Skunkiness, Coffee Chemistry, and Naturalism in Flavor

"Like flowers, but with garbage!" is how Roslyn, Jennifer Lawrence's character in American Hustle, describes her favorite Swiss topcoat. "It’s like perfumey but there’s also something rotten and I know that sounds crazy, but I can’t get enough of it. Smell it, it’s true. Historically, the best perfumes in the world, they’re all laced with something nasty."

Don't stop sniffing your nails, Roslyn, because you're onto something. The notion that the pleasant has to be laced with the foul to achieve its full effect has a long history in perfumery -- the term of art here is pudeur. Mary Gaitskill, in her 2006 novel Veronica, writing about the Paris runways in the early 1980s, describes the effect this way:

"Thumping music took you into the lower body, where the valves and pistons were working. You caught a dark whiff of shit, the sweetness of cherries, and the laughter of girls. Like lightning, the contrast cut down the center of the earth: We all eat and shit, screw and die. But here is Beauty in a white dress."

There's a satisfying, counterintuitive logic to this, even as the sentiment has become kind of a platitude: Your flaws make you beautiful, baby.

But this idea -- the putrid grace note -- seems a bit less appealing when it comes to flavor. Could there be something rotten or excremental undergirding the savoriness of our savories? Does vanilla flavor really come from the anal glands of a beaver? This might seem like one of the points where the flavor and fragrance industries diverge, where the logics of "good taste" differ depending on whether you're considering the aromatic and the edible. The history of the flavor chemistry of coffee, however, offers a more nuanced spin.   

Imagine for a moment the gorgeous, plush aroma of coffee. Wafting from the percolator, it eases you into the morning, cushioning the cruel shock of awakening, bringing the clan together around the breakfast table. Morning! Comfort! Optimism!

Now imagine a skunk trotting into the breakfast room, tail aloft, trailing the fumes of his distinctive parfum.

Is there any similarity between these two smells, the fair and the foul? A skunkiness in the Stumptown Hairbender? An element of Caffe Verona in yonder fair skunk?

Okay, by way of an answer, here's my story: in 1949, Cargille Scientific, a chemical and instrument supply company in New York, began selling something they called "Coffee-Captan."

"A smell is being made commercially available for the first time," toodled the Associated Press in 1949. "It is described as an essential constituent of the aroma of roasted coffee that provides a new scent for perfume and flavors." Food Industries also ran an item announcing that quantities of the synthetically produced furfuryl mercaptan were available for the first time manufacturing and for research. "In addition to its many uses in the food field for enriching flavors and aromas, it should also be useful as an intermediate in organic synthesis." Maison DeNavarre, in the June 1949 iteration of his monthly "Desiderata" column in the American Perfumer & Essential Oil Review, squealed: "The recent announcement of the availability of alpha furfuryl mercaptan, one of the essential constituents of the aroma of roasted coffee, has probably been read by everyone." He thought the powerful chemical could possibly help make the scent of formulas for "cold wave" permanents less offensive. Meanwhile, Chemical and Engineering News (March 28, 1949) noted its potential as a polymerization agent,and an accelerant in rubber vulcanization.

But what is furfuryl mercaptan? Also known as 2-furanmethanethiol, it is a sulfur-containing compound, not present in the green coffee bean, but created during roasting via the Maillard reaction. At very low concentrations (like, one part per million), it has a pleasantly familiar coffee aroma. At higher concentrations, it provides a... different sort of experience. Cargille's "Coffee-Captan," Kiplinger's noted in 1954, "is powerful stuff, having to be kept under double seal because in concentrated form it gives the impression that there has been an explosion involving a skunk about the size of an A-bomb." One flavor chemist remembers an entire facility being evacuated after an someone accidentally broke empty bottle had once contained the chemical.

How did this foul chemical become a commercial product?

Chemists had been trying to determine the constituents of the aroma of roasted coffee since the beginning of the nineteenth century. (There's a good technical account of this history in the textbook, Coffee Flavor Chemistry, written by two Firmenich chemists, Ivon Flament and Yvonne Bessiere-Thomas). Analyzing organic compounds was a painstaking and difficult process, demanding maximum skill and care. Chemists wondered: were the chemical changes that took place in green coffee beans specific to coffee, or were they common to other roasted things? Furthermore, was there a simple chemical "principle" that accounted for the smell of a substance -- a singular "essence" -- or instead, did a set of chemicals, interacting together in complex ways, produced what we recognize as an aroma?  

A minor tangent (file it under "Coffee, usefulness thereof"): In an 1832 article in the Leipzinger Zeitung entitled "Coffee Arabicae: Its Destructive Effect on Animal Emanations as a Protective Agent Against Contagion," the German chemist Christian Conrad Weiss described the power of roasted coffee aroma to neutralize stinks of all kinds: rotten eggs, putrid meats, animal musks, asafoetida. In an era before germ theory, when foul odors were thought to contribute to the spread of disease, Weiss believed that concentrated coffee extract or a pinch of finely ground coffee, burned in a lamp, could disinfect and purify a room for days. Coffee extract might also serve as a more pleasing alternative to the typical contents of the vinaigrette, the fashionable lady's dainty respite from intrusive odors. Weiss, however, did not make much progress in actually identifying the chemical components of roasted coffee aroma. At the beginning of the twentieth century, chemists had succeeded in provisionally identifying only ten volatile compounds in coffee.

The major leap in the understanding of the chemistry of roasted coffee aroma would have to wait until after the First World War. Starting in 1920, in a meticulous research project spanning more than a decade, two chemists working in Switzerland, Tadeus Reichstein and Hermann Staudinger -- both would later, separately, win the Nobel Prize -- definitively identified nearly thirty components in coffee that contributed to its aroma. One of these was furfuryl mercaptan, a previously unknown molecule. 

The Chemical Heritage Foundation, where I'm a fellow this year, has a 1985 oral history with Reichstein in its fantastic Beckman Center collection. In addition to kind of hilariously undermining his incendiary former PhD advisor Staudinger ("I didn't like his methods because... it's a kind of brutal chemistry. He liked everything which made noise and caused explosions. These were the things he liked." Whenever Staudinger worked in the laboratory, "afterwards everything was full of broken glass..."), Reichstein also pontificates about the role that small quantities of foul-smelling compounds play in flavor.

He tells the interviewer: "The sense for flavor is very delicate. If you have such a mixture and you take only one of the things out, the rest will go flat. For instance, what I realized at this time was that a very good smell in some flowers, jasmine or roses or violets -- the really good smell is only produced by some compounds present in very small quantities which smell awfully bad -- terrible -- if they are alone or concentrated. But without them, the good smell is not natural. It is like a cheap coiffure shop."

Producing a smell that was both "good" and "natural" was an important end goal of their research. Reichstein and Staudinger received funding from Kathreiner's Malzkaffee, a company that produced a sort of ersatz coffee from malted barley. After the miserable shortages of coffee (and other foods) in Europe during the First World War, Reichstein says: "they were interested because they thought they could add a little flavor to make their malt coffee smell like real coffee. They were very pleasant people. I worked through many tons of coffee to get only a few cubic centimeters of the flavor." Reichstein and Staudinger took out several patents in the 1920s in the UK and the US for their research, including for a "new or improved method of producing artificial coffee aroma."

After the coffee flavor project, Reichstein would go on to an illustrious career, doing important work on the synthesis of Vitamin C, and eventually being awarded the Nobel Prize in 1950 for his work on the chemistry of cortisone and other adrenal hormones. Staudinger would nab his own prize three years later, in honor of his visionary work on macromolecules and polymers.

But the significance of their work on the flavor chemistry of coffee does not seem to have been widely recognized before the late 1940s. Indeed, once Reichstein and Staudinger caught wind of Cargille's "Coffee-Captan," they cried foul about the company's claim to offer this synthetic chemical for sale "for the first time." They called attention to their work and their earlier patents, claiming priority for their discoveries. Indeed, Flament and Bessiere-Thomas note that furfuryl mercaptan was already one of the components of a flavor additive, "Cofarom," manufactured by the German flavor and fragrance firm Haarmann & Reimer. (Reichstein and Staudinger's research was not completely unknown, as it was respectfully cited in a pair of articles on coffee flavor by pioneering flavor chemist Morris B. Jacobs, which ran in the March and April 1949 American Perfumer & Essential Oil Review.)

Why did it take so long for this work to catch on? Part of it may be that flavor companies prior to the mid-1930s were not in the habit of using basic research into the flavor chemistry of foods to fuel product development. (There are some exceptions to this.)  Furthermore, much of their research and development focused on isolating and synthesizing organic compounds of Carbon, Hydrogen, and Oxygen -- aldehydes, ketones, ethyls, alcohols -- or, more rarely, Nitrogen-containing compounds such as methyl anthranilate (you know this one as the smell of a grape Jolly Rancher, or a Concord grape). Stinky sulfur-containing chemicals seem largely to have been shunned. Indeed, Alois von Isakovics, the founder of Synfleur, one of the earliest synthetic fragrance and flavor manufacturers in the U.S. called sulfur-containing compounds the "enemy of the perfume or flavor chemist." In a 1908 lecture to students at Columbia University, he advised "eliminating from perfume substances even the smallest traces of constituents that contain sulfur."

These early products may have been "good," but they did not necessarily also produce an impression that could be called "natural." However, by the late 1930s, flavor manufacturers were more and more interested in reproducing the effects of nature, creating "blended" flavors that had depth, delicacy, and complexity. And, as Bernard Smith, of the flavor company Virginia Dare put it in a speech to the landmark "Flavors in Foods" American Chemical Society Symposium in 1937: “It is a well-recognized principle that in minute traces compounds of even objectionable flavor or odor may greatly assist in producing a finished product of superior excellence." With an increasing number of volatile chemicals produced by organic chemical research, flavorists and flavor manufacturers had a growing field of materials with which to tailor specific, "naturalistic," effects.

Compounds like furfuryl mercaptan illustrated the complex way that flavor chemicals operated in foods and on the senses. Chemicals that at full strength were unambiguously foul, could also be the key to producing effects that were not just pleasant, but convincingly, compellingly "natural" -- whether or not they were actually materially identical to the "real thing."