Compounds that strengthen a fruity aroma do not necessarily smell fruity themselves. In a 2014 wine study, adding dimethyl sulfide (DMS), described as sulfurous on its own, to a model fruit aroma strengthened the impression of dark fruit and, at certain concentrations, enhanced blackcurrant character.
The study focused on both a quantitative role, making an aroma easier to detect, and a qualitative role, changing which fruit it evokes. It carefully examined how a compound can play a different role in a blend than it does when smelled alone.
Building a red-wine fruit aroma from 12 esters
The researchers combined 12 esters at average concentrations found in red wine. This reconstructed fruit-aroma model included ethyl butyrate, ethyl hexanoate, and isoamyl acetate, among other compounds.
They used two base liquids. One was a water–ethanol solution with 12% alcohol and a pH adjusted to 3.5. The other was red wine from which as much of the original aroma as possible had been removed. The preparation involved evaporation, treatment with an adsorbent resin, and restoration of the alcohol concentration and liquid volume, producing a weakly aromatic sample. This allowed the researchers to compare DMS in a simple solution and in a liquid that retained wine-derived components.
The assessors were staff at a University of Bordeaux research facility experienced in evaluating fruity aromas in red wine. A panel of 16 assessed discrimination, thresholds, and the intensity of aroma notes, while a panel of eight described the odors. Fruit syrups, liqueurs, and jams were presented as reference standards for the descriptors.
About 50 mL of each sample was served in a black glass and evaluated by smelling through the nose at 20°C. Because DMS is volatile, the researchers analyzed how its concentration changed over time and refreshed samples every 10 minutes. This was an experiment on aromas rising from a glass, not a comparison of taste or preference while drinking.
The background changes how easily DMS itself is detected
First, the researchers tested how dilute DMS could be while remaining detectable, using a task that required choosing the odd sample out of three. Thresholds corrected for chance responses were 1.74 µg/L in the simple alcohol solution, 2.80 µg/L when the 12-component fruit aroma was added, and 3.67 µg/L when that fruit aroma was added to dearomatized red wine.
The detectability of the same DMS thus changed with the other compounds present. The 5–70 µg/L concentrations used to examine its effects exceeded these thresholds. Treating this as an experiment in adding odorless DMS would misrepresent the design.
The researchers also examined descriptions of DMS alone. Samples at 10, 30, and 70 µg/L were described mainly in sulfur-related terms, changing from lightly sulfurous at low concentrations to strongly sulfurous or reductive at higher ones. DMS alone did not have a clear fruity aroma.
A smaller amount of fruit aroma became detectable
Next, the researchers diluted the 12-component fruit-aroma model itself to find its detection threshold. They compared an alcohol solution without DMS with an alcohol solution containing 5 µg/L DMS.
In the DMS condition, the reference background samples contained the same DMS. Choosing whichever sample contained DMS was therefore not enough to answer correctly. The task was to detect the fruit aroma added to that background.
The amount of fruit-aroma model corresponding to the threshold, in 50 mL of background liquid, was 1.80 mL without DMS and 0.55 mL with it. This was a statistically clear difference: roughly one-third as much was sufficient for detection. The ratio compares thresholds; it does not mean that an everyday aroma would simply smell three times as strong.
Blackcurrant character increased in particular
When DMS was added to the fruit-aroma model at 5, 10, 20, 30, 50, and 70 µg/L, the overall dark-fruit intensity in the simple alcohol solution was rated higher than without DMS at every concentration. However, ratings did not rise with increasing DMS. The degree of enhancement was broadly similar across the amounts added.
The changes were not uniform across individual fruits. Blackcurrant intensity was significantly higher than without DMS at 5, 10, 20, and 30 µg/L in the alcohol solution, and at 5, 10, and 20 µg/L in dearomatized red wine. No clear change was reported for blackberry, strawberry, or raspberry intensity.
It is important here to distinguish the paper’s broad dark-fruit category from the specific fruit, blackberry. The broader aroma family became stronger, but not every fruit within it increased in the same way. Specifically, the impression of blackcurrant stood out.
A compound’s role emerges through what it is mixed with
The authors interpreted DMS not as a compound that represents fruity aroma on its own, but as one that brings out fruitiness through combinations with other aroma compounds. Lowering the fruit-aroma threshold and shifting the impression toward blackcurrant were both demonstrated in the same study.
The authors also discussed the possibility that increases in DMS and certain esters during wine aging contribute to the development of fruity aroma. However, this study measured prepared mixtures; it was not an aging experiment that followed the same wine for years.
The assessors were small panels of experienced judges, and the fruit-aroma formulation was limited. Commercial wines contain many more odors, and the concentrations at which DMS had an effect changed with the background. The fact that blackcurrant ratings did not keep rising at higher concentrations suggests that simply adding more does not necessarily produce the intended aroma.
Summary
When added to a 12-component fruit-aroma model, sulfurous-smelling DMS made a smaller amount of fruit aroma detectable and enhanced blackcurrant character within a certain concentration range. The same change was not seen for blackberry, strawberry, or raspberry, and more DMS did not consistently mean a stronger blackcurrant impression.
A compound that supports fruitiness in a blend does not necessarily smell fruity on its own. That is an intriguing finding from this study.
The research received financial support from the Bordeaux Wine Council (CIVB). No separate competing-interest statement was found in the original paper.
Reference
Lytra G, Tempere S, Zhang S, Marchand S, de Revel G, Barbe JC. Olfactory impact of dimethyl sulfide on red wine fruity esters aroma expression in model solution. Journal International des Sciences de la Vigne et du Vin. 2014;48(1):75–85. doi:10.20870/oeno-one.2014.48.1.1660
