Quantitative Analysis by GC-MS/MS of 18 Aroma Compounds Related to Oxidative Off-Flavor in Wines

Article abstract of DOI:10.1021/jf505803u

A quantitation method for 18 aroma compounds reported to contribute to "oxidative" flavor in wines was developed. The method allows quantitation of the (E)-2-alkenals ((E)-2-hexenal, (E)-2-heptenal, (E)-2-octenal, and (E)-2-nonenal), various Strecker aldehydes (methional, 2-phenylacetaldehyde, 3-methylbutanal, and 2-methylpropanal), aldehydes (furfural, 5-methylfurfural, hexanal, and benzaldehyde), furans (sotolon, furaneol, and homofuraneol), as well as alcohols (methionol, eugenol, and maltol) in the same analysis. The aldehydes were determined after derivatization directly in the wine with O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydrochloride; the formed oximes along with the underivatized aroma compounds were isolated by solid-phase extraction and analyzed by means of GC-MS/MS. The method was used to investigate the effect of different closures (synthetic closures, natural corks, and screw cap) on the formation of oxidation-related compounds in 14 year old white wine. Results showed a significant increase in the concentration of some of the monitored compounds in the wine, particularly methional, 2-phenylacetaldehyde, and 3-methylbutanal.

Full text of DOI:10.1021/jf505803u

Article
pubs.acs.org/JAFC
Quantitative Analysis by GC-MS/MS of 18 Aroma Compounds
Related to Oxidative Off-Flavor in Wines
Christine M. Mayr,†,‡ Dimitra L. Capone,*,† Kevin H. Pardon,† Cory A. Black,† Damian Pomeroy,§
and I. Leigh Francis†
†The Australian Wine Research Institute, P.O. Box 197, Glen Osmond, South Australia 5064, Australia
§Agilent Technologies Australia Proprietary Limited Company Limited, 679 Springvale Road, Mulgrave, Victoria 3170, Australia
ABSTRACT: A quantitation method for 18 aroma compounds reported to contribute to “oxidative” flavor in wines was
developed. The method allows quantitation of the (E)-2-alkenals ((E)-2-hexenal, (E)-2-heptenal, (E)-2-octenal, and (E)-2-
nonenal), various Strecker aldehydes (methional, 2-phenylacetaldehyde, 3-methylbutanal, and 2-methylpropanal), aldehydes
(furfural, 5-methylfurfural, hexanal, and benzaldehyde), furans (sotolon, furaneol, and homofuraneol), as well as alcohols
(methionol, eugenol, and maltol) in the same analysis. The aldehydes were determined after derivatization directly in the wine
with O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydrochloride; the formed oximes along with the underivatized aroma
compounds were isolated by solid-phase extraction and analyzed by means of GC-MS/MS. The method was used to investigate
the effect of different closures (synthetic closures, natural corks, and screw cap) on the formation of oxidation-related compounds
in 14 year old white wine. Results showed a significant increase in the concentration of some of the monitored compounds in the
wine, particularly methional, 2-phenylacetaldehyde, and 3-methylbutanal.
KEYWORDS: wine, aroma, oxidation, aging, closures, PFBHA, aldehydes
table wine had higher concentrations of (E)-2-alkenals, whereas
Sherry contained large amounts of branched chain aldehydes.9
Beside those aldehydes, sotolon has been reported as a key
aroma compound of aged Sherry and Port wines.10−12 In other
studies, the presence of compounds such as furaneol and
homofuraneol in high concentrations have been shown to
increase the “caramel” character of red wines.6,13 Table 1
summarizes important aroma compounds associated with
“oxidized” aroma, along with their odor descriptors and odor
thresholds as determined in either model wine or water.
As shown in Table 1, all (E)-2-alkenals, with their “green”,
“fatty”, and “nutty” odor descriptors, have very low odor
thresholds, with (E)-2-nonenal as low as 0.17 μg/L. Similarly,
the Strecker aldehydes, in particular, methional and 2-
phenylacetaldehyde, as well as 3-methylbutanal and 2-
methylpropanal also have odor thresholds in the low μg/L
range.
To allow quantitation of these compounds at subthreshold
concentrations, the analytical method has to be highly sensitive,
selective, and robust. Different methods have been described in
the literature, mostly utilizing GC-MS techniques. Ferreira et al.
used a GC-ion trap method to quantitate various aldehydes
together with other compounds after a liquid−liquid micro-
extraction.8 In later works the same group described the
analysis of C5−C8 aldehydes after derivatization with O-
(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydrochloride
(PFBHA) on a solid-phase extraction (SPE) cartridge followed
by GC-MS analysis.19,20 For the analysis of sotolon different
INTRODUCTION
■
Oxygen plays an important role in the winemaking and aging
process of wines and can strongly influence the color and
aroma properties of the finished wine. “Oxidation” is a common
wine fault that can occur at different stages of the winemaking
process as well as after the wine has been bottled. On the other
hand, oxygen can also deliberately be used in winemaking and
storage to produce wines like Sherry or Port characterized by
desirable oxidative attributes.
Through oxidation, white wine color can change to dark
yellow or even brown, while red wines also become
progressively browner, due to reactions of phenols in wine
such as anthocyanins, catechins, and epicatechins.1 Besides
affecting its color, oxidation can affect the wine aroma well
before changes in color become noticeable, resulting in wines
characterized by a particular “oxidized” aroma or flavor. This
aroma has been described with a wide range of descriptors,
most commonly “honey-like”, “boiled potato”, “cardboard”,
“cooked vegetable”, “cider”, “woody”, and “hay-farm feed”.2−5
Given its importance in winemaking, several studies have
investigated the key volatile compounds responsible for
oxidized aromas, generally utilizing GC-olfactometry, and the
most important contributors to the oxidative off-flavor were
found to be methional and 2-phenylacetaldehyde.3,6,7 In
addition to the Strecker aldehydes, long-chained aldehydes
like (E)-2-nonenal, (E)-2-octenal, (E)-2-hexenal, as well as
benzaldehyde, furfural, and hexanal and some alcohols such as
1-octen-3-ol and eugenol were found in increased concen-
tration in wines exposed to oxygen.4,5,7,8
Cullere
́
et al. compared the amounts of (E)-2-alkenals,
Received: December 2, 2014
Revised: March 13, 2015
Accepted: March 14, 2015
Strecker aldehydes, and branched chained aldehydes in
different aged white and red wine as well as in fortified wines
such as Sherry and Port.9 Their results showed that oxidized
© XXXX American Chemical Society
A
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Ltd., Hornsby, NSW, Australia). 2-Phenyl-d5-acetaldehyde was
prepared from 2-phenyl-d5-ethanol using the method for formyl, α-
[13C2]-2-phenylacetaldehyde.25 d5-Benzaldehyde was present as a
byproduct (5%) in the 2-phenyl-d5-acetaldehyde sample and formed
as a result of C−C bond cleavage during the oxidation of 2-phenyl-d5-
ethanol as had been found previously.26
All chromatographic solvents were of HPLC grade; all chemicals
were of analytical reagent grade unless otherwise stated; water was
obtained from a Milli-Q purification system (Millipore, North Ryde,
NSW, Australia). Sulfuric acid (95−97%, synthesis grade) was from
Merck (Merck Pty. Ltd., Australia, Kilsyth, VIC, Australia). BondElut
EN resins (styrene-divinylbenzene polymer), prepacked in 500 mg
cartridges (6 mL total volume), were obtained from Agilent (Agilent
Technologies Australia Pty Ltd., Mulgrave, VIC, Australia). All
prepared solutions were % v/v with the balance made up with
redistilled ethanol and prepared freshly prior to analysis.
Synthesis of 3-(Methyl-d3-thio)-1-d2-propanol (d5-Methio-
nol). Methyl 3-mercapto-propionate (3.94 g, 33 mmol), K2CO3 (7.2 g,
52 mmol), and CD3I (5 g, 35 mmol) in THF (75 mL) were stirred for
7 days at room temperature in a sealed pressure vessel. The mixture
was filtered, dried (Na2SO4), and filtered again into a fresh round-
bottomed flask. The solution was cooled to 0 °C, LiAlD4 (1.25 g, 30
mmol) was added, and the solution was stirred at room temperature
overnight. The mixture was quenched with saturated NH4Cl, and the
organic layer was separated, dried (Na2SO4), and concentrated in
vacuo. The residue was purified by vacuum distillation (85−95 °C; 12
mmHg) to give d5-methionol (1.78 g, 54%). GC-MS, m/z (%): 94
(100), 110 (51), 64 (9), 33 (6), 78 (4).
Synthesis of 3-(Methyl-d3-thio)-propanal (d3-Methional). d3-
Methional was prepared using a modified version of the method of Sen
et al.27 d3-Methionol was prepared as described for d5-methionol above
using LiAlH4 in place of LiAlD4. d3-Methionol (2.6 g, 24 mmol),
pyridinium chlorochromate (7.7 g, 36 mmol), and CH2Cl2 (125 mL)
were stirred for 4 h. The mixture was loaded directly onto silica gel
(150 g) and eluted using 30% Et2O/pentane → 70% Et2O/pentane to
give d3-methional (0.4 g, 15%). GC-MS, m/z (%): 51 (100), 46 (41),
79 (30), 64 (30), 50 (24), 107 (20).
Table 1. Analyzed Compounds Associated with Oxidized
Aroma, Their Odor Descriptors and Odor Thresholds
compound
odor descriptor
odor threshold (μg/L)
(E)-2-alkenals
(E)-2-heptenal
(E)-2-hexenal
(E)-2-nonenal
(E)-2-octenal
Strecker aldehydes
methional
a
9
,
soapy, fatty
4.6
a
,
9
green apple
4
a
,
14
green, fatty, sawdust
fatty, nutty
0.17
9
a
,
3
b
,
7
2
cooked potato-like
malty
0.5
b
,
9
2-methylpropanal
3-methylbutanal
2-phenylacetaldehyde
furans
6
b
,
malty
4.6
b
,
9
honey, floral
1
b
,
b
,
b
,
15
15
12
furaneol
caramel
37
10
15
homofuraneol
sotolon
caramel
curry, seasoning
aldehydes
a
16
,
benzaldehyde
furfural
bitter almond-like
sweet, bread
2000
b
17
,
14 100
a
,
16
hexanal
grassy, green
20
a
16
,
5-methylfurfural
alcohols
sweet, bitter almond
2000
b
b
18
17
,
maltol
caramel
5000
1000
,
methionol
cooked potato-like
clove-like
b
,
17
eugenol
6
a
b
Odor detection threshold determined in water. Odor detection
threshold determined in model wine (10% aqueous ethanol, 7 g/L
glycerol, pH 3.2).
techniques were described: a HPLC and UPLC method21,22
and a liquid−liquid extraction followed by GC-MS analysis.23
In another paper, analysis of sotolon together with furaneol and
maltol was performed via GC-MS after isolation of the volatiles
via SPE.24
The aim of this work was to develop a highly sensitive GC-
MS/MS method which allows simultaneous quantitation of 18
compounds involved in wine oxidation in a single run. This
required quantifying a number of compounds which are present
in trace quantities including (E)-2-alkenals, various Strecker
aldehydes (methional, 2-phenylacetaldehyde, 3-methylbutanal,
2-methylpropanal), methionol, eugenol, maltol, as well as
furans (sotolon, furaneol, homofuraneol). The comprehensive
method was validated in an oxidized red, white, and model
wine, and its application was demonstrated through the analysis
of a 14 year old white wine which was stored under five
different closures to study the effects of the formation of these
flavor compounds.
Synthesis of 4-Hydroxy-2,5-dimethyl-d6-3(2H)-furanone (d6-
Furaneol). Furaneol (2.0 g, 16 mmol), D2O (12 mL), and 40 w/v%
NaOD/D2O (8 mL) were stirred for 4 days at 50 °C. The solution was
salted (NaCl), acidified to pH 2 using DCl/D2O, and extracted with
anhydrous Et2O. The combined organic layers were dried (Na2SO4)
and concentrated in vacuo to give d6-furaneol (870 mg, 41%). GC-MS,
m/z (%): 134 (100), 46 (62), 60 (32), 88 (26), 78 (4).
Synthesis of d5-5-Ethyl-4-hydroxy-2-methyl-3(2H)-furanone
(d5-Homofuraneol). Homofuraneol was synthesized using the
method of Blank et al.28 Homofuraneol (0.3 g, 2 mmol), D2O (6 g,
0.3 mol), and 40 wt % NaOD/D2O (1.0 g, 10 mmol) were stirred for 4
days in a sealed pressure vessel under nitrogen. The solution was
acidified to pH 3 using DCl (2.7 M), salted (NaCl), and extracted with
anhydrous Et2O. The combined organic layers were dried (Na2SO4)
and concentrated in vacuo to give d5-homofuraneol (160 mg, 54%).
GC-MS, m/z (%): 147 (100), 46 (62), 73 (53), 101 (30), 132 (26).
Synthesis of 3-Hydroxy-4-methyl-5-(methyl-d3)-2(5H)-fura-
none-5-d (d4-Sotolon). D2O (5 mL) and 37% w/w DCl/D2O (0.5
mL) was added to 2-ketobutyric acid (1.25 g, 12 mmol), and the
solution was stirred for 1 h. The solution was salted (NaCl), extracted
with anhydrous Et2O, dried (Na2SO4), filtered, and then concentrated
in vacuo. Additional D2O (5 mL) and 37% w/w DCl/D2O (5 mL) was
added to the resultant 3,3-d2-2-ketobutyric acid (0.9 g) in a 50 mL
glass high-pressure vessel with a Teflon screw cap seal. The solution
was chilled under N2, and d4-acetaldehyde (1.0 g, 21 mmol) was
added. The vessel was sealed and stirred for 7 days, after which further
D2O (5 mL) was added and the solution was salted (NaCl). The
product was extracted using Et2O and washed with brine (6 × 2 mL),
and the organic layer was dried (Na2SO4) and concentrated in vacuo
to give the product (1.5 g, 92%). GC-MS, m/z (%): 87 (100), 59 (75),
132 (45), 46 (39), 76 (14), 114 (5).
MATERIALS AND METHODS
■
Chemicals. 3-Hydroxy-4,5-dimethyl-2(5H)-furanone (sotolon)
(≥99%), 4-hydroxy-2,5-dimethyl-3(2H)-furanone (furaneol)
(≥98%), 2-phenylacetaldehyde (≥95%), furfural (≥98%), 5-methyl-
furfural (≥98%), hexanal (≥97%), (E)-2-hexenal (≥95%), (E)-2-
heptenal (≥97%), (E)-2-octenal (≥94%), (E)-2-nonenal (≥93%), 3-
(methylthio)-propanal (methional) (98%), 3-(methylthio)-propanol
(methionol) (≥98%), 3-methylbutanal (97%), 2-methylpropanal
(98%), 4-allyl-2-methoxyphenol (eugenol) (≥98%), benzaldehyde
(≥99%), and 3-hydroxy-2-methyl-4H-pyran-4-one (maltol) (≥99%)
were supplied by Sigma-Aldrich (Castle Hill, NSW, Australia). O-
(2,3,4,5,6-Pentafluorobenzyl)hydroxylamine hydrochloride (≥99%),
used as a derivatization reagent, was also purchased from Sigma-
Aldrich. d4-Furfural was purchased from CDN Isotopes (SciVac PTY.
B
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Synthesis of (E)-2-Octenal-5,6,7,8-d9 (d9-(E)-2-Octenal). d9-
(E)-2-Octenal was prepared in five steps from d9-bromobutane as per
Heading.29
Synthesis of (E)-2-Hexenal-3,4,4,5,5,6,6,6-d8 (d8-(E)-2-Hexe-
nal). d8-(E)-2-Hexenal was prepared from the reduction of d8-ethyl
hexanoate to d8-hexenol using lithium aluminum hydride and then
converted to the aldehyde using the Swern oxidation as described in
Grant-Preece et al.30
Wine. The wine studied was a Semillon from the Clare Valley
region in South Australia, processed entirely in a stainless steel tank
with no oak treatment. SO2 was added postfermentation as well as
dimethyl dicarbonate (Velcorin) at 10 μg/L, with the basic
composition at bottling of pH 3.16, titratable acidity (as tartaric
acid, pH 8.2) 6.1 g/L, volatile acidity 0.56 g/L as acetic acid, free sulfur
dioxide 28 mg/L, and total sulfur dioxide 100 mg/L. The wine was
bottled with the same closures used by an earlier closure trial31 and
had various flavor compound additions as described in Capone et al.32
The bottled wines were kept in an inverted position in a near constant
temperature storage facility (ca. 15 °C). The closures studied are
shown in Table 2.
additions of the reference compounds and standards (100 μg/L) were
injected using a ramp between 0 to 50 eV to determine the optimum
collision energies for each of the different transitions (see Table 3).
Sample Preparation for GC-MS/MS Analysis. Isolation of the
́
volatile analytes was performed according to Cullere et al. for
quantitation of sotolon, furaneol, and maltol with the following
modifications.19 An ethanolic solution (30 μL) containing the labeled
standards in concentrations from 1 to 60 μg/mL, depending on the
expected concentration, was added to the wine sample (20 mL) and
equilibrated for 15 min. To the spiked wine sample 1 mL of an
aqueous solution containing the derivatization agent O-(2,3,4,5,6-
pentafluorobenzyl)-hydroxylamine hydrochloride at 10 mg/mL was
added, and the sample was stirred with a magnetic stir bar for 15 min.
Ammonium sulfate (3 g) was then added to the samples prior to
loading on the Bond Elut ENV SPE cartridge (500 mg, 6 mL volume)
previously conditioned with 5 mL of dichloromethane, 5 mL of
methanol, and finally 5 mL of model wine (13% v/v ethanol solution
in water saturated with potassium hydrogen tartrate and adjusted to
pH 3.3 with aqueous tartaric acid solution). Excess reagent was
removed from the cartridge by loading 6 mL of a 0.05 M sulfuric acid
solution. Analytes were finally eluted with dichloromethane (6 mL),
and the organic layer was concentrated under a nitrogen stream to 200
μL at 40 °C using a turbovap (Turbo Vap LV Evaporator, Zymark).
For wine analysis three individual bottles of each wine were analyzed.
Table 2. Wine Samples Included in the Study, Bottled in
1999
code
closure
headspace
volume
Amp. N2
Amp. air
SC
glass ampule
N2, 3 mL
air, 3 mL
air, 3 mL
air, 3 mL
air, 3 mL
air, 3 mL
air, 3 mL
50 mL
RESULTS AND DISCUSSION
■
glass ampule
50 mL
Method Development. Sample Preparation. For iso-
lation of the different compounds various SPE materials (Bond
Elut C18, SDBL, Licholut EN) and wine volumes (2, 5, 10, and
20 mL) were tested (data not shown), with the highest
retention of the compounds achieved using a styrene-
divinylbenzene polymer cartridge (Bond Elut ENV 500 mg)
loaded with 20 mL of wine. The derivatization procedure with
PFBHA had been previously investigated thoroughly by Cullere
et al.,19 who showed that the derivatization performed directly
on the SPE cartridge was optimal, as derivatizing before the
extraction caused interference in the MS spectra with other
wine carbonyls that were also derivatized. However, since our
method utilizes MS/MS analysis and measures only the specific
mass transitions selected, the interference of other wine
carbonyls did not constitute a problem. This fact, combined
with the observation that derivatization conducted directly in
the wine, showed a greater response for the underivatized
compounds, allowing us to choose the latter procedure. The
issue of a possible incomplete derivatization, and therefore
inaccurate quantitation, was addressed by using isotopically
labeled analogues of the analytes as internal standards, prior to
the derivatization step. After loading the wine sample on the
cartridge, the excess derivatizing agent along with other
unwanted impurities were removed with 6 mL of a solution
of dilute sulfuric acid (0.05 M). The elution of the target
compounds was performed with dichloromethane (6 mL),
followed by concentration under a stream of nitrogen to a final
volume of 200 μL.
screw cap, ROTE
natural cork, 44 mm
natural cork, 38 mm
synthetic extruded
synthetic extruded
750 mL
750 mL
750 mL
750 mL
750 mL
NC 1
NC 2
Syn 1
Syn 2
Gas Chromatography−Mass Spectrometry. GC analysis was
performed on an Agilent 7000 Triple Quadrupole GC-MS/MS system,
equipped with an Agilent Multimode injector. The column used was a
VF-200 ms from J&W Scientific, 30 m × 0.25 mm i.d., with 0.25 μm
film thickness. The carrier gas was helium at a constant flow rate of 1.6
mL/min. The extract (2 μL) was injected in splitless mode, with a
pressure of 16.95 psi, septum purge flow was 3 mL/min, and splitless
time was 1 min. The injector temperature was 180 °C for 1 min and
then heated to 260 °C at 250 °C min−1. The oven temperature started
at 40 °C, was held for 1 min, then raised to 220 °C at 10 °C/min, and
finally heated to 270 °C at 100 °C/min. After each analytical run,
backflushing was performed at 270 °C for 7.24 min at a pressure of 25
psi (back inlet pressure 1 psi). The temperature of the transfer line was
240 °C, and nitrogen (1.5 mL/min) was used as the collision gas. The
mass spectrometer was operated in electron ionization mode at 70 eV
with multiple reaction monitoring (MRM), with the monitored
transitions listed in Table 3. Data acquisition and analyses were
performed using the MassHunter Workstation software version
B.05.01 supplied by the manufacturer (Agilent Technologies Australia
Pty Ltd., Mulgrave, VIC, Australia).
Method Development. In order to determine the retention time
and the characteristic mass fragments of the compounds, full scan
analysis (m/z 50−350) was performed individually on each of the
reference compounds as well as on the labeled standards. The
aldehydes (10 mL of a 10 mg/L aqueous solution) were separately
reacted with O-(2,3,4,5,6-pentafluorobenzyl)-hydroxyl-amine hydro-
chloride (1 mL of a 10 mg/mL aqueous solution) to obtain the oxime
derivatives, these were then extracted with a Bond Elut ENV (200 mg)
cartridge and eluted with 3 mL of dichloromethane (DCM), and after
solvent evaporation to 200 μL the concentrated extract was subjected
to GC-MS/MS analysis. The compounds that were quantified in their
free forms were individually injected directly as solvent extracts, all in a
concentration of 10 mg/L. The full scan experiments showed the
precursor ions, and then MS/MS experiments were conducted on the
precursor ions to determine their product ions. Once the selection of
product ions was established, extracts of wine samples containing
́
Isotopically Labeled Analogues. The synthesis of nine
deuterium-labeled analogues including d5-homofuraneol, d4-
sotolone, d6-furaneol, d5-methionol, d3-methional, d5-benzalde-
hyde, 2-phenyl-d5-acetaldehyde, (E)-2-d9-octenal, and d8-(E)-2-
hexenal was carried out in house for accurate quantitation of
the compounds of interest.
Isotopically labeled standards were not available for some
compounds; therefore, some standards were also used to
determine other structurally similar compounds as follows. For
the quantitation of 2-methylpropanal and 3-methylbutanal, d5-
benzaldehyde was used as an internal standard. The labeled
C
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 3. Mass Spectral Transitions and Collision Energies Selected for Analysis of the Compounds for Aldehydes
first transition parent ion (m/z) to product ion
(m/z) (quantifier)
collision
energy (V)
second transition parent ion (m/z) to product ion
(m/z) (qualifier)
collision
energy (V)
a
benzaldehyde
301−181
305−181
163.8−149
128−85
22
22
10
4
301−271
305−276
164−131
128−57
6
a
d5-benzaldehyde
eugenol
6
14
12
7
furaneol
d6-furaneol
134−88
5
134−60
a
furfural
291−181
295−181
307−181
295−181
293−181
301−181
142−127
147−132
126−97
22
22
22
18
20
20
4
291−248
295−251
307−250
295−252
293−250
301−251
142−99
147−101
126−97
299−102
302−105
106−88
6
a
d4-furfural
6
a
(E)-2-heptenal
2
a
hexanal
2
a
(E)-2-hexenal
2
a
d8-(E)-2-hexenal
homofuraneol
d5-homofuraneol
maltol
2
4
4
4
14
20
12
2
14
6
a
methional
299−181
302−181
106−59
a
d3-methional
4
methionol
2
d5-methionol
111−64
4
111−45
4
a
5-methylfurfural
2-methylpropanal
305−181
267−181
281−181
335−181
321−181
330−181
315−181
320−181
22
24
24
22
22
16
22
22
304.7−262
267−250
281−239
335−250
321−250
330−250
315−117
320−122
8
a
2
a
3-methylbutanal
2
a
(E)-2-nonenal
8
a
(E)-2-octenal
2
a
d9-(E)-2-octenal
2
a
2-phenylacetaldehyde
16
16
2-phenyl-d5-
acetaldehyde
a
sotolon
128−83
132−87
2
4
128−72
132−76
2
4
d4-sotolon
a
The parent ion is the derivative form.
Table 4. Method Validation and Calibration Data
a
slope
R2
linearity range (μg/L)
LOQ (μg/L)
LOD (μg/L)
benzaldehyde
eugenol
6.464 0.182
4.909 0.142
0.774 0.013
0.180 0.002
1.284 0.046
0.040 0.001
0.701 0.019
0.652 0.009
0.410 0.047
0.298 0.014
6.870 0.115
0.119 0.001
0.246 0.004
0.305 0.004
0.387 0.006
1.187 0.043
0.754 0.051
0.940 0.005
0.992
0.993
0.998
0.999
0.996
0.998
0.997
0.997
0.993
0.997
0.998
0.999
0.997
0.997
0.997
0.997
0.999
0.999
0.1−500
0.5−1000
5−1000
0.01
0.50
5.00
0.10
0.01
0.01
0.01
10.00
20.00
0.01
10.00
0.01
0.01
0.01
0.01
0.01
0.01
5.00
0.005
0.100
2.000
0.005
0.005
0.005
0.005
2.500
5.000
0.005
3.000
0.005
0.005
0.005
0.005
0.005
0.005
2.000
furaneol
furfural
0.1−1000
0.01−100
0.01−500
0.01−100
10−1000
20−1000
0.01−500
10−2000
0.01−1000
0.01−100
0.01−100
0.01−500
0.01−100
0.01−1000
5−1000
(E)-2-heptenal
hexanal
(E)-2-hexenal
homofuraneol
maltol
methional
methionol
5-methylfurfural
2-methylpropanal
3-methylbutanal
(E)-2-nonenal
(E)-2-octenal
2-phenylacetaldehyde
sotolon
a
Slope standard error.
standard of sotolon was also used for determination of eugenol
and maltol. d3-Methional was also used to quantitate hexanal.
For the determination of 5-methylfurfural the internal standard
d4-furfural was used, and d9-(E)-2-octenal was the standard for
(E)-2-nonenal and (E)-2-heptenal.
GC Parameters. Six GC columns with different polarity
(DB-35 ms, Solgel-Wax, DB-FFAP, DB-Wax, DB-5, DB-1701,
VF-200 ms) were investigated (data not shown), and the
majority of them gave unsatisfactory results for the compound
sotolon in terms of peak shape and thus detection limits. The
best chromatographic results, especially for sotolon, were
D
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 5. Concentrations (μg/L) of Volatile Compounds in Semillon White Wine Stored under Different Closures for 14 Years
NC 1 (natural
cork 1)
NC 2 (natural
cork 2)
Syn 1 (synthetic
closure 1)
Syn 2 (synthetic
closure 2)
a
compound (μg/L)
Amp. (N2)
Amp. (air)
SC (screw cap)
(E)-2-alkenals
(E)-2-heptenal
(E)-2-hexenal
(E)-2-nonenal
(E)-2-octenal
Strecker aldehydes
methional
0.104 (0.010)
0.024 (0.010)
0.054 (0.030)
0.014 (0.004)
0.101 (0.010)
0.017 (0.010)
<0.01
0.100 (0.002)
0.026 (0.010)
0.017 (0.006)
0.012 (0.004)
0.106 (0.001)
0.035 (0.010)
<0.01
0.112 (0.010)
0.038 (0.020)
0.032 (0.009)
0.018 (0.003)
0.105 (0.002)
0.072 (0.010)
0.016 (0.007)
0.017 (0.001)
0.104 (0.004)
0.087 (0.001)
0.012 (0.002)
0.014 (0.001)
<0.01
0.011 (0.005)
0.69 (0.05)
62.60 (1.97)
4.22 (0.97)
2.80 (0.10)
4.32 (0.35)
88.90 (6.68)
12.20 (1.03)
4.73 (0.43)
1.09 (0.06)
56.00 (1.48)
3.81 (0.33)
2.880 (0.003)
3.77 (4.76)
4.87 (3.47)
18.30 (0.66)
101.00 (9.65)
64.00 (2.80)
26.50 (0.55)
22.80 (0.41)
114.00 (0.01)
99.50 (8.1)
2-methylpropanal
3-methylbutanal
2-
72.80 (16.60)
12.80 (10.50)
9.16 (7.06)
91.00 (21.00)
27.50 (18.6)
12.60 (6.93)
18.80 (0.47)
phenylacetaldehyde
furans
furaneol
25.20 (6.43)
212.00 (72.68)
<2
16.70 (1.47)
61.40 (6.03)
<2
27.20 (1.16)
118.00 (4.97)
<2
23.40 (8.95)
94.70 (32.2)
<2
15.60 (6.77)
76.30 (30.59)
<2
6.01 (1.49)
20.30 (1.27)
6.90 (0.14)
3.45 (1.50)
4.67 (0.58)
16.80 (1.70)
homofuraneol
sotolon
aldehydes
benzaldehyde
furfural
24.6 (2.5)
1096.0 (121.0)
1.0 (0.3)
7.2 (1.0)
18.8 (0.6)
1000.0 (56.3)
0.7 (0.1)
25.6 (11.3)
1284.0 (50.4)
1.0 (0.3)
24.5 (3.5)
1082.0 (401.0)
1.5 (0.4)
7.8 (0.2)
23.0 (4.93)
1126.0 (76.7)
2.1 (0.2)
1149.0 (33.0)
0.9 (0.1)
1311.0 (77.6)
1.46 (0.1)
40.3 (1.3)
hexanal
5-methylfurfural
alcohols
46.6 (0.9)
42.2 (0.6)
41.9 (0.3)
42.1 (1.9)
44.8 (0.8)
33.4 (2.5)
maltol
150.0 (14.7)
483.0 (14.3)
1.8 (0.6)
149.0 (7.4)
466.0 (16.4)
1.5 (0.5)
160.0 (1.8)
494.0 (9.6)
2.0 (0.3)
165.0 (12.9)
470.0 (17.6)
1.4 (0.3)
158.0 (14.3)
470.0 (5.7)
1.9 (0.6)
117.0 (10.9)
471.00 (0.01)
1.3 (0.1)
79.0 (7.0)
351 (45.3)
1.4 (0.1)
methionol
eugenol
a
Mean values are shown, and standard deviation is in brackets (n ≥ 3). For each closure three different bottles were measured.
achieved on a VF-200 ms column (30 m × 0. 25 mm i.d., with
0.25 μm film thickness). The injection type selected was liquid
mode, and several inlet parameters were assessed, including
splitless, split, pulsed splitless, as well as programmable
temperature vaporization, with the splitless mode giving the
greatest signal-to-noise. The injection in splitless mode was
then tested at different temperatures ranging between 160 and
240 °C, and 180 °C was selected as the temperature giving the
greatest sensitivity.
MS Parameters. The full scan MS analysis of the individual
compounds showed a good signal for the molecular ion of each
of the compounds analyzed. MS/MS experiments revealed the
characteristic fragmentation of each of the compounds, and the
most intense fragment was chosen as a quantifier. The second
most intense fragment of each compound was chosen as a
qualifier, and the fragmentation of the ions, together with the
collision energies selected are shown in Table 3.
As the quantitation of 18 compounds and 10 internal
standards requires the recording of 56 mass transitions, the
sensitivity could potentially be compromised due to the
number of data points that are needed to be collected. To
minimize the number of recorded transitions and therefore
maximize the sensitivity, the MS/MS experiments were
grouped into three time segments over the GC-MS/MS run.
Method Validation. Linearity. Linearity was evaluated
across a series (15 points) of duplicate standard additions of
each of the compounds ranging over a concentration range
from 0.01 to 2000 μg/L, depending on the expected
concentration of the compound, and validated in red, white,
and model wine (13% v/v ethanol solution in water saturated
with potassium hydrogen tartrate and adjusted to pH 3.3 with
aqueous tartaric acid solution). As shown in Table 4, good
linearity was achieved for each of the compounds with
coefficient of determination (R2) values ranging between
0.992 and 0.999.
Repeatability. To assess the precision of the method, seven
identical samples were spiked with all of the 18 compounds at
two different concentrations (10 and 250 μg/L) in red, white,
and model wine, and the relative standard deviation of the
calculated concentrations was under 5% in all cases (data not
shown).
Recovery. Control samples were spiked with all of the
analytes at two different concentrations (10 and 100 μg/L) as
well as a nonspiked sample, to determine the concentration in
the blank, and the mean recoveries were between 92% and
105% (data not shown).
Limit of Detection. The LOD (S/N = 3) and LOQs (S/N =
10) were determined by calculating the signal-to-noise (S/N)
ratio in a spiked nonoxidized wine and a model wine matrix.
For some of the derivatized compounds the specific mass
transition measured showed hardly any noise, which is a
common occurrence for GC-MS/MS analysis. Therefore, the
LOD could not be calculated in the conventional way. The
limit of quantitation values were determined by measuring
seven replicates of unspiked samples and the same number of
replicates for samples containing increasing concentrations of
the analytes and assessing whether they were significantly
different (P < 0.05).
For all of the compounds included in the analysis the
detection limits (Table 4) were lower than their odor detection
threshold and ranged between 0.01 (for derivatized aldehydes)
and 10 μg/L (for sotolon).
Analysis of Wines. After validation was complete, the GC-
MS/MS method was applied to study a 14 year old white wine
(bottled in 1999) under different closures (two natural corks,
two synthetic corks, and a screw cap) to determine which of the
selected compounds would most be affected by oxidation.
E
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Wines stored under the same closures had been assessed by a
trained sensory descriptive analysis panel at a much earlier stage
of storage,31,33,34 with the two wines bottled with synthetic
closures being rated much higher in oxidized aroma than the
screw cap-sealed wine and the cork-sealed wines being
intermediate in scores. This set of samples also included
wines stored in glass ampules sealed under nitrogen or air, and
the color differences among the samples were clearly visible,
indicating the degree of oxidation that occurred. The color of
the wine ranged from light yellow in the case of the wines
sealed in glass ampules to a dark brown for wines under
synthetic corks. In order to account for possible bottle to bottle
variations with the same closures, the analysis for each type of
closure was carried out on three different bottles.
control. The (E)-2-alkenals have been reported to be important
contributors to oxidative off flavor.5 A positive correlation
between the (E)-2-octenal content and a “rotten apple” sensory
note of oxidized wines was shown by Escudero et al., and (E)-2-
decenal and (E)-2-nonenal were also found to be related to
other oxidized wine notes,4 whereas (E)-2-nonenal was
associated with a “sawdust” aroma.14 The concentration of
the C6 compound hexanal was higher only in the wine under
synthetic cork Syn 2, whereas in all other samples this
compound was not found in significantly different quantities.
The furanones, furaneol and homofuraneol, were found in
higher concentration in the control wines (N2) in comparison
to all of the other closures. Homofuraneol was found in very
high concentration in the control ampule sealed under nitrogen
(212 μg/L) and much lower in all other samples, with the
wines sealed with synthetic corks having the lowest
concentrations of 20 and 4 μg/L (Syn 1 and Syn 2). The
changes for furaneol were less drastic, but again the synthetic
corks resulted in significantly lower concentrations than the
ampouled controls, suggesting oxidative degradation of these
compounds.
The “caramel”-like smelling compound maltol, which is
commonly extracted from toasted oakwood used in the aging of
wines,37 was found to be present in approximately one-half the
concentration in the wine sealed with the synthetic closure
(Syn 2) compared to that of the glass ampules, and the wine
sealed in Syn 1 also contained a lower concentration. There was
no significant difference observed for all other closures, and the
concentration of maltol found was well below its odor detection
threshold of 5000 μg/L.13 Methionol was found in a
concentration of almost 500 μg/L in all samples (odor
threshold 1000 μg/L17), with the only significant difference
being with the wine sealed with the synthetic closure 2 (Syn 2),
which contained 350 μg/L.
Sotolon has been shown to be a key odorant of aged Sherry
wines, Ports, and botrytized wines and has also been reported
as a contributor to the oxidative off-flavor of table wines.5,10,22
In our study this compound was only found in increased
concentrations and above its aroma detection threshold in the
wines sealed with the synthetic closure 2 (Syn 2).
The compound furfural was present at a high concentration
in each of the wines, whereas the structurally similar compound
5-methylfurfural was found at the highest concentration in the
wine stored under nitrogen in the glass ampules (N2) and at the
lowest concentration in the wine under synthetic closure (Syn
2).
In previous studies using Maccabeo and Chardonnay wines it
was shown that the levels of eugenol increased upon wine
oxidation,38 and gas chromatography-olfactometry (GC-O)
investigations showed eugenol as an important odor linked to
oxidation.5 Our results with Semillon are in contrast to these
findings, as there were no significant differences in concen-
tration among the wines stored under different closures for this
compound even though the wines were clearly and visibly
affected by oxidation.
The role of different flavor compounds in contributing to
wine aroma was clearly observed through the analysis of 14 year
old white wines sealed under different closures and comparison
to the reported aroma detection thresholds. The wines had
naturally aged and oxidized via varied oxygen ingress over
extended bottle storage. In agreement with the literature, 2-
phenylacetaldehyde and methional were found in increased
concentrations in the wines exposed to high oxygen levels,
As expected from the immense differences in color
highlighted by visual inspection of the wines, most of the
compounds analyzed were found to be present in significantly
different (P < 0.05) concentrations depending on the closures
(Table 5). Among the Strecker aldehydes, the concentration of
2-phenylacetaldehyde, 3-methylbutanal, and methional was
significantly higher in the wines sealed with natural corks
than in those stored in glass ampules. Notably, the highest
concentration of these three compounds was found in the
wines stored under the synthetic closures, with 3-methylbutanal
reaching concentrations of almost 100 μg/L in Syn 1 compared
to around 4 μg/L found in the wines stored in glass ampules
under nitrogen. Similarly, for methional and 2-phenylacetalde-
hyde the wine sealed in glass ampules showed concentrations
below 5 μg/L, which increased by a factor of 4 in the wines
sealed with synthetic closures (Syn 1 and Syn 2). For each of
these flavor compounds (methional, phenylactaldehyde, and 3-
methylbutanal) the wines sealed with natural corks had the
second highest concentration. The wine sealed under screw cap
contained a very low concentration of Strecker aldehydes,
which were even lower than in the glass ampule sealed with air,
and was very similar to that of the glass ampule sealed under
nitrogen. This result is not surprising and is most likely
attributed to the differences in volume among the wine vessels:
the ratio of the head space to the wine volume was higher for
the ampules since the volume of the ampules was 50 mL with 3
mL of headspace, in contrast to that of the wine bottles, where
the ratio was much lower since the bottles had a 750 mL
volume. Escudero et al. showed that methional is an impact
odorant of oxidized wines, giving the wine a cooked vegetable
off-flavor.7 Their study showed that wines spiked with
methionol or methionine resulted in an increase in the
methional concentration and therefore explains that the
possible formation pathway during wine oxidation is the direct
peroxidation of methionol or a Strecker degradation of
methionine. The same pathway was reported for the formation
of 2-phenylacetaldehyde from phenylalanine, which also has a
low odor threshold and an aroma described as “honey-like”.35,36
Four (E)-2-alkenals with different chain length were
measured, and only (E)-2-hexenal and (E)-2-nonenal were
found in significantly different concentrations among closures,
with no significant differences observed in both (E)-2-octenal
and (E)-2-heptenal concentrations. The control ampule sealed
under nitrogen contained the highest concentration of (E)-2-
nonenal compared to all other closure samples, suggesting
nonoxygen-induced changes may be contributing to the
formation, whereas (E)-2-hexenal was about 3−4 times higher
in wines sealed with synthetic closures and 1.5 times higher in
wines sealed with natural corks compared with that of the
F
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
(3) Ferreira, S.; Hogg, T.; de Pinho, P. G. Identification of key
odorants related to the typical aroma of oxidation-spoiled white wines.
J. Agric. Food Chem. 2003, 51, 1377−1381.
(4) Escudero, A.; Asensio, E.; Cacho, J.; Ferreira, V. Sensory and
chemical changes of young white wines stored under oxygen. An
assessment of the role played by aldehydes and some other important
odorants. Food Chem. 2002, 77, 325−331.
(5) Escudero, A.; Cacho, J.; Ferreira, V. Isolation and identification of
odorants generated in wine during its oxidation: a gas chromatography
− olfactometric study. Eur. Food Res. Technol. 2000, 211, 105−110.
(6) Balboa-Lagunero, T.; Arroyo, T.; Cabellos, J. M.; Aznar, M.
Sensory and olfactometric profiles of red wines after natural and forced
oxidation processes. Am. J. Enol. Vitic. 2011, 62, 527−535.
which were significantly above their odor thresholds and could
therefore add to the oxidative “off-flavor” in Semillon wine.
In conclusion, validation and application of a novel GC-MS/
MS method is described that enables the simultaneous
quantitation of 18 oxidation-related compounds in wines at
mg/L, μg/L, and ng/L levels. The preparation of samples is
simple and requires only a small volume of wine. An added
advantage is the use of deuterated analogues as internal
standards which ensures accuracy of the method. The reliability
was established using validation criteria such as linearity,
repeatability, and reproducibility. This method, to the best of
our knowledge, is the first developed for assessing oxidation-
related compounds using GC-MS/MS, and its effectiveness was
proven through quantitation of 18 compounds in a white wine
closure study. Hence, the GC-MS/MS method will provide
new insights when used for future storage and shelf-life studies
of wines and related beverages.
́
(7) Escudero, A.; Hernandez-Orte, P.; Cacho, J.; Ferreira, V. Clues
about the role of methional as character impact odorant of some
oxidized wines. J. Agric. Food Chem. 2000, 48, 4268−4272.
́
(8) Ferreira, V.; Escudero, A.; Lopez, R.; Cacho, J. Analytical
characterization of the flavor of oxygen-spoiled wines through the gas
chromatography-ion-trap mass spectrometry of ultratrace odorants:
Optimization of conditions. J. Chromatogr. Sci. 1998, 36, 331−339.
AUTHOR INFORMATION
■
́
(9) Cullere, L.; Cacho, J.; Ferreira, V. An assessment of the role
Corresponding Author
played by some oxidation-related aldehydes in wine aroma. J. Agric.
Food Chem. 2007, 55, 876−881.
*Phone: +61 88313 6689. Fax: +61 88313 6601. E-mail:
dimitra.capone@awri.com.au.
́ ́ ́
(10) Martin, B.; Etievant, P. X.; Le Quere, J. L.; Schlich, P. More
clues about sensory impact of sotolon in some Flor Sherry wines. J.
Agric. Food Chem. 1992, 31, 475−478.
Present Address
‡Christine M. Mayr: Cara Technology Limited, Leatherhead
Enterprise Centre, Randalls Road, Leatherhead, Surrey, KT22
7RY, United Kingdom.
(11) Silva Ferreira, A. C.; Barbe, J.-C.; Bertrand, A. 3-Hydroxy-4,5-
dimethyl-2(5H)-furanone: a key odorant of the typical aroma of
oxidative aged Port wine. J. Agric. Food Chem. 2003, 51, 4356−4363.
(12) Collin, S.; Nizet, S.; Bouuaert, T. C.; Despatures, P. M. Main
odorants in jura flor-sherry wines. Relative contributions of sotolon,
abhexon, and theaspirane-derived compounds. J. Agric. Food Chem.
2012, 60, 380−387.
Funding
The Australian Wine Research Institute (AWRI), a member of
the Wine Innovation Cluster in Adelaide, is supported by
Australian grapegrowers and winemakers through their invest-
ment body, by the Australian Grape and Wine Authority
(AGWA), with matching funds from the Australian Govern-
ment.
(13) Ferreira, V.; Ortín, N.; Escudero, A.; Lop
Chemical characterization of the aroma of Grenache rose
́
ez, R.; Cacho, J.
wines: aroma
́
extract dilution analysis, quantitative determination, and sensory
reconstitution studies. J. Agric. Food Chem. 2002, 50, 4048−4054.
(14) Chatonnet, P.; Dubourdieu, D. Identification of substances
responsible for the ‘sawdust’ aroma in oak wood. J. Sci. Food Agric.
1998, 76, 179−188.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
(15) Kotseridis, Y.; Razungles, A.; Bertrand, A.; Baumes, R.
Differentiation of the aromas of Merlot and Cabernet Sauvignon
wines using sensory and instrumental analysis. J. Agric. Food Chem.
2000, 48, 5383−5388.
■
We thank our collaborators Agilent Technologies for their
generous support, in particular, Jim Tsiotinas and Nick B.
Harden for the loan of the GC-MS/MS system. We also thank
Dr. Dan Johnson for his efforts in ensuring the success of the
Agilent collaboration and Dr. Markus Herderich for manuscript
evaluation. We would also like to thank Dr. Mark A. Sefton and
Dr. Gordon Elsey for their contributions toward the deuterium-
labeled standard synthesis used in this study and Dr. Sefton and
Peter Godden for their contribution to the white wine closure
study.
́
(16) Etievant, P. X.. Wine. In Volatile Compounds in Foods and
Beverages; Maarse, H., Ed.; Marcel Dekker: New York, 1991; Vol. 14,
pp 483−546.
́
(17) Ferreira, V.; Lopez, R.; Cacho, J. F. Quantitative determination
of the odorants of young red wines from different grape varieties. J. Sci.
Food Agric. 2000, 80, 1659−1667.
(18) Cutzach, I.; Chatonnet, P.; Dubourdieu, D. Study of the
formation mechanisms of some volatile compounds during the aging
of sweet fortified wines. J. Agric. Food Chem. 1999, 47, 2837−2846.
́
(19) Cullere, L.; Cacho, J.; Ferreira, V. Analysis for wine C5−C8
ABBREVIATIONS USED
■
aldehydes through the determination of their O-(2,3,4,5,6-
pentafluorobenzyl)oximes formed directly in the solid phase extraction
cartridge. Anal. Chim. Acta 2004, 524, 201−206.
PFBHA, O-(2,3,4,5,6-pentafluorobenzyl)hydroxylamine hydro-
chloride; GC, gas chromatography; MS, mass spectrometry;
SPE, solid-phase extraction; THF, tetrahydrofuran; MRM,
multiple reaction monitoring; LOD, limit of detection; LOQ,
limit of quantitation; GC-O, gas chromatography-olfactometry
́
(20) Ferreira, V.; Cullere, L.; Loscos, N.; Cacho, J. Critical aspects of
the determination of pentafluorobenzyl derivatives of aldehydes by gas
chromatography with electron-capture or mass spectrometric
detection: Validation of an optimized strategy for the determination
of oxygen-related odor-active aldehydes in wine. J. Chromatogr. A
2006, 1122, 255−265.
REFERENCES
■
(1) Waterhouse, A. L.; Laurie, V. F. Oxidation of wine phenolics: A
critical evaluation and hypotheses. Am. J. Enol. Vitic. 2006, 57, 306−
313.
(21) Gabrielli, M.; Buica, A.; Fracassetti, D.; Stander, M.; Tirelli, A.;
du Toit, W. J. Determination of sotolon content in South African
white wines by two novel HPLC−UV and UPLC−MS methods. Food
Chem. 2015, 169, 180−186.
(2) Ugliano, M. Oxygen contribution to wine aroma evolution during
bottle aging. J. Agric. Food Chem. 2013, 61, 6125−6136.
G
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
(22) Guichard, E.; Pham, T. T.; Etievant, P. Quantitative
determination of sotolon in wines by high-performance liquid
chromatography. Chromatographia. 1993, 37, 539−542.
(23) Blank, I.; Sen, A.; Grosch, W. Potent odorants of the roasted
powder and brew of Arabica coffee. Z. Lebensm.-Unters. Forsch. 1992,
195, 239−245.
́
(24) Ferreira, V.; Jarauta, I.; Lopez, R.; Cacho, J. Quantitative
determination of sotolon, maltol and free furaneol in wine by solid-
phase extraction and gas chromatography-ion-trap mass spectrometry.
J. Chromatogr. A 2003, 1010, 95−103.
(25) Schuh, C.; Schieberle, P. Characterization of the key aroma
compounds in the beverage prepared from Darjeeling black tea:
quantitative differences between tea leaves and infusion. J. Agric. Food
Chem. 2006, 54, 916−924.
(26) Fernandes, R. A.; Kumar, P. PCC-mediated novel oxidation
reactions of homobenzylic and homoallylic alcohols. Tetrahedron Lett.
2003, 44, 1275−1278.
(27) Sen, A.; Grosch, W. Synthesis of six deuterated sulfur containing
odorants to be used as internal standards in quantification assays. Z.
Lebensm.-Unters. Forsch. 1991, 192, 541−547.
(28) Blank, I.; Fay, L. B.; Lakner, F. J.; Schlosser, M. Determination
of 4-Hydroxy-2,5-dimethyl-3(2H)-furanone and 2(or 5)-ethyl-4-
hydroxy-5(or 2)-methyl-3(2H)-furanone in pentose sugar-based
Maillard model systems by isotope dilution assays. J. Agric. Food
Chem. 1997, 45, 2642−2648.
(29) Heading, H. J. Development of SIDA protocols for
quantification of aldehydes in wine. Ph.D. Thesis, Flinders University
of South Australia, 2007; pp 31−34.
(30) Grant-Preece, P. A.; Pardon, K. H.; Capone, D. L.; Cordente, A.
G.; Sefton, M.; Jeffery, D. W.; Elsey, G. M. Synthesis of wine thiol
conjugates and labelled analogues: fermentation of the glutathione
conjugates of 3-mercaptohexan-1-ol yields the corresponding cysteine
conjugate and free thiol. J. Agric. Food Chem. 2010, 58, 1383−1389.
(31) Godden, P.; Francis, I. L.; Field, J.; Gishen, M.; Coulter, A.;
Valente, P.; Hoj, P.; Robinson, E. M. C. Wine bottle closures: physical
characteristics and effect on composition and sensory properties of a
Semillon wine 1. Performance up to 20 months post-bottling. Aust. J.
Grape Wine Res. 2001, 7, 64−105.
(32) Capone, D.; Sefton, M.; Pretorius, I.; Høj, P. Flavour “scalping”
by wine bottle closures-the “winemaking” continues post vineyard and
winery. Aust. N. Z. Wine Ind. J. 2003, 18, 18−20.
(33) Godden, P. AWRI trial of the technical performance of various
types of wine closure. Aust. Grapegrower Winemaker 1999, 425, 59−64.
(34) Godden, P.; Francis, L.; Field, J.; Gishen, M.; Coulter, A.;
Valente, P.; Høj, P.; Robinson, E. Results of an AWRI trial
investigating the technical performance of various types of wine
closure: Part 2-Wine composition up to 20 month post-bottling. Aust.
Grapegrower Winemaker 2001, 7, 89−98.
(35) 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.
(36) Grant-Preece, P.; Fang, H.; Schmidtke, L. M.; Clark, A.
Sensorially important aldehyde production from amino acids in model
wine systems: Impact of ascorbic acid, erythorbic acid, glutathione and
sulphur dioxide. Food Chem. 2013, 141, 304−312.
(37) Cutzach, I.; Chatonnet, P.; Henry, R.; Dubourdieu, D.
Identification of volatile compounds with a “ toasty ” aroma in heated
oak used in barrelmaking. J. Agric. Food Chem. 1997, 45, 2217−2224.
́
(38) Ferreira, V.; Escudero, A.; Fernandez, P.; Cacho, J. F. Changes
in the profile of volatile compounds in wines stored under oxygen and
their relationship with the browning process. Z. Lebensm.-Unters.
Forsch. A 1997, 205, 392−396.
H
DOI: 10.1021/jf505803u
J. Agric. Food Chem. XXXX, XXX, XXX−XXX