Authors: Molly Birnbaum
The industry changed in the middle of the next century with new technology: the gas chromatograph, a machine that separates a compound into its individual flavor components, using glass or metal tubes, releasing them one by one in vapor form. Flavorists, then, were able to smell each part of a whole, one volatile component at a time. This aided in identifying specific components of otherwise impermeable flavors that once were only replicable by trial and error. After all, an important part of a flavorist’s work is to duplicate the flavor of items found in nature or, today, to duplicate the flavors found in products already on the market. Before the gas chromatograph, chemists had to rely on guesswork, matching and mismatching chemicals, one at a time, never entirely sure, to reach their goal. This could be tough: a real strawberry, one that grows delicately on the vine, for example, contains more than three hundred different volatile molecules. It’s impossible for the human nose to pick up on every single one. In fact, researchers have found it to be almost impossible for individuals (both trained and not) to pick out and identify more than four individual scents within a larger whole.
In the 1970s, the gas chromatograph was coupled with another machine: the mass spectrometer. This changed the process of creating flavor yet again, as the mass spectrometer could give a specific chemical reading of each point of the flavor, printed off in graphs from a computer interface. Scientists no longer had to rely solely on their noses but could have a list of the chemical components of each flavor. It wasn’t perfect—“the mass spectrometer isn’t as accurate as a human nose,” Grosinger told me. But it was a start. Later, this was joined by headspace technology—portable devices capable of measuring the volatile molecules in the space around an item, able to catch olfactory moments in time, like the scent of my apple pie the moment it exits the oven.
At Citromax, I followed Grosinger as she walked inside the lobby and up the stairs. We entered the flavor lab, a large rectangle of a room on the second floor. When I stepped through the sealed glass door, I was immediately hit by smell: sweet and cloying, like caramel candy. Grosinger turned back toward me and smiled.
Inside, I saw two long counters jutting into the center of the space, each stacked with shelves holding dozens of small brown bottles. The counter space held glass beakers and tiny pipettes, metal scales and plastic jars. I could see a sink, a stove, and a refrigerator standing against the wall. A few women wearing white lab coats worked around the room.
“Those are the chemicals we use to make our flavors,” Grosinger explained, pointing to the bottles on the shelves. Those containers were filled with the raw materials of the flavor world—some of the more than three thousand components used in the industry. There are liquids and powders, solvents and sweeteners, and the caramel coloring needed to add to a cola. These chemicals are guarded from the light with their opaque bottles, their tightly sealed lids. It seemed strange that flavor—a term with such broad scope, with such powerful individualities—could originate in those containers, which looked so uniform, so tame.
Around the room I could see brand-name products, too. Cardboard boxes full of them sat on the shelves. Mainly beverages, familiar ones like sodas, teas, juices, and waters—ones that came in bottles and in cans, ones that I had sipped at my desk, at the kitchen counter, over dinner at home countless times. Grosinger and her small team had been hired to invent the flavors for many of these drinks. They had been hired to duplicate the flavors of many more, ones that were already on the market, produced by other brands. Citromax only creates sweet flavors, mainly because their facilities in New Jersey are too small, Grosinger told me. “You need to have separation of your plants. You don’t want your lemon flavors to taste like garlic.”
But before we had even entered the building, I promised not to name any names. The flavor world operates under strict secrecy. The mystery is due, in part, to the fact that most flavors cannot be patented. Even within the individual flavor companies, formulas are kept hidden from the majority of employees. But I wondered if the secrecy was also in place in order to keep reality one more step away from public consumption. Large food and beverage companies, producing products with household names, could not want the consumer to know that their sports drinks, their granola bars, their yogurts contain flavors made from chemicals, from machines, from faraway labs. They want to preserve the mystique. “Consumers don’t get that there are ghostwriters out there,” Grosinger said. “They wouldn’t have a clue that there are people like me sitting in a lab.”
Grosinger began to introduce me to her aromatic raw materials, one at a time. She opened bottle after bottle, waving the caps under my nose so that I could smell. Aroma, she explained, is tantamount to the creation of flavors. She assembles most of the ingredients that she will use for each flavor, which can be anywhere from a few to a hundred, with her previous knowledge and her sense of smell. Only later, when all of the elements are combined, will she taste. “Sometimes you pick up more with just your nose.”
But taste and smell work in different ways, combining in a unique manner, one that Grosinger needs to address with every single flavor she produces. To create a successful flavor, she needs to think about the way the product smells before it’s placed in the mouth. She needs to know how it tastes on the tongue, how that couples with the smell in the back of the throat. After all, taste and smell are important together and separate, providing important feedback both individually and in combination. Different areas of the brain are activated for the perception of taste and for the perception of smell, but when perceived
together
—the scent of chocolate combined with the taste of sweet on the tongue, for example—an additional, new area lights up as well. “These findings have caused some to believe that flavor is in itself a separate sense, the sixth sense, one all its own,” Johan Lundstrom, a scientist at Monell Chemical Senses Center once told me.
Likewise, it’s been found that there is a difference in the way the human brain processes scent
orthonasally,
or through the nose, and
retronasally,
which is through the mouth. Inhaling through the nose over a fragrant cup of coffee or exhaling through the mouth while chewing a bite of pizza both would send the odor molecules to the same receptors at the top of the nose, but each pathway actually causes the brain to light up in a different place, implying that there is a difference between smelling the external world and smelling the food that one ingests. Inhaling the scent of a chicken roasting in the oven for dinner is not the same as experiencing its flavor in the mouth.
I smelled Grosinger’s ingredients one by one. She began with benzaldehyde, a component of bitter almond oil, an ingredient widely used. I sniffed. It smelled sweet and familiar, its name just out of reach.
“What does it remind you of?” Grosinger asked.
I looked at her blankly. Just like during the scratch-and-sniff test in Philadelphia, I couldn’t think of a single word. I began to feel the familiar rise of panic. I tried to keep my breath slow. Normal.
“Cherry?” she suggested.
I inhaled again.
“Yes!”
A memory crystallized. It
did
smell of cherries, of ripe summer cherries, spitting the pit out into the palm of my hand. Of cherry sno-balls in New Orleans, of the cherry cola from the corner store I once loved to drink flat and warm.
“What else?” she asked.
Again, I froze.
“Uh . . .” I could smell
something,
that familiar something, on the edge of my consciousness. My frustration began to grow.
“Like some type of candy?” she nudged.
Only one word came to mind, like a broken record: cherry, cherry, cherry, cherry.
“Uh . . .”
“Marzipan?” she said.
Of course. There it was. It smelled of the thick almond paste, marzipan, which I had used again and again that winter to make dense, nutty cakes that came out of the oven smelling of almonds and lemon. Benzaldehyde smelled like baking, like dessert. I smiled.
“As soon as you said it, I could smell it,” I said.
“Suggestion is a powerful thing.”
Next came isoamyl acetate. I sniffed.
“Sweet,” I told Grosinger, and she shook her head.
“There’s more,” she said. For just a moment, a half-second blip, I considered throwing the bottle of chemical across the room, over the shelves, on the floor. That would give her more. But instead I sniffed, drawing blank.
“How can you so easily tell?” I asked.
“Practice. Years of practice.”
Isoamyl acetate, she told me, smelled like bananas. I sniffed again over the cap. Yes, there it was. Again, it became recognizable with her words. Bananas. Of course.
“Also, there’s a bit of cotton candy,” she said.
Cotton candy? I sniffed again, picking out the tropical mash, the powdery spins of sugar. Oh, yes. Wispy threads of cotton candy melting in my mouth at a town fair one summer near my aunt’s summer home in Pennsylvania.
The next bottle she cracked open was ethyl acetate, which to me smelled a bit harsh—like the edge of rubbing alcohol, like nail polish remover. “We use this to add a juicy note to flavor,” Grosinger explained. Although unpleasant to smell on its own, just a tiny bit of it could be added to a formula to make a fruit flavor taste just a bit riper, she said. “With some chemicals, it’s not about how it smells or tastes.” Then she handed me a container filled with white powder. “Maltol,” she said. “This can also make something taste more ripe.” It smelled like candy. Like inhaling over a cookie jar.
She decided to demonstrate and took two small plastic cups out of a drawer and placed them on the counter. She pulled a larger bottle of liquid off a shelf, one from the other side of the room. “This is a strawberry flavor,” she said, one already consisting of around twenty-five raw materials, mixed together to create a flavor that imitates the fruit. She poured a half inch of the flavor, which was water-colored clear, into each cup. We each took a sip. It tasted of strawberry, the strong, slightly strained strawberry found in many types of yogurt, like the kind I often ate for breakfast with a handful of granola. To the cups, she then added a bit of maltol powder, stirred it, and handed it back. We sipped. It tasted sweeter, richer, like the strawberry had spent a few more days getting juicier, riper on the vine.
Then Grosinger took out a bottle of cis-3-hexenol and waved it under my nose. I inhaled. This one was immediately familiar, and certainly not sweet. Definitely not pleasant. I wrinkled my nose. I sniffed again. And then I began to laugh. This was the smell of fresh-cut grass I had breathed in while walking into the building from outside. Not exactly the same: this one had a harsh edge, a chemical twang. But it was recognizable nonetheless.
Cis-3-hexenol, Grosinger explained, is classified in the green family of smells. She showed me a list of categories—descriptive words used to define all of the hundreds of chemical aromas used for flavor. These, she told me, make it easier to recognize and relate. There were more than fifteen categories on her list—among them fresh, juicy, acidic, jammy, sweet, creamy, nutty, citrus, smoky, floral, caramelized. In this list, colors also make an appearance. Brown is a category for the overripe, for the very sweet. It is cooked, like the sugared-fruit filling to my apple pie. Green, however, is fresh and grassy. It will make an apple flavor more Granny Smith than Red Delicious. It pulls the mouth back in time, transforming that strawberry from succulent to underripe.
Grosinger placed the bottles back on the shelf with a clink and walked me to the far side of the lab to the tasting booths. These booths had doors opening into a conference room on the other side of the wall but had small wooden covers that slid open toward the lab. Grosinger showed me how she could manipulate the lights within—reds and purples, bright and dim.
“We control the lights because the senses work together,” she said. The booths are soundproof and smellproof. The entire building, in fact, is controlled with negative pressure, an HVAC system that keeps each room pristine with its own airflow, its own contained scent. Subjects enter the booths to blindly taste new flavors, and Grosinger and her team do not want them to be influenced.
Flavor is indelibly tied to the other senses. It’s subject to suggestion. Contextual clues are key, especially the visual. In 2001, Frederic Brochet conducted a number of experiments for his doctoral dissertation (“Tasting: A Chemical Object Representation in the Field of Consciousness”), one with scandalous results. He had a group of wine experts taste a white wine and give a description, eliciting such words as “fresh, dry, honeyed, lively.” When he later had the experts taste the same wine, this time dyed red with food coloring, he got a very different response. “Intense,” they responded, “spicy, supple, deep.” Tasting is a form of representation, Brochet wrote. “Indeed, when our brain performs the task of ‘recognizing’ or ‘comprehending,’ it is manipulating representations. In reality, the taste of wine is a perceptual representation, because it manifests an interaction between consciousness and reality.”
Grosinger herself told me that when her kids were growing up, she helped them conduct a science project: give a group of people a plastic cup of unlabeled, dark brown carbonated soda and ask them what it is. They will say Coke or Pepsi. Next, give them a cup of soda with the same exact flavor but this time without the caramel coloring, the distinctive soda-pop-brown. They will say that it is the more citrus tasting Sprite or Mountain Dew. “Without fail,” Grosinger said. The visual plays a huge role in analyzing flavor.
Even sound can influence: in 2010, a study was published in the
Journal of Neuroscience
exploring the link between noise and scent perception in rats. “Smound,” as Daniel Wesson and Donald Wilson, colleagues at the Nathan S. Kline Institute in New York, called it, does indeed seem to exist. They found a change in reaction to certain scents in the brains of rats if paired with an auditory tone, hinting that humans experience a melding of senses on levels we are not remotely aware of. “It’s hard to find part of the brain that is purely one sense and not another sense too,” Wilson told me. “We’re blurring the lines between purely olfactory and purely auditory. The brain is, unfortunately, a lot more complex than we once thought.”