Characterization of the key aroma compounds in pink guawa (Psidium guajawa L.) by means of aroma Re-engineering experiments and Omission Tests

Article abstract of DOI:10.1021/jf803728n

Seventeen aroma-active volatiles, previously identified with high flavor dilution factors in fresh, pink Colombian guavas (Psidium guajava L), were quantified by stable isotope dilution assays. On the basis of the quantitative data and odor thresholds in

Full text of DOI:10.1021/jf803728n

2882 J. Agric. Food Chem. 2009, 57, 2882–2888
Characterization of the Key Aroma Compounds in
Pink Guava (Psidium guajava L.) by Means of Aroma
Re-engineering Experiments and Omission Tests
MARTIN STEINHAUS,† DIANA SINUCO,§ JOHANNES POLSTER,† CORALIA OSORIO,§
AND PETER SCHIEBERLE*,†
Deutsche Forschungsanstalt fu¨r Lebensmittelchemie, Lichtenbergstrasse 4, D-85748 Garching,
Germany, and Departamento de Química, Universidad Nacional de Colombia,
AA 14490, Bogota´, Colombia
Seventeen aroma-active volatiles, previously identified with high flavor dilution factors in fresh, pink
Colombian guavas (Psidium guajava L.), were quantified by stable isotope dilution assays. On the basis
of the quantitative data and odor thresholds in water, odor activity values (OAV; ratio of concentration to
odor threshold) were calculated. High OAVs were determined for the green, grassy smelling (Z)-3-hexenal
and the grapefruit-like smelling 3-sulfanyl-1-hexanol followed by 3-sulfanylhexyl acetate (black currant-
like), hexanal (green, grassy), ethyl butanoate (fruity), acetaldehyde (fresh, pungent), trans-4,5-epoxy-
(E)-2-decenal (metallic), 4-hydroxy-2,5-dimethyl-3(2H)-furanone (caramel, sweet), cinnamyl alcohol (floral),
methyl (2S,3S)-2-hydroxy-3-methylpentanoate (fruity), cinnamyl acetate (floral), methional (cooked potato-
like), and 3-hydroxy-4,5-dimethyl-2(5H)-furanone (seasoning-like). Studies on the time course of odorant
formation in guava puree or cubes, respectively, showed that (Z)-3-hexenal was hardly present in the
intact fruits, but was formed very quickly during crushing. The aroma of fresh guava fruit cubes, which
showed a very balanced aroma profile, was successfully mimicked in a reconstitute consisting of 13
odorants in their naturally occurring concentrations. Omission tests, in which single odorants were omitted
from the entire aroma reconstitute, revealed (Z)-3-hexenal, 3-sulfanyl-1-hexanol, 4-hydroxy-2,5-dimethyl-
3(2H)-furanone, 3-sulfanylhexyl acetate, hexanal, ethyl butanoate, cinnamyl acetate, and methional as
the key aroma compounds of pink guavas.
KEYWORDS: Pink guava; Psidium guajava L.; stable isotope dilution assay; aroma reconstitute; (Z)-3-
hexenal formation; omission tests
INTRODUCTION
dimethyl-2(5H)-furanone, trans-4,5-epoxy-(E)-2-decenal, and
methional, were reported for the first time as guava constituents.
The presence of 3-sulfanylhexyl acetate and 3-sulfanyl-1-
hexanol in guavas was very recently confirmed by Clery and
Hammond (2).
Guavas are the edible fruits of the guava tree Psidium guajaVa
L., native to the American tropics. However, although their
pleasant aroma, combining fruity, green, and tropical notes, is
mainly responsible for their popularity, systematic studies on
the key aroma compounds of guavas have been scarcely
performed. To fill this gap the aroma-active compounds of
Colombian pink guavas were recently screened by application
of the aroma extract dilution analysis (AEDA) (1). High FD
factors were found for 4-methoxy-2,5-dimethyl-3(2H)-furanone,
4-hydroxy-2,5-dimethyl-3(2H)-furanone, 3-sulfanylhexyl ac-
etate, and 3-sulfanyl-1-hexanol followed by 3-hydroxy-4,5-
dimethyl-2(5H)-furanone, (Z)-3-hexenal, trans-4,5-epoxy-(E)-
2-decenal, cinnamyl alcohol, ethyl butanoate, hexanal, methional,
and cinnamyl acetate. Among them, six odorants, namely,
3-sulfanylhexyl acetate, 3-sulfanyl-1-hexanol, 3-hydroxy-4,5-
To prove and rank the aroma contribution of individual aroma
compounds to the overall aroma of guavas, quantitative studies
and aroma re-engineering experiments are a necessary further
step. Thus, the aim of this study was to quantify the most odor-
active compounds recently characterized by AEDA, to re-
engineer the overall aroma based on the natural concentrations
in guava, and finally to elucidate the key aroma compounds by
means of omission tests.
MATERIALS AND METHODS
Fruits. Pink-fleshed guavas (P. guajaVa L.), variety Regional Roja,
were either from Puente Nacional (Santander, Colombia) and trans-
ported to Germany by air freight or of Colombian origin but purchased
at a local market in Munich, Germany.
Reference Odorants. The following compounds were purchased
from the commercial sources given in parentheses: 1 (Alfa Aesar,
* Corresponding author (e-mail Peter.Schieberle@ch.tum.de; tele-
phone +49 89 289 13265; fax +49 89 289 14183).
† Deutsche Forschungsanstalt fu¨r Lebensmittelchemie.
§ Universidad Nacional de Colombia.
10.1021/jf803728n CCC: $40.75  2009 American Chemical Society
Published on Web 03/02/2009

Key Aroma Compounds in Guava
J. Agric. Food Chem., Vol. 57, No. 7, 2009 2883
Figure 1. Structures of the deuterium- and 13C-labeled compounds used as internal standards in the stable isotope dilution assays: (9) positions of 13C atoms.
Karlsruhe, Germany); 2-7, 10, 11, 14-17 (Sigma-Aldrich Chemie,
Taufkirchen, Germany). Compounds 2, 8, 9, and 13 were synthesized
according to previously published procedures: 2, 8, and 9 (1); 13 (3).
(Z)-3-Hexenal (12) was synthesized by oxidation of (Z)-3-hexenol
according to the procedure reported in ref 1 for the synthesis of the
(E)-isomer.
Isotopically Labeled Odorants (Figure 1). The following com-
pounds were synthesized according to the literature given in parentheses:
d-14 (4); c-15 (5); c-16 (6); c-17 (7). Compounds d-1-d-9 and d-11-d-
13 were synthesized as detailed in the following section. C-10 was
purchased from Aldrich.
289 mg (11%); MS-EI, m/z (intensity in %) 55 (100), 41 (94), 61 (86),
58 (82), 56 (74), 57 (61), 102 (59), 69 (58), 85 (47), 42 (45), 101 (44),
83 (42), 47 (37), 43 (36), 39 (35), 68 (33), 84 (31), 89 (31), 75 (29),
59 (28), 45 (24), 136 (24); MS-CI (methanol), m/z (intensity in %)
119 (100; M + H+ - H2O).
[1,1-2H2]-3-Sulfanylhexyl Acetate (d-2). To a solution of [1,1-2H2]-
3-sulfanyl-1-hexanol (d-1; 0.52 mmol) in dichloromethane (10 mL)
was added dropwise acetyl chloride (1.31 mmol) in dichloromethane
(3 mL) at 0 °C. After 2 h of stirring, the solvent and the excess acetyl
chloride were removed by rotary evaporation.
The yield was determined by GC (9): 81 mg (87%); MS-EI, m/z
(intensity in %) 43 (100), 90 (42), 118 (40), 85 (39), 75 (32), 55 (25),
56 (23), 89 (19), 57 (18), 69 (18), 41 (16), 87 (15), 84 (14), 103 (12),
42 (11), 45 (10); MS-CI (methanol), m/z (intensity in %) 119 (100; M
+ H+ - H3C-COOH).
Syntheses. [1,1-2H2]-3-Sulfanyl-1-hexanol (d-1). Following the
reaction scheme used in ref 8 for the synthesis of [1,1-2H2]-3-sulfanyl-
2-methyl-1-pentanol from 2-methyl-2-pentenal, [1,1-2H2]-3-sulfanyl-
1-hexanol was prepared from (E)-2-hexenal as described below.
(a) 3-(Acetylthio)hexanal. In an argon atmosphere, thioacetic acid
(27 mmol) was slowly added to a mixture of (E)-2-hexenal (20 mmol)
and piperidine (0.25 mmol). After stirring for 18 h at ambient
temperature, diethyl ether (20 mL) was added and the mixture was
washed with hydrochloric acid (1 mol/L; 2 × 5 mL) followed by an
aqueous saturated sodium hydrogen carbonate solution (10 mL). After
drying over anhydrous sodium sulfate, the solvent was distilled off by
means of a Vigreux column.
[1,1-2H2]-Cinnamyl Alcohol (d-3). [1,1-2H2]-Cinnamyl alcohol was
prepared by reduction of ethyl cinnamate with lithium aluminum
deuteride following a procedure for the preparation of the unlabeled
compound (10). In an argon atmosphere, ethyl cinnamate (10 mmol)
was slowly added to lithium aluminum deuteride (5 mmol) at 0 °C
with stirring. After 20 min, hydrochloric acid (2 mol/L; 50 mL) was
added, followed by diethyl ether (50 mL). The organic phase was
separated, washed with an aqueous saturated sodium hydrogen carbon-
ate solution (10 mL), and dried over anhydrous sodium sulfate. The
solvent was removed by means of a Vigreux column, and the crude
product was purified by column chromatography to remove unreacted
ethyl cinnamate. The solution (1 mL) was applied onto a water-cooled
(12 °C) glass column (2 cm i.d.) filled with silica gel (25 g). The ester
was removed by elution with pentane/diethyl ether (90:10; 150 mL),
and [1,1-2H2]-cinnamyl alcohol was isolated with pentane/diethyl ether
(50:50; 150 mL). The solution was made up to 200 mL, and the
concentration of the target compound was determined by GC (9): yield,
1.11 g (82%); MS-EI, m/z (intensity in %) 93 (100), 136 (75), 92 (56),
78 (47), 91 (44), 79 (42), 117 (40), 106 (38), 77 (35), 94 (30), 107
(26), 51 (26), 80 (21), 116 (20), 135 (17), 118 (16), 104 (15), 119
(12), 63 (12), 103 (12), 39 (11), 105 (11); MS-CI (methanol), m/z
(intensity in %) 119 (100; M + H+ - H2O), 120 (9).
The intermediate showed the following mass spectrum (MS-EI): m/z
(intensity in %) 44 (100), 40 (16), 45 (12), 43 (8), 41 (6), 39 (5), 43
(5), 42 (4), 36 (4), 73 (4).
(b) 3-(Acetylthio)hexanoic Acid. 3-(Acetylthio)hexanal and ami-
dosulfonic acid (26 mmol) were dissolved in water/ethanol (2.5:1).
Sodium chlorite (27 mmol) in ethanol/water (1:1; 30 mL) was slowly
added, and the mixture was stirred for 5 h at ambient temperature. The
mixture was extracted with diethyl ether (2 × 80 mL), and the combined
etherial phases were filtered through a plug of silica gel and finally
dried over anhydrous sodium sulfate. The solvent was removed by
rotary evaporation.
The following mass spectrum (MS-EI) was obtained: m/z (intensity in
%) 43 (100), 45 (47), 70 (27), 44 (23), 130 (23), 55 (22), 97 (21), 41 (19),
69 (19), 89 (19), 73 (15), 87 (15), 60 (10), 71 (10), 88 (10).
(c) [1,1-2H2]-3-Sulfanyl-1-hexanol. 3-(Acetylthio)hexanoic acid was
dissolved in diethyl ether (5 mL) and dropwise added to a suspension
of lithium aluminum deuteride (30 mmol) in diethyl ether (25 mL)
under an argon atmosphere. After stirring of the refluxed mixture for
150 min, this was cooled to 0 °C, and saturated aqueous ammonium
chloride solution (15 mL) was slowly added, followed by hydrochloric
acid (2 mol/L; 10 mL). The organic phase was separated, washed with
a saturated aqueous sodium hydrogen carbonate solution (10 mL), dried
over anhydrous sodium sulfate, and made up to 100 mL. The
concentration of the target compound was determined by GC (9): yield,
[1,1-2H2]-Cinnamyl Acetate (d-4). Acetyl chloride (10 mmol)
dissolved in dichloromethane (5 mL) was dropwise added to a solution
of [1,1-2H2]-cinnamyl alcohol (4.1 mmol) in dichloromethane (5 mL)
at 0 °C with stirring. Stirring was continued for 2 h at room temperature,
and then the solvent and the excess acetyl chloride were removed by
rotary evaporation.
The yield was determined by GC (9): 469 mg (65%); MS-EI, m/z
(intensity in %) 43 (100), 116 (80), 117 (67), 135 (55), 136 (54), 119

2884 J. Agric. Food Chem., Vol. 57, No. 7, 2009
Steinhaus et al.
(49), 107 (37), 178 (34), 118 (34), 93 (27), 92 (12), 94 (12), 77 (12),
78 (10), 51 (8), 79 (8); MS-CI (methanol), m/z (intensity in %) 119
(100; M + H+- H3C-COOH).
2H4]-hexanal (54%); MS-EI, m/z (intensity in %) 44 (100), 59 (45), 60
(41), 43 (31), 45 (23), 57 (22), 46 (21), 58 (19), 61 (18), 41 (17), 76 (15),
42 (14), 47 (14); MS-CI (methanol), m/z (intensity in %) 87 (100; M +
H+ - H2O), 105 (93; M + H+), 86 (92; M + H+ - HDO).
[2,2,2-2H3]-Ethyl Butanoate (d-5). In a small tailor-made autoclave
(stainless steel, 80 mm, 10 mm i.d., 3 mm wall thickness) with screw
cap, PTFE sealing, and glass insert, butanoic acid (20 mmol) and [2,2,2-
2H3]-ethanol (2 mmol) were heated in the presence of sulfuric acid (50
µL) at 80 °C. After 30 min, the mixture was allowed to cool, and after
the addition of water (50 mL) followed by diethyl ether (80 mL), the
mixture was vigorously shaken. The aqueous phase was removed, and
the organic phase was washed with an aqueous sodium carbonate
solution (0.5 mol/L; 3 × 100 mL) and water (100 mL). After drying
over anhydrous sodium sulfate, the solution was made up to 100 mL.
The concentration of the target compound was determined by GC (9):
yield, 197 mg (83%); MS-EI, m/z (intensity in %) 71 (100), 91 (91),
43 (79), 41 (56), 61 (42), 42 (41), 74 (35), 48 (28), 39 (21), 72 (20),
104 (20), 73 (13), 90 (13); MS-CI (methanol), m/z (intensity in %)
120 (100; M + H+), 121 (10).
[5,5,6,6,6-2H5]-(Z)-3-Hexenal (d-12). 2-([5,5,6,6,6-2H5]-3-Hexyn-1-
yloxy)tetrahydro-2H-pyran was synthesized from 2-(3-butyn-1-yloxy)-
tetrahydro-2H-pyran and [2H5]-ethyl iodide and subsequently hydrolyzed
into [5,5,6,6,6-2H5]-3-hexyn-1-ol (19). Using a Lindlar catalyst modified
according to ref 20, [5,5,6,6,6-2H5]-3-hexyn-1-ol was hydrated to yield
[5,5,6,6,6-2H5]-(Z)-3-hexen-1-ol, which was finally converted into
[5,5,6,6,6-2H5]-(Z)-3-hexenal using Dess-Martin-periodinane (13).
(a) 2-([5,5,6,6,6-2H5]-3-Hexyn-1-yloxy)tetrahydro-2H-pyran. Butyl
lithium (15 mmol; 1.5 mL of 10 mmol/mL solution in THF) was added
to 2-(3-butyn-1-yloxy)tetrahydro-2H-pyran (10 mmol) in anhydrous
THF (20 mL) under argon at 0 °C and stirred for 1 h. Then, [2H5]-
ethyl iodide (25 mmol) was added. After 4 h of stirring, an aqueous
saturated ammonium chloride solution (10 mL) was added, and the
mixture was extracted with diethyl ether (3 × 50 mL). The combined
organic phases were washed with an aqueous saturated sodium chloride
solution (2 × 10 mL) and dried over anhydrous sodium sulfate to yield
2-([5,5,6,6,6-2H5]-3-hexyn-1-yloxy)tetrahydro-2H-pyran: MS-EI, m/z
(intensity in %) 85 (100), 86 (17), 41 (9), 83 (9), 57 (8), 67 (8).
(b) [5,5,6,6,6-2H5]-3-Hexyn-1-ol. The solution containing 2-([5,5,6,6,6-
2H5]-3-hexyn-1-yloxytetrahydro-2H-pyran was concentrated to 5 mL, and
methanol (100 mL) and p-toluenesulfonic acid monohydrate (5 mmol) were
added. After 2 h of stirring, diethyl ether (100 mL) and water (200 mL)
were added. The aqueous phase was removed, and the organic phase was
dried over anhydrous sodium sulfate and concentrated by means of a
Vigreux column. The [5,5,6,6,6-2H5]-3-hexyn-1-ol obtained was purified
by column chromatography by applying the solution (1 mL) onto a cooled
(12 °C) glass column (1 cm i.d.) filled with silica gel (7 g). Elution was
performed with pentane (50 mL), followed by pentane/diethyl ether (90:
10, 50 mL), pentane/diethyl ether (70:30, 50 mL), pentane/diethyl ether
(50:50, 50 mL), and finally diethyl ether (50 mL). The [5,5,6,6,6-2H5]-3-
hexyn-1-ol was eluted between 190 and 230 mL: MS-EI, m/z (intensity in
%): 73 (100), 72 (35), 71 (31), 55 (16), 103 (14), 56 (13), 44 (12), 45
(10), 57 (10), 42 (9), 41 (8), 74 (8).
Methyl [2H5]-Benzoate (d-6). Following a general procedure for ZnO-
catalyzed O-acylations (11), [2H5]-benzoyl chloride (1.8 mmol) was
added dropwise to a mixture of methanol (0.5 mL) and zinc oxide (0.62
mmol). After stirring for 30 min at 40 °C, the mixture was extracted
with dichloromethane (2 × 25 mL). The combined organic phases were
washed with an aqueous sodium hydrogencarbonate solution (10%; 3
× 60 mL) and dried over anhydrous sodium sulfate. Methyl [2H5]-
benzoate was obtained after rotary evaporation.
The yield was determined by GC (9): 199 mg (78%); MS-EI, m/z
(intensity in %) 110 (100), 82 (90), 52 (83), 141 (29), 54 (18), 52 (6),
111 (6), 83 (3), 142 (2); MS-CI (methanol), m/z (intensity in %) 142
(100; M + H+).
Ethyl [2H5]-Benzoate (d-7). This compound was synthesized as
detailed above for the methyl ester using ethanol instead of methanol:
yield, 183 mg (66%); MS-EI, m/z (intensity in %) 110 (100), 82 (40),
127 (127), 54 (15), 155 (12), 111 (7), 109 (4), 52 (4), 83 (3), 128 (2);
MS-CI (methanol), m/z (intensity in %), 156 (100; M + H+).
[2H3]-Methyl (2R,3S)-2-Hydroxy-3-methylpentanoate (d-8). The
compound was synthesized from L-isoleucine (0.45 mmol) and [2H4]-
methanol following the procedure described for the synthesis of the
unlabeled compound (1).
The yield was determined by GC-FID using butyl lactate as the
internal standard: yield, 15 mg (24%); MS-EI, m/z (intensity in %) 93
(100), 87 (69), 45 (69), 36 (43), 41 (40), 57 (28), 69 (24), 36 (17), 40
(13), 38 (12), 43 (11), 39 (10), 58 (8); MS-CI (methanol), m/z (intensity
in %) 150 (100; M + H+).
(c) [5,5,6,6,6-2H5]-(Z)-3-Hexen-1-ol. After removal of the solvent,
[5,5,6,6,6-2H5]-3-hexyn-1-ol was dissolved in pentane (25 mL) contain-
ing quinoline (200 µL) and was hydrated for 2 h in hydrogen
atmosphere using a Mn-modified Lindlar catalyst (20 mg) (20). The
quinoline was removed by column chromatography (for parameters see
[5,5,6,6,6-2H5]-3-hexyn-1-ol above). Pure [5,5,6,6,6-2H5]-(Z)-3-hexen-
1-ol was isolated in the effluent between 130 and 140 mL: MS-EI, m/z
(intensity in %) 105 (100), 86 (94), 43 (91), 74 (75), 57 (59), 71 (54),
69 (47), 45 (46), 70 (46), 68 (44), 60 (43), 42 (36), 87 (28), 41 (26),
85 (24), 58 (24), 72 (21), 40 (21), 59 (20), 46 (20), 47 (19), 56 (18),
39 (16), 55 (13).
[2H3]-Methyl (2S,3S)-2-hydroxy-3-methylpentanoate (d-9) was syn-
thesized accordingly from D-allo-isoleucine: yield, 21 mg (30%); MS-
EI, m/z (intensity in %) 93 (100), 87 (57), 45 (55), 41 (36), 36 (28), 57
(26), 69 (21), 36 (16), 38 (8), 43 (8), 58 (8), 40 (8), 39 (8); MS-CI
(methanol), m/z (intensity in %) 150 (100; M + H+).
(d) [5,5,6,6,6-2H5]-(Z)-3-Hexenal. After removal of the solvent,
[5,5,6,6,6-2H5]-(Z)-3-hexen-1-ol was dissolved in dichloromethane (30
mL) and oxidized with 1,1,1-tris(acetyloxy)-1,1-dihydro-1,2-benzio-
doxol-3-(1H)-one as described for (E)-3-hexenal (1). The concentration
of the analyte was determined by GC (9) using (E)-2-hexenal as the
internal standard: yield, 7.5 mg (0.8%); MS-EI, m/z (intensity in %)
44 (100), 43 (90), 74 (83), 45 (40), 46 (39), 57 (35), 42 (34), 103 (30),
85 (26), 41 (23), 60 (22), 40 (21), 39 (16), 47 (13), 72 (13), 84 (10),
56 (10), 55 (10), 73 (10), 58 (10); MS-CI (methanol), m/z (intensity in
%) 85 (100; M + H+ - HDO), 84 (35; M + H+ - D2O), 86 (17; M
+ H+ - H2O), 104 (17; M + H+).
[7,7,8,8-2H4]-trans-4,5-Epoxy-(E)-2-decenal (d-13) was synthesized
from [3,3,4,4-2H4]-hexanal following the procedure described in ref
18. [3,3,4,4-2H4]-Hexanal was synthesized from 3-hexyn-1-ol using the
same approach as applied for the synthesis of [5,5,6,6-2H4]-hexanal
(d-11) from 5-hexyn-1-ol.
[5,5,6,6-2H4]-Hexanal (d-11). Using Wilkinson catalyst, 5-hexyn-
1-ol was deuterated to obtain [5,5,6,6-2H4]-hexanol, which was
subsequently oxidized into [5,5,6,6-2H4]-hexanal using Dess-Martin-
periodinane (12, 13).
(a) [5,5,6,6-2H4]-Hexanol. Tris(triphenylphosphin)rhodium(I) chloride
(0.27 mmol) in toluene (15 mL) was stirred in a deuterium atmosphere
until the deep red suspension turned orange. 5-Hexyn-1-ol (25 mmol) in
toluene (15 mL) was added, and the brownish solution was stirred until
complete conversion. After dilution with pentane (50 mL), the toluene
was removed by column chromatography. To achieve this, the reaction
mixture was applied onto silica gel (20 g) filled into a glass column (1.5
cm i.d.), and after flushing with pentane (200 mL), the [5,5,6,6-2H4]-hexanol
was eluted with diethyl ether (100 mL). Residual catalyst was removed
by SAFE distillation (14), and the solvent was evaporated to yield [5,5,6,6-
2H4]-hexanol in a purity of 99.8%.
(b) [5,5,6,6-2H4]-Hexanal. [5,5,6,6-2H4]-Hexanol was oxidized into the
aldehyde as detailed in (1) for the oxidation of (E)-3-hexenol into (E)-3-
hexenal. The resulting solution was made up to 200 mL, and the
concentration of the target compound was determined by GC (9). As
reported earlier (15-17), the use of tris(triphenylphosphin)rhodium(I)
chloride avoided hydrogen-deuterium scrambling during deuteration,
frequently observed in heterogeneous catalysis (18). Yield, 1.34 g [5,5,6,6-
Quantitation of Aroma Compounds by Stable Isotope Dilution
Assays. Whole fruits were blended using a stainless steel blender, and
the puree was allowed to stand for exactly 5 min. Then, the labeled
standards (0.2-20 µg) dissolved in dichloromethane (50-1000 mL)
were added to portions of the puree (0.5-200 g) and further
homogenized. With continuous blending and cooling in an ice bath,
anhydrous sodium sulfate was added until the mixture became powdery

Key Aroma Compounds in Guava
J. Agric. Food Chem., Vol. 57, No. 7, 2009 2885
Table 1. Mass Fragments and Calibration Factors Used in the Stable
Isotope Dilution Assays for the Quantitation of 17 Aroma Compounds in
Guavas
stalt für Lebensmittelchemie and following the ASTM procedure for
the determination of odor and taste thresholds by a forced-choice
ascending concentration series method of limits in triangular tests
(22).
Aroma Reconstitution. Appropriate amounts (20-100 µL) of
aqueous or ethanolic stock solutions of the odorants were mixed and
made up to 1 L with water to yield the same concentrations as
determined in the guavas. Final ethanol concentration was kept below
1 g/L, that is, below the odor threshold of ethanol.
Aroma Profile Tests. Samples (20 g) were placed into cylindrical
ground neck glasses (height ) 7 cm; i.d. ) 3.5 cm) with lids and were
orthonasally evaluated by the sensory panel. Descriptors used were
determined in preliminary sensory experiments. Each descriptor used
was defined on the basis of the odor of a reference compound dissolved
in water at a concentration of 100 times above the respective threshold
value. Reference odorants used in the experiments were ethyl butanoate
(fruity), 4-hydroxy-2,5-dimethyl-3(2H)-furanone (caramel, sweet), cin-
namyl alcohol (flowery), trans-4,5-epoxy-(E)-2-decenal (metallic),
3-hydroxy-4,5-dimethyl-2(5H)-furanone (seasoning-like), (Z)-3-hexenal
(grassy), acetaldehyde (fresh), and 3-sulfanyl-1-hexanol (grapefruit).
Assessors were asked to rate each descriptor in the samples presented
on a seven -point scale from 0 to 3, with 0 ) not detectable, 1 )
weak, 2 ) moderate, and 3 ) strong.
m/zb
isotopic
internal response
no.
odorant
3-sulfanyl-1-hexanol
3-sulfanylhexyl acetate
cinnamyl alcohol
cinnamyl acetate
ethyl butanoate
methyl benzoate
ethyl benzoate
methyl (2R,3S)-2-
hydroxy-3-methylpentanoate
methyl (2S,3S)-2-
labelinga analyte standard factorc
1
2
3
4
5
6
7
8
2H2
2H2
2H2
2H2
2H3
2H5
2H5
2H3
117 119
117 119
117 119
117 119
117 120
137 142
151 156
147 150
0.98
0.99
0.90
0.89
0.97
0.83
0.67
0.98
9
2H3
147 150
0.98
hydroxy-3-methylpentanoate
10 acetaldehyded
11 hexanal
12 (Z)-3-hexenal
13C2
2H4
2H5
2H4
2H3
2H3
13C2
13C2
45
83
81
47
87
84-86
0.67
0.54
0.98
0.93
1.03
0.91
1.00
1.03
13 trans-4,5-epoxy-(E)-2-decenal
14 methional (3-(methylthio)propanal)
15 4-methoxy-2,5-dimethyl-3(2H)-furanone
16 4-hydroxy-2,5-dimethyl-3(2H)-furanone
17 3-hydroxy-4,5-dimethyl-2(5H)-furanone
139 143
105 108
143 146
129 131
129 131
Omission Tests. Models were prepared as detailed above. In
triangular tests with forced choice, models in which one odorant was
omitted were evaluated against two samples of the complete model.
a Labeling in the isotopologues used as internal standards in the stable isotope
dilution assays. b Mass traces obtained by GC-MS(CI) used for peak area evaluation
of analyte and standard, respectively. c Response factor determined from reference
mixtures of analyte and standard as described in ref 9. d Acetaldehyde was
previously not found during AEDA (1)1, but was detected as an aroma-active
compound during headspace analysis of Colombian pink guavas.
RESULTS AND DISCUSSION
Concentrations of the Most Aroma-Active Compounds in
Guavas. Sixteen odorous compounds that had shown high FD
factors in the previous study (1), namely, 4-methoxy-2,5-dimethyl-
3(2H)-furanone, 3-sulfanylhexyl acetate, 3-sulfanyl-1-hexanol, 4-hy-
droxy-2,5-dimethyl-3(2H)-furanone, 3-hydroxy-4,5-dimethyl-2(5H)-
furanone, (Z)-3-hexenal, trans-4,5-epoxy-(E)-2-decenal, cinnamyl
alcohol, ethyl butanoate, hexanal, methional, cinnamyl acetate,
methyl benzoate, methyl (R,S)-2-hydroxy-3-methylpentanoate, ethyl
benzoate, and methyl (S,S)-2-hydroxy-3-methylpentanoate, were
quantified by stable isotope dilution assays. In addition, acetalde-
hyde, which was found in this study as an additional aroma-active
compound during static headspace GC sniffing analyses applied
on a fresh pink Colombian guava puree (data not shown), was
included in the investigations.
and the supernatant organic phase became clear. Then, the mixture was
filtered through defatted cotton wool, and the dichloromethane extract
was submitted to a SAFE distillation (14) at 40 °C.
To measure the influence of time on enzymatic aroma compound
formation, samples (1 g) were allowed to stand for 24-600 s (puree) and
for 10-600 s (cubes). Then, a saturated aqueous calcium chloride solution
(1 mL) was added. To obtain a zero time value, a whole fruit was
submersed in calcium chloride solution and cutting was performed below
the surface of the solution. Further workup was done as described above.
Depending on the amount of standard added, SAFE distillates were
concentrated to a volume of 0.1-10 mL by means of a Vigreux column
and a microdistillation apparatus (21). Aliquots of the concentrates (0.5-4
µL) were analyzed by means of two-dimensional GC-GC-MS. The system
consisted of a Combi PAL autosampler (CTC Analytics, Zwingen,
Switzerland), a Trace Ultra GC (Thermo Scientific, Dreieich, Germany),
a CP 3800 GC, and a Saturn 2200 mass spectrometer (Varian, Darmstadt,
Germany). The Trace GC was equipped with a cold-on-column injector,
an FFAP capillary (30 m × 0.32 mm i.d., 0.25 µm film thickness), a
moving column stream switching system, and an FID (Thermo Scientific,
Dreieich, Germany). The moving column stream switching system was
connected to the CP 3800 via an uncoated but deactivated fused silica
transfer line (0.32 mm i.d) in a heated (250 °C) hose (Horst, Lorsch,
Germany). The GC hosted a cold trap (SGE, Griesheim, Germany) and a
DB-1701 column (30 m × 0.25 mm i.d., 0.25 µm film thickness), which
was connected to the mass spectrometer. Mass chromatograms were
recorded in the MS-CI mode using methanol as the reactant. Analyte
concentrations were calculated from the area counts of characteristic mass
traces of analyte and standard (Table 1) (9).
Quantitation of Acetaldehyde. Acetaldehyde was determined by
stable isotope dilution analysis of static headspace samples. The guava
puree (10 g) was diluted with water (50 mL) in a septum-sealed Erlenmeyer
flask (200 mL), and [13C2]-acetaldehyde (100 µg) in water (1 mL) was
added. After equilibration (30 min), aliquots of the headspace (5 mL) were
withdrawn using a gastight syringe and injected onto a DB-5 capillary
(25 m × 0.32 mm i.d., 1.2 µm film) held at 0 °C. The column was
connected to the mass spectrometer Incos XL (Finnigan MAT, Bremen,
Germany) running in the CI mode (methane, 115 eV).
For each of the 17 odorants selected (Table 1) a stable
isotopologue was synthesized (Figure 1) and used as internal
standard in the quantitations.
The results of the quantitation experiments (Table 2) revealed
concentrations ranging from the lower micrograms per kilogram
area to several milligrams per kilogram. In particular, high
amounts were found for (Z)-3-hexenal (6.9 mg/kg), acetaldehyde
(2.4 mg/kg), 4-hydroxy-2,5-dimethyl-3(2H)-furanone (1.4 mg/
kg), and cinnamyl alcohol (1.3 mg/kg), whereas methional (1.3
µg/kg) was 1000 times lower. To estimate the aroma potency
of the individual guava odorants, their concentrations were
correlated with the respective odor thresholds using the odor
activity value (OAV) concept (23).
The green grassy smelling (Z)-3-hexenal showed the highest
OAV (Table 3), exceeding its threshold by a factor of 57000.
Second in rank was the grapefruit-like smelling 3-sulfanyl-1-
hexanol with an OAV of 9300. High OAVs were also calculated
for 3-sulfanylhexyl acetate (black currant-like; OAV 570), hexanal
(green, grassy; OAV 360), and ethyl butanoate (fruity; OAV 170).
Somewhat lower OAVs were determined for acetaldehyde (fresh,
pungent), trans-4,5-epoxy-(E)-2-decenal (metallic), 4-hydroxy-2,5-
dimethyl-3(2H)-furanone (caramel, sweet), cinnamyl alcohol (flo-
ral), and methyl (2S,3S)-2-hydroxy-3-methylpentanoate (fruity). On
the other hand, the concentrations of cinnamyl acetate, methional,
Odor Thresholds. Odor thresholds were determined using a panel
of 15-20 trained panelists recruited from the Deutsche Forschungsan-

2886 J. Agric. Food Chem., Vol. 57, No. 7, 2009
Steinhaus et al.
Table 2. Concentrations of Aroma-Active Compounds in Colombian Pink
Guavas
concn
no. of
odorant
(Z)-3-hexenal
acetaldehyde
4-hydroxy-2,5-dimethyl-
3(2H)-furanone
cinnamyl alcohol
cinnamyl acetate
hexanal
3-sulfanyl-1-hexanol
ethyl butanoate
methyl (2S,3S)-2-
hydroxy-3-methylpentanoate
trans-4,5-epoxy-(E)-2-decenal
3-sulfanylhexyl acetate
methyl (2R,3S)-2-
hydroxy-3-methylpentanoate
4-methoxy-2,5-
(µg/kg)
replicates
SD (%)
6890
2440
1420
5
3
5
4
45
4
1290
862
857
555
126
27.2
5
5
3
2
3
3
11
9
6
2
13
12
16.8
11.4
9.7
3
2
3
24
5
16
Figure 2. Aroma profile of the Colombian pink guava puree model (gray)
in comparison with the original fruit puree (black).
9.4
7
8
prepared from fresh guavas by a sensory panel in an aroma
profile test. Despite the simplified matrix not including any
nonvolatile guava constituents, such as carbohydrates, acids, or
pectin, the results (Figure 2) showed a good agreement between
aroma model solution and the original guava puree. Both were
characterized by a strong green, grassy note, a moderate
grapefruit-like, some fruity and fresh notes, and rather weak
sweet, flowery, and metallic notes.
dimethyl-3(2H)-furanone
ethyl benzoate
3-hydroxy-4,5-
dimethyl-2(5H)-furanone
methyl benzoate
methional
7.2
4.3
2
5
0
39
2.6
1.3
3
3
14
12
Table 3. Orthonasal Odor Thresholds and Odor Activity Values (OAV) of
Aroma-Active Compounds in Guavas
Changes during Fruit Tissue Disruption. Despite the good
agreement of the aroma profiles of the model and the guava
puree, the panelists suggested that the green grassy note was
too predominant when compared to the perception detected
during guava fruit consumption. According to the OAVs, (Z)-
3-hexenal should undoubtedly be responsible for the green,
grassy odor note. The odorant is known to be formed upon plant
tissue disruption from R-linolenic acid (24) via the enzymatically
formed 13-(S)-hydroperoxide, which is finally cleaved by a
hydroperoxide lyase to yield (Z)-3-hexenal. Therefore, it might
be suspected that the homogenization of the guavas before
workup was responsible for the high (Z)-3-hexenal concentration
and, thus, for the high intensity of the green odor note in the
resulting puree. On the contrary, during consumption of whole
fruits (Z)-3-hexenal formation might be lower, either due to the
smaller time window of modification or a smaller degree of
tissue disruption during chewing. To follow these considerations,
the time course of (Z)-3-hexenal formation in a freshly prepared
guava puree in comparison to fruit cubes of approximately 0.5
cm edge length derived from the same fruit was determined.
To stop enzymatic reactions, the samples were poured into an
aqueous saturated calcium chloride solution, which is known
to inhibit the enzymes of the lipoxygenase pathway (25).
odor
threshold
odorant
quality
(µg/kg of water)
OAVa
(Z)-3-hexenal
grassy
0.12
0.06
0.02
2.4
0.76
25
57000
9300
570
360
170
98
3-sulfanyl-1-hexanolb
3-sulfanylhexyl acetateb
hexanal
grapefruit
black currant
grassy
ethyl butanoate
acetaldehyde
trans-4,5-epoxy-(E)-2-decenal
4-hydroxy-2,5-
dimethyl-3(2H)-furanone
cinnamyl alcohol
methyl (2S,3S)-2-
hydroxy-3-methylpentanoate
cinnamyl acetate
methional
3-hydroxy-4,5-
dimethyl-2(5H)-furanone
methyl (2R,3S)-2-
hydroxy-3-methylpentanoate
ethyl benzoate
4-methoxy-2,5-
dimethyl-3(2H)-furanone
methyl benzoate
fruity
fresh, pungent
metallic
caramel, sweet
0.22
40
76
36
floral
fruity
77
2.4
17
11
floral
cooked potato
seasoning-like
150
0.43
1.7
5.7
3.1
2.5
fruity
13
0.7
violet, floral
caramel, sweet
53
160
, 1
, 1
violet, floral
73
, 1
a OAV ) odor activity value ) concentration divided by threshold. b Because
enantiomeric distribution in guavas was close to 1:1, a racemic mixture was used
for threshold determination.
and 3-hydroxy-4,5-dimethyl-2(5H)-furanone were only slightly
above their threshold values. Because the OAVs of methyl (2R,3S)-
2-hydroxy-3-methylpentanoate, ethyl benzoate, 4-methoxy-2,5-
dimethyl-3(2H)-furanone, and methyl benzoate were below 1, these
compounds are assumed not to contribute to the overall aroma.
Aroma Reconstitution Experiments. It is a well-known
phenomenon that the overall aroma of a mixture of odor-active
compounds cannot be predicted. The reason for this is the fact
that the signals caused by the single interactions of the odorants
with the respective olfactory receptor neurons are combined by
the brain to generate the overall aroma perception. Thus, the
re-engineering of the aroma by using the natural concentrations
of the single odorants in a so-called “aroma reconstitute” is the
only available method to confirm that the identifications and
quantitations have led to the original blueprint of an aroma.
Thus, an aqueous solution containing the 13 odorants found to
exceed their respective thresholds (cf. Table 3) in the concentra-
tions determined (cf. Table 2) was compared to a fruit puree
The results showed clear differences between (Z)-3-hexenal
formation in the puree and in the cubes, respectively (Figure
3). As expected, the zero value, representing the (Z)-3-hexenal
content in the intact fruit tissue, was quite low. However, in
the puree, the (Z)-3-hexenal concentration quickly increased to
21 mg/kg after 24 s, just the time span needed for blending the
guavas. Upon standing, (Z)-3-hexenal concentration in the puree
showed a further increase, reaching 42 mg/kg after 10 min.
Compared to the first quantitations (Table 2), showing con-
centrations in the puree of 6.9 mg/kg, in this experiment (Z)-
3-hexenal formation was considerably higher, yielding almost
28 mg after 5 min.
A possible explanation for this observation could be a
different ripening history of the two fruit batches. Fruits used
to obtain the values listed in Table 2 were picked in Colombia
in a full ripe state and transported to Germany by air, whereas
the fruit used for the time course was purchased in Germany

Key Aroma Compounds in Guava
J. Agric. Food Chem., Vol. 57, No. 7, 2009 2887
Figure 4. Aroma profile of the Colombian pink guava cubes model (gray)
in comparison with the original fruit cubes (black).
droxy-2,5-dimethyl-3(2H)-furanone, cinnamyl alcohol, methyl
(2S,3S)-2-hydroxy-3-methylpentanoate, cinnamyl acetate, me-
thional, and 3-hydroxy-4,5-dimethyl-2(5H)-furanone was com-
pared in an aroma profile test to an original fruit sample, which
consisted of guava cubes. The concentrations of (Z)-3-hexenal
and hexanal were reduced as indicated in Table 4, but all other
compounds were used as given in Table 2. The results (Figure
4) again revealed a good match between the aroma model
solution and the fruit sample. As expected, the grassy note was
now assessed much lower as compared to the aroma profile of
the puree. In contrast, the grapefruit, fruity, and sweet notes
were rated higher. Obviously these notes were suppressed to a
certain degree in the puree and the reconstitute thereof by the
overwhelming green, grassy odor of (Z)-3-hexenal. Generally,
the orthonasal profiles of the guava fruit cubes and its
reconstituted model were more balanced than those of the puree
samples and were close to the aroma perception during
consumption of guava fruits. However, our approach to the use
of fruit cubes as sample for the quantitation of the aroma-active
C6-aldehydes was somewhat empirical. Therefore, further
investigations into the release behavior of these compounds
during fruit consumption, for example, by online measurements
during chewing, will be performed.
Omission Tests. Omission tests can be applied to assess the
contribution of individual odorants to the overall aroma (26)
and, thus, to define the key aroma compounds. In the present
case, omission tests were performed using the aroma reconstitute
of the guava cubes. In 13 independent experiments all 13
odorants were singly omitted from the model, and these
incomplete models were evaluated against the complete model
in orthonasal triangular tests by a sensory panel.
The results (Table 5) revealed significant differences in the
overall aroma when either (Z)-3-hexenal, 3-sulfanyl-1-hexanol,
4-hydroxy-2,5-dimethyl-3(2H)-furanone, 3-sulfanylhexyl ac-
etate, hexanal, ethyl butanoate, cinnamyl acetate, or methional,
respectively, was missing in the model. The most noticeable
aroma differences (level of significance ) 0.1%) were perceived
when (Z)-3-hexenal or 3-sulfanyl-1-hexanol was omitted, which
was not surprising because these compounds also showed by
far the highest OAVs. Omission of 4-hydroxy-2,5-dimethyl-
3(2H)-furanone, 3-sulfanylhexyl acetate, and hexanal was
detected at a level of significance of 1%, also corresponding to
their high OAVs. However, the clear perception of a difference
after omission of hexanal in the presence of an outstanding
amount of (Z)-3-hexenal is somewhat unclear. Yet at a level of
significance of 5%, the omission of ethyl butanoate, cinnamyl
acetate, and methional was detected, indicating also an essential
aroma contribution of these odorous compounds. In contrast,
omitting cinnamyl alcohol, trans-4,5-epoxy-(E)-2-decenal, ac-
Figure 3. Time course of (Z)-3-hexenal formation in guava puree (A)
and guava cubes (B) derived from the same fruit.
Table 4. Concentrations and Odor Activity Values (OAV) of Aroma-Active
C6-Aldehydes in Guava Cubes
odorant
concn (µg/kg)
threshold (µg/kg of water)
OAVa
(Z)-3-hexenal
hexanal
(E)-2-hexenal
1630
652
67
0.12
2.4
110
14000
270
0.6
a OAV ) odor activity value (concentration divided by odor threshold).
from a commercial source and therefore might have been
harvested at an earlier stage of maturity.
In comparison to the puree, (Z)-3-hexenal formation in the
cubes was by far lower. As in the puree, the initial increase
was fast, resulting in 1.3 mg/kg (Z)-3-hexenal after 10 s. A
further increase was then observed until a maximum of 2.2 mg/
kg was reached after 2 min. Then, the (Z)-3-hexenal concentra-
tion slowly decreased to 1.4 mg/kg after 10 min. These
experiments clearly showed that (Z)-3-hexenal formation taking
place upon crushing of guava fruits is a very rapid process and
that the final (Z)-3-hexenal concentrations are highly dependent
on the degree of tissue disruption.
Aroma Reconstitution Based on Data for Fruit Cubes. The
aroma reconstitution experiment was repeated, but using the
concentrations for (Z)-3-hexenal and hexanal obtained from
the fruit cubes after 5 min (Table 4) instead of the comparably
higher values found in the puree (Table 2). Despite the clearly
lower concentration, (Z)-3-hexenal was still the compound with
the highest odor activity value (14000) and by far more potent
than hexanal, exhibiting the same green, grassy odor charac-
teristics, but at a substantially lower OAV of 270. (E)-2-Hexenal,
which is formed from (Z)-3-hexenal by an enzymatic isomer-
ization reaction (24), was not present in aroma-active amounts
(OAV < 1) and was therefore omitted.
An aqueous solution containing the 13 odorants (Z)-3-hexenal,
3-sulfanyl-1-hexanol, 3-sulfanylhexyl acetate, hexanal, ethyl
butanoate, acetaldehyde, trans-4,5-epoxy-(E)-2-decenal, 4-hy-

2888 J. Agric. Food Chem., Vol. 57, No. 7, 2009
Steinhaus et al.
Table 5. Results of Omission Tests Applied on the Aroma Reconstitute of
Guava Cubes
in other Allium varieties using a stable isotope dilution assay. J.
Agric. Food Chem. 2004, 52, 2797–2802.
(9) Steinhaus, M.; Fritsch, H. T.; Schieberle, P. Quantitation of (R)-
and (S)-linalool in beer using solid phase microextraction (SPME)
in combination with a stable isotope dilution assay (SIDA). J.
Agric. Food Chem. 2003, 51, 7100–7105.
(10) Klemm, L. H.; Hsu Lee, D.; Gopinath, K. W.; Klopfenstein, C. E.
Intramolecular Diels-Alder reactions. III. Cyclizations of trans-
cinnamyl and phenylpropargyl phenylpropionates. J. Org. Chem.
1966, 31, 2376–2380.
(11) Tamaddon, F.; Amrollahi, M. A.; Sharafat, L. A green protocol
for the chemoselective O-acylation in the presence of zinc oxide
as a heterogeneous, re-usable and eco-friendly catalyst. Tetrahe-
dron Lett. 2005, 46, 7841–7844.
(12) Young, J. F.; Osborn, J. A.; Jardine, F. H.; Wilkinson, G. Hydride
intermediates in homogenous hydrogenation reactions of olefins
and acetylenes using rhodium catalysts. Chem. Commun. 1965,
73, 131–132.
odorant
omitted
no. of
correct
level of
panelists
answers
significance (%)
(Z)-3-hexenal
3-sulfanyl-1-hexanol
4-hydroxy-2,5-
dimethyl-3(2H)-furanone
3-sulfanylhexyl acetate
hexanal
ethyl butanoate
cinnamyl acetate
methional
cinnamyl alcohol
trans-4,5-epoxy-
(E)-2-decenal
19
16
15
16 ) 84%
12 ) 75%
11 ) 73%
0.1
0.1
1
14
16
14
16
16
14
19
10 ) 71%
11 ) 69%
9 ) 64%
10 ) 63%
9 ) 56%
7 ) 50%
8 ) 42%
1
1
5
5
5
nsa
ns
acetaldehyde
16
16
4 ) 25%
4 ) 25%
ns
ns
methyl (2S,3S)-2-
hydroxy-3-methylpentanoate
3-hydroxy-4,5-
19
3 ) 16%
ns
dimethyl-2(5H)-furanone
(13) Dess, D. B.; Martin, J. C. Readily accessible 12-I-5 oxidant for
the conversion of primary and secondary alcohols to aldehydes
and ketones. J. Org. Chem. 1983, 48, 4155–4156.
a Not significant.
etaldehyde, methyl (2S,3S)-2-hydroxy-3-methylpentanoate, or
3-hydroxy-4,5-dimethyl-2(5H)-furanone did not result in a
significant aroma change, although, for example, acetaldehyde
and trans-4,5-epoxy-(E)-2-decenal showed rather high OAVs.
Obviously, in the complex mixture, their aroma is somewhat
suppressed.
In conclusion, the omission tests clearly elucidated the key
aroma compounds of guava fruits. (Z)-3-Hexenal and, to a lower
extent, hexanal are responsible for the green, grassy odor note,
whereas 3-sulfanyl-1-hexanol and 3-sulfanylhexyl acetate ac-
count for the sulfury, tropical note and 4-hydroxy-2,5-dimethyl-
3(2H)-furanone, ethyl butanoate, and cinnamyl acetate predomi-
nantly make up the sweet, fruity, and flowery odor characteristics.
(14) 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.
(15) Osborn, J. A.; Jardine, F. H.; Young, J. F.; Wilkinson, G. The
preparation and properties of tris (triphenylphosphine) halogeno
rhodium(I) and some reactions thereof including catalytic ho-
mogenous hydrogenation of olefines and acetylenes and their
derivates. J. Chem. Soc. (A) 1966, 1711–1732.
(16) Birch, A. J.; Walker, K. A. M. Specific deuteration of some
unsaturated compounds. Tetrahedron Lett. 1966, 4939–4940.
(17) Morandi, J. R.; Jensen, H. B. Homogenous catalytic deuteration
of olefinic double bonds. J. Org. Chem. 1968, 34, 1889–1891.
(18) Guth, H.; Grosch, W. Deterioration of soya-bean oil: quantification
of primary flavour compounds using a stable isotope dilution
assay. Lebensm. Wiss. Technol. 1990, 23, 513–522.
ACKNOWLEDGMENT
(19) Brenstrum, T.; Brimble, M. A.; Rush, C. J.; Shaw, G. J.; Reid,
D. C. W. Syntheses of deuterium labelled hexenols. J. Labelled
Compd. Radiopharm 1993, 31, 629–635.
We thank Anja Mialki and Magdalena Uzunova for their
skillful technical assistance.
(20) Rajaram, J.; Narula, A. P. S.; Chawla, H. P. S.; Sukh, D.
Semihydrogenation of acetylenes. Modified Lindlar catalyst.
Tetrahedron 1983, 39, 2315–2322.
LITERATURE CITED
(1) Steinhaus, M.; Sinuco, D.; Polster, J.; Osorio, C.; Schieberle, P.
Characterization of the aroma-active compounds in pink guava
(Psidium guajaVa, L.) by application of the aroma extract dilution
analysis. J. Agric. Food Chem. 2008, 56, 4120–4127.
(2) Clery, R. A.; Hammond, C. J. New sulfur components of pink
guava fruit (Psidium guajaVa L.). J. Essent. Oil Res. 2008, 20,
315–317.
(3) Schieberle, P.; Grosch, W. Potent odorants of the wheat bread
crumb. Differences to the crust and effect of a longer dough
fermentation. Z. Lebensm. Unters. Forsch. 1991, 192, 130–135.
(4) 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.
(5) Schieberle, P.; Hofmann, T. Evaluation of the character impact
odorants in fresh strawberry juice by quantitative measurements
and sensory studies on model mixtures. J. Agric. Food Chem.
1997, 45, 227–232.
(21) Bemelmans, J. M. H. Review of isolation and concentration
techniques In Progress in FlaVour Research; Land, G. G., Nursten,
H. E., Eds.; Applied Science: London, U.K., 1979; pp 79-88.
(22) American Society of Testing and Materials. Standard E679-04.
Standard practice for determination of odor and taste thresholds
by a forced-choice ascending concentration series method of limits
In ASTM Book of Standards; American Society of Testing and
Materials: West Conshohocken, PA, 2005; Vol. 15.08, pp 38-
44.
(23) Schieberle, P. Recent developments in methods for analysis of
flavor compounds and their precursors In Characterization of
Food: Emerging Methods; Goankar, A., Ed.; Elsevier: Amsterdam,
The Netherlands, 1995; pp 403-431 (review).
(24) Hatanaka, A. The biogeneration of green odour by green leaves.
Phytochemistry 1993, 34, 1201–1218.
(25) Buttery, R. G.; Teranishi, R.; Ling, L. C. Fresh tomato aroma
volatiles: a quantitative study. J. Agric. Food Chem. 1987, 35,
540–544.
(26) Grosch, W. Evaluation of the key odorants of foods by dilution
experiments, aroma models and omission. Chem. Senses 2001,
26, 533–545.
(6) 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 suger-based
Maillard model systems by stable isotope dilution assays. J. Agric.
Food Chem. 1997, 45, 2642–2648.
(7) Blank, I.; Lin, J.; Fumeaux, R.; Welti, D. H.; Fay, L. B. Formation
of 3-hydroxy-4,5-dimethyl-2(5H)-furanone (sotolon) from 4-hy-
droxy-L-isoleucine and 3-amino-4,5-dimethyl-3,4-dihydro-2(5H)-
furanone. J. Agric. Food Chem. 1996, 44, 1851–1856.
(8) Granvogl, M.; Christlbauer, M.; Schieberle, P. Quantitation of
the intense aroma compound 3-mercapto-2-methylpentan-1-ol in
raw and processed onions (Allium cepa) of different origins and
Received for review December 1, 2008. Revised manuscript received
January 27, 2009. Accepted January 28, 2009. We appreciate financial
support from the Ministry of Agriculture of Colombia, ASOHOFRU-
COL, the German Academic Exchange Service (DAAD), and Colciencias.
JF803728N