Formation pathways of ethyl esters of branched short-chain fatty acids during wine aging

Article abstract of DOI:10.1021/jf048157o

The particular behavior during wine aging of fermentative branched fatty acid ethyl esters, related to yeast nitrogen metabolism, compared that of their straight-chain analogues, related to yeast lipid metabolism, was first checked in 1-5 year aged Muscadet wines. Quantitative SIDA measurements showed that the levels of the former increased, whereas those of the latter decreased. Then, three hypothetical pathways suggested in the literature to explain these variations of branched esters were investigated. Two Muscadet and Sylvaner wines were spiked with levels of deuterated isobutanoic acid and its ethyl ester, similar to those of their natural analogues, then they were submitted to model aging. Quantitative SIDA measurements on the formation of these natural and labeled ethyl esters from the corresponding acids revealed that the behavior of the natural and labeled compounds were similar. The acid levels were much higher than the ester levels in the initial young wine, and a significant upward trend of their esterification ratios to those of the acid-ester equilibrium was observed with aging. Thus, this equilibrium proved to be the most effective in generating the branched fatty acid ethyl esters during wine aging. In contrast, the formation of these acids by Strecker-type degradation of wine amino acids in the conditions of the model aging or by hydrolysis of their glycoconjugates proved to be ineffective.

Full text of DOI:10.1021/jf048157o

J. Agric. Food Chem. 2005, 53, 3503−3509 3503
Formation Pathways of Ethyl Esters of Branched Short-Chain
Fatty Acids during Wine Aging
M. CONSUELO D IÄ AZ-MAROTO,* R EÄ MI SCHNEIDER, AND RAYMOND BAUMES§
,
†
‡,§
AÄ rea de Tecnolog ´ı a de Alimentos, Facultad de Ciencias Qu ´ı micas (UCLM), Avenida de Camilo Jos e´
Cela, 10, 13071 Ciudad Real, Spain, and ITV France, 19 rue du G e´ n e´ ral Foy, 75008 Paris, France,
and INRA-UMR-Sciences pour l’œnologie, 2 place Viala, 34060 Montpellier, France
The particular behavior during wine aging of fermentative branched fatty acid ethyl esters, related to
yeast nitrogen metabolism, compared that of their straight-chain analogues, related to yeast lipid
metabolism, was first checked in 1-5 year aged Muscadet wines. Quantitative SIDA measurements
showed that the levels of the former increased, whereas those of the latter decreased. Then, three
hypothetical pathways suggested in the literature to explain these variations of branched esters were
investigated. Two Muscadet and Sylvaner wines were spiked with levels of deuterated isobutanoic
acid and its ethyl ester, similar to those of their natural analogues, then they were submitted to model
aging. Quantitative SIDA measurements on the formation of these natural and labeled ethyl esters
from the corresponding acids revealed that the behavior of the natural and labeled compounds were
similar. The acid levels were much higher than the ester levels in the initial young wine, and a significant
upward trend of their esterification ratios to those of the acid-ester equilibrium was observed with
aging. Thus, this equilibrium proved to be the most effective in generating the branched fatty acid
ethyl esters during wine aging. In contrast, the formation of these acids by Strecker-type degradation
of wine amino acids in the conditions of the model aging or by hydrolysis of their glycoconjugates
proved to be ineffective.
KEYWORDS: Wine; ethyl esters; branched short-chain fatty acids; stable isotope dilution assay
INTRODUCTION
from the amino acid metabolism of yeast, were also recognized
as important odorants of wine (7-11). The variation of their
levels during aging of wine is different to that of the ethyl esters
of straight-chain fatty acids, derived from the lipid metabolism
of yeast, and to that of the fusel alcohol acetates. The
concentrations of the two last classes of esters generally decrease
during wine aging. These variations are accelerated by high
temperature and low pH and depend on the equilibrium between
the esters and the corresponding acids, with an equilibrium
constant close to 4, as demonstrated previously (12-19).
In contrast, the levels of the branched fatty acid ethyl esters
are more or less stable or even can increase during the aging of
wine (10, 20), as well as of beer (21). Apart from their
production by flor yeasts during the biological aging process
of sherry wines (20), it is suggested in the literature that different
chemical mechanisms might explain these variations. They could
depend on the acid-ester equilibrium, if the esterification molar
ratios of the branched fatty acids were lower in young wines
than in aged wines at the equilibrium, as observed for the fixed
acids (1). They could also be formed by the shifting of this
acid-ester equilibrium, due to the formation of branched fatty
acids. In nonoxidative aging, the branched fatty acids could be
produced from the corresponding amino acids by Strecker
degradation or from some amino acid derivatives formed during
the fermentation, e.g., R-ketoacids (10). Similar hypothesis were
Wine volatiles are originated in part from grape and in part
from fermentation. The major fermentation aroma constituents
are ethanol, fusel alcohols, and esters. These last compounds
are essentially ethyl esters of organic acids, fusel alcohol
acetates, and ethyl esters of fatty acids (1). These esters arise
from alcoholysis of the corresponding acids activated as acyl-
S-CoA by yeast (2, 3).
Ethyl esters of straight-chain fatty acids with an even number
of carbon atoms, such as ethyl hexanoate, ethyl octanoate, or
ethyl decanoate, and fusel alcohol acetates, such as 2-methyl-
butyl, isopentyl, and 2-phenylethyl acetates, are considered
important contributors to young wine aroma and exhibit floral
and fruity odors (4, 5). In contrast, ethyl acetate can give an
unpleasant odor to the wine at concentrations higher than 100
mg/L (6).
However, it was generally admitted that the ethyl esters of
branched short-chain fatty acids, such as ethyl isobutanoate,
ethyl 2-methylbutanoate, and ethyl isopentanoate, were negli-
gible contributors to wine aroma (1). It is only in the past decade
that GC/olfactometry studies showed that these esters, derived
*
Corresponding author: MariaConsuelo.Diaz@uclm.es.
Facultad de Ciencias Qu ´ı micas (UCLM).
ITV France.
†
‡
§
INRA-UMR-Sciences pour l’œnologie.
1
0.1021/jf048157o CCC: $30.25
© 2005 American Chemical Society
Published on Web 04/02/2005

3504 J. Agric. Food Chem., Vol. 53, No. 9, 2005
D ´ı az-maroto et al.
Table 1. Levels of Some Fatty Acid Ethyl Estersa of Muscadet Wines of Different Vintages and Different Locations of the Muscadet Vineyard (Melon
B. grapes)
wineb
LIM 1999
agec
pH
2meC3
2meC4
3meC4
C4
C6
C8
C10
1
1
1
1
3.39
3.22
3.34
3.33
95
31
35
91
63
543
361
328
344.5
2970
4
3
4
6
1
3
3
3
141
100
143
178
140
1207
162
80
824
755
670
884
783
5438
602
424
513
4422
862
728
578
981
787
4576
506
218
362
3121
408
357
355
511
408
2040
153
150
151.5
1306
MON 1999
DRA 1999
LOR 1999
3
a
mean level 1999
3.5
27
7
21
14
121
d
molar level
23
30
135
82.5
711
DRA 1996
4
5
3.25
3.17
MON 1995
mean level 95
molar level
a
−96
121
1043
d
a
Levels in
µ
g/L; 2meC3: ethyl isobutanoate; 2meC4: ethyl 2-methylbutanoate; 3meC4: ethyl isopentanoate; C4: ethyl butanoate; C6: ethyl hexanoate; C8: ethyl
octanoate; C10: ethyl decanoate. b See experimental part. Natural aging at 12
c
°
C in years before analysis. d Molar level in nmol/L.
2
2
also mentioned previously to explain the formation of pheny-
lacetic acid in Spanish red wines (22). Technological parameters,
such as high temperature and pH, lack of water, and dissolved
oxygen and transition-metal contents, play an important role in
the rate of formation of the Strecker aldehydes and acids, and
consequently on aroma compounds formation (23, 24). Condi-
tions encountered during storage and aging of wine (aqueous
medium, low temperature, and low pH) are not optimal for this
reaction. However, in the last several years, it has been
demonstrated that the Strecker reaction can occur under condi-
tions close to those of wine (25, 26). Finally, the branched fatty
acid ethyl esters could occur through the slow hydrolysis at the
wine pH of fatty acid glycosides, as fatty acids have been
previously reported as aglycons of wine glycosides (27).
The purpose of the present work was first to check the
particular behavior of the branched fatty acids and their ethyl
esters during wine aging and then to investigate the hypothetical
pathways suggested in the literature to explain these variations.
purity), deuterium oxide (99.9% isotopic purity), [ H
3
]-methyl [ H
3
5
]-
2
methacrylate (99% isotopic purity), and L-[5,5,5- H
]leucine (99%
isotopic purity) were purchased from Aldrich (Saint-Quentin Fallavier,
France).
[2
H
]-Ethyl 2-methylbutanoate, [ H
2
]-ethyl isopentanoate, [ H
]-ethyl octanoate, and [ H
decanoate were synthesized by esterification of the corresponding acids
2
]-ethyl
]-ethyl
5
5
5
2
2
2
butanoate, [ H
5
]-ethyl hexanoate, [ H
5
5
2
with [ H
6
]-ethanol as reported previously (10).
2
Synthesis of [ H
5
]-Ethyl Isobutanoate. Isobutanoic acid (632.7 mg,
]-ethanol (158.1 mg, 3.1 mmol) were heated at 80
C during 20 h using sulfuric acid (35 µL) as catalyst. The solution
was cooled to room temperature, and then diluted with 30 mL of
pentane, washed with two 5 mL portions of 1 N aqueous sodium
hydroxide and with 5 mL water, dried over anhydrous sodium sulfate,
and filtered. The solution of isobutanoic acid [ H ]-ethyl ester in pentane
5
was kept at -20 °C. Its purity was checked by GC/EIMS (70 eV), m/z
(%): 43 (100), 34 (47), 71 (38), 41 (28), 121 (16), 42 (16), 30 (11), 90
2
7
°
.2 mmol) and [ H
6
2
(9), 39 (9), 74 (6).
2
Synthesis of [ H
7
]-Isobutanoic Acid (Figure 1). A solution of
isocyanomethyl p-tolyl sulfone (1035 mg, 5.3 mmol) in THF (2.7 mL)
was added dropwise to a stirred suspension of potassium tert-butoxide
MATERIALS AND METHODS
(
315 mg, 2.8 mmol) in THF (5.4 mL) at 7-10 °C. After cooling of the
Wines. Muscadet wines of the 1995, 1996, 1999, 2000 and 2002
vintages were produced using Vitilevure KD yeast (Martin-Vialatte,
Epernay, France) at pilot scale (35 L) according to standard wine-
making procedures by ITV France (Institut Technique du Vin, Unite´
de Nantes, France) from Melon B. grapes harvested in different
locations of the Muscadet vineyard: La Limouzini e` re (LIM), Monni e` res
2
sample to -10 °C, a solution of [ H
6
]-acetone (212 µL, 2.8 mmol) in
2
.7 mL of THF was added, and the resultant mixture was stirred for a
further 15 min. Glacial acetic acid (150 mg, 2.5 mmol) was then added,
and the solvent was removed by evaporation at 25 °C under vacuum.
The residue was washed with 11.3 mL of water and extracted with
5
6.3 mL of methylene chloride. The organic phase was dried over
(MON), Drain (DRA), and Le Loroux Bottereaux (LOR). After
sodium sulfate anhydrous, filtrated, and concentrated until dryness at
vinification, the wines were bottled with free sulfur dioxide levels
ranging from 30 to 43 mg/L and stored at 12 °C. The Muscadet wines
of the 1995, 1996, 1999 vintages (see pH in Table 1) were analyzed
in 2000 to determine their levels in ethyl esters of branched- and
straight-chain fatty acids. The 2000 Muscadet wine (40 mg/L of free
2
2
6
5 °C under vacuum to yield 787 mg (76%) of [ H ]-5,5-dimethyl-4-
1
p-tolylsulfonyl-2-oxazoline. Its H NMR spectrum was consistent with
that of the natural compound, reported previously (28).
2
To obtain the [ H
6
]-1-formylamino-1-p-tolylsulfonyl-2-methyl-1-
propene, a solution of the oxazoline obtained in the first step (253 mg,
1 mmol) in THF (1 mL) was added dropwise to a stirred suspension
of potassium tert-butoxide (120 mg, 1 mmol) in THF (2 mL) at -10
SO
with [ H
mg/L of free SO
25 mg/L of free SO
2
when bottled; pH 3.42) was used in the model aging experiment
2
3
]-leucine performed in 2001. The 2002 Muscadet wine (38
when bottled; pH 2.98) and a 2002 Sylvaner wine
when bottled; pH 3.38) were used in the model
2
°
C. The resultant mixture was stirred for 15 min. Glacial acetic acid
(
2
(60 mg, 1 mmol) was then added and the solvent was removed by
aging experiments carried out to investigate the equilibrium between
the branched fatty acids and their ethyl esters. The 2002 Sylvaner wine
was produced using the Levuline CHP yeast (Martin-Vialatte, Epernay,
France) according to standard wine-making procedures by ITV France
evaporation at 25 °C under vacuum. The residue was washed with 2.5
mL of water and extracted with 5.3 mL of methylene chloride. The
organic phase was dried over sodium sulfate anhydrous, filtrated, and
concentrated until dryness at 25 °C under vacuum to yield 126 mg
(Institut Technique du Vin, Unit e´ de Colmar, France).
2
(
49.8%) of [ H
6
]-1-formylamino-1-p-tolylsulfonyl-2-methyl-1-propene,
Chemicals. Sodium hydroxide, sodium sulfate, potassium hydroxide,
which was used without further purification in the next step.
acetic acid, and hydrochloric acid were all obtained from Merck
Darmstadt Germany). Diethyl ether, pentane, and dichloromethane
2
(
The hydrolysis of the crude [ H
6
]-1-formylamino-1-p-tolylsulfonyl-
were purchased from Riedel de H a¨ en (St Quentin Fallavier, France),
and ethanol (absolute) and methanol were from Carlo Erba (Rodano,
Italy). Water was purified with a Milli-Q system from Millipore S.A.
2-methyl-1-propene was carried out in 1 mL of a 2 N solution of
deuterated sulfuric acid in deuterium oxide. The solution was then
cooled, adjusted to pH 14 with aqueous sodium hydroxide 2 N, and
washed twice with diethyl ether. The aqueous solution was then adjusted
to pH 1 with sulfuric acid and extracted with diethyl ether. The organic
(Saint-Quentin Fallavier, France). The other chemicals were all
purchased from Sigma-Aldrich (St Quentin Fallavier, France).
2
2
Labeled Chemicals. [ H
6
]-Ethanol (99.5% isotopic purity), [ H
6
]-
phase was dried over anhydrous sodium sulfate and filtered. The
2
2
acetone (99.9% isotopic purity), [ H
2
]-sulfuric acid (99.5% isotopic
solution of [ H
7
]-isobutanoic acid in diethyl ether was kept at -20 °C.

Iso-Acid Ethyl Esters Formation during Wine Aging
J. Agric. Food Chem., Vol. 53, No. 9, 2005 3505
2
2
Figure 1. Synthesis of [ H
7
]-isobutanoic acid from [ H
6
]-acetone (28).
2
Table 2. Effect of Model Aging on the Levels (nmol/L) of Natural and [ H
5
]-Isobutanoic Acid and Their Ethyl Esters in a 2002 Muscadet Wine (pH
2
2
.98) Initially Spiked with [ H
5
]-Isobutanoic Acid and Its Ethyl Ester
before aging
after aging
mean levela
aging effect
128.4%
mean levela
%CVa
%CVa
level variationb
669.4 (p < 0.1%)
ester deficiencyc
96.2 (ns)
2
ethyl [ H
5
]-isobutanoate
521.5
5387.1
9.7%
3.0
6.2
1190.9
4621.5
25.8%
6.3
4.3
−
2
[
[
H
H
5
]-isobutanoic acid
]-ester/acid ratio
−765.6 (p < 5%)
−14.2%
2
5
ethyl isobutanoate
isobutanoic acid
ester/acid ratio
379.3
8883.0
4.3%
7.2
5.7
1031.0
7895.5
13.1%
1.2
3.0
651.7 (p < 0.1%)
987.5 (p < 5%)
171.8%
11.1%
−335.8 (ns)
−
−
a
)
3). b Difference between the levels after and before aging
Mean levels (nmol/L) and variation coefficient (%CV) of the acids and esters in the samples analyzed (n
c
(
nmol/L and %values) and level of significance of the Student test; ns
)
not significant. Difference between the ester levels (nmol/L) observed and expected (acid loss
totally esterified) after aging and level of significance of the Student test.
2
Table 3. Effect of Model Aging on the Levels (nmol/L) of Natural and [ H
5
]-Isobutanoic Acid and Their Ethyl Esters in a 2002 Sylvaner Wine (pH
2
3
.38) Initially Spiked with [ H
5
]-Isobutanoic Acid and Its Ethyl Ester
before aging
after aging
mean levela
aging effect
35.3%
mean levela
%CVa
%CVa
level variationb
381.0 (p < 1%)
ester deficiencyc
465.2 (p < 5%)
2
ethyl [ H
5
]-isobutanoate
1080.2
10862.4
9.9%
3.4
1.7
1461.2
10016.1
6.0
5.6
−
2
[
[
H
H
5
]-isobutanoic acid
]-ester/acid ratio
−846.2 (ns)
−7.8%
2
5
14.6%
ethyl isobutanoate
isobutanoic acid
ester/acid ratio
533.6
11609.1
4.6%
2.0
1.6
1050.0
11070.5
5.0
3.5
516.4 (p < 0.1%)
538.6 (ns)
96.8%
−4.6%
−22.3 (ns)
−
9.5%
a
)
3). b Difference between the levels after and before aging
Mean levels (nmol/L) and variation coefficient (%CV) of the acids and esters in the samples analyzed (n
c
(
nmol/L and %values) and level of significance of the Student test; ns
)
not significant. Difference between the ester levels (nmol/L) observed and expected (acid loss
totally esterified) after aging and level of significance of the Student test.
Its purity was checked by GC/EIMS (70 eV), m/z (%): 50 (100), 46
50), 30 (33), 77 (27), 42 (12), 45 (11), 49 (10), 48 (8), 95 (7), 33 (6).
was cooled to room temperature, and then diluted with 30 mL of
pentane, washed with two 5 mL portions of 1 N aqueous sodium
(
2
2
2
Synthesis of [ H
late (486 mg, 4.5 mmol) in 6 mL of methanol was hydrogenated over
0 mg of palladium on carbon (5%) in a two necked flask at room
5 3 5
]-Isobutanoic Acid. [ H ]-Methyl [ H ]-methacry-
hydroxide and with 5 mL water, dried over anhydrous sodium sulfate,
2
and filtered. The solution of [ H
5
]-isobutanoic acid ethyl ester in pentane
3
was kept at -20 °C. Its purity was checked by GC/EIMS (70 eV), m/z
(%): 46 (100), 47 (94), 48 (83), 45 (85), 44 (54), 75 (53), 76 (43), 74
(39), 43 (33), 73 (28).
temperature. When absorption of hydrogen ceased, the catalyst was
separated out through filtration, and methanol was eliminated by
concentration under vacuum at 40 °C from the filtrate. The residue
Model Aging of the Wines. A 2002 Muscadet wine and a 2002
Sylvaner wine (0.75 L of each wine) were spiked with the amounts of
(440 mg, 4 mmol) was subsequently heated at 60 °C for 1 h in a solution
2
2
of 448 mg (8 mmol) of potassium hydroxide in 35 mL of methanol.
Then, the solution was cooled to room temperature and concentrated
to dryness under vacuum, and the residue was added with water (8
mL). The aqueous solution was washed with pentane, and then it was
5 5
[ H ]-isobutanoic acid and ethyl [ H ]-isobutanoate mentioned under
“before aging” in Tables 2 and 3. Each wine was divided in two
portions of 0.6 and 0.15 L. The 0.6 L portions were aged in a laboratory
oven at 45 °C during 3 weeks and then analyzed as described below,
whereas the 0.15 L portions were analyzed at the starting time of the
aging.
adjusted to pH 1 with 1 N hydrochloric acid and extracted with diethyl
2
ether. The solution of [ H
5
]-isobutanoic acid in diethyl ether was dried
over anhydrous sodium sulfate, filtered, and kept at -20 °C. Its purity
was checked by GC/EIMS (70 eV), m/z (%): 45 (100), 46 (77), 47
In addition, 0.5 L of a 2000 Muscadet wine were spiked with 40
2
3
mg of L-[5,5,5- H ]leucine, aged as above, in a laboratory oven at 45
(
78), 48 (72), 44 (64), 49 (51), 43 (38), 75 (37), 76 (33), 42 (26).
°C during 3 weeks, and then analyzed as described below.
2
2
Synthesis of Ethyl [ H
111.6 mg, 1.2 mmol) and ethanol (55.2 mg, 1.2 mmol) were heated at
0 °C during 3 h using sulfuric acid (7 µL) as catalyst. The solution
5 5
]-Isobutanoate Acid. [ H ]-Isobutanoic
Isolation of Volatiles from Wines. General Procedure. The wine
was placed in a flask, and then the internal standards were added and
equilibrated under stirring. Then the samples were extracted with
(
8

3506 J. Agric. Food Chem., Vol. 53, No. 9, 2005
D ´ı az-maroto et al.
methylene chloride by stirring for 20 min under nitrogen. The emulsion
was resolved by centrifugation at 8000 rpm and 4 °C for 20 min. The
organic phases were blended, desiccated over anhydrous sodium sulfate,
filtered, and concentrated on a Vigreux column. Finally, the concentrates
were stored at -20 °C until analysis.
[2H
5
]-ethyl ester). The standard curves were obtained by linear
regression analysis.
Wines of the Aging Experiment with Deuterated Branched Fatty
Acids and Ethyl Esters. Eight solutions of both natural compounds,
ethyl isobutanoate and isobutanoic acid, and eight solutions of both
spiked [ H ]-analogues, ethyl [ H ]-isobutanoate and [ H ]-isobutanoic
5 5 5
acid, were made in methylene chloride with the following concentration
2
2
2
QuantitatiVe Determination of Fatty Acid Ethyl Esters of Muscadet
Wines of Different Vintages. The procedure used was reported previ-
2
ously to analyze these esters in Merlot wine (10). 13 µg of [ H
5
]-ethyl
ranges: 0.02-20.83 µg/mL of ethyl isobutanoate, 0.06-60.61 µg/mL
2
2
2
isobutanoate, 4.4 µg of [ H
ethyl isopentanoate, 7.5 µg of [ H
ethyl hexanoate, 33.9 µg of [ H
5
]-ethyl 2-methylbutanoate, 5.5 µg of [ H
5
]-
]-
of isobutanoic acid, 0.12-59.74 µg/mL of ethyl [ H
5
]-isobutanoate,
2
2
2
]-ethyl butanoate, 92.6 µg of [ H
]-ethyl octanoate, and 22.4 µg of [ H
5
5
and 0.77-48.20 µg/mL of [ H ]-isobutanoic acid, respectively. Then
5
2
2
2
2
5
]-
1.35 µg/mL of [ H
5
7
]-ethyl isobutanoate and 3.00 µg/mL of [ H ]-
5
ethyl decanoate were added as internal standards to 250 mL of wine,
which was extracted with 2 × 60 mL of methylene chloride. The final
volume of the extract was about 4 mL.
isobutanoic acid were added to each solution as internal standards. The
calibration curves were obtained from these solutions injected into the
GC-MS system in the same chromatographic conditions as the wine
extracts. For each compound, the peak area ratios (peak area of the
ion m/z 71/peak area of the ion m/z 71; peak area of the ion m/z 73/
peak area of the ion m/z 77; peak area of the ion m/z 76/peak area of
the ion m/z 71; and peak area of the ion m/z 47/peak area of the ion
m/z 50) were plotted against the concentration ratios (eight-point scale
Wines of the Aging Experiment with Deuterated Branched Fatty
2
Acids and Ethyl Esters. A total of 24.6 µg of [ H
7
]-isobutanoic acid
2
and 50.4 µg of [ H
5
]-ethyl isobutanoate were added as internal standards
to 50 mL of the different portions of wine, which were extracted with
2
× 15 mL of methylene chloride. The final volume of the extract was
mentioned above: micrograms/milliliter of ethyl isobutanoate/1.35 µg/
about 1 mL. Each sample was analyzed in triplicate.
2
mL of [ H
5
]-ethyl isobutanoate; micrograms/milliliter of isobutanoic
Wines of the Aging Experiment with Deuterated Leucine. 500 mL
of wine were extracted with 2 × 100 mL of methylene chloride. The
final volume of the extract was about 1 mL.
2
7
acid/3.00 µg/mL of [ H ]-isobutanoic acid; micrograms/milliliter of ethyl
]-isobutanoate/1.35 µg/mL of [ H ]-ethyl isobutanoate and micro-
5
grams/milliliter of [ H
2
2
[
H
5
2
2
5
7
]-isobutanoic acid/3.00 µg/mL of [ H ]-isobu-
Isolation of Glycosides from Wines and Enzymatic Hydrolysis
of the Glycosidic Extracts. The glycosides of 50 mL of a 2002
Muscadet and of a 2002 Sylvaner wines, diluted twice with Milli-Q
water (Millipore Corp.), were extracted using C18 reversed phase
cartridges, and the extracts were hydrolyzed using a hydrolase enzyme
AR2000 (DSM, France), as reported previously for Melon B. juice (29).
The aglycone extracts were kept at -20 °C until analyzed by GC/MS.
tanoic acid, respectively). The standard curves were obtained by linear
regression analysis.
RESULTS AND DISCUSSION
Levels of Fatty Acid Ethyl Esters during Aging of
Muscadet Wines. To evaluate their variations during aging,
the levels of the ethyl esters of straight-chain and branched fatty
acids were determined in 1-5 year aged (vintages 1999, 1996,
and 1995) experimental Muscadet wines (Table 1). These
different wines were made, using the same wine-making
procedure, from white Melon B. grapes grown in different
locations of the Muscadet vineyard (northwest France). The
quantitative determination of the straight-chain fatty acid ethyl
esters is easy (relatively high levels in wine), but that of the
branched esters is more difficult, due to their lower levels and
volatility. Thus, these levels were determined using the stable
isotope dilution assays previously reported for their quantifica-
tion in Merlot wines (10), but dichloromethane was used as the
Gas Chromatography-Mass Spectrometry. A Hewlett-Packard
890 series II gas chromatograph fitted with a DB-WAX capillary
5
column (30 m × 0.25 mm, 0.5 µm, J&W Scientific Inc., USA) was
used. The “on column” injector was heated from 30 to 250 °C at 180
°
C/min. Helium (Linde gaz, Marseille, France) at a flow rate of 1.3
mL/min was used as the carrier gas. The oven temperature program
was as follows: 25 °C, then increased at 70 °C/min to 50 °C (held for
3
min), then increased at 3 °C/min to 125 °C, then increased at 10
°
C/min to 245 °C (held for 20 min). The GC instrument was coupled
to a Hewlett-Packard 5989A mass spectrometer and a MS chemstation.
The transfer line was heated at 250 °C. The electron impact (EI) energy
was 70 eV, and the MS source and quadrupole temperatures were set
at 250 and 120 °C, respectively. EIMS spectra were recorded in full
scan mode in the range of 29 to 350 amu at 0.5 s intervals.
2
solvent of extraction instead of pentane. [ H5]-ethyl esters of
Calibration Curves. QuantitatiVe Determination of Fatty Acid Ethyl
Esters of Muscadet Wines of Different Vintages. Serial dilutions of each
the fatty acids, obtained by esterification of the corresponding
2
acids with [ H5]-ethanol, were used as internal standards (10).
fatty acid ethyl ester, added with the same amount of the corresponding
2
[
5
H ]-ethyl ester as internal standard, were made in methylene chloride
The differences observed in these Muscadet wines of different
vintages were consistent with those reported previously in the
literature during wine aging (see introduction): the levels of
the branched esters increased with age, whereas those of the
straight-chain esters decreased. They were also consistent with
the variations observed during the model aging of Merlot wines,
which consisted in controlled heating, at 45 °C for 24 days (10),
and was used previously as model aging for acid-catalyzed
reactions occurring during wine aging (30). Thus, we chose the
same model aging conditions, in the experiments undertaken
afterward to elucidate the pathways explaining the variations
of the branched esters during wine aging.
Formation of the Ethyl Esters of Branched Short-Chain
Fatty Acids from Their Corresponding Acids in Model
Aging Experiments. To study the equilibrium between the ethyl
esters of branched short-chain fatty acids and their corresponding
acids in wine, two one-year-old wines, a 2002 Muscadet and a
with the following concentration ranges: 0.025-5.1 µg/mL of ethyl
2
isobutanoate (1.3 µg/mL of [ H
mL of ethyl 2-methylbutanoate (0.44 µg/mL of [ H
ylbutanoate), 0.031-1.0 µg/mL of ethyl isopentanoate (0.55 µg/mL of
5
]-ethyl isobutanoate), 0.042-2.68 µg/
2
5
]-ethyl 2-meth-
2
[
5
H ]-ethyl isopentanoate), 0.14-4.31 µg/mL of ethyl butanoate (0.75
2
µg/mL of [ H
9.26 µg/mL of [ H
octanoate (3.39 µg/mL of [ H
mL of ethyl decanoate (2.24 µg/mL of [ H
5
]-ethyl butanoate), 0.73-23.5 µg/mL of ethyl hexanoate
2
(
5
]-ethyl hexanoate), 0.08-24.0 µg/mL of ethyl
2
5
]-ethyl octanoate), and 0.20-19.6 µg/
2
5
]-ethyl decanoate),
respectively. The calibration curves were obtained from these solutions
injected into the GC-MS system in the same chromatographic conditions
as the wine extracts. The ions used as quantifiers for the natural ethyl
esters were m/z 116 for ethyl isobutanoate, m/z 102 for ethyl
2
-methylbutanoate, and m/z 88 for the other esters. The ions used as
2
2
quantifiers for the [ H
isobutanoate, m/z 107 for [ H
5
]-ethyl esters were m/z 121 for [ H
5
]-ethyl
2
5
]-ethyl 2-methylbutanoate, and m/z 93
for the other internal standards. For each compound and for the
concentrations mentioned above, the peak area ratios (peak area of the
ion used as quantifier ion for the natural ester/peak area of the ion
2
002 Sylvaner wines, were spiked with a deuterated branched
2
fatty acid and its ethyl ester, and submitted to the same model
aging as above. Then, the levels of the spiked deuterated
compounds were compared to those of the corresponding natural
5
used as quantifier ion for the corresponding [ H ]-ethyl ester) were
plotted against the concentration ratios (micrograms/milliliter of each
dilution of each natural ester/micrograms/milliliter of the corresponding

Iso-Acid Ethyl Esters Formation during Wine Aging
J. Agric. Food Chem., Vol. 53, No. 9, 2005 3507
compounds, before and after aging. Isobutanoic acid and its ethyl
ester were chosen, as this ester was the most abundant of the
branched fatty acid ethyl esters quantified in the Muscadet wines
analyzed above (Table 1). To carry out these experiments, four
deuterium labeled derivatives were necessary and were synthe-
used were their relative ion abundance and their least interfer-
ence from each other and from other wine volatile components.
Analysis of the Target Compounds in the Model Aging
Experiments. Tables 2 and 3 show the levels of the natural and
labeled ethyl isobutanoate and their corresponding acids in the
spiked Muscadet and Sylvaner wines, before and after aging at
45 °C for 3 weeks. In both wines, the natural acid levels were
much higher than the corresponding ester levels in the initial
young wine, leading to esterification molar ratios lower than
5%. As these ratios were expected to be higher when the acid-
ethyl ester equilibrium was brought about, assuming values
similar to those of straight-chain fatty acids (18, 32), those
chosen for the spiked deuterated compounds were doubled
2
2
sized. [ H5]-Isobutanoic acid and ethyl [ H5]-isobutanoate were
2
2
used to spike the wines, and [ H7]-isobutanoic acid and [ H5]-
ethyl isobutanoate were used as internal standards to determine
the exact amounts of the target compounds by means of stable
isotope dilution assays.
2
Synthesis of the Deuterated Compounds. [ H7]-Isobutanoic
2
acid was synthesized by conversion of [ H6]-acetone into the
next higher carboxylic acid (Figure 1), using the procedure
reported previously for the unlabeled compound (28). However,
as the first step of the procedure reported did not convert totally
the deuterated acetone to the propene adduct, it was slightly
(∼10%). Despite this doubling, the model aging brought about
significant increases (p < 1% and p < 0.1%) in the molar
concentrations of the deuterated ethyl isobutanoate, which were
also observed for the natural ester, as expected according to
the acid-ethyl ester equilibrium model. Simultaneously, de-
creases in the molar concentrations of their corresponding acids
were observed, but they were significant (p < 5%) in the
Muscadet wine only. As a matter of fact, the relative variations
in the acid molar levels (ranging from -4.6 to -14.2%) were
much lower than that of the ethyl esters (ranging from +35.3
to +171.8%), due the much higher acid levels (about 10-20
times). Thus, despite similar fair coefficients of variation for
the quantitative determination of the acids (1.6-6.2%) and ethyl
esters (1.2-7.2%), the level of significance of the acid decreases
in both wines (ns or p < 5%) was much lower than that of the
increases in the molar concentrations of the corresponding ethyl
esters (p < 1% or p < 0.1%). Furthermore, assuming the acid-
ethyl ester equilibrium model, the absolute variations of the
molar levels of each acid should be equal to that of the
corresponding ethyl ester, but no ester in both wines reached
these theoretical values. That was particularly evident in the
cases of the deuterated acid-ester pair in Sylvaner wine (381
nmol vs 846.2 nmol) and of the natural acid-ester pair in
Muscadet wine (651.7 nmol vs 987.5 nmol). These deficiencies
in ethyl ester levels after model aging could be partly explained
by the consumption of the acids by other constituents of wine
than ethanol. However, that was not consistent with the high
content in ethanol, by far the most abundant wine constituent
reacting with the acids. In addition, the deficiencies observed
for the natural and the labeled esters in both wines were different
and in different value orders. Thus, these discrepancies could
be attributed mainly to the difference in the magnitude orders
of the ester and acid levels, as discussed above. That explained
why these ester deficiencies were not statistically significant
2
modified to isolate the [ H6]-5,5-dimethyl-4-p-tolylsulfonyl-2-
2
oxazoline intermediate, as shown in Figure 1. Thus, [ H6]-
2
acetone was first converted into [ H6]-5,5-dimethyl-4-p-
tolylsulfonyl-2-oxazoline by reaction with R-metalated isocyano-
methyl p-tolyl sulfone in strong basic conditions. The isolated
2
oxazoline was then converted into [ H6]-1-formylamino-1-tosyl-
2
-methyl-1-propene in the same basic conditions. Finally, the
hydrolysis of this adduct with a solution of deuterated sulfuric
2
acid in deuterium oxide gave rise to [ H7]-isobutanoic acid,
obtained as a pure isotopomer, as shown by GC-MS (molecular
ion m/z 95).
[2
H5]-Isobutanoic acid was obtained by hydrogenation of the
2
2
double bond of commercial [ H3]-methyl [ H5]-methacrylate,
followed by saponification of the deuterated methyl isobutyrate
obtained. However, alkenes and even alkanes are well-known
to exchange hydrogens during heterogeneous catalytic hydro-
genation (31). Thus, as shown by GC-MS analysis, the partial
exchange of deuterium by hydrogen atoms gave rise to a mixture
2
of deuterated isobutanoic acid isotopomers, instead of [ H5]-
isobutanoic acid, which was named, by convenience, [ H5]-
2
isobutanoic acid throughout this text. As it contained only a
trace of the natural isotopomer (molecular ion m/z 88 was less
than 1% of the molecular ion cluster), it was used, with its ethyl
ester, to spike the wines in the model aging experiment. Finally,
2
2
[
H5]-isobutanoic acid ethyl ester and isobutanoic acid [ H5]-
ethyl ester were obtained by acid-catalyzed esterification of the
2
corresponding acids with ethanol and [ H5]-ethanol, respectively.
The former was a mixture of isotopomers similar to the
corresponding acid, whereas the latter was a pure isotopomer
(molecular ion m/z 121).
SIDA Assays Using GC-MS. The quantification of the natural
(except for the deuterated ester in the Sylvaner wine), not
isobutanoic acid and its ethyl ester on one hand, and that of the
allowing rejection of the acid-ethyl ester equilibrium model.
Finally, assuming this model with first-order esterification
and hydrolysis reactions and constant water and ethanol
concentrations, the rate of the formation of the ethyl esters could
be written:
2
deuterated analogues spiked in the wine, [ H5]-isobutanoic acid
2
and its ethyl ester on the other hand, were achieved using [ H7]-
2
isobutanoic acid and isobutanoic acid [ H5]-ethyl ester as internal
standards. Extraction of these compounds from wine samples
with methylene chloride and analysis of the extracts by GC/
EIMS, used in full scan mode, gave sufficiently sensitive
detection. In the GC conditions used, the three target esters
de/dt ) k (a + e - e)[EtOH] - k e[H O]
1
0
0
2
2
2
appeared as two resolved peaks. [ H5]-Isobutanoic acid ethyl
2
which gave at the equilibrium (de/dt ) da/dt ) 0):
k (a + e - e )[EtOH] ) k e [H O]
ester and isobutanoic acid [ H5]-ethyl ester were coeluted in
the first eluted one and the natural ester was in the second eluted
peak. The three target acids gave a first eluted peak with a
shoulder, corresponding to the deuterated acids, and a second
eluted peak, separated from the first one, corresponding to the
natural acid. Reconstructed ion chromatograms (RIC) were used
for the quantification of the different isotopomers of isobutanoic
acid and its ethyl ester, and the requirements to choose the ions
1
0
0
E
2
E
2
where k1 and k2 are rate constants of the esterification and
hydrolysis reactions, respectively, e, e0, and eE are the concen-
trations of the ester at t time, starting time and equilibrium,
respectively, a and a0 are the concentrations of the corresponding

3508 J. Agric. Food Chem., Vol. 53, No. 9, 2005
D ´ı az-maroto et al.
acid at t time and starting time, respectively, and [EtOH] and
To verify this hypothesis, the same model aging as above
was performed on a Muscadet wine added with 40 mg/L of
[H2O] are the concentrations of ethanol and water, respectively.
2
2
Assuming that the isotope effect was negligible (k1 and k2
L-[5,5,5- H3]leucine, which would give rise to [4,4,4- H3]-
isopentanoic acid by the Strecker degradation, and subsequently
to its ethyl ester. The amount of deuterated L-leucine added to
the wine was similar to the highest levels of natural L-leucine
found previously in white wines, ranging from 21 to 43 mg/L
(33). GC-MS detection of the wine volatiles, after the model
aging, was performed with the method described above to detect
the different isotopomers of isobutanoic acid and its ethyl ester.
It revealed only natural isopentanoic acid and its ethyl ester,
but not their trideuterated isotopomers, which would be expected
in high amounts (theoretical amounts of 30.6 and 38.8 mg/L).
Thus, in our conditions of medium and temperature, the
excessive levels of deuterated leucine here added allowed us to
conclude that the levels of isopentanoic acid ethyl ester produced
during Muscadet wine aging cannot be explained by the Strecker
degradation of leucine, but the formation of branched fatty acids
from some amino acid derivatives formed by yeast, e.g.,
R-ketoacids, cannot be excluded.
Formation of Branched Short-Chain Fatty Acid from
Grape Glycoconjugates. One of the sources of branched fatty
acid ethyl esters during wine aging could be the release of
isobutanoic acid through the slow hydrolysis at the wine pH of
its glycosilated forms (27) and subsequently the esterification
in its corresponding ethyl ester. To elucidate the importance of
this route, the glycosides of a 2002 Muscadet and of a 2002
Sylvaner wines were extracted and the extracts were hydrolyzed
using an appropriate glycosidase enzymatic preparation, as
reported previously (29). The aglycones released were semi-
quantified as 4-nonanol equivalents, used as an internal standard.
That showed that the Sylvaner wine contained a low level of
isobutanoic acid aglycone (4.5 µg/L of 4-nonanol equivalent),
whereas it was not even detected in the Muscadet wine extract,
which was consistent with the low levels of acid aglycons
reported previously in Melon B. grapes (29). Thus, this route
could not explain the much higher amounts of ethyl esters of
branched fatty acids produced during wine aging.
equal for the natural and deuterated compounds), the same
equation could be written for the natural and deuterated
analogues, which led to the following equation:
H
E
D
E
H
D
e /e ) (a + e ) /(a + e )
0
0
0
0
where the H and D superscripts referred to the natural and
deuterated compounds, respectively.
H
E
D
E
Hence, the ratio e /e should be equal to 1.57 in the
Muscadet wine and 1.02 in the Sylvaner wine. These values
H
D
were compared to those of the ratio e /e at the starting time
0.73 and 0.49) and at the end of the model aging time (0.87
(
and 0.72, respectively) in the corresponding wines, which
showed that the equilibrium was not reached at the end of the
model aging time. At this time, the esterification molar ratios
were 13.1 and 25.8% for the natural and deuterated isobutanoic
acid in the Muscadet wine, and 9.5 and 14.6% in the Sylvaner
wine, respectively. The ester levels reached were lower than
the levels of ethyl isobutanoate measured in the 1996 and 1995
Muscadet wines (natural aging of 4 and 5 years, respectively)
(Table 1), but no comparison with the esterification ratios of
these wines could be made, as the levels of isobutanoic acid
were not determined in these wines. However, these esterifi-
cation molar ratios were consistent with those measured at the
equilibrium in Port wine (20% ethanol content V:V), for the
straight-chain C-4, C-6, C-8, and C-10 fatty acid-ethyl ester
couples, which ranged from 28 to 40% (32), that is, from ∼18
to 26% for the ethanol content of the wines used in our
experiments (∼13% V:V).
Whatever these values are, as the labeled acid and ester were
the only sources of labeled compounds in the wines used in the
model experiments, the increase of their esterification ratios
could be reasonably explained only under the assumption that
the acid-ester equilibrium was the major process involved in
the generation of branched ethyl esters during wine aging,
although it was not possible to exclude the concomitant
occurrence of other chemical processes shifting this equilibrium.
Of course, the acid-ester equilibrium setting up would have
afforded further insight into the balance between the branched
acids and their corresponding ethyl esters. However, even if
the 3-week aging period was shorter to reach the equilibrium,
the model (wine medium, 45 °C) allowed promotion of
variations of the levels of the branched acid and its ethyl ester
in the same direction as those observed under natural aging
CONCLUSION
The results of the model experiments performed with addition
of labeled compounds to wine have shown that the acid-ester
equilibrium was the most effective in generating the branched
fatty acid ethyl esters from their corresponding acids during
wine aging. Their esterification molar ratios, very low in young
wine, increased significantly with aging, but this upward trend
did not reach the equilibrium at the end of the model aging
experiment. On the other hand, the formation of branched short-
chain fatty acids by the Strecker degradation of their corre-
sponding amino acids or from grape glycoconjugates could not
explain the much higher amounts of their corresponding ethyl
esters produced during wine aging.
(Table 1). In contrast, harder conditions for model aging, such
as an increase of temperature above 45 °C, could favor reactions
requiring higher energy, as Strecker degradation. Thus, this
reaction, which could also occur during wine aging, was tested
using the same model aging conditions.
Formation of Branched Short-Chain Fatty Acid by
Strecker Degradation in Model Aging Experiment. The
increase of ethyl esters of branched short-chain fatty acids during
the model aging of the Muscadet and Sylvaner wines could be
explained by the formation of these fatty acids from the
corresponding amino acids by Strecker degradation. Indeed, the
excess of acid formed in this process should shift the acid-
ester equilibrium toward the ester formation. Similar hypothesis
were also mentioned previously to explain the formation of
phenylacetic acid in Spanish red wines (22) and that of ethyl
esters of branched short-chain fatty acids during the model aging
of Merlot wines (10).
ACKNOWLEDGMENT
M.C. D ´ı az-maroto gratefully acknowledges the postdoctoral
fellowship provided by the Autonomous Government of Castilla-
La Mancha (Spain).
LITERATURE CITED
(
1) Eti e´ vant, P. X. Wine. In Volatile Compounds in Food and
BeVerages; Maarse, H. Ed.; Marcel Dekker: New York, USA,
1991; pp 483-546.

Iso-Acid Ethyl Esters Formation during Wine Aging
J. Agric. Food Chem., Vol. 53, No. 9, 2005 3509
(
(
(
(
2) Nordstrom, K. Studies on the formation of volatile esters in
(20) Cort e´ s, M.; Moreno, J.; Zea, L.; Moyano, L.; Medina, M.
Changes in aroma composition of sherry wines during their
biological aging carried out by Saccharomyces cereVisiae races
bayanus and capensis. J. Agric. Food Chem. 1988, 46, 2394-
2398.
(21) Schieberle, P.; Komarek, D. Changes in key aroma compounds
during natural beer aging. In Freshness and Shelf Life of Foods;
ACS Symposium series 836, American Chemical Society:
Washington, 2003; pp 70-79.
(22) Ferreira, V.; Aznar, M.; Lopez, R.; Cacho, J. F. Quantitative
GC-olfactometry carried out at different dilutions of an extract.
Key differences in the odor profiles of four high quality Spanish
aged red wines. J. Agric. Food Chem. 2001, 49, 4818-4824.
(23) Hofmann, T.; M u¨ nch, P.; Schieberle, P. Quantitative model
studies on the formation of aroma-active aldehydes and acids
by Strecker-type reactions. J. Agric. Food Chem. 2000, 48, 434-
440.
fermentation with brewer’s yeast. SVen. Kem. Tidskr. 1964, 76,
5
10-543.
3) Bertrand, A. Use of gas chromatography to analyze wine volatile
constituents. Thesis, Universit e´ Victor Segalen Bordeaux II,
1
968.
4) Ribereau-Gayon, P. Wine aroma. In FlaVour of Foods and
BeVerages; Charalambous, G., Inglett, G. E. Eds.; Academic:
New York, 1978; pp 362-371.
5) Van Der Merwe, C. A.; Van Wyk, C. J. The contribution of
some fermentation products to the odor of dry white wines. Am.
J. Enol. Vitic. 1981, 32, 41-46.
(6) Dubois, P. Wine aromas and off-flavors. ReV. Fr. Oenol. 1994,
1
45, 27-40.
(7) Guth, H. Identification of character impact odorants of different
white wine varieties. J. Agric. Food Chem. 1997, 45, 3022-
3
026.
(
8) Guth, H. Quantitation and sensory studies of character impact
(24) Hofmann, T.; Schieberle, P. Formation of aroma-active Strecker-
aldehydes by a direct oxidative degradation of Amadori com-
pounds. J. Agric. Food Chem. 2000, 48, 4301-4305.
(25) Ferreira, V.; Escudero, A.; Cacho, J. F. Flavours and aromas
formed during oxidative storage of wines: clues about their
formation and some technological consequences. Ital. Food
BeVerage Technol. 2000, 22, 19-24.
(26) Pripis-Nicolau, L.; de Revel, G.; Bertrand, A.; Maujean, A.
Formation of flavor components by the reaction of amino acid
and carbonyl compounds in mild conditions. J. Agric. Food
Chem. 2000, 48, 3761-3766.
(27) Cabaroglu, T.; Canbas, R.; Baumes, R.; Bayonove, C.; Lepoutre,
J. P.; G u¨ nata, Z. Aroma composition of a white wine of Vitis
Vinifera L. cv. Emir as affected by skin contact. J. Food Sci.
1997, 62, 680-683.
(28) Sch o¨ llkopf, U.; Schr o¨ der, R.; Blume, E. N-(1-Arylsulfonyl-1-
alkenyl)-formamide aus R-metallierten Arylsulfonyl-methylliso-
cyaniden und Carbonylverbindungen, Umwandlung einer Car-
bonylverbindung in die n a¨ chsth o¨ here Carbons a¨ ure. Liebigs Ann.
Chem. 1972, 766, 130-141.
(29) Schneider, R.; Charrier, F.; Moutounet, M.; Baumes, R. Rapid
analysis of grape aroma glycoconjugates using Fourier transform
infrared spectrometry and chemometric techniques. Anal. Chim.
Acta 2004, 513, 91-96.
(30) Francis, I. L.; Sefton, M. A.; Williams, P. J. The sensory effects
of pre- or post-fermentation thermal processing on Chardonnay
and Semillon wines. Am. J. Enol. Vitic. 1994, 45, 243-251.
(31) March, J. Hydrogenation of double and triple bonds. In AdVanced
Organic Chemistry, 3rd ed.; John Wiley & Sons Eds., New York,
USA, 1985; pp 691-699.
odorants of different white wine varieties. J. Agric. Food Chem.
1
997, 45, 3027-3032.
(
9) Ferreira, V.; Lopez, R.; Escudera, A.; Cacho, J. F. The aroma
of Grenache red wine: hierarchy and nature of its main odorants.
J. Sci. Food Agric. 1998, 77, 259-267.
(
10) Kots e´ ridis, K. Study on aroma of Merlot and Cabernet Sauvignon
wines from Bordeaux region. Thesis, Universit e´ Victor Segalen
Bordeaux II, 1999.
11) Aznar, M.; L o´ pez, R.; Cacho, J. F.; Ferreira, V. Identification
and quantification of impact odorants of aged red wines from
Rioja. GC-olfactometry, quantitative GC-MS, and odor evalu-
ation of HPLC fractions. J. Agric. Food Chem. 2001, 49, 2924-
(
2
929.
(
12) Marais, J. The effect of pH on esters and quality of Colombar
wine during maturation. Vitis 1978, 17, 396-403.
(
13) Marais, J.; Pool, H. J. Effect of storage time and temperature on
the volatile composition and quality of dry white table wines.
Vitis 1980, 19, 151-154.
(
(
(
14) Ramey, D.; Ough, C. S. Volatile ester hydrolysis or formation
during storage of model solutions of wines. J. Agric. Food Chem.
1
980, 28, 928-934.
15) Shinohara, T.; Watanabe, M. Effects of fermentation conditions
and ageing temperature on volatile ester contents in wine. Agric.
Biol. Chem. 1981, 45, 2645-2651.
16) Marais, J. Effect of storage time and temperature on the volatile
composition and quality of South African Vitis Vinifera L. cv.
Colombar wines. In The Shelf Life of Foods and BeVerages;
Charalambous, G. Ed.; Elsevier: Amsterdam, 1986; pp 169-
1
85.
(
17) Garofolo, A.; Piracci, A. Evolution of fatty acid esters during
wine storage. Equilibrium constants and activation energies. Bull.
O. I. V. 1994, 757-758, 225-245.
(32) Silva Ferreira, A. C. Chemical and statistical approach to the
characterization of Porto wine aging. Sotolon as aroma impact
odorant. Thesis, Universit e´ Victor Segalen Bordeaux II, 1998.
(33) Drawert, F.; Leupold, G.; Lessing, V. GC determinations of
organic acids, neutral compounds (sugars), amino acids and
phenolic compounds in wines. Wein Wiss. 1978, 33, 54-70.
(
18) Bertrand, A.; Anocibar-Beloqui, A.; Guesdes de Pinho, P.;
Kots e´ ridis, Y. Varietal and fermentation aromas. In 20i e´ me
Congr e´ s Mondial de la Vigne et du Vin. O. I. V. Ed.; Punta del
Este, Uruguay, 1995; pp 63-94.
(
19) P e´ rez-Coello, M. S.; Gonz a´ lez-Vi n˜ as, M. A.; Garc ´ı a-Romero,
E.; D ´ı az-Maroto, M. C.; Cabezudo, M. D. Influence of storage
temperature on the volatile compounds of young white wines.
Food Control 2003, 14, 301-306.
Received for review November 5, 2004. Revised manuscript received
February 23, 2005. Accepted March 3, 2005.
JF048157O