Chocolate with a Hint of Yuck
Could a new chemical test that quantifies off flavors transform the chocolate industry?
For all the delectable flavors added to all the confections in the world, the chocolate industry goes to equally great lengths to keep unwanted tastes and aromas out.
Dried and fermented cacao beans, the $103 billion industry’s most basic raw ingredient, can be spoiled with mold, fouled by bacteria, or contaminated with the chemical byproducts of too much or too little fermentation. Sometimes randomly cutting open and reaching into a 100 pound bag of cacao will yield a handful of gross, moldy beans.
Other times the spoilage is less obvious to the eye, but bad batches are noteworthy for their grotesque and almost scary off flavors, including urine, smoke, mushrooms, sweat, or dung. And when those beans nevertheless make it into chocolate bars, their rank foulness can draw consumer complaints, damage brands, and even trigger recalls.
Companies seek to avoid those costly risks by employing panels of expert tasters to screen the beans they buy for quality. But some experts say it’s more of an art than a science.
Now a team of researchers in the Leibniz-Institute for Food Systems Biology at the Technical University of Munich in Germany has developed a new way to quantitatively screen samples as a way of complementing taste panels.
Starting with a single sample of untainted cacao and nine foul-flavored samples obtained from German chocolate manufacturers, the scientists applied the analytical techniques of gas chromatography and mass spectrometry to detect trace amounts of chemicals that were generating moldy-musty notes in the beans.
From the data, they identified two specific metabolites associated with those off flavors, including a chemical compound called geosmin that appeared to be the prime moldy-taste culprit.
“Our approach really was the first one to definitely show that it’s the key compound,” says Martin Steinhaus, head of the sensory systems chemistry unit at TU Munich, who led the research.
Identifying the chemical culprit behind a widespread flavor problem for chocolate manufacturers suggests the technology could be a new commercial screening tool for the industry.
“If a low-cost system for measuring these molecules is developed, this could be used for screening cocoa bean shipments to identify lots with levels above a specified threshold which would be more precise than current methods,” says Pennsylvania State University plant biology expert Mark Guiltinan, who was not involved in the research.
The work could also help identify ways of improving early steps in cacao processing, Guiltinan says. “This new knowledge can be used as a foundation for studies to develop fermentation, drying, and storage methods to eliminate the formation of these off flavors.”
The long road from bean to bar
The complete list of ingredients on the back of a chocolate bar wrapper is often deceptively short—not much more than cocoa, sugar, cocoa butter (or sometimes substituted vegetable oil), vanilla, and usually an emulsifier like lecithin. Some specialty chocolate makers even push this to the extreme, using only two ingredients in their dark chocolate bars (cocoa and sugar).
But that seeming simplicity belies what a complex concoction chocolate really is. The way it shines, the way it snaps, its nuanced flavors, the exact melting point that turns it liquid in your mouth—these are all achievements of a carefully controlled manufacturing. Chocolate may be an emulsion of just a few ingredients, but it’s the product of a dizzying array of industrial processes to roast, crush, winnow, agitate, conch, mix, temper, mold, cool, and otherwise torture cacao beans into those sweet squares that tickle the tongue.
Unless something goes wrong.
A thousand things can ruin a batch of good chocolate, from complicated chemical processes that take place during the processing of the beans to just plain bad luck, like having too many rainy days in a row when a farmer is trying to dry their crop.
But if chocolate companies covet uniformity in their final products, they can’t really expect to find it in their supply chains, because they buy cacao by the bag, crate, and cargo container—almost a trillion pounds a year—largely from middlemen bean brokers far removed from the farmers at the point of origin. And ultimately, as an agricultural product, even cacao from the same farm will differ from year to year or season to season.
“Cacao, I think it’s fair to say, is at least centuries behind the cultivation of other orchard crops.”
So sensory panels, which are composed of people with some chocolate tasting expertise, play a key quality control role by checking for off flavors. The problem is that sometimes the people on those panels disagree, and this goes far beyond the subjectivity of human taste. Physiology is also an issue, and people can suffer from a reduced ability to smell or taste particular compounds.
“I’ve been consistently disappointed at the lack of consensus among different sensory panels,” says Daniel O’Doherty, an independent expert and consultant within the cacao industry.
O’Doherty says he has submitted the exact same samples from a wide range of manufacturers, ranging from Guittard and Valrhona to smaller specialty producers such as Dandelion, only to find some panels detect off flavors in the beans while others find them perfect. It would be great to have an objective test for particular compounds corresponding to off-flavor perception by the larger population, he says.
An oddball fruit on smallholder farms
Among the world’s fruit orchards, the cocoa tree Theobroma cacao is an oddball. It only thrives in a narrow range of tropical places straddling the equator, and it’s still grown in more or less the same way it was in centuries past—almost never on a massive, industrial scale but rather in a patchwork of tiny, individual “smallholder” plots of a couple hectares with a few thousand trees.
“You don’t really see that with anything else,” says O’Doherty. “Cacao, I think it’s fair to say, is at least centuries behind the cultivation of other orchard crops.”
Besides being grown on smaller spreads, cacao is also unlike oranges, apples, pears, mangoes, or any other orchard tree you can think of in that it’s still generally grown from seed rather than fields of genetically identical trees made from cut branches grafted onto other trees to yield a uniform, consistent fruit. As a result, cacao pods are strange and beautiful in diverse green, yellow, purple, red, and orange colors that conceal a sweet-sour gloppy pulp known as “baba” that you can buy on the side of the road in parts of Ecuador or Brazil and that people in the industry say tastes something like a cross between a soursop and a mangosteen—a description that makes it seem even more exotic.
But the cacao fruit itself is just a footnote. What really matters are the seeds. Harvesting them is a laborious process that involves cutting the fruit by hand with a machete, scooping the seeds into wet piles, and tending to them as they spontaneously ferment, a process that kills the seed but develops its flavor.
Fermented beans are then spread out to air dry. Cacao beans contain about 50 percent fat, an evolutionary adaptation that helps them grow in thick shade by giving them the energy-dense burst they need to germinate and grow in the low light that exists below tropical canopies. But the fat also readily absorbs flavors, which is why you never want to store burlap bags of cacao next to burlap sacks of coffee.
One of the major flavor problems with cacao comes from the fact that farmers sometimes try to shortcut the drying process by using wood fires to dry the beans. But smoke and other off flavors may also be picked up before drying, developing biochemically within the beans during fermentation, especially if the process goes on too long, is not managed well, or allows the beans to become contaminated with microbes.
In an earlier work published last year, Steinhaus and his colleague Daniela Füllemann used their new technique to identify different phenol compounds that gave smoky off flavors to cacao. Then they homogenized those compounds in cocoa butter and employed taste panels of 15 people or more to determine acceptable mean thresholds for levels of the off-flavor molecules.
Steinhaus says their approach could increase the chocolate quality on the market if implemented. “I think it reduces the risk of recalls—definitely,” he says.