agents, also tend to produce interesting sulfur-related characteristics. The successful ones do, that is; natural fermentations that run amok rarely get bottled and tasted.

Parts per gazillion

The problem with trying to micro-manage the positive VSCs is that they only work their magic in truly tiny concentrations, and our methods for controlling the formation of these compounds are blunt instruments indeed. If we made wine by taking a 5,000-gallon tank of a model solution and then tweaking it with eyedroppers full of dimethyl sulfide, or terpenes, or whatever, maybe we could plot all this out--but that's not how wine is made. Real-world methods are just not geared to producing something only in the range of 4-7 parts per billion, no more, no less.

There's a pretty straight line from yeast choices to some of these compounds, but even those effects occur in a much broader, more complicated context. Vintage variation in the levels of VSCs suggests something happens in the vineyard, but we don't know exactly what. These compounds compete for a taster's attention with hundreds of other volatiles, a matrix big enough to make your head swim.

Sue Ebeler says, "I don't think we know enough to control a fermentation to get a target. There are so many variables that could potentially influence levels, and we don't know how all these variables interact with each other. We can't yet predict aroma based on compositional information alone, based simply on concentrations of individual compounds."

Ken Fugelsang acknowledges that some of these compounds can be beneficial some of the time, but he leans toward a conservative, stay-out-of-trouble regimen. "There's no question that it's hard to engineer these compounds into your wine, even though they can contribute to complexity. I usually approach the question as, 'How do we avoid the problems?' With nutrition, yeast, fermentation management protocols, ways to minimize it--I think the whole raft are agents to avoid."

Bruce Zoecklein, Fugelsang's former colleague at Fresno, is probably the most adventuresome of the group. "We know how processing can control sulfur-like off-odors," he says, "but less about processing and the positive ones. Fine wine crafting is about managing subtle nuances. The capstone message is that the problem is complex, and people who want to handle it need a HACCP-style plan (Hazard Analysis and Critical Control Point), methodology common in the food and beverage industries. You want to optimally manage and steer away from the bad, see what that does to the good."

My conclusion, once again, is that the more science we know, the more winemaking remains an art.

Tim Patterson 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. Contact him through edit@winesandvines.com.