How Sweet Coffee Tastes! Towards an Understanding of Coffee Sweetness | 25, Issue 22
Researchers Dr. NANCY CORDOBA and Dr. DEVIN PETERSON, joined by PETER GIULIANO, share early results of a multi-year Coffee Science Foundation research project, undertaken at the Flavor Research and Education Center at Ohio State University, to establish foundational knowledge about the phenomenon of sweetness in coffee.
Today, sweetness is considered an extremely important attribute of specialty coffee. In a 2021 study by the Specialty Coffee Association (SCA) about the perceived strengths and weaknesses of the 2004 SCA Cupping Protocol, a majority of the 1,500 survey respondents said they would “add a sweetness scale.” Furthermore, 95 percent of approximately 50 experts interviewed agreed that “sweetness as merely presence does not work, as it is too important an attribute to be assessed in this way,” and many of them advocated for the addition of an intensity scale. Many experts mentioned that processing variations in the last 20 years have made sweetness a more important attribute than ever before, and they did not understand why it was not rated quantitatively in the 2004 protocol. Some of them also mentioned that, though sweetness is clearly a key attribute, they lack understanding about what causes sweetness in coffee.
However, the importance of sweetness as a coffee attribute is not new. The Coffee Cupper’s Handbook, a seminal work on specialty coffee cupping, written by Ted Lingle and published in 1984,[1] mentions the words “sweet” or “sweetness” no less than 140 times, and the words “sugar” or “sugars” 80 times. Lingle defined sweetness as the taste “characterized by solutions of sugars,” and proposed it was caused by “caramelized sugars and amino acid complexes.” He said that “sweetness refers to a pleasing fullness of flavor as well as any obvious sweetness and its perception is the result of the presence of certain carbohydrates. The opposite of sweetness in this context is sour, astringency or ‘green’ flavors.”
Lingle suggested that sweetness is key to the coffee sensory experience due to its interactions with the other taste modalities and believed that sugars were responsible for coffee sweetness. Although scientific knowledge at the time would have sufficed to refute such hypotheses, there was little will to challenge Lingle’s theories. In fact, Lingle’s work remained the basis for all our cupping systems until the Coffee Value Assessment. Scientists published research that challenged Lingle’s ideas but failed to change the widely accepted narrative about what causes coffee sweetness. For example, in 1985 (just one year after Lingle’s handbook was published), Clarke and Macrae published a very popular book about coffee chemistry.[2] They found that medium-roasted coffee contains 0.9 percent of the sucrose content found in green coffee, while light-roasted coffee contains 2.9 percent. This means that there is a concentration in the final coffee brew of just 0.014 grams of sucrose, 0.014 grams of glucose, and 0.009 grams of fructose per 100 millilitres.[3] Experts have known since at least 1965 that these concentrations of sugars are far below the human sensory threshold for sweetness.
Thus, the biggest mystery to me is not why coffee may be sweet, but why it was always assumed that coffee sweetness was due to sugars, when we’ve long known that cannot be possible. For 40 years, we had not challenged that assumption, and had built our industry systems, such as the 2004 Cupping Protocol, on it. I am so grateful that the researchers at Ohio State University have accepted the challenge and started to uncover the mystery of what causes coffee sweetness.[4]
Dr. MARIO R. FERNÁNDEZ-ALDUENDA
Technical Officer
For at least 300 years, coffee drinkers have perceived coffee as being “sweet”: Bach’s eighteenth-century Coffee Cantata exclaims, “Ah! How sweet coffee tastes! Lovelier than a thousand kisses, sweeter far than muscatel wine!”
This perception is backed up by research. In one 2019 study, filter-coffee consumers perceived plain filter coffee as sweet, and, in fact, sweetness was the most powerful factor in driving liking.[5] Other consumer studies show that sweetness is an overall positive attribute of coffee.[6] Likewise, coffee cuppers and buyers see sweetness as a positive—the attribute drives both impressions of quality and prices of coffee.[7] Anecdotally, most coffee drinkers have an intuitive explanation for this: coffee is the seed of a fruit, they reason, and fruits have sugar; therefore, sweetness must be related to sugars in coffee.
It sounds logical, but this reasoning for coffee’s sweetness has come under fire. Scientists have known for some time that roasted coffee has almost no sugar in it. In a recent paper, Batali et al. found that although sugars exist in coffee, they are below human sensory thresholds. This means that tasters were not capable of perceiving the sugar in the coffees tested.[8] This creates a mystery—what could explain the phenomenon of sweetness in coffee if it’s unlikely to be sugar?
As researchers who study flavor perception in coffee, we embrace a mystery like this and treat it as a challenge. In this case, the Coffee Science Foundation reached out to the Flavor Research and Education Center at Ohio State University with a proposal: to work together to try to solve the mystery of coffee sweetness. In 2023, we launched a two-year project called "Sweetness in Coffee: Sensory Analysis and Identification of Key Compounds," aimed at answering the question of what drives tasters’ perception of sweetness.[9] This is fundamental work—we’re investigating the phenomenon of coffee sweetness and its origins within the chemistry of the coffee bean itself.
As scientists, we turned to the scientific method, in which we develop hypotheses that might explain a phenomenon. We then test our hypotheses using the best methods we can find. In this article we’ll walk through five hypotheses, how we tested them, and what we learned.
Hypothesis One: Coffee Sweetness Is Real and Perceptible
Even though our review of the existing evidence strongly suggested that coffee sweetness is a real phenomenon (something that can be perceived by people when they drink black coffee), we began our work by demonstrating the phenomenon for ourselves. To do this, our research team worked with a specialty coffee roasting company—Intelligentsia Coffee—to specially select a group of coffees to represent a range of flavors and, ideally, various levels of sweetness. The coffees and their codes are listed in the table on the right.
We then asked a group of 125 professional coffee tasters to taste the coffees and rank their sweetness in order of intensity, as well as rate the sweetness on a 15-point intensity scale (see figure 1). As our project aims to understand the sensory experience of coffee as sweet, we did not investigate the impact of specific variables like coffee roasting, cherry ripeness, processing methods, or grind size on perceived sweetness. Instead, our investigation focused on the phenomenon itself.
According to this test, there was a statistically significant difference in sweetness intensity among the coffees, with a total range of 6 points of difference on a 15-point intensity scale. Furthermore, several of the coffees were statistically significantly different (in sweetness intensity) from other coffees,[10] and two coffees in particular—coffees 5 and 6—were less sweet than all the others. This demonstrated that coffee sweetness is indeed real, and our tasters tended to agree on which coffees were sweet and which were not.
On to the next hypothesis!
Figure 1. Average sweetness intensity for each coffee sample by professional coffee tasters. The graph shows the mean sweetness intensity, rated by 125 professional tasters on a 15-point scale. There is a 6-point difference between the coffee with the lowest sweetness intensity score (Sample 6, score 3.6) and the highest (Sample 2, score 9.6). Samples 5 and 6 were perceived as significantly less sweet than the others.
Hypothesis Two: Sweetness in Coffee Is an Aromatic Phenomenon
Coffee is an intensely aromatic beverage, and it’s well established that aromas—whether experienced orthonasally (through the nose) or retronasally (rising up from the throat into the back of the nasal cavity)— have a profound effect on the perception of flavor, including sweetness.[11] To test this phenomenon in coffee, we asked the same tasters who took part in the first test to follow the same ranking and rating procedure to evaluate the sweetness of the coffees again—this time wearing a nose clip (see figure 2). If they lost their ability to discern sweetness in the coffees, we would know it was aroma that drove sweetness perception. If the rankings remained the same, we would know aroma wasn’t a factor.
Interestingly, overall perception of sweetness intensity went down. Also, there was a smaller difference between the sweetest and least sweet coffee—only 3.6 points instead of 6. However, the rankings were similar, and most crucially, the “least sweet” coffees were the same in both tests. This suggests that aroma compounds have an influence, but there also must be some nonvolatile compounds, tasted on the tongue, that lead to the perception of sweetness.
Figure 2. Average intensity of sweetness for each sample as rated by professional coffee tasters wearing a nose clip. The difference between the lowest ranking (Sample 6, score 4.4) and highest ranking (Sample 7, score 8.0) samples is reduced to 3.6. Samples 5 and 6 are still ranked the least sweet.
Hypothesis Three: Could It Be Sugars After All?
Now that we had established that nonvolatile compounds tasted on the tongue were involved in creating coffee’s sweetness, we had to reconsider the sugar hypothesis. Though previous research strongly suggested that sugars were not involved, we had to check for ourselves. We used liquid chromatography– mass spectrometry—a chemical analysis technique— to evaluate five sugars in coffee: fructose, mannose, sucrose, and the combination of glucose and galactose. Sucrose is thought to be the most abundant sugar in roasted coffee, and our tests confirmed that: most of the coffees we tested had some sucrose. However, in the coffees we tested, all sugars were far below the human threshold of detection, supporting the results of the Batali et al. study.[12] (Humans can detect sucrose at a concentration of about 2,000 milligrams per liter, and the coffees we tested maxed out at about 100 milligrams per liter.) In fact, the coffees with the highest total amounts of sugars were the least sweet of the coffees according to the tasters, suggesting that sugars were not directly impacting coffee sweetness.
Figure 3. Principal component analysis plot showing the relationship between the coffee professionals' recorded tastes, aromas, and flavors and the chemical presence of sugars.
Hypothesis Four: Sweetness Perception Is Modulated by Other Flavors
To explain this hypothesis, we need to take a short detour into an explanation of tastes, aromas, and flavors in coffee. “Taste” is the technical term we use for a stimulus perceived in the mouth by the taste buds as a response to dissolved compounds. Most tastes fall into five categories: sweet, sour, salty, bitter, and umami. “Aromas” are stimuli perceived in the nasal cavity as a response to volatile compounds in the air. This air can enter the nasal cavity orthonasally or retronasally. “Flavor” is another matter entirely—when the brain gets taste and aroma signals, it processes these signals together with visual and touch stimuli, sound, and memory to create an image in the brain we call “flavor.” Because coffee is so chemically complex, aromas and tastes can combine in unique ways to create flavor images, including ones we think of as “sweet.” This is different from the simple sweet taste perceived on the tongue but is instead a flavor image associated with sweetness that our brains create. Interestingly, flavor images can also work in the opposite way—associating certain flavor images with “anti-sweetness” and suppressing the sweet flavor image. Because we’re dealing with so many stimuli here, we have to turn to complex multivariate statistics that take into account all the tastes, aromas, and flavors that tasters perceive. But the result is telling. Figure 3 is a principal component analysis of tastes, aromas, and flavors along with the coffees we tested.
In this analysis, the sweetest coffees were associated with fruity and fermented aromas and flavors and some floral aromatics like vanilla. The two least-sweet coffees were associated with bitter taste, roasted and papery flavors, and the highest level of sugars. This makes some sense—as sugars caramelize under heat, they lose their sweetness and create bitter by-products. It’s possible that this could be happening in these coffees. Regardless, it seems that sweetness is in fact influenced by the perception of other flavors, aromas, and tastes. But, speaking of taste, we had one more hypothesis to test.
Hypothesis Five: Sweet Taste Is Caused by Non-Sugar Compounds
Back to the mouth and tongue for a second. Our taste buds perceive the “sweet” stimulus when exposed to enough sugar, but they also can perceive other chemicals as sweet. Sugar substitutes like stevia and aspartame work on this principle: that our sweet perception can be stimulated by non-sugar compounds. Does coffee contain compounds like this? It’s a difficult question to answer. When you have a chemically complex ingredient like coffee, finding individual compounds that create specific tastes is a lot like finding a needle in a haystack. Fortunately, the team at the Flavor Research and Education Center has developed an exciting new technique called flavoromics that uses data mining techniques to analyze immense amounts of chemical and sensory data and help solve mysteries just like this one. The research team will focus on this in the coming year, potentially identifying specific compounds that, individually or working in combination, can cause the sweet taste in coffee.
Conclusions
In short, we’ve learned that the phenomenon of sweetness in coffee is far from simple. It includes both taste and aroma, associations with other flavors, and previously unknown sweet-tasting compounds. There’s lots more to learn—as we mentioned earlier, this research did not investigate the role of roasting in causing or contributing to sweetness, or the role of coffee cherry ripeness. Our findings serve as an essential first step in identifying and characterizing what causes us to find coffee sweet, so that future researchers will be able to investigate how agricultural, roasting, and processing techniques might help make coffee sweeter and more enjoyable.
It’s even possible that by improving coffee’s natural sweetness, consumers might use less sugar in their drinks, leading to positive health outcomes. We’re incredibly excited to share this knowledge about coffee sweetness and eager to learn even more in the coming years.
Dr. NANCY CORDOBA is a former member of the Flavor Research and Education Center at Ohio State University and currently the Research and Development Principal Scientist at Westrock Coffee Company. Dr. DEVIN PETERSON is Distinguished Professor of Food, Agricultural, and Environmental Sciences at Ohio State University. PETER GIULIANO is the Executive Director of the Coffee Science Foundation.
References
[1] Ted Lingle, The Coffee Cupper’s Handbook (Long Beach, CA: Specialty Coffee Association of America, 1984).
[2] R. J. Clarke and R. Macrae, eds., Coffee: Volume 1, Chemistry (Dordrecht: Springer Netherlands, 1985), https://link.springer.com/book/10.1007/978-94-009-4948-5.
[3] Assuming total extraction of sugars, at a ratio of 55 grams per liter of light-roasted coffee.
[4] Maynard A. Amerine, Rose Marie Pangborn, and Edward B. Roessler, Principles of Sensory Evaluation of Food (New York: Academic Press, 1965).
[5] Scott C. Frost, William D. Ristenpart, and Jean-Xavier Guinard, “Effect of Basket Geometry on the Sensory Quality and Consumer Acceptance of Drip Brewed Coffee,” Journal of Food Science 84, no. 8 (2019): 2297–312, https://doi.org/10.1111/1750-3841.14696.
[6] L. Geel, M. Kinnear, and H. L. de Kock, “Relating Consumer Preferences to Sensory Attributes of Instant Coffee,” Food Quality and Preference 16, no. 3 (2005): 237–44, https://doi.org/10.1016/j.foodqual.2004.04.014; M. M. Chayan Mahmud, Robert A. Shellie, Mohammadreza Mohebbi, Dipendra Kumar Mahato, and Russell Keast, “The Effect of Fat and Coffee Concentration on the Consumer Acceptance of Iced-Coffee Beverages,” Food Science 86, no. 11 (2021): 5004–15, https://doi.org/10.1111/1750-3841.15932.
[7] Togo M. Traore, Norbert L. W. Wilson, and Deacue Fields III, “What Explains Specialty Coffee Quality Scores and Prices: A Case Study from the Cup of Excellence Program,” Journal of Agricultural and Applied Economics 50, no. 3 (2018): 349–68, https://doi.org/10.1017/aae.2018.5.
[8] Mackenzie E. Batali, Scott C. Frost, Carlito B. Lebrilla, William D. Ristenpart, and Jean-Xavier Guinard, “Sensory and Monosaccharide Analysis of Drip Brew Coffee Fractions versus Brewing Time,” Journal of the Science of Food and Agriculture 100, no. 7 (2020), https://doi.org/10.1002/jsfa.10323.
[9] Coffee Science Foundation, “Sweetness in Coffee: Sensory Analysis and Identification of Key Compounds,” https://coffeescience.foundation/sweetness-in-coffee.
[10] We used compact letter display to measure statistical difference. Any coffees that share letters with other coffees, e.g., AB or BCD, share statistical similarities with them. The fewer letters that a coffee shares with others, the more statistically different it is.
[11] Qian Janice Wang, Line Ahm Mielby, Jonas Yde Junge,
Anne Sjoerup Bertelsen, Ulla Kidmose, Charles Spence, and Derek Victor Byrne, “The Role of Intrinsic and Extrinsic Sensory Factors in Sweetness Perception of Food and Beverages: A Review,” Foods 8, no. 6 (2019): 211, https://doi.org/10.3390/foods8060211.
[12] Batali et al., “Sensory and Monosaccharide Analysis.”
We hope you are as excited as we are about the release of 25, Issue 22. This issue of 25 is made possible with the contributions of specialty coffee businesses who support the activities of the Specialty Coffee Association through its underwriting and sponsorship programs. Learn more about our underwriters here.