Changes in Wine Ethanol Content Due to Evaporation from Wine

Sep 17, 2016 - Australian Research Council Training Centre for Innovative Wine Production, PMB 1, Glen Osmond, South Australia 5064, Australia ... Res...
4 downloads 11 Views 1MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Changes in wine ethanol content due to evaporation from wine glasses David Wollan, Duc-Truc Pham, and Kerry L. Wilkinson J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b02691 • Publication Date (Web): 17 Sep 2016 Downloaded from http://pubs.acs.org on September 19, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26

Journal of Agricultural and Food Chemistry

Title: Changes in Wine Ethanol Content due to Evaporation from Wine Glasses and Implications for Sensory Analysis

Authors: David Wollan1,2, Duc-Truc Pham1 and Kerry Leigh Wilkinson1,*

Research Affiliations: 1

The University of Adelaide, School of Agriculture, Food and Wine, PMB 1, Glen Osmond, SA

5064, Australia and The Australian Research Council Training Centre for Innovative Wine Production. 2

Memstar Pty. Ltd., 712 Research Road, Nuriootpa SA 5355, Australia

* Corresponding author: Dr Kerry Wilkinson, facsimile + 61 8 8313 7116, email [email protected]

1 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

1

Changes in Wine Ethanol Content due to Evaporation from

2

Wine Glasses and Implications for Wine Sensory Analysis

Page 2 of 26

3 4

Abstract

5

The relative proportion of water and ethanol present in alcoholic beverages can significantly

6

influence the perception of wine sensory attributes. This study therefore investigated changes in

7

wine ethanol concentration due to evaporation from wine glasses. The ethanol content of

8

commercial wines exposed to ambient conditions whilst in wine glasses was monitored over time.

9

No change in wine ethanol content was observed where glasses were covered with plastic lids, but

10

where glasses were not covered, evaporation had a significant impact on wine ethanol content, with

11

losses from 0.9 to 1.9% alcohol by volume observed for wines that received direct exposure to

12

airflow for 2 hours. Evaporation also resulted in decreases in the concentration of some

13

fermentation volatiles (determined by gas chromatography-mass spectrometry) and a perceptible

14

change in wine aroma. The rate of ethanol loss was strongly influenced by exposure to airflow (i.e.

15

from the laboratory air-conditioning unit), together with certain glass shape and wine parameters;

16

glass headspace in particular). This is the first study to demonstrate the significant potential for

17

ethanol evaporation from wine in wine glasses. Research findings have important implications for

18

the technical evaluation of wine sensory properties; in particular, informal sensory trials and wine

19

show judging, where the use of covers on wine glasses is not standard practice.

20 21 22

Key words: alcohol, aroma, ethanol, evaporation, sensory evaluation, wine, wine glasses

23

2 ACS Paragon Plus Environment

Page 3 of 26

Journal of Agricultural and Food Chemistry

24

Introduction

25

Besides the aesthetic appeal of wine glasses, knowledgeable consumers believe the shape, size,

26

weight, clarity and/or color of glasses can profoundly affect their appreciation of wine sensory

27

properties.1 Not surprisingly, there is a well-established market for wine glasses which

28

manufacturers promote as having been optimized for particular wine styles. Differences in glass

29

architecture can supposedly enhance the aroma intensity, flavor balance and texture of wine.2

30

Certainly the shape and size of a wine glass are thought to influence the development of favorable

31

changes to wine sensory properties, for example, the process of wine ‘breathing’ or ‘opening up’ in

32

the glass.3 The Australian wine show system has come to recognize these effects and has

33

progressively moved away from ISO XL5 tasting glasses, to larger glassware options from Riedel

34

or other manufacturers.4

35 36

The literature comprises several studies and reviews that describe the key matrix constituents of

37

alcoholic beverages, such as whiskey, sake, shochu and wine, together with their effects on the

38

volatility of aroma compounds, and their sensory impact.5–11 The relative proportion of water and

39

ethanol present in alcoholic beverages, and their interactions with other matrix components (e.g.

40

sugars, acids and tannins) can also have significant effects on the behavior of volatile compounds

41

responsible for aroma and flavor, both in the liquid phase and the headspace.9,12,13 However,

42

changes in ethanol concentration of wine exposed to air in wine glasses have not previously been

43

investigated. There are reports on the partitioning of aroma compounds between the liquid phase

44

and the headspace14,15 and in a subsequent study,16 Tsachaki and colleagues reported the effect of

45

ethanol, temperature and headspace gas dilution on the mass transfer coefficients of a range of

46

volatile aroma compounds. Contrary to the decrease in volatile concentrations that might be

47

expected in the headspace above a water-ethanol solution (i.e. compared to water alone), the

48

presence of ethanol actually promoted evaporation of volatiles. The authors attributed this to the

49

‘Marangoni effect’ (also called ‘tears of wine’17), which involves evaporation of alcohol from a thin 3 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

50

film on the inner surface of a wine glass. Presumably this resulted in a reduction in ethanol

51

concentration, yet there was no obvious attempt to measure changes in alcohol content with time.16

Page 4 of 26

52 53

It is standard practice in formal sensory studies to cover wine glasses, especially if samples are not

54

being assessed immediately.18 However the reason behind this recommendation is to maintain the

55

equilibrium of volatiles in the headspace; there is no specific mention of the potential for alcohol

56

evaporation. Importantly, any change in alcohol concentration that occurs over time in wine

57

exposed to ambient conditions has relevance in its own right, besides the loss of aroma volatiles. It

58

remains contentious, but some winemakers believe there exists an alcoholic sweet spot

59

phenomenon, where relatively small changes in alcohol concentration can have a disproportionate

60

effect on the perceived quality and style of a wine.19 Some authors dispute the ability of sensory

61

panels to discriminate differences in alcohol concentration below 0.5% alcohol by volume

62

(abv).20,21 Nevertheless, the belief and associated practices continue.

63 64

The phenomenon of alcohol evaporation from wine glasses could have the most profound

65

consequences for wine show judging. The Australian wine show system has been a feature of wine

66

production and promotion for almost 200 years and has often influenced the quality and style of

67

Australian wines.22 In this regard, it is notably different to more formal sensory evaluations, as

68

noted by Lawless and Heymann.18 While the Best Practice Recommendations for the conduct of

69

Australian wine shows suggests tasting brackets for individual judges should be limited to 30

70

samples,4 it is not unusual for brackets to be larger, and therefore, to take over an hour to judge.

71

Given wines are poured prior to judging and remain exposed to air during this period, samples

72

judged late in a bracket may have changed significantly. Clearly this is a very different situation to

73

the more rigorous protocols of formal sensory evaluations, yet the phenomenon may introduce

74

random results to the wine show award system with significant commercial consequences. As such,

75

this study aimed to measure changes in wine ethanol concentration due to evaporation from wine 4 ACS Paragon Plus Environment

Page 5 of 26

Journal of Agricultural and Food Chemistry

76

glasses as a function of time; i.e. to determine to what extent ethanol evaporation occurs and the

77

potential impact on wine sensory properties.

78 79

Materials and Methods

80

Chemicals. Sodium chloride was purchased from Sigma Aldrich (Castle Hill, NSW, Australia).

81

Deuterated internal standards (d5-ethyl propanoate, d5-ethyl-2-methylpropanoate, d9-2-methylpropyl

82

acetate, d5-ethyl butanoate, d5-ethyl-2-methylbutanoate, d5-ethyl-3-methylbutanoate, d5-3-

83

methylbutyl acetate, d5-2-methylbutyl acetate, d13-hexanol, d5-ethyl hexanoate, d13-hexyl acetate,

84

d3-2-phenylethanol, d5-ethyl octanoate, d3-2-phenylethyl acetate and d5-ethyl decanoate) were

85

synthesized, as previously reported.23

86 87

Ethanol evaporation trials

88

A series of experiments were performed to investigate changes in the ethanol concentration of

89

(commercial) wines in glasses exposed to ambient conditions over time. In the first trial, a 2014

90

Cabernet Sauvignon wine was poured into ISO standard XL5 wine glasses (50 mL per glass) which

91

were either: (i) immediately covered with plastic lids (hereafter referred to as ‘covered’ wines); or

92

(ii) placed (without covers) on a laboratory bench in a position that (a) avoided direct exposure to

93

airflow (from a ducted air conditioner vent in the ceiling; hereafter referred to as ‘uncovered A’) or

94

(b) received direct exposure to air flow (hereafter referred to as ‘uncovered B’). The air flow at

95

these positions, being < 5.0 and 30.5 L/s, was measured with an AccuBalance Plus Air Capture

96

Hood (TSI, Shoreview MN, USA); with the airflow from the air conditioner vent being 221 L/s.

97

Ambient temperature and humidity conditions were 23 ± 1 °C and 35 ± 5%, respectively. At regular

98

intervals, being t = 0, 15, 30, 45, 60, 75, 90, 105, 120, 240 and 360 min (Table 1), the ethanol

99

content and density of wines (two technical replicates per treatment per time point) were measured

100

with an Alcolyzer (Anton Paar, Graz, Austria). Changes in mass were also determined, as the mass

101

of two technical replicates (per treatment), measured repeatedly with an analytical balance 5 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 6 of 26

102

(AUW220D, Shimadzu, Rydalmere, NSW, Australia), at each time point. Trial 2 involved two

103

white table wines, two red table wines and two fortified wines (Table 2) being subjected to the same

104

treatment conditions as in trial 1 (i.e. covered, and uncovered with and without exposure to direct

105

airflow), with the ethanol content of wines (two technical replicates per treatment) determined

106

before and after (i.e. at t = 0 and 120 min). In trial 3, the influence of wine volume and glass type

107

was studied (Table 4). Wine (the 2014 Cabernet Sauvignon from trial 1) was poured into XL5,

108

sparkling and Riedel (Ouverture Magnum, Kufstein, Austria) wine glasses (50 mL per glass), which

109

were then subjected to treatment conditions as above; with an additional treatment comprising 100

110

mL of wine in XL5 glasses, also included. The ethanol content of wines (two technical replicates

111

per treatment) was again determined before and after (i.e. at t = 0 and 120 min). Wine glass shape

112

parameters, i.e. glass height, internal wine diameter and opening diameter (Figure 1), were

113

measured with calipers.

114 115

Sensory analysis

116

A triangle test24 was conducted with a panel of 18 judges to establish whether or not evaporation

117

(for t = 120 min, as per trial 1 above) resulted in a perceptible difference in wine aroma. Wines

118

were presented in a balanced, randomized presentation order, comprising all possible configurations

119

(i.e. AAB, ABA, BAA, BBA, BAB and ABB, where A denotes covered wine and B denotes

120

uncovered B wine) an equal number of times. Wines (50 mL) were served at 22–24°C, in covered

121

ISO tasting glasses with randomly assigned four-digit codes, and panelists were asked to smell each

122

sample and to identify the sample which was different.

123 124

GC-MS analysis

125

The concentration of a range of fermentation volatiles (esters and alcohols) were determined in

126

covered and uncovered B wines, to determine the impact of evaporation (for t = 120 min, as per

127

trial 1 above) on wine composition. Measurements were performed by Metabolomics Australia 6 ACS Paragon Plus Environment

Page 7 of 26

Journal of Agricultural and Food Chemistry

128

(Australian Wine Research Institute) using gas chromatography-mass spectroscopy (GC/MS)

129

according to stable isotope dilution analysis (SIDA) methods reported previously.23 This

130

publication describes the preparation of internal standards and method validation. The conditions

131

for headspace sampling by solid phase micro-extraction (SPME) were as follows: wine (1 mL) was

132

diluted with buffer solution (9 mL, pH 3.7) and saturated with sodium chloride (2 g), prior to the

133

addition of internal standards. A polyacrylate SPME fiber was subsequently exposed to the sample

134

headspace for 10 min, prior to desorption (splitless mode), at an injector temperature of 200 °C,

135

onto an Agilent 7890A gas chromatograph equipped with a Gerstel MPS2 multi-purpose sampler,

136

and coupled to an Agilent 5975C mass selective detector. Separation was achieved with a

137

Phenomenex wax column (60 m x 0.25 mm i.d. x 0.25 µm film thickness), with helium (Ultra High

138

Purity) as the carrier gas (in constant flow mode). The initial oven temperature was 35 °C (held for

139

3 min), then increased to 220 °C (at 5 °C/min, held for 3 min). The mass spectrometer quadrupole

140

temperature was 150 °C, the source was set at 230 °C and the transfer line was held at 250 °C.

141

Positive ion electron impact spectra (at 70 eV) were recorded in selective ion monitoring (SIM)

142

mode with a solvent delay of 5 min. Raw data from Agilent ChemStation software (v E.02.02.1431)

143

were converted into MassHunter data files and processed using MassHunter Workstation Software

144

for Quantitative Analysis (v B.04.00). Fermentation volatiles were identified by comparing mass

145

spectral data with the NIST mass spectral database and subsequently quantified against their

146

corresponding isotopically labelled internal standard.

147 148

Data analysis

149

Chemical data were analyzed by one- and/or two-way analysis of variance (ANOVA) using

150

GenStat (15th Edition, VSN International Limited, Herts, UK). Mean comparisons were performed

151

by least significant difference (LSD) multiple comparison test at P< 0.05.

152 153 7 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

154

Results and Discussion

155

Ambient conditions were measured during the evaporation trials, and while temperature and

156

humidity remained constant, at 23 ± 1°C and 35 ± 5% respectively, the air flow at positions A and

157

B differed considerably, being < 5.0 and 30.5 L/s, respectively. These measurements were

158

calculated to correspond to air speeds of 0.01 and 0.08 m/s respectively, which tend towards the

159

lower end of wind speeds previously reported for indoor workplaces, which ranged from 0.01 to

160

1.80 m/s and averaged 0.3 m/s, based on anemometer measurements.25 Nonetheless, uncovered

161

wine glasses were clearly exposed to different airflows at the two positions, A and B, used in the

162

current study.

Page 8 of 26

163 164

Influence of evaporation on wine ethanol content, density and mass

165

No significant changes were observed in the ethanol content, density or mass of covered wines

166

throughout the 360 minute duration of trial 1 (Table 1). However, where wine glasses were not

167

covered, evaporation resulted in statistically significant differences in ethanol content being

168

observed within just 15 minutes. After 6 hours, the ethanol content of wine exposed to moderate

169

airflow (i.e. uncovered A wine) had decreased by almost 1% abv; which corresponded to a slight

170

increase in wine density, i.e. from 0.9948 to 0.9959 g/mL, and the loss of almost 1 g of total wine

171

mass. Where evaporation was exacerbated by direct exposure to airflow for 6 hours (i.e. in the case

172

of uncovered B wine), ethanol content decreased by 3.2% abv, i.e. from 15.1 to 11.9% abv. In this

173

time, wine density increased from 0.9948 to 0.9994 g/mL, with a corresponding loss of almost 3.5 g

174

(approximately 7%) of total wine mass.

175 176

Similar results were obtained in trial 2, which involved monitoring changes in ethanol concentration

177

of different wines over 2 hours (Table 2). The ethanol content of covered wines remained constant,

178

while the concentration of ethanol in uncovered wines decreased by between 0.2 and 1.9% abv,

179

depending on both exposure to air flow and wine type. The largest change in ethanol content was 8 ACS Paragon Plus Environment

Page 9 of 26

Journal of Agricultural and Food Chemistry

180

observed for the wine with the highest initial ethanol content, such that after two hours, the alcohol

181

content of (uncovered B) fortified wine 2, a non-vintage Muscat, had decreased from 21.7 to 19.8%

182

abv. However, with the exception of this example, the initial wine alcohol content did not seem to

183

influence the extent of ethanol evaporation, with the remaining wines each losing similar quantities

184

of ethanol, i.e. 0.2 to 0.4% abv for uncovered A wines and 0.9 to 1.2% abv for uncovered B wines.

185

Certainly, red wine 2 and fortified wine 1, which both had relatively high initial ethanol

186

concentrations, did not lose substantially higher amounts of ethanol, relative to the lighter-bodied

187

white wines 1 and 2, which had the lowest initial ethanol concentrations.

188 189

Although changes in alcohol concentration were more pronounced for uncovered wines exposed to

190

direct airflow (i.e. uncovered B wines), it still took almost 2 hours for wine alcohol content to

191

decrease by approximately 1% abv. For consumers, evaporation is therefore unlikely to be of any

192

concern, since in most consumer-related wine consumption scenarios, it would be uncommon for a

193

glass of wine to be exposed to ambient conditions for such a long time. However, more importantly,

194

there are a number of professional wine tasting scenarios in which this could conceivably occur,

195

including wine show judging. It is possible that the consistency of wine evaluation could be

196

significantly affected by changes in alcohol content arising from ethanol evaporation, particularly if

197

it is accepted that relatively small changes in wine alcohol content can profoundly influence aroma

198

volatile partitioning9 and/or wine quality and style.

199 200

Variation in airflow/air speed within a given tasting area will exacerbate the evaporation rates

201

experienced by different judges, in addition to the varying times taken by different judges to

202

complete evaluation of a bracket of wines (especially large brackets, which can comprise more than

203

30 samples). This could be substantially overcome by ensuring wines are presented to judges in

204

covered wine glasses. Whereas standard practice for formal, controlled wine sensory evaluations

205

require wine glasses to be covered,18 informal sensory trials may be undertaken under far less 9 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 10 of 26

206

rigorous conditions. Where evaluations are conducted over an extended period of time to

207

accommodate panelist availability, and where wine glasses are not covered, there is again potential

208

for wine alcohol content to significantly change due to evaporation, such that the sensory properties

209

of the wines being evaluated may differ significantly between the time they are poured and when

210

they are eventually assessed.

211 212

Influence of ethanol evaporation on wine aroma and the concentration of fermentation volatiles

213

A triangle test24 was performed to establish whether or not there was a perceptible difference in the

214

aromas of covered and uncovered wines, following exposure to ambient conditions for 120 min.

215

Covered and uncovered B wines were presented to a panel of 18 judges, 13 of whom correctly

216

identified the different sample, demonstrating evaporation had a significant impact on wine aroma

217

(at the 99% confidence level). GC-MS analysis of samples was subsequently performed to

218

determine the impact of evaporation on the concentration of a range of fermentation volatiles (Table

219

3). Of the volatiles measured, most (i.e. 11 of 15) were found at significantly lower concentrations

220

in uncovered B wine compared with covered wine; with losses ranging from 64% for ethyl

221

octanoate, to 100% for 2-methylpropyl acetate and hexyl acetate (albeit the levels of these two

222

compounds were already quite low). In contrast, moderate losses of just 21 and 38% were observed

223

for hexanol and ethyl decanoate respectively, while concentrations of 2-phenyl ethanol and 2-

224

phenylacetate were similar (i.e. varied by ≤ 10%). These results support the observations of

225

Tsachaki and colleagues16 who investigated the dynamics of these interactions and noted some

226

significant enhancement of overall mass transfer from liquid to gas phase (headspace) for various

227

volatile constituents in ethanolic solutions compared to aqueous solutions. Notably, this effect

228

varied from compound to compound. Importantly, it should also be recognized that other wine

229

volatiles (i.e. in addition to those measured) would similarly have evaporated. The combined

230

evaporation of ethanol and volatile compounds explains the apparent difference observed in the

231

aromas of covered and uncovered B wines. Given the previously reported dynamic interaction of 10 ACS Paragon Plus Environment

Page 11 of 26

Journal of Agricultural and Food Chemistry

232

ethanol solution concentration with volatile headspace concentration,15 the loss of ethanol from

233

uncovered B wine could enhance this effect.

234 235

Influence of glass shape and sample volume on the evaporation of ethanol from wine.

236

Several previous studies have attempted to investigate the influence of glass shape on wine sensory

237

properties. Glass shape (‘beaker’, ‘tulip’ vs. ‘bulbous’) was found to influence the hedonic ratings

238

given by untrained panelists to white and red wine aroma;26 but no such effect was apparent when

239

visual and tactile cues were removed.27 In contrast, a recent study found glass shape and/or

240

equilibration time affected both wine headspace composition and the intensity of aroma sensory

241

descriptors, albeit no correlations were found between various wine glass shape parameters and the

242

intensity of wine aroma attributes.28

243 244

In the third evaporation trial, the influence of glass shape/size and sample volume (Figure 1, Table

245

4) on the evaporation of ethanol from wine was investigated (Table 5). As expected, there was no

246

change in the ethanol concentration of covered wines; nor was a significant change observed when

247

uncovered A wine volume was doubled to 100 mL (in the XL5 glass). Relatively small (i.e. ≤ 0.2%

248

abv), but statistically significant changes in wine ethanol content were observed for uncovered A

249

wines in XL5 and sparkling wine glasses. In contrast, the ethanol concentration of uncovered A

250

wine decreased substantially, i.e. by 0.7% abv, in the Riedel glass. The most significant changes in

251

wine ethanol content occurred for uncovered B wines, which can be attributed to direct exposure of

252

wine glasses to greater ambient airflow. The larger dimensions of Riedel glasses resulted in the

253

greatest evaporation of ethanol, such that wine ethanol content decreased from 15.1 to 13.2% abv.

254

This observation is consistent with wine evaporation results reported by Venturi and colleagues,29

255

who found changes in wine volume loss due to evaporation (based on changes in weight of glass

256

content) was greatest for glassware with the largest opening diameters.

257 11 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 12 of 26

258

In this study, the main parameters directly related to the shape and size of the glass were volume

259

and opening diameter; both of which significantly influenced changes in wine ethanol concentration

260

(Table 5). However, differences in ethanol concentrations were evident for XL5 wine glasses

261

containing different volumes of wine. In this comparison, wine glass dimensions were identical and

262

only wine volume, wine surface area, wine surface circumference, and glass headspace volume

263

differed; albeit surface areas were similar, being 31.7 and 33.2 cm2 (Table 4). This suggests that by

264

themselves, or together, glass volume and opening diameter did not explain the differences

265

observed. Correlation coefficients (R2) were therefore calculated for changes in wine ethanol

266

content vs. various wine glass dimensions (volume and opening diameter) and/or wine parameters

267

(wine and headspace volumes, surface circumference and surface area) parameters (Table 6); with

268

the strongest correlations (R2 values of 0.999 and 0.970) being observed for headspace volume and

269

the ratio of headspace volume to wine surface area, for uncovered A and uncovered B wines,

270

respectively.

271 272

Headspace volume, calculated as the glass volume less the volume of wine sample, seemed to be

273

strongly correlated with changes in ethanol concentration, particularly when other parameters, i.e.

274

wine volume, glass opening diameter, surface circumference and surface area, were considered

275

(Table 6). These correlations point to the influence of larger glass dimensions on the rate of ethanol

276

evaporation. However, wine volume also needs to be taken into account; correlation coefficients of

277

0.990 and 0.940 were calculated for changes in wine ethanol content of uncovered A and uncovered

278

B wines respectively, vs. the ratio of headspace volume to wine volume (Table 6). Interestingly, the

279

larger Riedel style wine glass had the greatest headspace to wine volume ratio, so it is worth

280

considering whether its increasing use in wine shows may exacerbate the effects of ethanol

281

evaporation, particularly under conditions similar to those used in the current study.

282

12 ACS Paragon Plus Environment

Page 13 of 26

Journal of Agricultural and Food Chemistry

283

The ethanol concentration of (uncovered) wine was found to decrease significantly over time as a

284

consequence of evaporation; with the rate of ethanol loss being strongly influenced by exposure to

285

airflow, together with glass shape, headspace and wine volume. However, the evaporation of

286

ethanol was prevented by simply placing covers over wine glasses. These findings have important

287

implications for the technical evaluation of wine, since even by the conservative assessments made

288

in previous studies, small changes in alcohol content (i.e. ~1% abv) can significantly influence our

289

perception of wine sensory attributes. In some instances, e.g. sensory trials (including those

290

employed for ‘alcohol sweet-spotting’) and wine show judging where the use of covers on wine

291

glasses is not standard practice, there is therefore the potential for significant sample variation as a

292

consequence of evaporation. The extent to which evaporation occurs will be influenced by ambient

293

conditions (temperature and airflow, in particular) and the glassware used. Evaporation can also

294

result in the loss of wine volatile compounds and can have a marked impact on wine aroma.

295 296

ACKNOWLEDGEMENTS

297

The authors thank Tony Robinson (Treasury Wine Estates) for his considered and constructive

298

comments.

299 300

FUNDING SOURCES

301

This research was conducted by the Australian Research Council Industrial Transformations

302

Training Centre for Innovative Wine Production (project number IC130100005).

13 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

303

REFERENCES

304

1.

Page 14 of 26

Spence, C.; Wan, X. Beverage perception and consumption: The influence of the container on the perception of the contents. Food Qual. Pref., 2015, 39, 131–140.

305 306 307

2.

Riedel. The content determines the shape, 2015. http://www.riedel.com/all-about-riedel/shapespleasure/content-commands-shape/ [accessed 19/12/15].

308 309 310

3.

Jackson, R. S. (2009) Chapter 6 - Qualitative Wine Assessment. In: Wine Tasting, 2nd Edition. Jackson, R. S., Ed. Academic Press: San Diego, CA, USA, 2009, pp 303–348.

311 312 313

4.

Jordan, T.; Robinson, A.; Bulleid, N. The 2015 ASVO Wine Show Best Practice

314

Recommendations. http:// http://www.asvo.com.au/2015-asvo-best-practice-recommendations/

315

[accessed 11/02/16].

316 317

5.

Conner, J. M.; Paterson, A.; Piggott, J. R. Interactions between ethyl esters and aroma compounds in model spirit solutions. J. Agric. Food Chem., 1994, 42, 2231–2234.

318 319 320

6.

Conner, J. M.; Birkmyre, L.; Paterson, A.; Piggott, J. R. Headspace concentrations of ethyl esters at different alcoholic strengths. J Sci. Food Agric., 1998, 77, 121–126.

321 322 323

7.

Nose, A.; Hojo, M. Interaction between water and ethanol via hydrogen bonding in alcoholic

324

beverages. In: Alcoholic Beverage Consumption and Health, Mazzei, A; D’Arco, A., Ed. Nova

325

Science Publishers Inc.: Hauppage, New York, USA, 2009, pp 37–91.

326

14 ACS Paragon Plus Environment

Page 15 of 26

327

Journal of Agricultural and Food Chemistry

8.

Goldner, M. C.; Zamora, M. C.; Di Leo Lira, P.; Gianninoto, H.; Bandoni, A. Effect of ethanol

328

level in the perception of aroma attributes and the detection of volatile compounds in red wine.

329

J. Sens Stud., 2009, 24, 243–257.

330 331

9.

Robinson, A. L.; Ebeler, S. E.; Heymann, H.; Boss, P. K.; Solomon, P. S.; Trengove, R. D.

332

Interactions between wine volatile compounds and grape and wine matrix components

333

influence aroma compound headspace partitioning. J. Agric. Food Chem., 2009, 57, 10313–

334

10322.

335 336 337

10. Secor, A. C. Effects of ethanol, tannin and fructose on the sensory and chemical properties of Washington State Merlot. Masters Thesis, Washington State University, 2012.

338 339 340

11. Villamor, R. R.; Ross, C. F. Wine matrix compounds affect perception of wine aromas. Ann. Rev. Food Sci. Tech., 2013, 4, 1–20.

341 342

12. Fontoin, H.; Saucier, C.; Teissedre, P.-L.; Glories, Y. Effect of pH, ethanol and acidity on

343

astringency and bitterness of grape seed tannin oligomers in model wine solution. Food Qual.

344

Pref., 2008, 19, 286–291.

345 346

13. McRae, J. M.; Ziora, Z. M.; Kassara, S.; Cooper, M. A.; Smith, P. A. Ethanol concentration

347

influences the mechanism of wine tannin interactions with poly(1-proline) in model wine. J.

348

Agric. Food Chem., 2015, 63, 4345–4352.

349 350 351

14. Jung, D.-M.; Ebeler, S. E. Headspace solid-phase microextraction method for the study of the volatility of selected flavor compounds. J. Agric. Food Chem., 2002, 51, 200–205.

352 15 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

353

Page 16 of 26

15. Tsachaki, M.; Linforth, R. S. T.; Taylor, A. J. Dynamic headspace analysis of the release of

354

volatile organic compounds from ethanolic systems by direct APCI-MS. J. Agric. Food Chem.,

355

2005, 53, 8328–8333.

356 357

16. Tsachaki, M.; Gady, A.-L.; Kalopesas, M.; Linforth, R. S. T.; Athès, V.; Marin, M.; Taylor, A.

358

J. Effect of ethanol, temperature, and gas flow rate on volatile release from aqueous solutions

359

under dynamic headspace dilution conditions. J. Agric. Food Chem., 2008, 56, 5308–5315.

360

361

17. Neogi, P. Tears-of-wine and related phenomena. J. Colloid Interface Sci., 1985, 105, 94–101.

362 363 364

18. Lawless, H. T.; Heymann, H. Sensory Evaluation of Food: Principles and Practices, 2nd Edition. Springer: New York, USA, 2010.

365 366 367

19. Wollan, D. Alcohol sweet spot seminars in McLaren Vale. Aust. N. Z. Grapegrow. Winemak., 2005, 502, 70–71.

368 369 370

20. King, E. S.; Heymann, H. The effect of reduced alcohol on the sensory profiles and consumer preferences of white wine. J. Sens. Stud., 2014, 29, 33–42.

371 372 373

21. Yu, P.; Pickering, G. J. Ethanol difference thresholds in wine and the influence of mode of evaluation and wine style. Am. J. Enol. Vitic., 2008, 59, 146–152.

374 375 376

22. Dunphy, R;.Lockshin, L. A history of the Australian wine show system. J. Wine Res., 1998, 9, 87–105.

377

16 ACS Paragon Plus Environment

Page 17 of 26

Journal of Agricultural and Food Chemistry

378

23. Siebert, T. E.; Smyth, H. E.; Capone, D. L.; Neuwöhner, C.; Pardon, K. H.; Skouroumounis, G.

379

K.; Herderich, M. J.; Sefton, M. A.; Pollnitz, A. P. Stable isotope dilution analysis of wine

380

fermentation products by HS-SPME-GC-MS. Ana. Bioanal. Chem., 2005, 381, 937–947.

381 382 383

24. Meilgaard, M.; Civille, G. V.; Carr, B. T. Sensory Evaluation Techniques, 3rd Edition. CRC: New York, USA, 1999.

384 385 386

25. Baldwin, P. E. J.; Maynard, A. D. A survey of wind speeds in indoor workplaces. Ann. Occup. Hyg., 1998, 42, 303–313.

387 388

26. Hummel, T.; Delwiche, J. F.; Schmidt, C.; Hüttenbrink, K. B. Effects of the form of glasses on

389

the perception of wine flavors: a study in untrained subjects. Appetite, 2003, 41, 197–202.

390 391 392

27. Delwiche, J. F.; Pelchat, M. L. Influence of glass shape on wine aroma. J. Sens. Stud., 2002, 17, 19–28.

393 394 395

28. Hirson, G. D.; Heymann, H.; Ebeler, S. E. Equilibration time and glass shape effects on chemical and sensory properties of wine. Am. J. Enol. Vitic., 2012, 63, 515–521.

396 397

29. Venturi, F.; Andrich, G.; Sanmartin, C.; Scalabrelli, G.; Ferroni, G.; Zinnai, A. The expression

398

of a full-bodied red wine as a function of the characteristics of the glass utilized for the tasting.

399

CyTA – J. Food, 2014, 12, 291–297.

400 401

17 ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

402

FIGURE CAPTIONS

403

Figure 1. Dimensions of XL5, sparkling and Riedel wine glasses (± 1 mm).

Page 18 of 26

18 ACS Paragon Plus Environment

Page 19 of 26

Journal of Agricultural and Food Chemistry

Table 1. Changes in ethanol concentration, density and mass of Cabernet Sauvignon wine in covered and uncovered wine glasses exposed to ambient conditions over time (from 0 to 360 min); with uncovered glasses positioned to either (A) avoid or (B) receive exposure to direct airflow. Massa (g) Ethanol Content (% abv) Density (g/mL) Time Point (min) covered uncovered A uncovered B covered uncovered A uncovered B covered uncovered A uncovered B 0 15.1 15.1 a 15.1 a 0.9948 0.9948 a 0.9948 a 0.000 0.000 a 0.000 a 15 15.0 15.0 b 14.9 b 0.9948 0.9948 a 0.9950 b 0.008 0.049 ab 0.237 b 30 15.0 15.0 b 14.7 c 0.9948 0.9949 b 0.9952 c 0.014 0.095 abc 0.412 c 0.9948 0.9949 b 0.9953 d 0.019 0.133 bcd 0.565 d 45 15.0 15.0 b 14.6 d 60 15.0 14.9 c 14.5 e 0.9948 0.9949 b 0.9955 e 0.024 0.182 bcde 0.702 e 75 15.0 14.9 c 14.3 f 0.9948 0.9949 b 0.9958 f 0.028 0.221 cde 0.825 f 90 15.0 14.9 c 14.2 g 0.9948 0.9950 c 0.9959 g 0.033 0.264 def 0.953 g 105 15.0 14.9 c 14.2 g 0.9948 0.9950 c 0.9960 h 0.038 0.310 ef 1.082 h 120 15.0 14.8 d 14.0 h 0.9948 0.9950 c 0.9963 i 0.043 0.353 f 1.237 i 240 15.0 14.6 e 13.0 i 0.9948 0.9954 d 0.9978 j 0.075 0.672 g 2.328 j 360 15.0 14.2 f 11.9 j 0.9948 0.9959 e 0.9994 k 0.105 0.986 h 3.452 k ns P ns < 0.001 < 0.001 ns < 0.001 < 0.001 < 0.001 < 0.001 LSD 0.030 (treatment); 0.057 (time point) 0.000042 (treatment); 0.00008 (time point) 0.068 (treatment); 0.131 (time point) Values are means of two replicate measurements (n=2). Standard errors were ≤ 0.07 and ≤ 0.0005 for ethanol and density, respectively. Values followed by different letters within columns are significantly different; ns = not significant. a reported as a decrease in mass relative to t = 0.

19

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 20 of 26

Table 2. Changes in ethanol concentration of wines in covered and uncovered glasses exposed to ambient conditions over time (0 to 120 min); with uncovered glasses positioned to either (A) avoid or (B) receive exposure to direct airflow. Ethanol Content (% abv) Wine initial covered uncovered A (t = 0 min) (t = 120 min) (t = 120 min) white wine 1 2011 Sauvignon Blanc Semillon 12.6 a 12.6 a 12.4 b white wine 2 2013 Sauvignon Blanc 11.3 a 11.3 a 11.1 a red wine 1 2009 Cabernet Sauvignon Merlot 13.7 a 13.6 a 13.3 b red wine 2 2014 Cabernet Sauvignon 15.0 a 15.0 a 14.8 b fortified wine 1 non-vintage Apera 16.3 a 16.3 a 16.0 b fortified wine 2 non-vintage Muscat 21.7 a 21.7 a 21.4 b Values are means of two replicate measurements (n=2). Standard errors were ≤ 0.15. Values followed by different letters within rows are significantly different.

uncovered B (t = 120 min) 11.4 c 10.4 b 12.5 c 14.0 c 15.1 c 19.8 c

P

LSD

< 0.001 0.002 < 0.001 < 0.001 < 0.001 < 0.001

0.072 0.267 0.054 0.165 0.200 0.256

20

ACS Paragon Plus Environment

Page 21 of 26

Journal of Agricultural and Food Chemistry

Table 3. Concentrations (µg/L) of fermentation volatiles in covered and uncovered B wine following exposure to ambient conditions (for 120 min); with uncovered glasses positioned to receive exposure to direct airflow. Threshold23 (µg/L)

Concentration (µg/L) P covered uncovered B ethyl propanoate fruity 1,840 385 a 115 b 0.001 ethyl 2-methylpropanoate fruity 15 200 a 66 b < 0.001 2-methylpropyl acetate banana, fruity 1,600 13 nd – ethyl butanoate acid fruit 20 291 a 79 b 0.001 ethyl 2-methylbutanoate sweet fruit 1 56 a 16 b < 0.001 ethyl 3-methylbutanoate berry 3 66 a 19 b < 0.001 3-methylbutyl acetate banana 30 339 a 100 b < 0.001 2-methylbutyl acetate banana, fruity 1,600 183 a 46 b < 0.001 hexanol green, grass 4,000 3,693 a 2,923 b < 0.01 ethyl hexanoate green apple 5 46 a 15 b < 0.001 hexyl acetate sweet, perfume 670 3 nd – 2-phenylethanol roses 10,000 75,746 b 79,628 a < 0.01 ethyl octanoate sweet, soap 2 39 a 14 b 0.002 2-phenylacetate floral 250 33 a 30 b < 0.001 ethyl decanoate pleasant, soap 200 8 5 < 0.1 Values are means of two replicate measurements (n=2). Relative standard errors were ≤ 5%. nd = not detected Values followed by different letters within rows are significantly different. Volatile Compound

Descriptor23

21

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 22 of 26

Table 4. Changes in ethanol concentration of wines in covered and uncovered glasses exposed to ambient conditions over time (0 to 120 min); with uncovered glasses positioned to either (A) avoid or (B) receive exposure to direct airflow. Ethanol Content (% abv) Wine Glass P initial covered uncovered A uncovered B (t = 0 min) (t = 120 min) (t = 120 min) (t = 120 min) XL5 (100 mL) 15.0 a 15.0 a 14.9 a 14.3 b 0.002 XL5 (50 mL) 15.0 a 15.0 a 14.8 b 14.0 c < 0.001 sparkling 15.0 a 15.0 a 14.9 b 13.9 c < 0.001 Riedel 15.1 a 15.0 a 14.4 b 13.2 c < 0.001 Values are means of two replicate measurements (n=2). Standard errors were ≤ 0.15. Values followed by different letters within rows are significantly different.

LSD 0.228 0.165 0.092 0.300

22

ACS Paragon Plus Environment

Page 23 of 26

Journal of Agricultural and Food Chemistry

Table 5. Parameters of XL5, sparkling and Riedel wine glasses. Glass Wine Wine Wine Headspace Headspace Headspace Headspace Glass Wine Headspace Opening Surface Surface Surface Volume: Volume: Volume: Volume: Wine Glass Volume Volume Volume Diameter Diameter Circumference Area Wine Opening Wine Wine (mL) (mL) (mL) (mm) (mm) (mm) (cm2) Volume Diameter Circumference Surface Area XL5 (100 mL) 215 100 115 44 65 204 31.7 3.30 3.75 0.81 4.97 XL5 (50 mL) 215 50 165 44 64 199 33.2 1.15 2.61 0.58 3.63 sparkling 155 50 110 46 51 160 20.4 2.10 2.31 0.66 5.14 Riedel 560 50 480 68 72 226 40.7 10.20 7.50 2.25 12.53 Values are means of two replicate measurements (n=2). Standard errors were ≤ 0.15. Values followed by different letters within rows are significantly different.

23

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 24 of 26

Table 6. Correlations between changes in wine ethanol content and various wine glass dimensions and/or wine parameters. Correlation Coefficient (R2) uncovered A uncovered B Changes in wine ethanol content vs. glass volume 0.975 0.805 Changes in wine ethanol content vs. wine volume 0.165 0.382 Changes in wine ethanol content vs. headspace volume 0.999 0.884 Changes in wine ethanol content vs. glass opening diameter 0.960 0.913 Changes in wine ethanol content vs. wine surface circumference 0.570 0.259 Changes in wine ethanol content vs. wine surface area 0.624 0.308 Changes in wine ethanol content vs. headspace volume/wine volume 0.990 0.940 Changes in wine ethanol content vs. headspace volume/glass opening diameter 0.989 0.842 0.996 0.921 Changes in wine ethanol content vs. headspace volume/wine surface circumference Changes in wine ethanol content vs. headspace volume/wine surface area 0.968 0.970

24

ACS Paragon Plus Environment

Page 25 of 26

Journal of Agricultural and Food Chemistry

Figure 1.

25

ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry

Page 26 of 26

TOC Graphic.

26

ACS Paragon Plus Environment