Ars sine scientia nihil est. Art without science is nothing, the old Latin motto goes. Flavor creation requires both. Veteran flavorist and consultant Frank Fischetti recently highlighted these dualities during the 9th Annual West Coast Flavor Industry Forum in Anaheim, California.
Flavor Components and Challenges
Flavors, Fischetti noted, serve three purposes: the simulation of named flavors, character fixation/maintenance, and the enhancement of flavor impressions and acceptability. And all flavor compounds, whatever the application, comprise just two parts—the flavor portion and the diluent. “The solvent system should be the most important part,” Fischetti said, “because it makes your flavor usable. It’s not the flavor materials that make the flavor work, it’s the solvent system.” The longstanding problem for flavorists, he added, concerns the descriptors used for the flavor materials in the flavor portion. “We have no vocabulary,” he said. “We can’t talk to each other. What does ‘green’ mean? cis-3-Hexenol? Galbanum? There are different kinds of green; which green is it?”
During his talk, Fischetti outlined five key concepts that serve to better elucidate flavor creation: characteristic threshold value, flavor impact item, flavor contributory item, flavor differential item and flavor impact group.
Characterization of Materials
“Why do we characterize materials?” Fischetti asked. “For greater retention. Greater retention means greater creativity.” Here, he said, it is crucial for flavorists to sustain a ppm-oriented frame of mind. “How many of you in your laboratories talk about ppm?” he asked the audience. “Most flavor chemists talk in percentages. At use level, no material is used at percentage quantity. We don’t think in these terms, and we should.” In addition, he stressed the importance of tasting materials multiple times at various levels and in various media. “You can’t taste a chemical one time in one meeting,” he said. “You’ll never define that chemical. You have to taste it in various media and then take the collective characterization. Benzaldehyde has a character of cherry and … it’s a character impact item of almond. How do you distinguish between the two? ”
Fischetti then presented the concept of a characteristic threshold value procedure for categorizing materials. This concept considers flavor materials at “realistic” levels that take into account flavor materials appearing in finished products at ppm levels, not percentage quantities. Materials can be tasted at different levels in various media to determine both character and the “contribution it makes under what condition of use.” In some cases, a flavorist will realize that certain materials fall below threshold upon application. This process of evaluation also allows the flavorist to separate materials into the categories mentioned earlier: flavor impact item, flavor contributory item and flavor differential item.
Character Impact Item
The first two ingredient categories can be considered the science end of the flavorist’s job. Fischetti explained that these characteristic ingredients are essential and necessary to the named flavor. By Fischetti’s own definition, a character impact item “is a material when smelled and/or tasted is reminiscent of the named flavor. Flavorists derive from these materials a great deal of the target flavor’s organoleptic effect. In other words, they are “characteristic, essential and necessary for the flavor.”
Among the character impact materials cited by Fischetti were: 3-methoxy-2-isobutyl pyrazine for bell or pea, ethyl-2-methylbutyrate for apple, 4-methyl nonanoic acid and 4 methyl octanoic acid for lamb, 5-methyl thiophenecarboxaldehyde for almond and isoamyl acetate for banana.
Contributory items, meanwhile, help create, enhance or potentiate flavors—though on their own they do not necessarily taste or smell like the named flavor target. While these materials may not embody the primary organoleptic character, they can bring the overall flavor closer to it. “A contributory item is necessary to produce a desired effect,” Fischetti explained, “but not essential for the flavor.” He continued, “They are not characteristic, but they are necessary, because it acts with the impact items to produce a definite character.”
Among the contributory items Fischetti cited were: ethyl butyrate and 2,5-dimethyl-4-hydroxy-3(2H)- furanone in grape, ethyl safronate for blueberry, acetophenone for black cherry, and furanthiol and γ-decalatone for chicken/chicken broth.
Finally, differentiating ingredients present the flavorist with the greatest opportunity for creativity. “This is the most imaginative part of the flavor,” Fischetti said. Differentiating materials are not characteristic, essential or necessary to achieve a named flavor. Adding these materials is not about building a better or more characteristic flavor, said Fischetti, but rather about creating a different flavor, building signature. It is the use of these materials, he said, that truly marks a flavor chemist.
Among the differential items he highlighted were: the Liederkranz cheeselike methylthio butyrate in strawberry for “seediness,” Exaltolide* for bringing muskiness to blackberry, and 2,4-octadienal and 4-phenyl-4-pentenal in walnut.
Character Impact Groups
“I think you should learn one flavor at a time rather than one chemical,” Fischetti said. “Why? Because you can extrapolate an apple flavor to a strawberry quite easily. If you’re familiar with the apple, you have a lot of components that could quite easily go into strawberry.” For this reason, he said, Fischetti prefers to present materials in character impact groups. “In every character impact group you have the start of any flavor. Ask any flavor chemist, ‘What’s the hardest thing to do?’ It’s hard to get started making the first compounds. This is a better way to do it than teaching one compound at a time.”
Fischetti’s impact group examples included blue cheese (methyl ketones), fish/seafood (amines such as trimethyl amine), butter/caramel (diones), citrus (aliphatic, alicyclic aldehydes), honey (phenylacetic acid esters) and coconut (δ-lactones).
Fischetti presented a diagramming system for organizing flavor materials with the following data fields: name of the flavor; application off the flavor; diluents/solvent for the flavor; character impact items/group; contributory flavor items; and differential flavor items. “The age of 25 items in a formula is ridiculous,” he said. “Most of them, at use level, drop below their threshold. It costs companies money.” Using these systems, Fischetti stressed, a flavorist can and must be able to articulate why any particular ingredient appears in a formula. “If they can’t,” he said, “then take it out!"
*Exaltolide is a trademark of Firmenich.