Characterization of the key aroma compounds in apricots (Prunus armeniaca) by application of the molecular sensory science concept

Article abstract of DOI:10.1021/jf0705015

An aroma extract dilution analysis applied on an aroma distillate prepared from fresh apricots revealed (R)-γ-decalactone, (E)-β-damascenone, δ-decalactone, and (R/S)-linalool with the highest flavor dilution (FD) factors among the 26 odor-active compounds identified. On the basis of quantitative measurements performed by application of stable isotope dilution assays, followed by a calculation of odor activity values (OAVs), β-ionone, (Z)-1,5-octadien-3-one, γ-decalactone, (E,Z)-2,6-nonadienal, linalool, and acetaldehyde appeared with OAVs >100, whereas in particular certain lactones, often associated with an apricot aroma note, such as γ-undecalactone, γ-nonalactone, and δ-decalactone, showed very low OAVs (<5). An aroma recombinate prepared by mixing the 18 most important odorants in concentrations as they occurred in the fresh fruits showed an overall aroma very similar to that of apricots. Omission experiments indicated that previously unknown constituents of apricots, such as (E,Z)-2,6-nonadienal or (Z)-1,5-octadien-3-one, are key contributors to the apricot aroma.

Full text of DOI:10.1021/jf0705015

J. Agric. Food Chem. 2007, 55, 5221−5228 5221
Characterization of the Key Aroma Compounds in Apricots
(Prunus armeniaca) by Application of the Molecular Sensory
Science Concept
†
VERONIKA GREGER AND PETER SCHIEBERLE*,†,#
Chair for Food Chemistry, Technical University of Munich, and German Research Center for Food
Chemistry, Lichtenbergstrasse 4, D-85748 Garching, Germany
An aroma extract dilution analysis applied on an aroma distillate prepared from fresh apricots revealed
(R)-γ-decalactone, (E)-â-damascenone, δ-decalactone, and (R/S)-linalool with the highest flavor
dilution (FD) factors among the 26 odor-active compounds identified. On the basis of quantitative
measurements performed by application of stable isotope dilution assays, followed by a calculation
of odor activity values (OAVs), â-ionone, (Z)-1,5-octadien-3-one, γ-decalactone, (E,Z)-2,6-nonadienal,
linalool, and acetaldehyde appeared with OAVs >100, whereas in particular certain lactones, often
associated with an apricot aroma note, such as γ-undecalactone, γ-nonalactone, and δ-decalactone,
showed very low OAVs (<5). An aroma recombinate prepared by mixing the 18 most important
odorants in concentrations as they occurred in the fresh fruits showed an overall aroma very similar
to that of apricots. Omission experiments indicated that previously unknown constituents of apricots,
such as (E,Z)-2,6-nonadienal or (Z)-1,5-octadien-3-one, are key contributors to the apricot aroma.
KEYWORDS: Aroma; apricot; molecular sensory concept; (E,Z)-2,6-nonadienal; (Z)-1,5-octadien-3-one
INTRODUCTION
The human odorant receptors present in the nasal cavity show
considerable selectivity and high differences in sensitivity toward
organic volatiles, and, thus, the evaluation of odor thresholds
is a good tool to select potent aroma compounds from the bulk
of odorless or less odor-active volatiles. GC-olfactometry uses
the concept of “ranking by odor thresholds” and, thus, studies
using GC-olfactometry on serial dilutions of aroma distillates
(10), such as CHARM analysis or aroma extract dilution analysis
(AEDA), have clearly shown that only if the concentration of
a given volatile exceeds its odor threshold will this compound
interact with the human odorant receptors and, consequently,
participate in the creation of the respective aroma impression
in the brain. Before 1980, however, no attempts were made to
investigate the volatiles of apricot with regard to their aroma
activity and, surprisingly, no application of dilution to odor
threshold techniques, such as the AEDA, on apricots is available
in the current literature.
Like many stone fruits, apricots are appreciated by consumers
all over the world and, consequently, have gained great
economic importance. The fruits are presently cultivated in all
Mediterranean countries and in South Africa as well as in South
and North America, mainly in California. The crop is either
marketed fresh, but also used as canned and dried fruit, or is
manufactured into juice.
Apart from color, sweetness, and texture, apricots are
especially popular for their characteristic aroma. However,
although flavor is one of the most important criteria in the
evaluation of fruit quality, investigations on the volatile fraction
and, in particular, the odor-active compounds of apricot are
rather scarce. The first comprehensive studies on apricot
volatiles were performed by Tang and Jennings about 40 years
ago (1, 2) and led to the identification of γ-octalactone,
γ-decalactone, γ-dodecalactone, δ-decalactone, linalool, R-ter-
pineol, geraniol, myrcene, and limonene as the predominant
volatile constituents. Further studies undertaken by Rodriguez
et al. (3), Guichard and Souty (4), and Toth-Markus et al. (5)
as well as Bolzoni et al. (6), Takeoka et al. (7), and Go´mez and
Ledbetter (8) have considerably increased the number of
volatiles identified in apricot, and >200 compounds have been
listed so far (9).
In a few studies quantitative data were related to the odor
thresholds of individual compounds (5, 7). On the basis of the
calculation of odor activity values (OAVs; ratio of concentration
to odor thresholds in water), these authors suggested that, in
particular, â-ionone, linalool, γ-decalactone, hexanal, (E)-2-
hexenal, and geraniol should be considered as key odorants of
apricot. However, these results are somewhat preliminary
because quantitative data were obtained by GC-FID measure-
ments either without the use of internal standards or using only
one internal standard for the entire set of volatiles. Moreover,
no aroma reconstitution experiments based on the results of
quantitative data have been performed, which are, however,
* Corresponding author (telephone +49 89 289 13265; fax +49 89 289
14183; e-mail Peter.Schieberle@ch.tum.de).
† Chair for Food Chemistry, Technical University of Munich.
# German Research Center for Food Chemistry.
10.1021/jf0705015 CCC: $37.00
© 2007 American Chemical Society
Published on Web 05/27/2007

5222 J. Agric. Food Chem., Vol. 55, No. 13, 2007
Greger and Schieberle
4-methoxyphenol (Merck, Darmstadt, Germany); geraniol (Roth,
Karlsruhe, Germany). (E)-â-Damascenone and â-ionone were gifts from
Symrise (Holzminden, Germany).
necessary to confirm the correctness of the quantitative data
(10). Such aroma recombination studies based on the real
concentrations of the respective odorants in the food itself are
among the most reliable tools to evaluate whether a mixture of
odorants with a variety of odor qualities will finally be able to
mimic the odorant receptor responses. This approach, which
can be assigned as “molecular sensory science” or “sensomics”,
has already successfully been used to confirm the key odorants
of many foods and, most recently, to identify the aroma and
taste compounds of a black tea beverage (11, 12). It is a valuable
tool, in particular, to address the challenge that single key
odorants of a certain food do not smell like the food itself, but
a distinct mixture may perfectly match the overall food aroma.
Consequently, the aim of this study was, first, to locate and
characterize the most odor-active compounds in an extract
isolated from fresh apricots by application of the AEDA. In a
second step, the key odorants were quantified by means of stable
isotope dilution analyses (SDIAs) and their OAVs were
calculated. Finally, aroma recombinates based on the concentra-
tions of each aroma compounds as occurring in apricot were
prepared and evaluated by aroma profile analysis. On the basis
of this approach the sensory active volatiles are selected from
the total volatile metabolites (the metabolome) of the fruit, and
a pattern of key aroma compounds becomes available, repre-
senting the blueprint of compounds evoking the human odorant
receptor responses and, finally, creating the overall aroma
impression in the brain.
Synthesis of trans-4,5-Epoxy-(E)-2-decenal. 3-Chloroperoxybenzoic
acid (1.8 g, 10 mmol) acid was dissolved in dichloromethane (80 mL)
and added in nine equal portions to a solution of (E,E)-2,4-decadienal
(936 mg, 6.2 mmol in 10 mL of dichloromethane). The mixture was
stirred at room temperature for 140 min and finally left overnight (15
h) at -20 °C. The solution was dried over sodium sulfate and
concentrated to 2 mL at 45 °C using a Vigreux column. The target
compound was purified by column chromatography (water-cooled
column, 20 cm × 1 cm) using modified silica gel 60 (13).
MS-EI (m/z in %): 68 (100), 81 (35), 41 (23), 39 (17), 55 (15), 43
(14), 69 (12), 57 (8). MS-CI (m/z in %): 153 (100), 154 (10), 169 (7).
The following reference compounds were synthesized according to
the literature cited: (Z)-3-hexenal (14), 3-methyl-2,4-nonandione (15),
and (Z)-1,5-octadien-3-one (14).
Isotopically Labeled Internal Standards. [2H3]-â-Ionone. â-Ionone
(39 mg, 0.2 mmol), dissolved in dry tetrahydrofuran (1 mL), was added
to sodium hydride (10 mg, 0.4 mmol) maintained in an atmosphere of
nitrogen. Deuterium oxide (1 mL, 0.056 mol) was added, and the
mixture was stirred for 24 h in a closed flask at room temperature.
The mixture was extracted with diethyl ether (2 × 20 mL), dried over
sodium sulfate, and subjected to high vacuum distillation (16) to remove
nonvolatile compounds.
MS-EI (m/z in %): 180 (100), 46 (21), 181 (12), 91(8), 138 (7), 93
(7), 162 (6).
MS-CI (m/z in %): 196 (100), 197 (14).
[2H2]-δ-Decalactone. 4-Bromo-1-butene (3.37 g, 25 mmol), 1,3-
cyclohexandione (2.88 g, 25 mmol), and potassium hydroxide (1 g, 25
mmol) were dissolved in dioxane/water (1:1, v/v, 25 mL) and refluxed
for 7 h (17). Aqueous potassium hydroxide (25 mL; 30 g/L) was added,
and the mixture was extracted with diethyl ether (3 × 50 mL). The
organic layer was discarded, and the aqueous layer was adjusted to pH
4.0 with hydrochloric acid and extracted with diethyl ether (3 × 50
mL). The combined organic layers were dried over sodium sulfate and,
after filtration, the solvent was distilled off. The residue (3.65 g) was
refluxed for 30 h in an aqueous Na2CO3 (300 g/L; 37.5 mL). The
mixture was adjusted to pH 3.0 with hydrochloric acid (5 mol/L) and
extracted with diethyl ether (3 × 50 mL). The combined organic layers
were dried over sodium sulfate, and the solvent was distilled off,
yielding 5-oxo-dec-9-enoic acid (0.2 g, 1.1 mmol). Sodium borohydride
(0.85 g, 25 mmol) was added in small portions, followed by aqueous
potassium hydroxide (50 mL; 30 g/L). The mixture was stirred at 40
°C for 9 h, adjusted to pH 1.0 with HCl (30%), and refluxed for 30
min. The aqueous solution was cooled to room temperature and
extracted with diethyl ether (3 × 50 mL). The combined organic layers
were dried over sodium sulfate, and the solvent was distilled off,
yielding δ-dec-9-enolactone. The crude product was purified by column
chromatography (water-cooled column, 20 cm × 1 cm i.d.) using silica
gel 60 adjusted to 7% water content. After the column had been flushed
with pentane (150 mL), the target compound was eluted with pentane/
diethyl ether (60:40, v/v). The δ-dec-9-enolactone was dissolved in
toluene (20 mL), mixed with Wilkinson’s catalyst [tris(triphenylphos-
phine)rhodium(I) chloride] (300 mg, 0.32 mmol) and deuterated in an
autoclave at 5 bar for 90 min. The target compound was purified by
column chromatography (water-cooled column, 20 cm × 1 cm) on silica
gel 60. Elution was performed with pentane/diethyl ether (60:40, v/v).
MS-EI (m/z in %): 99 (100), 71 (33), 42 (25), 70 (22), 55 (16), 43
(12), 41 (12), 114 (10).
MATERIALS AND METHODS
Apricots. Fully ripe French apricots (cv. Bergeron) were purchased
from a local store (Olympia Fruchthaus, Munich, Germany). For the
aroma profile analysis and the AEDA as well as for the quantification
of (Z)-3-hexenal and acetaldehyde, a new batch of fresh fruits of the
same variety was always used.
For the quantitation of the remaining odorants, fruits were frozen
with liquid nitrogen, shrink-wrapped in plastic film, and stored at -20
°C. All fruits, either fresh or deep-frozen, were taken from the same
variety of apricots, but from different batches. To avoid enzymatic
reactions, care was taken to not use thawed fruits for volatile isolation
before aqueous CaCl2 was added (see below).
Chemicals. 4-Bromo-1-butene, 3-chloroperoxybenzoic acid, deute-
rium oxide, 1,4-dichloro-2-butyne, ethyl magnesium bromide (1 mol/L
in tetrahydrofuran), 3-hexyn-1-ol, and sodium hydride were obtained
from Aldrich (Taufkirchen, Germany). Methyl octanoate was obtained
from Fluka. Dess-Martin-periodinane and Wilkinson’s catalyst were
obtained from Lancaster Synthesis (Ward Hill, MA). Deuterium and
liquid nitrogen were obtained from Linde (Munich, Germany). Calcium
chloride, citric acid, 1,3-cyclohexanedione, dichloromethane, diethyl
ether, ethanol, fructose, glucose, hydrochloric acid (37%), lithium
aluminum hydride, pentane, potassium bromide, potassium iodide,
potassium hydroxide, sucrose, sea sand, silica gel, sodium boronhydride,
sodium carbonate, sodium hydrogen carbonate, anhydrous sodium
sulfate, sodium thiosulfate-pentahydrate, sorbitol, sulfuric acid, tet-
rahydrofuran, and toluene were obtained from VWR (Darmstadt,
Germany). Malic acid and pectin were obtained from Roth (Karlsruhe,
Germany).
Diethyl ether, dichloromethane, and pentane were freshly distilled
prior to use.
MS-CI [m/z in %): 173 (100), 174 (12), 155 (2).
[2H4]-trans-4,5-Epoxy-(E)-2-decenal. 3-Hexyn-1-ol (2.0 g, 20.4
mmol), dissolved in toluene (15 mL), was deuterated in the presence
of Wilkinson’s catalyst (750 mg, 0.81 mmol) under a slight pressure
of deuterium for 24 h, yielding [2H4]hexanol. The crude product was
purified by column chromatography (water-cooled column; 20 cm ×
1 cm i.d.) on silica gel in pentane. Elution was performed with diethyl
ether (300 mL). To remove nonvolatile compounds, the orange solution
was subjected to high-vacuum distillation (16) and the solvent was
finally distilled off. The [2H4]hexanol (2.1 g, 19.8 mmol) was dissolved
in dichloromethane (20 mL) and added dropwise to a solution of Dess-
Reference Aroma Compounds. The following reference compounds
were obtained from the suppliers given in parentheses: γ-decalactone,
δ-decalactone, γ-dodecalactone, hexanal, (E)-2-hexenal, (R)-linalool,
3-methylbutanoic acid, (E,Z)-2,6-nonadienal, γ-nonalactone, (E)-2-
nonenal, γ-octalactone, (E)-2-octenal, (E,E)-2,4-decadienal (Aldrich,
Sigma-Aldrich Chemie, Taufkirchen, Germany); butanoic acid, hexanoic
acid, hexyl acetate, (R,S)-linalool, 2-methylbutanoic acid, pentanoic acid
(Fluka, Neu-Ulm, Germany); (Z)-3-hexenyl acetate, 1-octen-3-one
(Lancaster, Mu¨hlheim am Main, Germany); acetic acid, eugenol,

Aroma Compounds in Apricots
J. Agric. Food Chem., Vol. 55, No. 13, 2007 5223
Martin-periodinane (9 g, 21.2 mmol, in 20 mL of dichloromethane).
After 16 h of stirring, diethyl ether (100 mL) was added and the solution
was treated with sodium thiosulfate (100 mL, 1 mol/L, saturated with
sodium hydrogen carbonate) until the organic layer was clear. The
organic layer was washed again with a solution of sodium thiosulfate
(50 mL, 1 mol/L, saturated with sodium hydrogen carbonate), followed
by a saturated solution of sodium hydrogen carbonate (100 mL) and,
finally, with water (100 mL). The organic layer was dried over sodium
sulfate, and the solvent was distilled off, yielding [2H4]hexanal.
For the synthesis of 1-methoxy-but-1-en-3-yne, a solution of
potassium hydroxide (10 g, 0.18 mol) in methanol (33 mL, 0.8 mol)
was heated to 65 °C. With continuous stirring, 1,4-dichloro-2-butyne
(10.3 g, 83.7 mmol) was dropwise added to the solution, and the mixture
was stirred for another 60 min at room temperature. Water (20 mL)
was added, and the solution was extracted with diethyl ether (3 × 50
mL). The organic layer was dried over calcium chloride and subjected
to high-vacuum distillation for purification. The solvent was finally
distilled off, and the residue was taken up in tetrahydrofuran. The
1-methoxy-but-1-en-3-yne (1 g, 12 mmol) obtained was dissolved in
THF (5 mL) and dropwise added to ethyl magnesium bromide (12 mmol
dissolved in 12 mL of THF) held at 40 °C. The mixture was stirred for
60 min. Then, [2H4]hexanal (0.86 g, 8 mmol, dissolved in 5 mL of
THF) was dropwise added, and the solution was stirred for another 2
h. After the addition of ethanol (0.8 mL, 13.5 mmol) and further stirring
for 20 min, lithium aluminum hydride (0.4 g, 10 mmol) was added in
small portions. The mixture was stirred for 15 h, and ethyl acetate (1.2
mL), followed by water (5.6 mL) and sulfuric acid (28 mL, 4 mol/L),
was added. The organic layer was separated, and the aqueous phase
was extracted with pentane/dichloromethane (30 mL, 1:1, v/v). The
combined organic layers were washed with aqueous sodium carbonate
(10% w/w, 250 mL), then with water (50 mL) and finally dried over
sodium sulfate. The solvent was distilled off to yield [2H4]-(E,E)-2,4-
decadienal (1.1 g). The [2H4]-(E,E)-2,4-decadienal was finally oxidized
into [2H4]-trans-4,5-epoxy-(E)-2-decenal with 3-chloroperoxybenzoic
acid (1.8 g, 10 mmol) as described above for the unlabeled compound.
MS-EI (m/z in %): 68 (100), 81 (31), 39 (11), 69 (10), 43 (8), 45
(8), 44 (7).
aqueous sodium bicarbonate (0.5 mol/L, 4 × 50 mL). The combined
aqueous solutions were adjusted to pH 2 with hydrochloric acid (2 mol/
L) and extracted with diethyl ether (4 × 50 mL) to obtain the acidic
compounds (AF).
The solutions containing either the acidic (AF) or the neutral-basic
(NBF) fraction were concentrated to 2 mL at 40 °C using a Vigreux
column. The extracts were transferred into a microflask and further
concentrated to 200 µL using a microdistillation column.
High-Resolution Gas Chromatography (HRGC) and HRGC-
Olfactometry (HRGC-O). HRGC was performed by means of a gas
chromatograph type 8000 series (Fisons Instruments, Mainz, Germany)
using the following capillary columns: DB-5 (30 m × 0.25 mm i.d.;
0.25 µm film thickness; J&W Scientific, Folsom, CA); DB-FFAP (30
m × 0.32 mm i.d.; 0.25 µm film thickness; J&W Scientific); OV-
1701 (30 m × 0.25 mm i.d.; 0.25 µm film thickness; J&W Scientific);
BGB-176 (30 m × 0.25 mm i.d.; 0.25 µm film thickness; J&W
Scientific). The temperature programs used were as follows:
DB-5: 40 °C/1 min 7 °C/1 min8 110 °C 5 °C/min8
180 °C 10 °C/min8 240 °C/10 min
DB-5: 40 °C/1 min 7 °C/1 min8 180 °C 10 °C/min8 240 °C/10 min
OAV-1701: 40 °C/1 min 8 °C/min8 240 °C/10 min
BGB-176: 40 °C/1 min 10 °C/min8 100 °C 6 °C/min8 225 °C/10 min
For HRGC-O, the effluent was split evenly at the end of the column
between a flame ionization detector (FID) and a heated (180 °C) sniffing
port using a Y-shaped glass splitter and uncoated deactivated fused
silica capillaries (50 cm, 0.1 mm i.d. as transfer lines). Helium, adjusted
to a flow rate of 1.5 mL/min, served as the carrier gas. Retention indices
(RI) were calculated from the retention times of n-alkanes by linear
interpolation.
AEDA. Flavor dilution (FD) factors were determined by AEDA (7)
of fractions NBF and AF, respectively, using capillary columns DB-5,
FFAP, or OV-1701. Extracts were diluted stepwise with solvent in a
1:1 ratio, and each dilution was analyzed by HRGC-O (injection volume
) 1 µL). The AEDA was performed by three experienced assessors.
These underwent a GC-O training including sniffing of about 50
reference compounds in amounts adjusted to 5-fold above their odor
thresholds in air.
HRGC-Mass Spectrometry (HRGC-MS). For compound iden-
tification, mass spectra were generated by means of an MAT 95 S mass
spectrometer (Finnigan, Bremen, Germany) at 70 eV in the electron
ionization (EI) mode and at 115 eV in the chemical ionization (CI)
mode (reagent gas ) isobutane) using the FFAP and DB-5 capillaries.
SIDA. Extracts of the fruits (fresh or frozen) were prepared as
described under Isolation of Volatiles. Various amounts of fruit pure´e
(20-600 g) were used, depending on the amounts of the target
compounds estimated in preliminary experiments. After the fruit
samples had been spiked with defined amounts of the labeled standards
(resulting in a concentration of 1-5 µg/mL of each compound in the
aroma extract), the samples were stirred for 60 min for equilibration.
To isolate the volatile fraction, a SAFE distillation (16) and solvent
extraction were performed as described above.
For the quantitation of acetaldehyde by headspace gas chromatog-
raphy-mass spectrometry, the fruit samples (50 g) and the labeled
acetaldehyde were added to a 200 mL Erlenmeyer flask, which was
tightly sealed with a septum, and the mixture was stirred for 45 min at
room temperature until equilibrium was reached.
Quantitation of aroma compounds was performed by means of three
different HRGC-MS systems using the FFAP, DB-5, and OV-1701
capillaries: Acids and lactones (except γ-nonalactone and γ-undeca-
lactone) were quantified using a gas chromatograph Varian GC 3800
(Varian, Darmstadt, Germany) coupled to an ion trap mass spectrometer
Saturn 2000 (Varian). Quantitation of the remaining compounds was
performed by means of a two-dimensional HRGC-MS system consisting
of a gas chromatograph Trace 2000 series (Thermo Quest, Egelsbach,
Germany) coupled to a gas chromatograph Varian GC 3800. Mass
MS-CI (m/z in %): 157 (100), 158 (10).
The following isotopically labeled internal standards were synthe-
sized according to the literature cited: [2H2]-butanoic acid (17), [2H6]-
(E)-â-damascenone (18), [2H2]-γ-decalactone (19), [2H2]-γ-dodecalac-
tone (19), [2H2]-geraniol (A. Fischer, personal communication), [2H4]-
hexanal (20), [2H2]-(E)-2-hexenal (21), [2H2]-(Z)-3-hexenal (22), [13C2]-
(Z)-3-hexenyl acetate (23), [2H2-4]-hexyl acetate (23), [2H2]-linalool
(24), [2H3]-2-methoxyphenol (25), [2H2]-3-methylbutanoic acid (26),
[2H3]-3-methyl-2,4-nonandione (22), [2H2]-(E,Z)-2,6-nonadienal (22),
[2H2]-γ-nonalactone (19), [2H2]-(E)-2-nonenal (22), [2H2]-1-octen-3-
one (22), [2H2]-(Z)-1,5-octadien-3-one (22), [2H2]-γ-octalactone (27),
[2H2-4]-4-propyl-2-methoxyphenol (28).
[13C2]-Acetaldehyde and [2H3]-acetic acid were purchased from
Aldrich (Steinheim, Germany).
Concentrations of Labeled Compounds. Because most of the
syntheses were performed at a microscale level, common purification
procedures, such as distillation or crystallization, were not possible.
To determine the yields and exact concentrations, the following
approach was used: First, an FID response factor was determined by
GC analysis of a solution containing defined amounts of the respective
unlabeled compound and methyl octanoate as a reference standard. In
a second step, a defined amount of methyl octanoate was added to a
defined volume of the solution containing the labeled compound. The
resulting mixture was analyzed by GC-FID, and the concentration of
the labeled compound was calculated from the peak areas of the GC
chromatogram, using the FID response factor determined for the
unlabeled compound.
Isolation of Volatiles. Fresh apricots (250 g) were sliced and blended
with saturated calcium chloride solution (250 mL) to inhibit enzymic
reactions (29). After the suspension had been adjusted to pH 3.3 (natural
pH of apricots), volatiles were isolated using the solvent-assisted flavor
evaporation technique (SAFE) (16). The aqueous distillate was extracted
with diethyl ether (4 × 50 mL), and, to separate the acidic volatiles
from the neutral-basic fraction, the solvent extract was treated with

5224 J. Agric. Food Chem., Vol. 55, No. 13, 2007
Greger and Schieberle
Table 1. Isotopically Labeled Standards, Ions Selected, and Response Factors Used in the Isotope Dilution Assays of the 30 Key Apricot Aroma
Compounds
odorant
acetaldehyde
butanoic acid
(E)- -damascenone
ion (m/z)
labeled standard
13C2]-acetaldehyde
[2H2]-butanoic acid
ion (m/z)
RFa
45
89
[
47
91
196−198
173
173
201
143
64
0.88
0.83
0.73
0.60
1.03
0.62
0.98
0.78
1.02
0.92
0.76
0.99
0.80
0.62
0.88
0.91
0.87
0.92
1.04
0.86
0.83
0.97
0.74
0.90
0.91
0.61
0.96
0.49
0.82
0.48
â
191
171
171
199
139
61
165
137
101
117
99
[2H5 7]-(E)-
â
-damascenone
-
γ
δ
γ
-decalactone
-decalactone
-dodecalactone
[2H2]-
[2H2]-
[2H2]-
γ
δ
γ
-decalactone
-decalactone
-dodecalactone
trans-4,5-epoxy-(E)-2-decenal
acetic acid
eugenol
geraniol
hexanal
hexanoic acid
(E)-2-hexenal
(Z)-3-hexenal
(Z)-3-hexenylacetate
hexylacetate
[2H4]-trans-4,5-epoxy-(E)-2-decenal
[2H3]-acetic acid
[2H4 6]-4-propyl-2-methoxyphenol
169−171
-
[2H2]-geraniol
139
105
105
101
83
[2H4]-hexanal
[2H2]-3-methylbutanoic acid
[2H2]-(E)-2-hexenal
[2H2]-(Z)-3-hexenal
81
143
145
193
137
125
103
171
139
157
141
127
127
125
143
103
185
[
13C2]-(Z)-3-hexenylacetate
145
[2H2 5]-hexylacetate
147−150
-
â
-ionone
linalool
[2H3]-
â
-ionone
196
139
128
105
174
141
159
143
143
129
127
145
105
187
[2H2]-linalool
2-methoxyphenol
[2H3]-2-methoxyphenol
[2H2]-3-methylbutanoic acid
[2H3]-3-methyl-2,4-nonandione
[2H2]-(E,Z)-2,6-nonadienal
2- and 3-methylbutanoic acid
3-methyl-2,4-nonandione
(E,Z)-2,6-nonadienal
γ
-nonalactone
(E)-2-nonenal
[2H2]-
γ-nonalactone
[2H2]-(E)-2-nonenal
(E)-2-octenal
1-octen-3-one
(Z)-1,5-octadien-3-one
[2H2]-(E)-2-nonenal
[2H2]-1-octen-3-one
[2H2]-(Z)-1,5-octadien-3-one
γ
-octalactone
pentanoic acid
-undecalactone
[2H2]-
γ
-octalactone
[2H2]-3-methylbutanoic acid
[2H2]-
-undecalactone
γ
γ
a MS response factor determined by analyzing defined mixtures of the analyte and the internal standard.
spectra were recorded using a Varian Saturn 2000 mass spectrometer.
Acetaldehyde was quantified HRGC-MS using headspace samples and
a gas chromatograph CP-9001 (Chrompack, Frankfurt, Germany)
coupled to a quadrupole mass spectrometer INCOS XL (Finnigan MAT,
Bremen, Germany). All mass spectra were recorded in the CI mode
with methanol as the reagent gas and using the temperature programs
mentioned above.
For each compound, a calibration factor was determined by analyzing
mixtures of defined amounts of the labeled and unlabeled compound
in five different mass ratios (1:5, 1:3, 1:1, 3:1, 5:1) by HRGC-MS.
The MS response factors, which were calculated from the peak areas
and the amounts of labeled and unlabeled compound, are summarized
in Table 1.
containing the following compounds in concentration levels equal to
those determined in apricots were prepared: acetaldehyde, (E)-â-
damascenone, γ-decalactone, δ-decalactone, γ-dodecalactone, trans-
4,5-epoxy-(E)-2-decenal, geraniol, hexanal, (Z)-3-hexenal, hexyl acetate,
â-ionone, linalool, 3-methyl-2,4-nonandione, (E,Z)-2,6-nonadienal, (E)-
2-nonenal, (Z)-1,5-octadien-3-one, γ-octalactone, 1-octen-3-one. As the
matrix an aqueous solution of 1.7% glucose, 0.9% fructose, 5.1%
sucrose, 0.8% sorbitol, 1% malic acid, 0.4% citric acid, and 1% pectin
in tap water was prepared to simulate the natural content of carbohy-
drates and acids in apricots (30).
The model mixture and the freshly mashed apricots were presented
to the sensory panel and were orthonasally evaluated. In one session,
the aroma profile of the apricots was evaluated, in another session the
aroma profile of the recombinate was judged. In a third session, the
similarity of the recombinate and the fruit puree was compared. All
evaluations were performed in triplicates.
Omission Experiments. A second model mixture omitting (Z)-1,5-
octadien-3-one, (E,Z)-2,6-nonadienal, 1-octen-3-one, trans-4,5-epoxy-
(E)-2-decenal, and (E)-2-nonenal was prepared (model II). Together
with the original recombinate (model I), it was presented to the sensory
panel in a triangle test in which each panelist had to evaluate two
triangle rows. The panelists were asked to identify the odd sample,
and the significance of the difference was calculated (31).
Determination of Enantiomeric Distributions. The enantiomeric
ratios in linalool, γ-octalactone, γ-decalactone, and γ-dodecalactone
were determined by two-dimensional gas chromatography-mass
spectrometry using the chiral BGB-176 capillary as the second column
according to the described procedure (24).
Aroma Profile Analysis. Aroma profile analyses were performed
by a trained sensory panel consisting of 16 panelists as previously
described (11). The following aroma descriptors, represented by the
compounds given in parentheses, were chosen for sensory evaluation,
and their intensities were ranked on a seven-point scale (steps of 0.5)
from 0 (not perceivable) to 3 (strongly perceivable): fruity/fresh
(acetaldehyde); peach-like (γ-decalactone, γ-dodecalactone); fruity/
apple-like [(E)-â-damascenone]; banana-like (hexyl acetate); coconut-
like (γ-octalactone, δ-decalactone); green/grassy [hexanal, (Z)-3-
hexenal]; mushroom-like (1-octen-3-one); cucumber-like [(E,Z)-2,6-
nonadienal]; flowery/citrus-like (linalool); geranium-like [(Z)-1,5-
octadien-3-one], metallic [trans-4,5-epoxy-(E)-2-decenal]; violet-like
(â-ionone); and rose-like (geraniol). The judgments of the panelists
were averaged.
RESULTS AND DISCUSSION
Odor-Active Compounds in Apricots. The entire volatiles
from fresh apricots, isolated by extraction with diethyl ether
and SAFE distillation, were evaluated by the sensory panel by
smelling a drop of the etherial extract on a strip of filter paper
as done by perfumers. After evaporation of the solvent, all 10
panelists agreed that the distillate evoked the characteristic odor
of apricot, thereby indicating that the method used for aroma
isolation was appropiate.
Aroma Reconstitution Experiments. As the reference, fresh
apricots were blended and the samples (20 mL each) were placed in
glass vessels (total volume ) 45 mL). The sensory evaluation was
finished about 20 min after blending of the fruits. Aroma models
Application of HRGC-olfactometry on an aliquot of the
aroma extract equal to ∼1.25 g of fresh fruits (250 g of fruits

Aroma Compounds in Apricots
J. Agric. Food Chem., Vol. 55, No. 13, 2007 5225
g 8) Identified in an Aroma Distillate Prepared from Fresh Apricots
Table 2. Important Aroma Compounds (FD
RIa on
DB-5
earlier identified as
volatile in apricots
no.
aroma compoundb
(Z)-3-hexenal
hexyl acetate
1-octen-3-one
(Z)-3-hexenyl acetate
(Z)-1,5-octadien-3-one
unknown
(E)-2-octenal
acetic acid
odor qualityc
FFAP
OV-1701
FDd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
green/grassy
banana-like
mushroom-like
banana-like
geranium-like
fruity/citrus-like
fatty, green
vinegar-like
fatty
flowery/citrus-like
sweaty
sweaty
leather-like
green, hay-like
metallic
fruity/apple-like
rose-like
smoky, sweet
coconut-like
coconut-like
violet-like
metallic
peach-like
clove-like
1144
1271
1295
1312
1370
1393
1420
1443
1533
1540
1619
1662
1685
1711
1732
1802
1835
1860
1875
1910
1930
2000
2140
2156
2196
2380
800
1010
975
885
1086
1070
1081
1099
nd
1165
775
1280
1200
990
64
8
32
8
128
16
8
16
8
256
32
32
16
8
16
512
16
32
16
128
16
8
4
4
1010
980
ndf
1093
600
1160
1105
821
33
1
7
1
5
(E)-2-nonenal
(R/S)-linalool
butanoic acid
2- and 3-methylbutanoic acide
unknown
878
1030
nd
1380
nd
1
1190
1247
nd
3-methyl-2,4-nonandione
unknown
(E)-
geraniol
â
-damascenone
1380
1256
1095
1321
1268
1480
1390
1471
1360
1490
1680
1498
1590
1225
nd
1480
1630
1570
1700
1505
1740
1910
34
3
2-methoxyphenol
unknown
(R/S)-
-ionone
trans-4,5-epoxy-(E)-2-decenale
(R)- -decalactone
eugenol
-decalactonee
(R)- -dodecalactone
γ
-octalactone
1
34
â
γ
2048
8
256
8
1
δ
coconut-like
peach-like
2
2
γ
a Retention index. b Identification was performed on the basis of a comparison with reference compounds using the following criteria: retention index on three different
capillary columns, odor quality and odor threshold perceived at the sniffing port, mass spectra in the EI- and CI-mode on two different capillary columns. c Odor quality
perceived at the sniffing port. d Flavor dilution factor. e The stereochemistry was not determined. f Not determined.
f 200 µL of distillate f 1 µL for HRGC-O) led to the detection
of 26 odor-active regions in the FD factor range of 8-2048
(Table 2). A great variety of odor qualities, such as green/grassy,
flowery, fruity/peach-like, coconut-like, mushroom-like, or
cucumber-like were detected, but no aroma quality of a single
odorant resembled the overall apricot aroma. Sniffing of serial
dilutions of the aroma extract revealed the highest FD factor
(2048) for a fruity/peach-like smelling compound (no. 23).
Somewhat lower FD factors were determined for compounds
with fruity/apple-like (no. 16), flowery/citrus-like (no. 10),
coconut-like, geranium-like, and green/grassy odor qualities
(Table 2).
For preliminary identifications, the odor quality perceived at
the sniffing port and the retention index on three stationary
columns are important parameters. Comparison of these data
with an in-house database containing >600 food aroma com-
pounds suggested a structure for the respective target compound.
To unequivocally identify the compounds responsible for the
odors detected in the respective area, the neutral-basic volatiles
isolated from 1 kg of fresh apricots were separated by column
chromatography on silica according to polarity (13). The
odorants were then located by GC-olfactometry in the six
fractions obtained, and mass spectra in the MS-EI and the MS-
CI mode were recorded at the respective retention index. In
addition to the mass spectra, and to avoid wrong identifications
caused, for example, by coeluting compounds, also the odor
potency of a target molecule (ratio of peak area to FD factor)
was used as further criterion in the comparison with the
reference compounds. The compounds with the highest FD
factors could, thus, be identified as γ-decalactone (no. 23), (E)-
â-damascenone (no. 16), linalool (no. 10), and δ-decalactone
(no. 25) followed by γ-octalactone (no. 20) and (Z)-1,5-octadien-
3-one (no. 5) (Table 2). Apart from these six substances, (Z)-
3-hexenal, 1-octen-3-one, butanoic acid, and 2- and 3-methyl-
butanoic acid as well as 2-methoxyphenol were identified as
further odor-active constituents in apricots, but with somewhat
lower FD factors.
The results of the entire identification experiments are
summarized in Table 2. Of the 26 compounds with FD factors
between 8 and 2048, 22 odorants could be identified. Seven of
them, namely, (Z)-3-hexenal, 1-octen-3-one, (Z)-1,5-octadien-
3-one, 3-methyl-2,4-nonandione, 2-methoxyphenol, trans-4,5-
epoxy-(E)-2-decenal, and eugenol, are reported here for the first
time as constituents of fresh apricot.
Analysis of the chiral constituents on a special stationary
phase revealed that linalool and γ-octalactone occurred as
mixtures of the (R)- and (S)-isomers in ratios of 50.7:49.3 and
85.4:14.6, respectively. As far as γ-decalactone and γ-dodeca-
lactone are concerned, only the (R)-isomer was present. These
findings are in agreement with studies of Guichard et al. (32),
who investigated the enantiomeric distribution of γ-lactone
isomers in six apricot cultivars and also found a considerable
excess of the (R)-isomer for all lactones investigated. By
contrast, no studies exist so far on the enantiomeric distribution
of (R)- and (S)-linalool in apricots.
The results of the AEDA suggested that γ-decalactone, (E)-
â-damascenone, linalool, and δ-decalactone might be of special
importance for the aroma of apricot. However, whereas most
literature studies on apricot aroma are consistent in their findings
in that γ-decalactone, linalool, and, to a somewhat smaller
degree, δ-decalactone are important contributors to the overall
aroma of apricot, no literature data previously existed pointing
to a similar importance of (E)-â-damascenone. On the other
hand, â-ionone, with a rather low FD factor (Table 2), was
considered as an essential constituent of apricot aroma (7),
whereas some other compounds with high FD factors, such as
(Z)-1,5-octadien-3-one or 1-octen-3-one, were not characterized
in previous studies. Like (E)-â-damascenone, these compounds
showed high odor activities, but did not produce an FID signal
and, therefore, might have been overlooked in investigations

5226 J. Agric. Food Chem., Vol. 55, No. 13, 2007
Greger and Schieberle
Table 3. Concentrations of 31 Important Aroma Compounds in Fresh
Apricots (Listed in Decreasing Concentration)
Table 4. Orthonasal Odor Thresholds (OOT) and Odor Activity Values
(OAV) of Apricot Aroma Compounds
concn
g/kg)
range
g/kg)
OOT (µg/L
in water)
aroma compound
acetic acid
acetaldehyde
-decalactone
pentanoic acid
-decalactone
linalool
-dodecalactone
(µ
na
(µ
aroma compound
-ionone
(Z)-1,5-octadien-3-one
-decalactone
(E,Z)-2,6-nonadienal
linalool
ref
OAV
8780
3560
509
267
245
183
181
160
146
114
4
2
6
2
6
6
6
4
4
4
4
2
4
6
3
5
3
4
6
3
4
2
2
4
2
3
3
2
4
2
8190
3540
407
262
210
136
146
147
136
100
56.3
56.2
42.9
32.4
32.9
20.0
9.5
−9410
−3580
−561
−272
−269
−218
−215
−170
−154
−126
−67.5
−60.4
−60.0
−48.1
−39.4
−37.9
−11.0
−6.1
−
−
−
−
−
−
−
−
−
−
−
−
â
0.20
0.0012
2.6
a
35
a
11
11
25
a
36
a
37
11
a
11
38
a
a
a
a
a
a
a
11
a
a
308
250
196
190
155
142
91
57
55
41
30
15
11
10
4.8
1.6
1.1
1.0
<1
<1
<1
<1
<1
<1
<1
<1
<1
γ
γ
0.03
δ
0.60b
acetaldehyde
25
2.0
2.0
0.04
0.25
0.01
10
3.2
γ
γ
-dodecalactone
hexyl acetate
2- and 3-methylbutanoic acid
hexanal
hexyl acetate
1-octen-3-one
(Z)-3-hexenal
3-methyl-2,4-nonandione
hexanal
geraniol
â
-ionone
61.6
hexanoic acid
butanoic acid
(E)-2-hexenal
geraniol
58.3
50.6
39.6
35.2
27.4
10.2
5.7
5.0
2.7
2.2
1.4
1.2
1.0
1.0
0.7
0.7
0.4
0.3
0.3
trans-4,5-epoxy-(E)-2-decenal
0.12
δ
-decalactone
(E)- -damascenone
-octalactone
51
0.43
24
0.69
180000
24
1200
890
7700
190
8.0
27
4.0
γ
-octalactone
(Z)-3-hexenal
â
γ
(E,Z)-2,6-nonadienal
(Z)-3-hexenyl acetate
5.1
(E)-2-nonenal
3.6
7.8
acetic acidc
γ
-nonalactone
1-octen-3-one
2.6
2.8
pentanoic acidc
3-methylbutanoic acidc
hexanoic acidc
butanoic acidc
1.9
2.6
(E)-2-octenal
trans-4,5-epoxy-(E)-2-decenal
1.3
1.5
1.1
1.2
γ
-undecalactone
eugenol
0.9
1.2
(E)-2-hexenal
(Z)-3-hexenyl acetate
0.9
1.1
39
a
36
a
40
24
(E)-2-nonenal
0.6
0.7
γ
-nonalactone
(E)-2-octenal
-undecalactone
(E)-
â
-damascenone
0.5
1.0
2-methoxyphenol
0.3
0.4
γ
4.2
6.0
2.5
<1
<1
<1
(Z)-1,5-octadien-3-one
3-methyl-2,4-nonandione
0.2
0.3
eugenol
0.2
0.3
2-methoxyphenol
a Number of different fruits analyzed.
a Odor thresholds were determined by a group of 30 panelists at the German
Research Center for Food Chemistry (Czerny et al., 2007; unpublished data). b The
OAV of linalool was determined using the amount of (R)-linalool (50.7%) and the
odor threshold of (R)-linalool (0.60 µg/L) for calculation. The amount of (S)-linalool,
the odor threshold of which is about 80 times higher than that of the (R)-isomer
(41), was neglected. c The odor threshold was determined at pH 4.5.
on apricot aroma performed without the application of GC-O.
However, also the different cultivars used in previous studies
or fruits of different ripening stages might be an explanation
for these differences.
Quantitation of Aroma Compounds; Calculation of OAVs.
The application of dilution to odor threshold techniques, such
as the AEDA, does not provide immediate information about
the contribution of single odorants to the overall aroma. The
reason is that each volatile is completely vaporized during
AEDA, so that this method is based on odor thresholds in air.
Moreover, possible losses of odorants during the workup
procedure are not fully taken into account. Therefore, to
investigate the contribution of individual compounds to apricot
aroma, 30 odorants were quantified in the fresh fruit pure´e by
means of SIDAs.
The highest concentrations of 8.8 or 3.6 mg/kg, respectively,
were determined for acetic acid and acetaldehyde, followed by
γ-decalactone, pentanoic acid, δ-decalactone, linalool, γ-dode-
calactone, 2- and 3-methylbutanoic acid, hexanal, and hexyl
acetate (Table 3). Whereas these 11 compounds were present
in concentrations >100 µg/kg, very low concentrations (<1 µg/
kg) were found for (E)-2-nonenal, (E)-â-damascenone, 2-meth-
oxyphenol, and 3-methyl-2,4-nonandione.
An important challenge in the analysis of fruit aroma
compounds is fruit-to-fruit variations due to, for example, the
different ripening stages. To address this problem, two to six
different extracts, prepared in separate experiments from the
same batch of fruits, were analyzed for the 30 compounds.
Although the concentrations varied among the fruits analyzed
(Table 3), for most odorants the differences amounted to about
30%, and only for (Z)-3-hexenyl acetate and γ-octalactone was
a larger variation observed.
To correlate the quantitative data with the aroma of the food
itself, the odor activity values (OAV ) ratio of concentration
to odor threshold) of the 30 compounds were calculated on the
basis of odor threshold determined in water (35). Although it
is known that the odor thresholds might be influenced by
nonvolatile fruit compounds, water is the main constituent of
apricots. The odor threshold of γ-undecalactone was determined
for the first time in this study (cf. Table 4). The highest OAV
was calculated for â-ionone (308), followed by (Z)-1,5-octadien-
3-one (250) and γ-decalactone (196; Table 4). Surprisingly,
(E,Z)-2,6-nonadienal, which did not show a very high FD factor,
reached a high OAV of 190. Also, for 1-octen-3-one, an OAV
of 55 was calculated. This is all the more remarkable because
the odor qualities of (Z)-1,5-octadien-3-one (geranium-like),
(E,Z)-2,6-nonadienal (cucumber-like), and 1-octen-3-one (mush-
room-like) were not detectable in the overall aroma profile.
Further compounds with high OAVs were linalool (155),
acetaldehyde (142), γ-dodecalactone (91), hexyl acetate (57),
and (Z)-3-hexenal (41). However, it is interesting to note that
12 compounds assigned as odor-active on the basis of AEDA
results (Table 2), among them all acids and γ-nonalactone as
well as γ-undecalactone, showed OAVs below 1 and should,
consequently, not contribute significantly to apricot aroma.
The reason for the differences between FD factors and OAVs
is simply that the entire amount of the odorants present in the
aroma extract is vaporized during GC-olfactometry, whereas
OAVs are calculated using odor thresholds in a matrix, thus
considering only the amount of an individual odorant present

Aroma Compounds in Apricots
J. Agric. Food Chem., Vol. 55, No. 13, 2007 5227
in the headspace above the matrix. Usually polar compounds
are overestimated by AEDA, because they are quite soluble in
water and, thus, their vapor pressure is comparatively low.
As already mentioned, all previous studies have considered
γ-decalactone as the key aroma compound of apricot, and this
lactone is suggested to be responsible for the typical sweet and
fruity sensory properties of this fruit. Our results confirmed the
importance of γ-decalactone in apricot aroma, because the
lactone showed the third highest OAV of all compounds.
However, although most studies discussed the additional
importance of γ-decalactone, comparatively little attention has
been paid to γ-dodecalactone, which has quite the same sensory
properties. The present study shows that γ-dodecalactone, which
is no. 7 in rank of the list of apricot volatiles sorted by their
OAV (Table 4), is of higher importance than, for example,
δ-nona-, δ-deca-, and δ-undecalactone. Obviously, in certain
varieties this lactone is of greater importance for apricot aroma
than previously assumed.
Figure 1. Aroma profile analysis of apricots
recombinate ( ).
(s) and the aroma
)
Although it can be suggested that the sweet, fruity aroma of
apricot is mostly created by lactones and not by esters as is the
case with other fruits, such as pineapple or banana, our data
show that lactones are not entirely responsible for the typical
apricot flavor. Although it has to be taken into account that the
OAVs were calculated on the basis of odor thresholds in water,
the very high OAVs of 301 and 155 determined for â-ionone
and linalool, respectively, clearly show that also the metabolism
of terpenes has a significant influence on the apricot aroma.
These data suggest that the key aroma compounds of apricot
were successfully identified in this study.
Omission Experiment. Apricots contain a number of aroma
compounds, which are products of an enzymatic lipid peroxi-
dation. Although some of them, namely, (Z)-1,5-octadien-3-
one, (E,Z)-2,6-nonadienal, and 1-octen-3-one, have rather high
OAVs, others [e.g., trans-4,5-epoxy-(E)-2-decenal and (E)-2-
nonenal] were of somewhat lower importance. What they have
in common, though, is that their odor qualities (geranium-like,
cucumber-like, mushroom-like, metallic, and fatty-cucumber-
like) seem to be quite unusual for the apricot aroma. To find
out whether these compounds are of any greater importance in
apricot aroma, a recombinate containing all aroma compounds
with OAVs g 1, except (Z)-1,5-octadien-3-one, (E,Z)-2,6-
nonadienal, 1-octen-3-one, trans-4,5-epoxy-(E)-2-decenal, and
(E)-2-nonenal, was prepared and presented to the sensory panel
in a triangle test in comparison with the original recombinate.
The omission of the five aroma compounds turned out to be
highly significant, because the deviating sample was recognized
by 23 of 32 panelists. Most panelists confirmed that the original
recombinate was closer to the aroma of fresh apricots than the
sample in which the lipid peroxidation products were omitted.
Until now, no studies exist reporting (Z)-1,5-octadien-3-one,
(E,Z)-2,6-nonadienal, trans-4,5-epoxy-(E)-2-decenal, and 1-octen-
3-one as volatile constituents of apricot. It can be assumed that
these aldehydes or ketones, respectively, are formed in the fruit
by an enzymatic degradation of either linolenic or linoleic acid.
(E,Z)-2,6-Nonadienal is known to be formed from 9-hydro-
peroxy-10,12,15-octadecatrienoic acid (42), whereas (Z)-1,5-
octadien-3-one and 1-octene-3-one are formed from linolenic
or linoleic acid, respectively, via the 10-hydroperoxide (14).
Furthermore, acetaldehyde, which had been neglected in most
previous studies, is suggested as an important contributor to
apricot flavor based on its high OAV. It makes an impression
of freshness to the rather sweet, creamy odor note of the lactones
and might, therefore, be essential for apricot aroma.
Aroma Reconstitution Experiments. In most cases, none
of the odor-active compounds present in a food bear the typical
aroma of the food itself. In a certain number of studies (e.g.,
refs 7, 10, 15, 17, 19, 20, and 23), our group has been able to
show that food aromas can be mimicked by a mixture of key
odorants in their natural concentrations. However, as previously
shown by us, for example, for the aroma of butter, in mixtures
containing more than three odorants their typical odor notes
can no longer be singly perceived (17). Because it is difficult
to predict the overall aroma of a complex mixture of compounds,
aroma reconstitution experiments are a useful tool to verify the
correctness of the quantitative data. For this purpose, all 18
aroma compounds with OAVs g 1 (Table 4) were mixed in
the same concentrations as they occur in the fruits using pure
reference compounds. Differences or similarities were then
determined by comparing the odor of the model mixture with
the original fruit puree. The results of this experiment are
illustrated in Figure 1. The aroma profiles of the fresh fruit
and that of the recombinate 1 showed a good agreement apart
from some differences in the fruity/fresh, green/grassy, and
flowery/citrus-like odor notes, which were perceived a bit more
intensely in the apricots. The overall similarity between the fruits
and the recombinate was marked 2.6 on a scale from 0 to 3.
In summary, the results show that the responses of the human
odorant receptors toward apricot aroma can be closely mimicked
by a mixture of 18 volatiles being identical in their concentra-
tions to those present in the fruit. Interestingly, also compounds
with odor qualities not resembling fruity notes, such as (Z)-
1,5-octadien-3-one, are also needed as aroma constituents. The
data confirm again that odor-active components in a complex
aroma profile can be elucidated using approaches of molecular
sensory science. The set of such key odorants can be assigned
as “blueprint” of a mixture interacting with the human odorant
receptors and finally evoking the overall odor perception in the
brain.
LITERATURE CITED
(1) Tang, C. S.; Jennings, W. G. Volatile compounds of apricot. J.
Agric. Food Chem. 1967, 15, 24-28.
(2) Tang, C. S.; Jennings, W. G. Lactonic compounds of apricot. J.
Agric. Food Chem. 1968, 16, 252-254.
(3) Rodriguez, F.; Seck, S.; Crouzet, J. Constituants volatils de
l’abricot varie´te´ rouge du Roussillon. Lebensm. Wiss. Technol.
1980, 13, 152-155.

5228 J. Agric. Food Chem., Vol. 55, No. 13, 2007
Greger and Schieberle
(4) Guichard, E.; Souty, M. Comparison of the relative quantities
of aroma compounds found in fresh apricot (Prunus armeniaca)
from six different varieties. Z. Lebensm. Unters. Forsch. 1988,
186, 301-307.
(5) To´th-Markus, M.; Boross, F.; Blazso´, M.; Kerek, M. Volatile
flavour substances of different apricot cultivars. Nahrung 1989,
33, 433-442.
(6) Bolzoni, L.; Careri, M.; Mangia, A. Characterization of volatile
components in apricot pure´es by gas chromatography-mass
spectrometry. J. Chromatogr. 1990, 518, 221-229.
(7) Takeoka, G. R.; Flath, R. A., Mon. T. R.; Teranishi, R.; Guentert,
M. Volatile constituents of apricot (Prunus armeniaca). J. Agric.
Food Chem. 1990, 38, 471-477.
(8) Go´mez, E.; Ledbetter, C. A. Development of volatile compounds
during fruit maturation: characterization of apricot and plum ×
apricot hybrids. J. Sci. Food Agric. 1997, 74, 541-546.
(9) Nijssen, L. M.; Visscher, C. A.; Maarse, H.; Willemsens, L. C.;
Boelens, M. H. Volatile Compounds in FoodsQualitatiVe and
QuantitatiVe Data; TNO Nutrition and Food Research Institute:
Zeist, The Netherlands; on-line version 9.2, 2007.
(25) Cerny, C.; Grosch, W. Quantification of character-impact odour
compounds of roasted beef. Z. Lebensm. Unters. Forsch. 1993,
196, 417-422.
(26) Guth, H.; Grosch, W. Identification of the character impact
odorants of stewed beef juice by instrumental analyses and
sensory studies. J. Agric. Food Chem. 1994, 42, 2862-2866.
(27) Fukuzawa, S.; Nakanishi, A.; Fujinami, T.; Sakai, S. Samarium-
(II) di-iodide induced reductive coupling of R,â-unsaturated
esters with carbonyl compounds leading to a facile synthesis of
γ-lactone. J. Chem. Soc., Perkin Trans. 1 1988, 1669-1675.
(28) Schmitt, R. On the role of ingredients as sources of key aroma
compounds in crumb chocolate. Ph.D. thesis, Technical Uni-
versity of Munich, 2005.
(29) Buttery, R. G.; Teranishi, R.; Ling, L. C.; Flath, R. A.; Stern,
D. J. Fresh tomato volatiles: a quantitative study. J. Agric. Food
Chem. 1987, 35, 540-544.
(30) Souci, S. W.; Fachmann, W.; Kraut, H. Die Zusammensetzung
der Lebensmittel - Na¨hrwert-Tabellen. 6. Auflage; Scientific
Publishers: Stuttgart, Germany, 2000.
(10) Schieberle, P. Recent developments in methods for analysis of
flavor compounds and their precursors. In Characterization of
Food: Emerging Methods; Goankar, A., Ed.; Elsevier: Am-
sterdam, The Netherlands, 1995; pp 403-431.
(11) 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.
(12) Scharbert, S.; Hofmann, T. Molecular definition of black tea taste
by means of quantitative studies, taste reconstitution, and
omission experiments. J. Agric. Food Chem. 2005, 53, 5377-
5384.
(13) Schieberle, P.; Grosch, W. Potent odorants of the wheat bread
crumb. Z. Lebensm. Unters. Forsch. 1991, 192, 130-135.
(14) Ullrich, F.; Grosch, W. Identification of the most intense odor
compounds formed during autoxidation of methyl linolenate at
room temperature. J. Am. Oil Chem. Soc. 1988, 65, 1313-1317.
(15) Guth, H.; Grosch, W. 3-Methylnonane-2,4-dionesan intense
odour compound formed during flavour reversion of soya-bean
oil. Fat Sci. Technol. 1989, 91, 225-230.
(16) Engel, W.; Bahr, W.; Schieberle, P. Solvent assisted flavour
evaporationsa new and versatile technique for the careful and
direct isolation of aroma compounds from complex food
matrices. Eur. Food Res. Technol. 1999, 209, 237-241.
(17) Schieberle, P.; Gassenmeier, K.; Guth, H.; Sen, A.; Grosch, W.
Character impact odour compounds of different kinds of butter.
Lebensm. Wiss. Technol. 1993, 26, 347-356.
(18) Sen, A.; Laskawy, G.; Schieberle, P. Quantitative determination
of â-damascenone in foods using a stable isotope dilution assay.
J. Agric. Food Chem. 1991, 39, 757-759.
(19) Poisson, L. Characterization of the key odorant in bourbon
whisky (in German). Ph.D. thesis, Technical University of
Munich, 2003.
(20) Guth, H.; Grosch, W. Odorants of extrusion products of oat
mealschanges during storage. Z. Lebensm. Unters. Forsch. 1993,
196, 22-28.
(31) Jellinek, G. Sensory EValuation of Food; VCH Verlagsgesell-
schaft: Weinheim, Germany, 1985.
(32) Guichard, E.; Kustermann, A.; Mosandl, A. Chiral flavour
compounds from apricots. Distribution of γ-lactone enantiomers
and stereodifferentiation of dihydroactinidiolide using multi-
dimensional gas chromatography. J. Chromatogr. 1990, 498,
396-401.
(33) Nitz, S.; Kollmannsberger, H. Changes in flavour composition
during thermal concentration of apricot pure´e. Z. Lebensm.
Unters. Forsch. 1993, 197, 541-545.
(34) Chairote, G.; Rodriguez, F.; Crouzet, J. Characterization of
additional volatile flavor compounds of apricot. J. Food Sci.
1981, 46, 1898-1906.
(35) Rychlik, M.; Schieberle, P.; Grosch, W. Compilation of Odor
Thresholds, Odor Qualities and Retention Indices of Key Food
Odorants; Deutsche Forschungsanstalt fu¨r Lebensmittelchemie:
Garching, Germany, 1998; ISBN 3-9803426-5-4.
(36) Takeoka, G. R.; Buttery, R. G.; Flath, R. A. Volatile constituents
of Asian pear (Pyrus serotina). J. Agric. Food Chem. 1992, 40,
1925-1929.
(37) Buttery, R. G.; Teranishi, R.; Ling, L. C.; Turnbaugh, J. G.
Quantitative and sensory studies on tomato paste volatiles. J.
Agric. Food Chem. 1990, 38, 336-340.
(38) Kerler, J.; Grosch, W. Odorants contributing to warmed-over
flavor (WOF) of refrigerated cooked beef. J. Food Sci. 1996,
61, 1271-1274, 1284.
(39) Masanetz, C.; Grosch, W. Key odorants of parsley leaves
(Petroselinum crispum [Mill.] Nym. ssp. crispum) by odor
activity values. FlaVour Fragrance J. 1998, 13, 115-124.
(40) Buttery, R.G. Quantitative and sensory aspects of flavor of tomato
and other vegetables and fruits. In FlaVor SciencesSensible
Principles and Techniques in FlaVor Science; Acree, T., Ter-
anishi, R., Eds.; American Chemical Society: Washington, DC,
1993; pp 259-286.
(41) Jagella, T. Studies on the flavor and off-flavor of black and white
pepper. Ph.D. thesis, Technical University of Munich, 1999.
(42) Heiler, C.; Schieberle, P. Model studies on the precursors and
formation of the metallic smelling (E,Z)-2,6-nonadienol during
the manufacture and storage of buttermilk. Int. Dairy J. 1997,
7, 667-674.
(21) Guth, H.; Grosch, W. Quantification of potent odorants of virgin
olive oil by stable isotope dilution assay. J. Am. Oil Chem. Soc.
1993, 70, 513-518.
(22) Guth, H.; Grosch, W. Deterioration of soy-bean oil: quantifica-
tion of primary flavour compounds using a stable isotope dilution
assay. Lebensm. Wiss. Technol. 1990, 23, 513-522.
(23) Fuhrmann, E.; Grosch, W. Character impact odorants of the apple
cultivars Elstar and Cox orange. Nahrung 2002, 46, 187-193.
(24) Steinhaus, M.; Fritsch, H. T.; Schieberle, P. Quantitation of (R)-
and (S)-linalool in beer using a solid-phase microextraction
(SPME) in combination with a stable isotope dilution assay
(SIDA). J. Agric. Food Chem. 2003, 51, 7100-7105.
Received for review February 20, 2007. Revised manuscript received
April 17, 2007. Accepted April 17, 2007.
JF0705015