Aroma Compounds in Glucose/- and Rhamnose/Cysteine
J. Agric. Food Chem., Vol. 45, No. 3, 1997 901
preparation); no. 16 (Cerny and Grosch, 1992). Ethyl 2-mer-
captopropionate MS/EI data, m/z (%): 61 (100); 134 (40); 88
(26); 60 (22); 59 (18); 45 (15); 43 (14); 55 (12); 73 (8); 106 (4),
were obtained by reacting 2-mercaptopropanoic acid and
ethanol as recently reported for the synthesis of ethyl cyclo-
hexanoate (Guth and Grosch, 1991). 2-Oxobutanal was pre-
pared by oxidation of 1-hydroxy-2-butanone using bismutoxide
in acetic acid.
Mod el Rea ction . L-Cysteine (3.3 mmol) and either D-
glucose (10 mmol) or L-rhamnose (10 mmol) were dissolved in
phosphate buffer (100 mL; 0.5 M; pH 5.0) and allowed to react
in a laboratory autoclave (model II, 200 mL total volume; Roth,
Karlsruhe, Germany) by raising the temperature within 20
min from 20 to 145 °C.
Isola tion of th e Vola tiles a n d Ar om a Extr a ct Dilu tion
An a lysis. The volatile reaction products formed were isolated
by extraction with diethyl ether and sublimation in vacuo as
recently described for the ribose/cysteine mixture (Hofmann
and Schieberle, 1995). The distillates of both model solutions
were each separated into an acidic volatile fraction (I in Tables
1 and 2) and a neutral/basic volatile fraction (II in Tables 1
and 2) by treatment with sodium bicarbonate, and then the
odor-active compounds were evaluated by aroma extract
dilution analysis (AEDA; cf. review by Schieberle, 1995) as
recently described (Hofmann and Schieberle, 1995).
High -Resolu tion Ga s Ch r om a togr a p h y (HRGC)/Olfa c-
tom etr y a n d HRGC/Ma ss Sp ectr om etr y (MS; Ap p r oxi-
m a tion of Od or Th r esh old s). HRGC was performed with
a type 5160 gas chromatograph (Fisons, Mainz, Germany) by
using capillary FFAP (30 m × 0.32 mm fused silica capillary,
free fatty acid phase, 0.25 µm; J &W Scientific, Fisons, Mainz,
Germany) and capillary SE-54 (30 m × 0.32 mm fused silica
capillary DB-5, 0.25 µm; J &W Scientific, Fisons, Mainz,
Germany). The samples were applied by the “cold on-column”
technique at 40 °C. After 2 min, the temperature of the oven
was raised at 40 °C/min to 50 °C (SE-54) or 60 °C (FFAP),
respectively, held for 5 min isothermally, and then raised at
6 °C/min to 230 °C and held for 15 min. The flow of the carrier
gas helium was between 2.3 and 2.5 mL/min. At the end of
capillary, the effluent was split 1:1 (by vol) into an FID and
an odor port using deactivated but uncoated fused silica
capillaries (50 cm × 0.32 mm). The FID and the odor port
were maintained at 180 °C. Calculation of the linear retention
indices (RI) and mass spectral measurements were performed
as recently described (Hofmann and Schieberle, 1995). Odor
thresholds were approximated by HRGC/O as previously
reported (Hofmann and Schieberle, 1995).
F igu r e 8. Mass spectrum (MS/EI) of 4-hydroxy-2,5-dimethyl-
3(2H)-thiophenone.
I (Figure 7) was isolated by extraction with diethyl ether. I
was taken up in a mixture of acetone (51.5 mL), sodium
hydrogencarbonate (0.88 mmol), magnesium sulfate (2.84
mmol), and water (30.9 mL) and then oxidized by addition of
potassium permanganate (10 mmol). After 2 h of stirring at
25 °C and maintaining the pH between 7.0 and 8.0 by dropwise
addition of hydrochloric acid (1 M), the 2,5-bisacetoxyhexane-
3,4-dione formed (II; Figure 7), which was isolated by extrac-
tion with diethyl ether. II (3.9 mmol) was then dissolved in
ethanol (20 mL) containing sodium sulfide (8 mmol) and stirred
for 12 h at room temperature. After addition of dilute
hydrochloric acid (50 mL; pH 3.0), the mixture was extracted
with diethyl ether (total volume: 100 mL). To purify the
target compound from byproducts, the organic phase was then
extracted with sodium bicarbonate (0.5 M; total volume: 60
mL), the combined organic phases were discarded, and the
target compound was re-extracted into diethyl ether after
adjusting the pH of the combined aqueous phases to 3.0.
The HDMT (III; Figure 7) was further purified by flash
chromatography using the same equipment as described
recently (Hofmann and Schieberle, 1995). After the column
was flushed with pentane (150 mL), followed by pentane/
diethyl ether (150 mL, 7 + 3 by vol) and pentane/diethyl ether
(150 mL, 1 + 1 by vol), the target compound, which was
obtained in a yield of 6%, was eluted with diethyl ether (150
mL).
The MS/EI (Figure 8) data were in agreement with a
spectrum published by Martin (1988). In addition, the MS/
1
CI (isobutane), m/z (%) [145 (100; M+ + 1)], and the H-NMR
data confirmed the structure for HDMT displayed in Figure
8. 1H-NMR, δ in ppm [multiplicity, coupling constant (Hz),
RESULTS
3
intensity, relevant carbon in Figure 8]: 1.55 (d, J 7,2 ) 7.08,
4
4J ) 1.77, 3 H, C-7); 2.51 (s, J ) 1.77, 3 H, C-6); 3.70 (quart,
Glu cose/Cystein e. A solution of glucose and cys-
teine was thermally treated for 20 min at 145 °C, and,
after cooling, the overall odor was characterized by a
six-membered sensory panel. The intensity of ten odor
qualities had to be evaluated on a seven-point scale
ranging from 0 (no odor) to 3.0 (most intense odor) as
previsouly described (Hofmann and Schieberle, 1996).
In Figure 10A, mean values of intensities of each odor
note are displayed as a spider web diagram. Sulfury,
roasty, and caramel-like notes were the most intense,
and the overall odor was described as somewhat resem-
bling roast chicken.
To identify the odorants responsible for these odors,
the volatiles were isolated by extraction with diethyl
ether followed by high-vacuum distillation. The distil-
late was separated into the acidic volatiles (fraction I;
Table 1) and the neutral/basic volatiles (fraction II;
Table 1), and then the most odor-active compounds in
both fractions were selected by aroma extract dilution
analysis (AEDA).
3J 2,7 ) 7.08, 4J ) 1.77, 1 H, C-2); 5.25 (broad s, 1 H, OH group).
Bis(1-(2′-fu r yl)-1-et h yl) Disu lfid e (ME F -ME F ) a n d
2-Fu r fu r yl-(1-(2′-fu r yl)-1-eth yl) Disu lfide (FFT-MEF). 2-(1-
Mercaptoethyl)furan and 2-furfurylthiol (0.1 mmol each) were
dissolved in diethyl ether (15 mL) and oxidized to the corre-
sponding disulfides by copper(II) sulfate as previously de-
scribed (Hofmann and Schieberle, 1995). The mixture of the
three disulfides formed was separated by high-resolution
column chromatography (HPLC) using a Beckman model 100
A pump and a Rheodyne valve (type 7125). The sample was
applied onto a stainless steel column (25 cm × 0.46 cm i.d.)
filled with ODS-Hypersil (Shandon, Eastmore, GB; 5 µm), and
the target compounds were eluted with a mixture of acetoni-
trile/water (65 + 35 by vol). The effluent was monitored by
an UVIDEC-100-III spectrophotometer (J asco, Tokyo, J apan)
at 254 nm. Eluates of 20 repeated injections were collected
(MEF-MEF, 7.8 mL to 9.2 mL; MEF-FFT, 9.4 mL to 12.2
mL) and then diluted with water, and the mixture was
extracted with diethyl ether. The mass spectra of MEF-MEF
and FFT-MEF are displayed in Figure 9A and 9B. The
molecular weights were established by MS/CI (isobutane).
The following odorants (cf. Tables 1 and 2) were synthesized
as reported in the literature: nos. 1, 4, and 13 (Hofmann and
Schieberle, 1995); nos. 28 and 30 (Hofmann et al., 1995); no.
3 (Asinger et al., 1964); no. 14 (Zehentbauer and Grosch, in
In fraction I, 17 odor-active compounds were detected
in the flavor dilution (FD) factor range of 4-1024 (first
column in Table 1). The highest FD factors were found