Graduate Thesis Or Dissertation


Tropical Fruit Aroma: Relevance to Oregon White Wines, the Effect of Winemaking Processes on Fermentation Esters and Volatile Thiol Levels, and the Relationship Between Sensory Perception and Volatile Chemistry Public Deposited

Downloadable Content

Download PDF


Attribute NameValues
  • Aroma is one of the major components that determines the quality of wine and consumer acceptability. Fruitiness is a highly desirable aroma quality in white wines and has been well accepted by consumers across the globe. Tropical fruit aroma has been substantially studied in Sauvignon Blanc wines due to its higher levels of volatile thiol precursors present in the grapes. These volatile thiols are thought to cause the tropical fruit aroma of Sauvignon blanc wines. To date, there is limited work investigating the levels of free volatile thiols found in wines made from other white grapes. Research on tropical fruit aroma in wine has been exclusively linked to volatile thiols. However, wine contains hundreds of aroma compounds and compound interactions need to be investigated to determine the cause of tropical fruit aroma in wine. Therefore, this dissertation aimed to investigate the effect of esters and thiols to tropical fruit aroma perception. To do so, we considered the levels of volatile thiols and esters found in Chardonnay wines. Specific winemaking processes focused on increasing or decreasing these aroma families. The esters and volatiles thiols of the resulting wines were determined, and tropical aroma was evaluated using sensory analysis. The first study aimed to explore the fruitiness aroma perception of commercially available varietal white wines from Oregon. Chardonnay, Pinot gris and Viognier wines were evaluated by wine expert and wine consumer panels using an adaptation of the Polarized Projective Mapping (PPM) sensory method. Normally, the poles in PPM are products (in this case, wine) that are representative of the sensory space. This method requires products to be stable over time if comparisons between samples from different sensory panels are desired. Considering the complexity and lack of stability of wine, the adaptation of the method involved the use of aroma standards as poles instead of products. Aroma standards are stable products and easy to reproduce. The results showed that white wines from Oregon have tropical fruit, citrus, pear, and stone fruit aromas. Wine experts produced a more detailed description of the aroma profile of the wines compared to the wine consumer panel. These wine experts have a superior sense of smell as they regularly evaluate wine, therefore, they make a better panel for wine sensory methods that do not require training, such as PPM. It was concluded that the wine consumer panel would produce better results using the PPM method if they are trained both in the methodology and in wine aroma identification. Much of the work investigating the causes of tropical fruit aroma in white wines attribute it exclusively to the volatile thiols 3-sulfanylhexan-1-ol (3-SH), 3-sulfanylhexyl acetate (3-SHA), and 4-methyl-4-sulfanylpentan-2-one (4-MSP). King et al. (2011), showed evidence that a combination of esters and thiols resulted in stronger positive correlations to tropical fruit aromas compared to thiols themselves. Therefore, the second study assessed the effect of esters and volatile thiol levels found in Chardonnay wines on fruitiness perception. Three levels of esters (none, low, and medium) and four levels of thiols (none, low, medium, and high) were added to dearomatized white wine using a full factorial design. Check-All-That-Apply (CATA) utilizing white wine consumers was used to gather descriptors that most differentiated wine samples. Sensory descriptive analysis with a trained panel, employed 100 mm visual analog scales (line scales) to explore the aroma perception and attribute intensity of the wines. The results showed that thiols by themselves did not contribute to tropical fruit aromas for the concentrations studied. Thiols, regardless of their concentrations, imparted grass and earthy aromas. Tropical fruit aromas were associated with esters and esters+thiols combinations, where changes in the quality of tropical fruit aroma perception (e.g. passionfruit, guava, grapefruit, or citrus) changed depending on compound levels and combinations. Given that esters and thiols are important to tropical fruit aroma perception in white wines, the third study investigated specific winemaking processes and could increase or decrease the concentrations of both aroma families in Chardonnay wine. The effect of pre-fermentative skin contact (10˚C for 18 hours), β-lyase enzyme (40 μL/L), and two fermentation gradient temperature regimes FG1 (started at 20˚C and after 96 hours dropped to 13˚C), and FG2 (started at 20˚C and after ~ 11.5˚Brix dropped to 13˚C) were studied using a full factorial design. A total of 30 esters were measured in the wines using a HS-SPME GC-MS method. The volatile thiols 3-SH, 3-SHA, and 4-MSP were derivatized and concentrated, followed by quantification using LC-MS/MS. β-lyase did not alter the volatile thiol concentrations of the wines. Skin contact, followed by FG1 and FG2 showed increased the levels of both esters and thiols. The combinations of skin contact and FG1 or skin contact and FG2 resulted in the greatest levels of both aroma families. For example, a 2-fold increase in 3-SH was observed in both SC+FG1 and SC+FG2 treatments. This study showed that skin contact was effective at extracting thiol precursors and nutrients for ester production from grape skins. Fermentation gradient showed to be efficient in reducing volatile compound loss during fermentation due to evaporation. A future study should include the addition of β-lyase in higher levels, both alone and in combination with SC and FG, to evaluate the esters and thiol composition of the wines. The fourth study aimed to investigate whether the concentrations of esters and volatile thiols found in commercial Sauvignon blanc and Chardonnay wines could be used as predictors of tropical fruit aroma perception. Thirty wines having specific aroma characteristics were selected for this study after a preliminary tasting. A sensory descriptive analysis panel (n=19) was fully trained with standards representing the aromas found in these wines. Esters and volatile thiols were quantified using HS-SPME GC-MS and LC-MS/MS methods, respectively. As expected, higher levels of volatile thiols were found in the Sauvignon blanc wines, while esters were found in greater concentrations in Chardonnay wines. Principal component analysis showed that most Sauvignon blanc wines were split into two groups and were described by either tropical fruit or grass and green bell pepper aromas. Chardonnay wines were also grouped in two clusters and were associated with fruity (citrus, stone fruit, pineapple, and pome) or oaky, butter, and honey aromas. Partial Least Square Regression (PLS-R) revealed that tropical fruit aroma is positively associated with volatile thiols and the presence of specific ethyl esters, such as ethyl hexanoate, ethyl butanoate, ethyl propanoate, ethyl trans-2-hexenoate, ethyl 3-hydroxyhexanoate, and also neryl acetate, thus showing that the presence of these compounds is important to impart this aroma perception in white wines. Finally, the wine industry will benefit from the results of this work as they will know the causes of tropical fruit aromas in Chardonnay wine. The outcomes of this study will also provide insights of which winemaking processes can achieve or not tropical fruit aroma. Lastly, this work will inform winemakers of existing tools to adapt winemaking protocols to achieve specific wine styles desired by consumers.
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Academic Affiliation
Rights Statement
Peer Reviewed



This work has no parents.

In Collection: