Influence of the Production Process on the Key Aroma Compounds of Rum: From Molasses to the Spirit

Article abstract of DOI:10.1021/acs.jafc.6b04046

The production of rum consists of fermentation, distillation, and aging. To check the influence of each step on the final rum aroma, molasses, mash, distillate, and the final rum were analyzed using the sensomics concept. The changes in key aroma compounds were determined by application of aroma extract dilution analysis (AEDA) in combination with gas chromatography-mass spectrometry for identification and by stable isotope dilution assays (SIDAs) for quantitation. Odor activity values (OAVs; ratio of concentration to respective odor threshold) were calculated for the compounds determined in the rum and, finally, the rum aroma was successfully simulated by recombination. (E)-β-Damascenone showed by far the highest OAV (3280) in rum. Although this compound was determined already in molasses, its concentration increased significantly during distillation, indicating a thermolabile precursor. Vanillin, 4-ethylphenol, 2-methoxyphenol, 4-ethyl-2-methoxyphenol, and 2-methoxy-4-propylphenol are well-known compounds mainly stemming from the wood barrels used for aging and showed an OAV ≥ 1. Another important group of aroma-active compounds in rum were ethyl esters, for which a significant increase was determined during fermentation but also to a lesser extent during aging. Altogether, the concentrations of 68% of the aroma-active compounds increased during the process, demonstrating its influence on the overall rum aroma.

Full text of DOI:10.1021/acs.jafc.6b04046

Subscriber access provided by UNIV NEW ORLEANS
Article
Influence of the Production Process on the Key Aroma
Compounds of Rum: from Molasses to the Spirit
Laura Franitza, Michael Granvogl, and Peter Schieberle
J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b04046 • Publication Date (Web): 27 Oct 2016
Downloaded from http://pubs.acs.org on November 1, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Agricultural and Food Chemistry is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 53
Journal of Agricultural and Food Chemistry
1
Influence of the Production Process
on the Key Aroma Compounds of Rum:
from Molasses to the Spirit
Laura Franitza, Michael Granvogl,* and Peter Schieberle
Lehrstuhl für Lebensmittelchemie,
Technische Universität München, Departement für Chemie,
LiseꢀMeitnerꢀStraße 34, Dꢀ85354 Freising, Germany
*
Corresponding Author
Phone:
+49 8161 71 2987
Fax:
+49 8161 71 2970
Eꢀmail:
michael.granvogl@ch.tum.de
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 2 of 53
2
1
ABSTRACT: The production of rum consists of fermentation, distillation, and
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
ageing. To check the influence of each step on the final rum aroma, molasses, mash,
distillate, and the final rum were analyzed using the sensomics concept. The
changes in key aroma compounds were determined by application of aroma extract
dilution analysis (AEDA) in combination with gas chromatographyꢀmass spectrometry
for identification and by stable isotope dilution assays (SIDAs) for quantitation. Odor
activity values (OAVs; ratio of concentration to respective odor threshold) were
calculated for the compounds determined in the rum and, finally, the rum aroma was
successfully simulated by recombination. (E)ꢀβꢀDamascenone showed by far the
highest OAV (3280) in rum. Although this compound was determined already in
molasses, its concentration increased significantly during distillation indicating a
thermoꢀlabile precursor. Vanillin, 4ꢀethylphenol, 2ꢀmethoxyphenol, 4ꢀethylꢀ2ꢀ
methoxyphenol, and 4ꢀpropylꢀ2ꢀmethoxyphenol are wellꢀknown compounds mainly
stemming from the wood barrels used for ageing and showed all an OAV ≥ 1.
Another important group of aromaꢀactive compounds in rum were ethyl esters, for
which a significant increase was determined during fermentation, but also to a lesser
extent during ageing. Altogether, the concentrations of 68% of the aromaꢀactive
compounds increased during the process, demonstrating its influence on the overall
rum aroma.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
KEYWORDS: rum, production, fermentation, distillation, ageing, aroma extract
dilution analysis, stable isotope dilution analysis, odor activity value, aroma
recombination
ACS Paragon Plus Environment

Page 3 of 53
Journal of Agricultural and Food Chemistry
3
24
INTRODUCTION
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
The production of rum traditionally starts with the fermentation of molasses, the
remainder after sugar production, using yeasts like Saccharomyces cerevisiae.
During fermentation, many aromaꢀactive compounds are formed, including alcohols,
aldehydes, and ethyl esters. Besides for rum production, molasses is used for many
foods because of its sweetness and special aroma. First studies on molasses by gas
1
chromatography were done by Hashizume et al., who identified anisole, benzoic
2
acid, vanillin, and vanillic acid. Abe et al. analyzed for the first time the soꢀcalled
“sugary flavor” by gas chromatographyꢀmass spectrometry (GCꢀMS), postulating 5ꢀ
methylꢀ2ꢀacetylfuran, cyclotene, γꢀnonalactone, γꢀdecalactone, δꢀdecalactone, and
vanillin as potential key compounds. In a further study, nine new compounds were
identified, e.g., damascenone (no information about stereochemistry was mentioned),
3
but, a mixture of all compounds did not show the characteristic aroma of molasses.
Later, 3ꢀhydroxyꢀ4,5ꢀdimethylfuranꢀ2(5H)ꢀone was characterized as a key compound
and was named “sotolon”, a combination of “soto”, the Japanese word for cane
4
sugar, and “olon” (“enolic lactone”). In addition, vanillin, maltol, δꢀnonalactone, 4ꢀ
vinylphenol, and 2,6ꢀdimethoxyphenol were characterized as key compounds for the
„sugary flavor“ of molasses.
The next step of rum production is the distillation of the mash. Thereby,
undesired compounds, but also solids, are separated from the distillate.
Unfortunately, the discrimination of compounds with a lower or higher boiling point
compared to ethanol can lead to a loss of aroma characteristics.
The final steps are the ageing of the distillate, which is commonly performed in
oak barrels firstly used in whisky or brandy production, followed by a dilution step
adjusting the desired alcohol content of 40% (by vol.). Numerous reactions take
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 4 of 53
4
5
,6
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
place and influence the aroma of the final product. Due to the thermal treatment
during barrel production, many aroma compounds are generated and, finally,
extracted during ageing. Cisꢀ and transꢀwhisky lactone (βꢀmethylꢀγꢀoctalactone) are
marker compounds for spirits, which were stored in barrels. These compounds were
7
firstly isolated from different oak woods by Masuda and Nishimura. Although four
stereoisomers are possible, up to now, only two isomers (3S,4S and 3S,4R) were
8
9,10
6
characterized in oak woods and, consequently, in spirits, e.g., in rum,
whisky, or
1
1
cognac. Due to the high ethanol content, the concentrations of ethyl esters are
increasing. The reaction equilibrium of the acids and ethanol strongly depends on the
ageing conditions. A continuous increase of ethyl esters in a model solution
6
simulating ageing conditions was already shown.
12
First studies on rum volatiles were done by Bober and Haddaway using gas
1
3
chromatography, followed by Maarse and Ten Noever de Brauw, who identified
over 100 volatile compounds in a Jamaican rum by gas chromatography, mass
1
4
spectrometry, and infrared spectrometry. In 2006, de Souza et al. analyzed rum
odorants for the first time via gas chromatographyꢀolfactometry (GCꢀO). Using Charm
analysis, βꢀdamascenone (no information about stereochemistry was mentioned),
1,1ꢀdiethoxyethane, ethyl 2ꢀmethylbutanoate, whisky lactone (no information about
stereochemistry was mentioned), vanillin, and 2ꢀmethoxyphenol were characterized
9
as most potent odorants. Pino et al. quantitated 98 compounds using external
calibration methods and estimated the potency of the odorants by aroma extract
dilution analysis (AEDA) and odor activity values (OAVs; ratio of concentration to
respective odor threshold) determining ethanol, (E)ꢀβꢀdamascenone, ethyl butanoate,
ethyl
hexanoate,
vanillin,
2ꢀmethoxyphenol,
cisꢀwhisky
lactone,
ethyl
methylpropanoate, ethyl octanoate, ethyl decanoate, 4ꢀpropylꢀ2ꢀmethoxyphenol,
eugenol, 1,1ꢀdiethoxyethane, ethyl 2ꢀmethylbutanaote, γꢀnonalactone, 2ꢀ
ACS Paragon Plus Environment

Page 5 of 53
Journal of Agricultural and Food Chemistry
5
7
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
phenylethanol, 4ꢀethylꢀ2ꢀmethoxyphenol, 3ꢀmethylbutyl acetate, and 2ꢀphenylethyl
acetate as key odorants. Very recently, the key aroma compounds of two commercial
rums were characterized using the sensomics concept revealing ethanol, ethyl (S)ꢀ2ꢀ
methylbutanoate, (E)ꢀβꢀdamascenone, 3ꢀmethylbutanal, 2,3ꢀbutanedione, ethyl
butanoate, 1,1ꢀdiethoxyethane, ethyl 3ꢀmethylbutanoate, ethyl pentanoate, and ethyl
hexanoate with OAVs ≥ 1 in both rums. In addition, both rum aromas were
7
7
7
7
8
8
8
8
8
8
8
8
8
8
9
9
9
9
9
9
10
successfully simulated using 14 and 12 reference compounds, respectively.
As briefly outlined, the production process significantly influences the chemical
composition of the final spirit. Due to the fact that an aromatization of rum and an
15
addition of ethanol is forbidden, the production process is, besides the raw material,
the only way to improve the aroma.
Thus, the aim of the present study was to investigate the influence of the
16
production process on the aroma of rum by the sensomics concept, consisting of (i)
the identification of the key aroma compounds in molasses, mash, distillate, and rum
(representing the four main steps of rum production) by aroma extract dilution
analysis (AEDA) in combination with gas chromatographyꢀmass spectrometry, (ii) the
quantitation of the most potent odorants by stable isotope dilution assays (SIDAs) in
molasses, mash, distillate, and rum, (iii) the calculation of odor activity values (OAVs)
for the rum odorants, and (iv) the simulation of the rum aroma. These data will
provide rum producers knowledge about possibilities to positively influence the
aroma of the final spirit.
96
MATERIALS AND METHODS
97
98
99
Samples. Molasses, mash, distillate, and rum were obtained from a Bavarian
distillery (Feinbrennerei Simon, Alzenau, Germany) ensuring traceability of molasses,
mash, distillate as well as the analyzed final spirit. Molasses was from the Dominican
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 6 of 53
6
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
00 Republic. The raw material was diluted with water (1+3; v+v) and used for
01 fermentation performed at room temperature for about 2.5 weeks. Distillation was
02 carried out in a pot still. Each batch yielded 20 L of distillate, whereat the first 2 L
03 (foreshots) were withdrawn and the heart was stopped if the alcohol content reached
04 65% (by vol.) merging into the feints. The distillate obtained was stored at room
05 temperature for 3 years in heavy toasted oak barrels (225 L; wood originating from
06 the Spessart, Germany), which were already used for red wine ageing 30 months
07 before. Before bottling, a dilution step adjusting the desired alcohol content of 40%
08 (by vol.) was performed. There was no blending leading to single cask rum of high
09 quality, a specialty of the smallꢀsized premium distillery. Prior to analysis, molasses
10 and mash were vacuumꢀtight packed and stored at ꢀ18 °C in the dark; distillate and
11 rum were stored at room temperature in the dark.
12
Chemicals. The following compounds used as references for identification and
13 quantitation were obtained from commercial sources: acetic acid, 2,3ꢀbutanedione,
14 (E,E)ꢀ2,4ꢀdecadienal, decanoic acid, 1,1ꢀdiethoxyethane, ethyl butanoate, 3ꢀethylꢀ
15 2,5ꢀdimethylpyrazine, ethyl hexanoate, ethyl 2ꢀmethylbutanoate, ethyl 3ꢀ
16 methylbutanoate, ethyl octanoate, ethyl pentanoate, 4ꢀethylphenol, ethyl 3ꢀ
17 phenylpropanoate, (Z)ꢀ4ꢀheptenal, 3ꢀhydroxyꢀ4,5ꢀdimethylfuranꢀ2(5H)ꢀone, 4ꢀ
18 hydroxyꢀ2,5ꢀdimethylfuranꢀ3(2H)ꢀone, linalool, (R)ꢀlinalool, 2ꢀmethoxyphenol, 2ꢀ
19 methoxyꢀ4ꢀpropylphenol, 2ꢀmethylbutanal, 3ꢀmethylbutanal, 3ꢀmethylbutanoic acid,
20 2ꢀmethylꢀ1ꢀbutanol, (S)ꢀ2ꢀmethylꢀ1ꢀbutanol, 3ꢀmethylꢀ1ꢀbutanol, 3ꢀmethylbutyl
21 acetate,
4ꢀmethylꢀ2ꢀmethoxyphenol,
3ꢀ(methylthio)propanal,
methyl
octanoate,
4ꢀmethylphenol,
22 methylpropanol,
γꢀnonalactone,
phenylacetaldehyde,
23 phenylacetic acid, 2ꢀphenylethanol, and cisꢀ/transꢀwhisky lactone (SigmaꢀAldrich
24 Chemie, Taufkirchen, Germany); 2ꢀmethylbutanoic acid and 1ꢀoctenꢀ3ꢀone (Alfa
25 Aeasar, Karlsruhe, Germany); ethyl cyclohexanoate and 4ꢀethylꢀ2ꢀmethoxyphenol
ACS Paragon Plus Environment

Page 7 of 53
Journal of Agricultural and Food Chemistry
7
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
26 (Lancaster, Mühlheim/Main, Germany); butanoic acid and vanillin (Merck, Darmstadt,
27 Germany); and (R)ꢀ2ꢀmethylbutanoic acid (Akos, Steinen, Germany). 1ꢀ(2,6,6ꢀ
28 Trimethylꢀ1,3ꢀcyclohexadienꢀ1ꢀyl)ꢀ2ꢀbutenꢀ1ꢀone ((E)ꢀβꢀdamascenone) was kindly
29 provided by Symrise (Holzminden, Germany).
17
30
31
Ethyl (S)ꢀ2ꢀmethylbutanoate was synthesized as recently described.
Ammonium chloride, hydrochloric acid, lithium aluminum hydride, sodium
32 carbonate, sodium chloride, sodium hydrogen carbonate, sodium sulfate, and sodium
33 thiosulfate were from Merck. 1,1,1ꢀTrisꢀ(acetyloxy)ꢀ1,1ꢀdihydroꢀ1,2ꢀbenziodoxolꢀ3ꢀ
34 (1H)ꢀone (DessꢀMartin periodinane) was obtained from SigmaꢀAldrich. Diethyl ether,
35 dichloromethane, and pentane (Merck) were freshly distilled prior to use. Liquid
36 nitrogen was obtained from Linde (Munich, Germany). All chemicals were at least of
37 analytical grade.
2
2
38
Stable Isotopically Labeled Standards. [ H ]ꢀ4ꢀMethylphenol, [ H ]ꢀ2ꢀmethylꢀ
7 9
2
39 butanoic acid, and [ H
5
]ꢀ2ꢀphenylethanol were obtained from C/D/N Isotopes
13
2
13
40 (Quebec, Canada); [ C
2
]ꢀacetic acid, [ H
9
]ꢀmethylpropanol, and [ C
2
]ꢀphenylacetic
13
41 acid were from SigmaꢀAldrich; and [ C
2
]ꢀ4ꢀhydroxyꢀ2,5ꢀdimethylfuranꢀ3(2H)ꢀone was
42 from Aromalab (Planegg, Germany).
43
The following stable isotopically labeled standards were synthesized according to
13
18
2
19
2
44 the literature: [ C
4
]ꢀ2,3ꢀbutanedione;
[ H
2
]ꢀbutanoic acid;
[ H4ꢀ7]ꢀ(E)ꢀβꢀ
20
2
21
2
22
13
45 damascenone;
[ H2ꢀ4]ꢀ(E,E)ꢀ2,4ꢀdecadienal;
[ H
2
]ꢀdecanoic acid;
2
[ C ]ꢀ1,1ꢀ
23
2
18
2
3 3
46 diethoxyethane; [ H ]ꢀethyl butanoate; [ H ]ꢀethyl cyclohexanoate analogue to
2
18
2
23
2
24
47 [ H ]ꢀethyl butanoate; [ H ]ꢀethyl hexanoate; [ H ]ꢀ4ꢀethylꢀ2ꢀmethoxyphenol;
3
3
2
3
2ꢀ4
2
25
2
26
2
27
48 [ H
]ꢀethyl 2ꢀmethylbutanoate; [ H
3
]ꢀethyl octanoate; [ H
5
]ꢀethyl pentanoate;
2
23
2
27
13
49 [ H
]ꢀ3ꢀethylphenol;
[ H
5
]ꢀethyl
3ꢀphenylpropanoate;
2
[ C ]ꢀ3ꢀhydroxyꢀ4,5ꢀ
28
2
29
2
30
2
50 dimethylfuranꢀ2(5H)ꢀone;
[ H
2
]ꢀlinalool;
[ H
3
]ꢀ2ꢀmethoxyphenol;
[ H2ꢀ4]ꢀ2ꢀ
31
2
2
32 2
51 methoxyꢀ4ꢀpropylphenol; [ H
2
]ꢀ2ꢀmethylbutanal and [ H
2
]ꢀ3ꢀmethylbutanal; [ H
2
]ꢀ3ꢀ
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 8 of 53
8
3
3
2
26
2
34
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
52 methylbutanoic acid;
[ H
2
]ꢀ3ꢀmethylꢀ1ꢀbutanol;
[ H11]ꢀ3ꢀmethylbutyl acetate;
2
35
2
21
2
24
2
53 [ H
3
]ꢀ3ꢀ(methylthio)propanal; [ H2ꢀ4]ꢀ1ꢀoctenꢀ3ꢀone; [ H
3
]ꢀvanillin; and [ H
2
]ꢀcisꢀ
36
54 whisky lactone.
2
2
55
[ H
9
]ꢀEthyl 3ꢀmethylbutanoate was obtained by esterification of [ H
9
]ꢀ3ꢀ
25
56 methylbutanoic acid with ethanol following the procedure previously described.
57 Concentrations of the isotopically labeled standards were determined using
58 methyl octanoate as internal standard. Solutions of each unlabeled reference
59 compound and methyl octanoate were analyzed by means of a TRACE 2000 gas
60 chromatograph equipped with a flame ionization detector (FID) (ThermoQuest,
61 Egelsbach, Germany) to determine the FID response factor. Next, the labeled
62 standard was spiked with methyl octanoate and analyzed by GCꢀFID. The
63 concentration was calculated by the peak areas of the labeled compound and methyl
64 octanoate, the concentration of methyl octanoate, and the determined FID response
65 factor for the unlabeled compound.
66
Synthesis. (R)ꢀ2ꢀMethylꢀ1ꢀbutanol. A solution of (R)ꢀ2ꢀmethylbutanoic acid
67 (2 mmol) in anhydrous diethyl ether (10 mL) was added to a solution of LiAlH
4
68 (5 mmol) in anhydrous diethyl ether (10 mL) and stirred for 120 min at room
69 temperature. The solution was cooled (0 °C) using an iceꢀbath and a saturated
70 ammonium chloride solution (15 mL) was added, followed by hydrochloric acid
71 (2 mol/L; 10 mL). The aqueous phase was extracted with diethyl ether (3 x 30 mL),
72 the organic phases were combined, washed with a saturated aqueous solution of
73 sodium hydrogen carbonate (50 mL), and dried over anhydrous sodium sulfate.
74 Yield: 91%.
+
75 MSꢀEI, m/z (%): 70 (57, [M ꢀ H
2
O] ), 58 (5), 57 (94), 56 (100), 55 (19), 43 (9), 42 (12),
76 41 (75).
+
77 MSꢀCI, m/z (%): 71 (100, [(M ꢀ H
2
O) + H] ), 70 (15), 57 (17), 56 (25), 55 (16).
ACS Paragon Plus Environment

Page 9 of 53
Journal of Agricultural and Food Chemistry
9
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
78 Retention index (RI) on DBꢀFFAP: 1213; RI on BGBꢀ174E: 1105.
79 (R)ꢀ2ꢀ and (S)ꢀ2ꢀMethylbutanal. DessꢀMartin periodinane (1.4 mmol) was added
80 to a solution of (R)ꢀ2ꢀ or (S)ꢀ2ꢀmethylꢀ1ꢀbutanol (1.2 mmol) dissolved in dichloroꢀ
81 methane (15 mL) and stirred for 90 min at room temperature. The mixture was
82 washed with an aqueous solution of sodium thiosulfate (1 mol/L), which was
83 saturated with sodium hydrogen carbonate (15 mL), and the organic phase was dried
84 over anhydrous sodium sulfate. Yield: 88% for (R)ꢀ2ꢀMethylbutanal and 85% for (S)ꢀ
85 2ꢀMethylbutanal, respectively.
+
86
(R)ꢀ2ꢀMethylbutanal: MSꢀEI, m/z (%): 86 (15, M ), 71 (5), 58 (66), 57 (100), 56
87 (10), 55 (6), 43 (12), 41 (82).
+
88 MSꢀCI, m/z (%): 87 (100, [M + H] ), 86 (20), 84 (6), 69 (70), 58 (47), 57 (60), 55(5).
89 RI on OVꢀ1701: 731; RI on BGBꢀ174E: 1061.
+
90
(S)ꢀ2ꢀMethylbutanal: MSꢀEI, m/z (%): 86 (16, M ), 71 (5), 58 (67), 57 (100), 56
91 (6), 55 (7), 43 (9), 41 (79).
+
92 MSꢀCI, m/z (%): 87 (100, [M + H] ), 85 (7), 69 (65), 58 (33), 57 (37), 55 (5).
93 RI on OVꢀ1701: 731; RI on BGBꢀ174E: 1067.
94
Isolation of the Volatiles. Molasses and Mash. To the molasses (35 g), water
95 (25 mL) and diethyl ether (70 mL) were added and the mixture was stirred for 30 min
96 at room temperature. To the mash (50 mL), diethyl ether (50 mL) was added and
97 also stirred for 30 min at room temperature. After the addition of sodium sulfate to
98 improve phase separation, the organic phase was separated and the extraction was
99 repeated twice. The organic phases were combined and dried over anhydrous
00 sodium sulfate.
01
Distillate and Rum. To the distillate (20 mL), water (20 mL) was added due to the
02 high amount of ethanol improving phase separation, followed by an extraction with
03 diethyl ether (3 x 40 mL) in a separating funnel. The rum (50 mL) was directly
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 10 of 53
10
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
04 extracted with diethyl ether (3 x 50 mL). The organic phases were washed with an
05 aqueous solution of sodium chloride (1 mol/L; 3 x 150 mL) to remove the high
06 amount of ethanol and dried over anhydrous sodium sulfate.
07
All four organic extracts were subjected to high vacuum distillation using solvent
37
08 assisted flavor evaporation (SAFE) technique to separate the volatiles from the
09 nonꢀvolatiles. The distillates obtained were concentrated to a final volume of ~200 ꢁL
38
10 using a Vigreuxꢀcolumn (50 cm x 1 cm i.d.) and microꢀdistillation.
11 High-Resolution Gas Chromatography-Olfactometry (HRGC-O). HRGCꢀO
12 was performed as recently described. RIs for each compound were determined by
10
39
13 a series of nꢀalkanes (C6ꢀC26 (DBꢀFFAP) and C6ꢀC18 (DBꢀ5), respectively).
14 Aroma Extract Dilution Analysis (AEDA). For AEDA, molasses, mash,
15 distillate, and rum were extracted and the volatiles were isolated by SAFE
37
16 distillation as described above. The distillates obtained were concentrated to the
17 same final volume and diluted stepwise 1+1 (v+v) with diethyl ether. The original
18 distillate and each dilution were analyzed by HRGCꢀO to determine the flavor dilution
19 (FD) factors. The original distillate was analyzed by three experienced panelists to
20 avoid overlooking of odorꢀactive compounds.
21
High-Resolution Gas Chromatography-Mass Spectrometry (HRGC-MS) for
22 Identification. A gas chromatograph 5890 series II (HewlettꢀPackard, Waldbronn,
23 Germany) coupled to a sector field mass spectrometer type MAT 95 S (Finnigan,
24 Bremen, Germany) was used for HRGCꢀMS. An aliquot of the sample (0.5 ꢁL) was
25 manually injected by the cold onꢀcolumn technique at 40 °C. The initial temperature
26 was held for 2 min, then raised at 6 °C/min to 230 °C and held for 5 min. Mass
27 spectra were generated in the electron ionization (EI) mode at 70 eV and in the
28 chemical ionization (CI) mode using isobutane as reactant gas at 115 eV.
ACS Paragon Plus Environment

Page 11 of 53
Journal of Agricultural and Food Chemistry
11
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
29
Quantitation by Stable Isotope Dilution Analysis (SIDA). Water (in case of
30 molasses and distillate), diethyl ether (1+1 by vol., but a minimum of 10 mL), and the
31 internal standards (dissolved in ethanol or dichloromethane; amounts depending on
32 the concentration of the analytes determined in preliminary experiments) were added
33 to the respective samples. After stirring for 30 min at room temperature for
34 equilibration, the workꢀup procedure was performed as described above for the
35 isolation of the volatiles prior to HRGCꢀMS or twoꢀdimensional highꢀresolution gas
36 chromatographyꢀmass spectrometry (HRGC/HRGCꢀMS). All quantitation experiments
37 (complete workꢀup procedure) were performed in triplicates.
38
f
For determination of the respective response factors (R ) (Table 1), mixtures of
39 known amounts of the unlabeled and labeled compounds were analyzed in five
40 different ratios (5:1, 3:1, 1:1, 1:3, 1:5) in the same way as described below.
41
Quantitation of butanoic acid, decanoic acid as well as 2ꢀ and 3ꢀmethylbutanoic
10
42 acid were performed as recently described.
43
High-Resolution Gas Chromatography-Mass Spectrometry (HRGC-MS) for
44 Quantitation. For quantitation experiments, a gas chromatograph type 431 (Varian,
45 Darmstadt) equipped with a DBꢀFFAP column (30 m x 0.25 mm i.d., 0.25 ꢁm film
46 thickness; J&W Scientific) coupled to a 220 ion trap mass spectrometer (Varian) was
47 used. An aliquot of the sample (2 ꢁL) was injected by means of a Combi PAL
48 autosampler (CTC Analytics, Zwingen, Switzerland) using a cold onꢀcolumn injector
49 at 40 °C. The initial temperature of 40 °C was held for 2 min, and then raised at
50 230 °C using a gradient of 6 °C/min. The final temperature was held for 5 min. Mass
51 spectra were generated in CI mode (70 eV) using methanol as reactant gas.
52
Two-Dimensional
High-Resolution
Gas
Chromatography-Mass
53 Spectrometry (HRGC/HRGC-MS). In case that a trace compound was overlapped
54 by a major volatile, HRGC/HRGCꢀMS was performed by a TRACE 2000 series GC
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 12 of 53
12
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
55 (ThermoQuest) equipped with a DBꢀFFAP column (30 m x 0.32 mm i.d., 0.25 ꢁm film
56 thickness) coupled to a CP 3800 gas chromatograph (Varian) equipped with an OVꢀ
57 1701 column (30 m x 0.25 mm i.d., 0.25 ꢁm film thickness) (both J&W Scientific). An
58 aliquot of the sample (2 ꢁL) was injected by means of a Combi PAL autosampler
59 using a cold onꢀcolumn injector at 40 °C. Heart cuts were transferred to the second
60 column by a moving capillary stream switching system (ThermoQuest). Thereby, the
61 volatiles were cryoꢀfocused in a cold trap cooled to ꢀ100 °C with liquid nitrogen, and
62 separated on the second column after increasing the trap temperature to 230 °C. A
63 Saturn 2000 ion trap mass spectrometer (Varian) was used to generate mass spectra
64 in CI mode (70 eV) using methanol as reactant gas.
65
Quantitation of Acetic acid and Ethanol. Acetic acid (in the distillate and rum)
66 and ethanol (in the mash, distillate, and rum) were determined using a commercial
67 enzymeꢀkit (RꢀBiopharm, Darmstadt). Analyses were done according to the
68 instructions of the manufacturer and measuring was performed by means of a UVꢀ
69 2401PC photometer (Shimadzu, Duisburg, Germany).
70
Separation of Ethyl (R)-2- and Ethyl (S)-2-methylbutanoate, (R)- and (S)-
71 Linalool as well as (R)-2-, (S)-2-, and 3-Methylbutanoic acid. Ethyl (R)ꢀ2ꢀ and
72 ethyl (S)ꢀ2ꢀmethylbutanoate as well as (R)ꢀ2ꢀ, (S)ꢀ2ꢀ, and 3ꢀmethylbutanoic acid were
73 separated by HRGC/HRGCꢀMS using a DBꢀFFAP column (30 m x 0.32 mm i.d.,
74 0.25 ꢁm film thickness; J&W Scientific) in the first dimension and a chiral BGBꢀ176
75 column (30 m x 0.25 mm i.d., 0.25 ꢁm film thickness; BGB Analytik, Böckten,
17
76 Switzerland) in the second dimension as recently reported. The separation of (R)ꢀ
29
77 and (S)ꢀlinalool was performed using the same column combination.
78 Separation of (R)-2- and (S)-2-Methylbutanal as well as (R)-2-, (S)-2-, and 3-
79 Methyl-1-butanol. (R)ꢀ2ꢀ and (S)ꢀ2ꢀmethylbutanal as well as (R)ꢀ2ꢀ, (S)ꢀ2ꢀ, and 3ꢀ
80 methylꢀ1ꢀbutanol were separated by HRGC/HRGCꢀMS using a DBꢀFFAP column
ACS Paragon Plus Environment

Page 13 of 53
Journal of Agricultural and Food Chemistry
13
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
81 (30 m x 0.32 mm i.d., 0.25 ꢁm film thickness; J&W Scientific) in the first dimension
82 and a chiral BGBꢀ174E column (30 m x 0.25 ꢁm i.d., 0.25 ꢁm film thickness; BGB
17
83 Analytik) in the second dimension as recently reported.
84 Determination of Orthonasal Odor Thresholds. Orthonasal odor thresholds
85 were determined in an aqueous solution containing 40% of pure ethanol (by vol.) by
40
86 our inꢀhouse sensory panel (at least 20 assessors) as previously described.
87 Aroma Profile Analysis (APA). For APA, a sensory panel rated the intensities of
88 selected odor attributes (ethanolic, malty, butterꢀlike, fruity, cloveꢀlike/phenolic,
89 caramelꢀlike, vanillaꢀlike, and baked appleꢀlike) on a linear sevenꢀpoint scale in steps
90 of 0.5 from 0 (not perceivable) to 3 (strongly perceivable). The panel consisted of 20
91 experienced assessors participating in weekly sensory sessions to train their abilities
92 to recognize and describe different aroma qualities. The samples (15 g of molasses;
93 15 mL of mash, distillate, and rum) were presented in covered glass vessels (40 mm
94 i.d., total volume = 45 mL). Sensory experiments were performed in a sensory room
95 at 21 ± 1 °C equipped with single booths.
96
Aroma Recombination. For aroma recombination experiments, all analyzed
97 aromaꢀactive compounds with an OAV ≥ 1 were prepared in aqueous solutions
98 containing 40% of pure ethanol (by vol.) and mixed in the natural occurring
99 concentrations determined in the rum. For an odorless (proven by the sensory panel
00 performing APA) matrix, rum (200 mL) was extracted with pentane (3 x 200 mL) and
01 diethyl ether (3 x 200 mL), and the remaining liquid was lyophilized. The mixture of
02 aroma compounds was added to the lyophilisate simulating the original ratio. The
03 original rum and the recombinate were evaluated in parallel by the sensory panel as
04 explained above for the APA.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 14 of 53
14
305 Results and Discussion
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
06
To compare the overall aroma of the four different process steps (molasses,
07 mash, distillate, and rum), aroma profile analyses were performed (Figure 1). As
08 expected, the ethanolic odor quality showed no intensity in the molasses, clearly
09 increased in the mash, and showed the highest intensity in the distillate and in the
10 rum. A clear difference was observed for the fruity and malty attributes revealing the
11 highest intensities in the mash. In contrast, the butterꢀlike odor quality was evaluated
12 with the highest intensity in the distillate and at the same level in the molasses, the
13 mash, and the rum. The molasses as starting material showed the highest rating for
14 the caramelꢀlike attribute, whereas the rum elicited the highest intensities in the
15 cloveꢀlike/phenolic, vanillaꢀlike, and baked appleꢀlike attributes. To identify the aroma
16
16 compounds responsible for these different overall aromas, the sensomics concept
17 was applied.
18
Identification of Aroma-Active Compounds. After workꢀup of all samples
37
19 including solvent extraction and high vacuum distillation using SAFE technique,
20 droplets of the distillates on strips of filter paper elicited the typical aromas of
21 molasses, mash, distillate, and rum, proving a successful extraction and cleanꢀup of
22 the volatile fraction of the raw material, the intermediates, and the final rum. The
23 SAFE distillates were subjected to aroma extract dilution analysis (AEDA) by highꢀ
24 resolution gas chromatographyꢀolfactometry (HRGCꢀO). For identification of the
25 compounds responsible for the single odor impressions, retention indices of the
26 aromaꢀactive regions determined on two GC columns of different polarity were
27 compared to an inꢀhouse database containing ~1000 aromaꢀactive reference
28 compounds. In addition, the aroma quality and intensity of the odor impressions
29 detected at the sniffing port were compared to those of the suggested compounds at
ACS Paragon Plus Environment

Page 15 of 53
Journal of Agricultural and Food Chemistry
15
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
30 similar concentration ranges. For an unequivocal identification, mass spectra (EI and
31 CI mode) were recorded for the odorants and compared to those of the references.
32
AromaꢀActive Compounds in the Molasses. Twentyꢀnine aromaꢀactive regions
33 were detected in the flavor dilution (FD) factor range between 8 and 2048 in the
34 SAFE distillate of the molasses (Table 2). Following the above mentioned procedure
35 for identification, 2ꢀmethoxyphenol (27; gammonꢀlike, smoky) and vanillin (42; vanillaꢀ
36 like) were identified with the highest FD factor of 2048, followed by acetic acid (16;
37 vinegarꢀlike), 3ꢀ(methylthio)propanal (17; cooked potatoꢀlike), butanoic acid (20;
38 sweaty), 2ꢀ and 3ꢀmethylbutanoic acid (22a,b; sweaty), 4ꢀhydroxyꢀ2,5ꢀdimethylfuranꢀ
39 3(2H)ꢀone (34; caramelꢀlike; all FD factor of 1024), 2,3ꢀbutanedione (2; butterꢀlike), 2ꢀ
40 and 3ꢀmethylꢀ1ꢀbutanol (10a,b; malty), and 3ꢀhydroxyꢀ4,5ꢀdimethylfuranꢀ2(5H)ꢀone
41 (39; seasoningꢀlike, spicy; all 512). Using a chiral column in the second dimension of
42 the twoꢀdimensional HRGC/HRGCꢀMS system to determine the ratio of (R)ꢀ2ꢀ and
43 (S)ꢀ2ꢀmethylbutanoic acid as well as of (R)ꢀ2ꢀ and (S)ꢀ2ꢀmethylꢀ1ꢀbutanol, a
44 separation of 2ꢀ and 3ꢀmethylbutanoic acid as well as of 2ꢀ and 3ꢀmethylꢀ1ꢀbutanol
45 was also achieved. 2ꢀMethylbutanoic acid and 2ꢀmethylꢀ1ꢀbutanol showed 81% and
46 84% of the (S)ꢀenantiomer, respectively.
47
AromaꢀActive Compounds in the Mash. AEDA of the mash SAFE distillate
48 showed 38 aromaꢀactive regions in the FD factor range between 8 and 2048
49 (Table 2). The highest FD factor of 2048 was determined for 2ꢀ and 3ꢀmethylꢀ1ꢀ
50 butanol (10a,b; >99% of the (S)ꢀenantiomer), followed by ethyl 2ꢀmethylbutanoate (5;
51 fruity; 95% of the (S)ꢀenantiomer), 3ꢀ(methylthio)propanal (17), butanoic acid (20), 2ꢀ
52 and 3ꢀmethylbutanoic acid (22a,b; 94% of the (S)ꢀenantiomer), 2ꢀmethoxyphenol
53 (27), 2ꢀphenylethanol (30; flowery, honeyꢀlike; all 1024), ethyl butanoate (4; fruity),
54 and acetic acid (16; both 512).
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 16 of 53
16
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
55
AromaꢀActive Compounds in the Distillate. Thirtyꢀseven aromaꢀactive compounds
56 were determined in the FD factor range between 8 and 2048 during AEDA of the
57 distillate (Table 2). Among them, 2ꢀ and 3ꢀmethylꢀ1ꢀbutanol (10a,b; >99% of the (S)ꢀ
58 enantiomer) and 2ꢀmethoxyphenol (27) were determined with the highest FD factor of
59 2048, followed by 2,3ꢀbutanedione (2; butterꢀlike), 3ꢀmethylbutyl acetate (8; bananaꢀ
60 like; both 1024), ethyl 2ꢀmethylbutanoate (5; 98% of the (S)ꢀenantiomer), ethyl
61 cyclohexanoate (14; fruity, sweet), ethyl octanoate (15; fruity, green), and 3ꢀ
62 (methylthio)propanal (17; all 512).
63
AromaꢀActive Compounds in the Rum. During AEDA of the rum, 38 aromaꢀactive
64 regions were determined in the FD factor range between 8 and 2048 (Table 2). 2ꢀ
65 and 3ꢀmethylꢀ1ꢀbutanol (10a,b; >99% of the (S)ꢀenantiomer), 2ꢀmethoxyphenol (27),
66 and cisꢀwhisky lactone (32; coconutꢀlike) showed the highest FD factor of 2048,
67 followed by ethyl cyclohexanoate (14), ethyl octanoate (15), transꢀwhisky lactone (29;
68 coconutꢀlike), 4ꢀethylꢀ2ꢀmethoxyphenol (35; smoky, gammonꢀlike), vanillin (42; all
69 1024), ethyl 2ꢀmethylbutanoate (5; 97% of the (S)ꢀenantiomer), 3ꢀmethylbutyl acetate
70 (8), ethyl 3ꢀphenylpropanoate (28; flowery), 2ꢀphenylethanol (30), 2ꢀmethoxyꢀ4ꢀ
71 propylphenol (37; phenolic), 4ꢀethylphenol (38; phenolic), and decanoic acid (40;
72 soapy, musty; all 512).
73
Quantitation of the Aroma-Active Compounds. AEDA is a screening method
74 to reduce the huge amount of volatiles to a reduced set of aromaꢀactive substances,
75 which have to be quantitated for the characterization of the overall aroma of a certain
76 food. Thus, 44 odorants, showing an FD factor ≥ 64 in at least one of the analyzed
77 process steps, were quantitated by means of SIDA and enzyme kits (acetic acid and
78 ethanol).
79
Concentrations of the Odorants in Molasses. The highest concentration was
80 determined for acetic acid (2.6 g/kg), followed by butanoic acid (8.12 mg/kg), 3ꢀ
ACS Paragon Plus Environment

Page 17 of 53
Journal of Agricultural and Food Chemistry
17
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
81 methylꢀ1ꢀbutanol (6.57 mg/kg), methylpropanol (3.40 mg/kg), phenylacetic acid
82 (2.79 mg/kg), (S)ꢀ2ꢀmethylꢀ1ꢀbutanol (2.15 mg/kg), and vanillin (1.09 mg/kg). The
83 acids 3ꢀmethylbutanoic acid (764 ꢁg/kg), (S)ꢀ2ꢀmethylbutanoic acid (571 ꢁg/kg),
84 decanoic acid (242 ꢁg/kg), and (R)ꢀ2ꢀmethylbutanoic acid (108 ꢁg/kg), the aldehydes
85 (S)ꢀ2ꢀmethylbutanal (519 ꢁg/kg), (R)ꢀ2ꢀmethylbutanal (449 ꢁg/kg), 3ꢀmethylbutanal
86 (388 ꢁg/kg), and 3ꢀ(methylthio)propanal (255 ꢁg/kg) as well as the furanones 4ꢀ
87 hydroxyꢀ2,5ꢀdimethylfuranꢀ3(2H)ꢀone (734 ꢁg/kg) and 3ꢀhydroxyꢀ4,5ꢀdimethylfuranꢀ
88 2(5H)ꢀone (216 ꢁg/kg) showed concentrations > 100 ꢁg/kg (Table 3).
89
Concentrations of the Odorants in the Mash. In the mash, ethanol showed by far
90 the highest concentration (75.5 g/L), followed by acetic acid (3.21 g/L), 3ꢀmethylꢀ1ꢀ
91 butanol (248 mg/L), methylpropanol (86.3 mg/L), (S)ꢀ2ꢀmethylꢀ1ꢀbutanol (54.4 mg/L),
92 and 2ꢀphenylethanol (43.6 mg/L). Lower amounts were found for the acids butanoic
93 acid (3.48 mg/L), 3ꢀmethylbutanoic acid (1.55 mg/L), (S)ꢀ2ꢀmethylbutanoic acid
94 (1.37 mg/L), phenylacetic acid (1.14 mg/L), decanoic acid (721 ꢁg/L), and (R)ꢀ2ꢀ
95 methylbutanoic acid (93.3 ꢁg/L) (Table 3).
96
Concentrations of the Odorants in the Distillate. By far the highest concentration
97 was again determined for ethanol (620 g/L), followed by 3ꢀmethylꢀ1ꢀbutanol
98 (2.86 g/L), methylpropanol (732 mg/L), (S)ꢀ2ꢀmethylꢀ1ꢀbutanol (586 mg/L), 1,1ꢀ
99 diethoxyethane (79.9 mg/L), and acetic acid (43.6 mg/L). Besides, 2,3ꢀbutanedione
00 (9.65 mg/L), ethyl octanoate (7.86 mg/L), 3ꢀmethylbutyl acetate (6.12 mg/L), 2ꢀ
01 phenylethanol (3.54 mg/L), 3ꢀmethylbutanal (3.49 mg/L), and (S)ꢀ2ꢀmethylbutanal
02 (1.68 mg/L) were determined with high concentrations (Table 3).
03
Concentrations of the Odorants in the Rum. As expected, ethanol showed by far
04 the highest concentration (317 g/L). In a concentration range between 1 mg/L and
05 1 g/L, 3ꢀmethylꢀ1ꢀbutanol (909 mg/L), methylpropanol (183 mg/L), (S)ꢀ2ꢀmethylꢀ1ꢀ
06 butanol (148 mg/L), acetic acid (121 mg/L), 1,1ꢀdiethoxyethane (20.8 mg/L), ethyl
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 18 of 53
18
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
07 octanoate (3.61 mg/L), 2ꢀphenylethanol (2.67 mg/L), 3ꢀmethylbutanal (2.08 mg/L), 3ꢀ
08 methylbutyl acetate (1.86 mg/L), 4ꢀethylphenol (1.45 mg/L), decanoic acid
09 (1.24 mg/L), and vanillin (1.20 mg/L) were determined. Somewhat lower amounts
10 were found for the ethyl esters (e.g., ethyl butanoate (441 ꢁg/L), ethyl hexanoate
11 (424 ꢁg/L), and ethyl pentanoate (36.6 ꢁg/L)) and the phenols (e.g., 2ꢀ
12 methoxyphenol (399 ꢁg/L), 4ꢀethylꢀ2ꢀmethoxyphenol (110 ꢁg/L), and 4ꢀmethylphenol
13 (32.3 ꢁg/L)). In addition, cisꢀ and transꢀwhisky lactone were analyzed in the rum with
14 quite high concentrations (cis: 210 ꢁg/L; trans: 191 ꢁg/L) (Table 3).
15
Calculation of Odor Activity Values of the Odorants in Rum. To evaluate
16 which odorants contribute to the overall aroma of rum, odor activity values (OAVs;
17 ratio of concentration to respective odor threshold) were calculated (Table 4). By far
18 the highest OAV in rum was determined for (E)ꢀβꢀdamascenone (3280), followed by
19 3ꢀmethylbutanal (743), ethanol (320), 2,3ꢀbutanedione (188), ethyl (S)ꢀ2ꢀ
20 methylbutanoate (97), vanillin (55), ethyl butanoate (46), (S)ꢀ2ꢀmethylbutanal (44),
21 and 2ꢀmethoxyphenol (43). Only 11 of 41 compounds showed an OAV < 1, e.g.,
22 butanoic acid, transꢀwhisky lactone, and 4ꢀmethylphenol, and thus, these compounds
23 should not contribute to the overall aroma. The results were in good agreement with
9,10
24 previously reported key odorants in rum.
25 Aroma Simulation. Recombination experiments were performed to verify that all
26 key odorants contributing to the overall aroma of rum were successfully identified and
27 quantitated. Therefore, all 30 aroma compounds showing an OAV ≥ 1 were added to
28 an odorless rum matrix in their determined naturally occurring concentrations in a
29 mixture of ethanol/water (40/60 by vol.). The recombinate was evaluated in parallel
30 with the original rum by a trained sensory panel using aroma profile analysis and
31 showed a very good similarity (Figure 2).
ACS Paragon Plus Environment

Page 19 of 53
Journal of Agricultural and Food Chemistry
19
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
32
Changes of the Concentrations of the Key Odorants during the Single
33 Processing Steps. To enable a comparison of the concentrations of the aromaꢀ
34 active compounds during the process, data are shown on the basis of the production
35 of 1 L of rum, for which 0.72 L of molasses were needed, resulting in 2.9 L of mash
36 and 0.4 L of distillate, respectively.
37
Changes during Fermentation. 0.72 L of molasses was used for the production of
38 1 L of rum, resulting in 2.9 L of mash. A significant change in the concentrations was
39 determined for (R)ꢀ2ꢀmethylꢀ1ꢀbutanol, (S)ꢀ2ꢀmethylꢀ1ꢀbutanol, 3ꢀmethylꢀ1ꢀbutanol,
40 methylpropanol, and 2ꢀphenylethanol. Except of (R)ꢀ2ꢀmethylꢀ1ꢀbutanol, the
41 concentrations of the alcohols clearly increased (Table 5). Following the Ehrlich
41
42 pathway, amino acids can react to the corresponding odorꢀactive alcohols. First, an
43 enzymatic deamination leads to the respective αꢀketo acid, followed by a
44 decarboxylation and a reduction step to the respective alcohols. Further reactions
45 can also lead to the corresponding aldehydes and acids. As the reaction is catalyzed
46 by enzymes, the degradation is stereo specific. Hence, the degradation of
Lꢀ
47 isoleucine during fermentation led to an increased concentration of (S)ꢀ2ꢀmethylꢀ1ꢀ
48 butanol, whereas (R)ꢀ2ꢀmethylꢀ1ꢀbutanol decreased due to secondary reactions. This
49 finding was in agreement with a recent study also revealing 99% of (S)ꢀ2ꢀmethylꢀ1ꢀ
17
50 butanol in spirits.
51 The differences in FD factors during AEDA found for esters were confirmed by
52 increased concentrations during fermentation. The fatty acid metabolism of yeasts
42
53 leads to the formation of ethyl esters due to the reaction of acylꢀCoA and ethanol.
54 Depending on the yeast strain, different concentrations of the esters 3ꢀmethylbutyl
43
55 acetate, ethyl butanoate, ethyl hexanoate, and ethyl decanoate were formed. An
56 increase of the branched chain esters ethyl (R)ꢀ2ꢀmethylbutanoate, ethyl (S)ꢀ2ꢀ
57 methylbutanoate, and ethyl 3ꢀmethylbutanoate was also shown during fermentation
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 20 of 53
20
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
58 in the production of Williams Christ pear brandy. Ethanol can react with 2ꢀ and 3ꢀ
59 methylbutanoic acid, stemming from amino acid degradation following the Ehrlich
60 pathway. Consequently, (R)ꢀ2ꢀ and (S)ꢀ2ꢀmethylbutanoic acid and the corresponding
61 ethyl esters showed almost the same ratio. An increase by a factor of 8 was shown
62 for the concentration of ethyl cyclohexanoate, which was already determined in
4
5
63 wines, brandy, and whiskys.
64 Interestingly, also an increase of the phenols (2ꢀmethoxyphenol, 2ꢀmethoxyꢀ4ꢀ
65 propylphenol, 4ꢀmethylphenol, 4ꢀethylꢀ2ꢀmethoxyphenol, and 4ꢀethylphenol) was
4
6
66 found. Steinke and Paulson postulated the formation of 4ꢀethylphenol and 4ꢀethylꢀ
47
67 2ꢀmethoxyphenol from pꢀcoumaric acid and ferulic acid. Chatonnet et al. suggested
68 a formation pathway by yeasts: after decarboxylation of the acids, the phenols are
69 reduced by vinylphenol reductase. The formation of 2ꢀmethoxyphenol, 4ꢀethylꢀ2ꢀ
70 methoxyphenol, and 4ꢀmethylphenol was also shown in a synthetic must containing a
71 precursor extract obtained from four different nonꢀfloral grape varieties. However, the
48
72 formation and the concentrations were strongly dependent on the used yeast.
73 The concentration of (R)ꢀlinalool increased by a factor of 48 compared to only 8
74 for the (S)ꢀenantiomer, indicating that the enzymeꢀcatalyzed release may occur
75 stereo specifically from the respective precursors.
76
Changes during Distillation. 2.9 L of mash yielded 0.4 L of distillate and, thereby,
77 the concentrations of 61% of the quantitated odorants decreased (Table 6).
78 Compounds showing a higher or lower boiling point compared to ethanol might be
79 discriminated during the distillation process leading to a loss of aroma. A significant
80 decrease was shown for the acids (butanoic acid, (R)ꢀ2ꢀmethylbutanoic acid, (S)ꢀ2ꢀ
81 methylbutanoic acid, acetic acid, phenylacetic acid, and decanoic acid) and the
82 phenols (2ꢀmethoxyphenol, 4ꢀethylꢀ2ꢀmethoxyphenol, 2ꢀmethoxyꢀ4ꢀpropylphenol, 4ꢀ
83 methylphenol, 4ꢀethylphenol, and vanillin), but also for the both furanones (3ꢀ
ACS Paragon Plus Environment

Page 21 of 53
Journal of Agricultural and Food Chemistry
21
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
84 hydroxyꢀ4,5ꢀdimethylfuranꢀ2(5H)ꢀone and 4ꢀhydroxyꢀ2,5ꢀdimethylfuranꢀ3(2H)ꢀone),
85 and 2ꢀphenylethanol (Table 6). A significant decrease of 2ꢀphenylethanol and vanillin
86 during simultaneous distillation/extraction (SDE) was already shown for a model
6
87 solution.
88
The highest increase was determined for the concentration of 1,1ꢀdiethoxyethane
89 (factor of 20). In a model experiment, a solution of acetaldehyde (pH = 4.5; 10 Vol.ꢀ
90 % EtOH) was distilled for 60 min. However, no significant increase of the
6
91 concentration of 1,1ꢀdiethoxyethane was determined, although the conditions for the
92 formation of the acetal of acetaldehyde and ethanol were given. Thus, 1,1ꢀ
93 diethoxyethane is probably released during distillation from precursors of the mash or
94 catalysts from the mash promote its formation.
95
The concentration of 2,3ꢀbutanedione increased by a factor of 12 indicating a
96 thermal induced formation. A possible pathway was postulated by an aldol reaction
49
97 from acetaldehyde and hydroxyacetaldehyde (glycolaldehyde).
98 An increase by a factor of 4 was determined for (E)ꢀβꢀdamascenone, which
50
99 confirmed data obtained for unheated and heated grapes. The formation rate was
00 higher if heat treatment was performed under acidic conditions. Therefore, the
01 formation or the release of (E)ꢀβꢀdamascenone from precursors present in the mash
02 during the distillation is very likely. Interestingly, the concentrations of (R)ꢀ and (S)ꢀ
03 linalool increased by ~30 ꢁg/L each, showing that the acidic conditions and the heat
04 treatment enabled a nonꢀstereo specific release of linalool from precursors.
05
In addition, the aldehydes (R)ꢀ2ꢀ, (S)ꢀ2ꢀ, and 3ꢀmethylbutanal showed higher
06 concentrations in the distillate compared to the mash. The heatꢀinduced formation of
07 these Strecker aldehydes by oxidative decarboxylation of amino acids is wellꢀknown.
08 An increase of the concentration of the amino acids revealed an increase of the
51
09 concentration of the aldehydes. During distillation, the concentrations of (R)ꢀ2ꢀ and
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 22 of 53
22
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
10 (S)ꢀ2ꢀmethylbutanal rose almost by the same factor indicating that the formation was
11 thermally induced and, consequently, not stereo specific. However, the concentration
12 of 3ꢀ(methylthio)propanal decreased, which was already shown in another study on
6
13 American whisky, concluding that the higher boiling point leads to a discrimination
14 during distillation.
15
Changes during Ageing. 0.4 L of the distillate are necessary for the production of
16 1 L of rum. Cisꢀ and transꢀwhisky lactone were detected during AEDA of the rum for
17 the first time in the process and were determined with quite high concentrations (cisꢀ
18 whisky lactone: 210 ꢁg/L; transꢀwhisky lactone: 191 ꢁg/L; Table 7). These aroma
52
19 compounds derived from the oak barrels, which are commonly used for ageing. The
20 concentrations strongly depended on the used wood, on the ageing conditions, and
53ꢀ55
21 on the working life of the barrel.
Lower concentrations were found for cisꢀ and
53
22 transꢀwhisky lactones in wines, which were stored in used barrels. In contrast,
23 higher concentrations were found in Merlot (pH = 3.7; 13.6 Vol.ꢀ% ethanol) compared
24 to Cabernet Sauvignon (pH = 3.45; 12.3 Volꢀ% ethanol) differing in pH value and
54
25 ethanol content. In addition, wines aged in European wood barrels showed a ratio
26 of cisꢀ and transꢀwhisky lactone between 1 and 1.5, whereas the ratio of wines aged
55
27 in American wood barrels was between 5 and 8. In the actual study, a ratio of 1.1
28 was determined confirming on a molecular level the ageing in barrels produced of
29 European wood (Spessart).
30
A significant increase was also shown for the phenols 4ꢀethylꢀ2ꢀmethoxyphenol,
31 2ꢀmethoxyphenol, 4ꢀpropylꢀ2ꢀmethoxyphenol, 4ꢀmethylphenol, 4ꢀethylphenol, and
32 vanillin. During barrel making, the formation of these phenolic compounds occurs due
56,57
33 to the thermal treatment and are extracted during ageing.
To a lesser extent, an
34 increase of the phenol concentrations might also be caused by contaminations of the
35 yeasts used. Microbial activity can lead to the degradation of ferulic acid and pꢀ
ACS Paragon Plus Environment

Page 23 of 53
Journal of Agricultural and Food Chemistry
23
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
36 coumaric acid resulting in the formation of ethylphenols, which was already shown for
47
37 wine.
38 The concentration of 3ꢀhydroxyꢀ4,5ꢀdimethylfuranꢀ2(5H)ꢀone increased by a
39 factor of 49. The formation from acetaldehyde and αꢀketobutyric acid was already
58
40 shown in a synthetic wine by aldol condensation, followed by cyclization. Thereby,
41 αꢀketobutyric acid is formed from threonine by enzymeꢀcatalyzed dehydratization and
42 deamination.
43
Also increased concentrations of ethyl esters, like ethyl butanoate, ethyl (S)ꢀ2ꢀ
44 methylbutanoate, ethyl pentanoate, ethyl 3ꢀphenylpropanoate, ethyl (R)ꢀ2ꢀ
45 methylbutanoate, ethyl 3ꢀmethylbutanoate, and ethyl cyclohexanoate were found.
46 Only the concentrations of ethyl hexanoate and ethyl octanoate were more or less
47 the same after ageing and 3ꢀmethylbutyl acetate was present even to a lesser extent,
59,60
48 which was already reported in studies on wines.
In a model study, 2ꢀ and 3ꢀ
49 methylbutanoic acid, methylpropanoic acid, butanoic acid, and hexanoic acid were
50 stored for 479 days in an ethanolic solution (40 Vol.ꢀ%) spiked with white oak wood
51 chips. Due to the reaction of ethanol and the respective acids, a continuous increase
52 of the ethyl esters was determined. In contrast, in a solution without white oak wood
6
53 chips the concentrations only marginally increased.
54
A small increase of (R)ꢀ2ꢀ, (S)ꢀ2ꢀ, and 3ꢀmethylbutanal was found, whereas 3ꢀ
55 (methylthio)propanal showed an increase by a factor of 8. In an ethanolic solution
56 (12 Vol.ꢀ%) of amino acids (valine, leucine, isoleucine, methionine, and
61
57 phenylalanine) and αꢀdicarbonyls (pH 3.5; 25 °C; in the dark), PripisꢀNicolau et al.
58 reported about the formation of the respective aldehydes following the Strecker
59 reaction, even at these moderate temperatures. In the actual study, a nonꢀstereo
60 specific formation (decarboxylation of isoleucine) of 2ꢀmethylbutanal probably
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 24 of 53
24
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
61 occurred ((R)ꢀenantiomer: + 243 ꢁg; (S)ꢀenantiomer: + 213 ꢁg; Table 7) leading to a
62 changed enantiomeric ratio of 24/76 (R/S) in the distillate to 34/66 (R/S) in the rum.
63
In contrast, a significant decrease by almost 90% was determined for the highly
64 volatile 2,3ꢀbutanedione, already proven in a model solution also showing a decrease
6
65 of 90% after five months.
66
Changes of the Key Odorants during the Whole Process. In the following, a
67 short overview is given about the processing steps mostly influencing the formation
68 of the specific aroma compounds. The concentration trends of selected compounds
69 during the whole production process are summarized in Table 8. In addition, a factor
70 representing the ratio of the concentration in the rum (1 L) divided by the respective
71 concentration in the molasses (0.72 L, initial amount needed for the production of 1 L
72 of rum) was calculated.
73
(E)ꢀβꢀDamascenone was determined as key aroma compound in the rum and
74 showed by far the highest OAV (3280). A concentration leading to a high OAV was
75 already determined in the molasses (87 ꢁg/0.72 L). The fermentation and the ageing
76 showed almost no influence, but during distillation a significant increase of the
77 concentration was determined.
78
For (R)ꢀ2, (S)ꢀ2ꢀ, and 3ꢀmethylbutanal, 2,3ꢀbutanedione, and 1,1ꢀdiethoxyethane,
79 an OAV ≥ 1 was determined in rum. The aldehydes increased during distillation and
80 ageing, whereas 2,3ꢀbutanedione and 1,1ꢀdiethoxyethane increased during
81 fermentation and distillation.
82
Vanillin, 2ꢀmethoxyphenol, 4ꢀethylꢀ2ꢀmethoxyphenol, 2ꢀmethoxyꢀ4ꢀpropylphenol,
83 and 4ꢀethylphenol were already determined in the molasses. Their concentrations
84 increased during fermentation and significantly decreased during distillation.
85 However, during ageing in the wood barrels, their concentrations increased again.
ACS Paragon Plus Environment

Page 25 of 53
Journal of Agricultural and Food Chemistry
25
5
5
5
5
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
86
Most of the ethyl esters (e.g., ethyl butanoate, ethyl 2ꢀmethylbutanoate, ethyl
87 hexanoate) and the alcohols (e.g., (S)ꢀ2ꢀ, 3ꢀmethylꢀ1ꢀbutanol, methylpropanol, and 2ꢀ
88 phenylethanol) were formed during fermentation in very high concentrations.
89 Although they were discriminated during distillation, an OAV ≥ 1 was determined in
90 rum.
91
4ꢀHydroxyꢀ2,5ꢀdimethylꢀ3ꢀ(2H)ꢀfuranone and 3ꢀhydroxyꢀ4,5ꢀdimethylfuranꢀ2(5H)ꢀ
92 one showed a relative high concentration in the molasses, but, a quantitative transfer
93 from the raw material into the rum was not found.
94
In summary, the present study demonstrated for the first time the importance
95 and the high influence of the single processing steps on the key aroma compounds in
96 rum. The concentrations of 68% of the quantitated aromaꢀactive compounds
97 increased, e.g., ethyl esters, phenols, and alcohols. In contrast, a decrease was
98 shown for the acids. With these data at hand, manufacturer can draw conclusions at
99 which step they have potential to optimize their product in matters of its overall
00 aroma, which is an important quality criteria for consumers. However, also the raw
01 material might have a significant influence on the key aroma compounds due to the
02 fact that many compounds might be released or formed from precursors present in
03 the raw material. Thus, further studies on aromaꢀactive compounds in rums produced
04 from molasses or sugar cane juice are currently undertaken.
05
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 26 of 53
26
606 REFERENCES
6
07 (1)
Hashizume, T.; Yamagami, T.; Sasaki, Y. Constituents of cane molasses. Part
II. Separation and identification of the phenolic compounds. Agri. Biol. Chem.
1967, 31, 324ꢀ329.
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
08
09
10 (2)
11
Abe, E.; Nakatani, Y.; Yamanishi, T.; Muraki, S. Studies on the “sugary flavor”
of raw cane sugar. I. Proc. Japan Acad. Ser. B 1978, 54, 542ꢀ547.
Tokitomo, Y.; Kobayashi, A.; Yamanishi, T.; Muraki, S. Studies on the "sugary
flavor" of raw cane sugar. II. Proc. Japan Acad. Ser. B 1980, 56, 452ꢀ456.
Tokitomo, Y.; Kobayashi, A.; Yamanishi, T.; Muraki, S. Studies on the "sugary
flavor" of raw cane sugar. III. Key compound of the sugary flavor. Proc. Japan
Acad. Ser. B 1980, 56, 457ꢀ462.
12 (3)
13
14 (4)
15
16
17 (5)
18
Nishimura, K.; Matsuyama, R. Maturation and maturation chemistry. In The
Science and Technology of Whiskies. Pigott, J. R., Sharp, R., Duncan, R. E.
B., Eds.; Longman Scientific and Technical: Harlow, UK, 1989, pp 253ꢀ263.
Vocke, M. The influence of processing steps on the formation of important
aroma compounds in American whisky (in German). Ph.D. thesis, Technical
University of Munich, Munich, Germany, 2008.
19
20 (6)
21
22
23 (7)
24
Masuda, M.; Nishimura, K. Branched nonalactones from some Quercus
species. Phytochemistry 1971, 10, 1401ꢀ1402.
25 (8)
26
Günther, C.; Mosandl, A. Stereoisomeric aroma substances XV. Chirospezific
analysis of natural aroma components: 3ꢀmethylꢀ4ꢀoctanolidꢀ„quercusꢀ, whisky
lactone" (in German). Z. Lebensm.ꢀUnters. Forsch. 1987, 185, 1ꢀ4.
Pino, J. A.; Tolle, S.; Gök, R.; Winterhalter, P. Characterisation of odourꢀactive
compounds in aged rum. Food Chem. 2012, 132, 1436ꢀ1441.
27
28 (9)
29
ACS Paragon Plus Environment

Page 27 of 53
Journal of Agricultural and Food Chemistry
27
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
30 (10) Franitza, L.; Granvogl, M.; Schieberle, P. Characterization of the key aroma
31
32
compounds in two commercial rums by means of the sensomics approach. J.
Agric. Food Chem. 2016, 64, 637ꢀ645.
33 (11) Uselmann, V.; Schieberle, P. Decoding the combinatorial aroma code of a
34
35
36
commercial cognac by application of the Sensomics concept and first insights
into differences from a German brandy. J. Agric. Food Chem. 2015, 63, 1948ꢀ
1956.
37 (12) Bober, A.; Haddaway, L. W. Gas chromatographic identification of alcoholic
38 beverages. J. Chrom. Sci. 1963, 1, 8ꢀ13.
39 (13) Maarse, H.; Ten Noever de Brauw, M. C. Analysis of volatile components of
40 Jamaica rum. J. Food Sci. 1966, 31, 951ꢀ955.
41 (14) De Souza, M. D. C. A.; Vasquez, P.; Del Mastro, N. L.; Acree, T. E.; Lavin, E.
42
43
H. Characterization of cachaҫa and rum aroma. J. Agric. Food Chem. 2006,
54, 485ꢀ488.
44 (15) Regulation (EC) No 110/2008 of the European parliament and of the council of
45
46
47
15 January 2008 on the definition, description, presentation, labeling and the
protection of geographical indications of spirit drinks and repealing council
regulation (EEC) no. 1576/89.
48 (16) Schieberle, P.; Hofmann, T. Mapping the combinatorial code of food flavors by
49
50
51
52
means of molecular sensory science approach. In Chemical and Functional
Properties of Food Components Series. Food Flavors. Chemical, Sensory and
Technological Properties; Jelen, H., Ed.; CRC Press: Boca Raton, FL, USA,
2012, pp 413ꢀ438.
53 (17) Matheis, K.; Granvogl, M.; Schieberle, P. Quantitation and enantiomeric ratios
54
55
of aroma compounds formed by an Ehrlich degradation of Lꢀisoleucine in
fermented foods. J. Agric. Food Chem. 2016, 64, 646ꢀ652.
ACS Paragon Plus Environment

Journal of Agricultural and Food Chemistry
Page 28 of 53
28
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
56 (18) Schieberle, P.; Hofmann, T. Evaluation of the character impact odorants in
57
58
fresh strawberry juice by quantitative measurements and sensory studies on
model mixtures. J. Agric. Food Chem. 1997, 45, 227ꢀ232.
59 (19) Schieberle, P.; Gassenmeier, K.; Guth, H.; Sen, A.; Grosch, W. Character
60
61
impact odour compounds of different kinds of butter. Lebensm.ꢀWiss. Technol.
1993, 26, 347ꢀ356.
62 (20) Sen, A.; Laskawy, G.; Schieberle, P.; Grosch, W. Quantitative determination of
63
64
βꢀdamascenone in foods using a stable isotope dilution assay. J. Agric. Food
Chem. 1991, 39, 757ꢀ759.
65 (21) Guth, H.; Grosch, W. Deterioration of soyaꢀbean oil: quantification of primary
66
67
flavor compounds using a stable isotope dilution assay. Lebensm.ꢀWiss.
Technol. 1990, 23, 513ꢀ522.
68 (22) Czerny, M.; Schieberle, P. Influence of the polyethylene packaging on the
69
70
adsorption of odourꢀactive compounds from UHTꢀmilk. Eur. Food Res.
Technol. 2007, 225, 215ꢀ223.
71 (23) Guth, H. Quantitation and sensory studies of character impact odorants of
72 different white wine varieties. J. Agric. Food Chem. 1997, 45, 3027ꢀ3032.
73 (24) Semmelroch, P.; Laskawy, G.; Blank, I.; Grosch, W. Determination of potent
74
75
odourants in roasted coffee by stable isotope dilution assays. Flavour
Fragrance J. 1995, 10, 1ꢀ7.
76 (25) Guth, H.; Grosch, W. Quantitation of potent odorants of virgin olive oil by
77 stableꢀisotope dilution assays. J. Am. Oil Chem. Soc. 1993, 70, 513ꢀ518.
78 (26) Gassenmeier, K.; Schieberle, P. Potent aromatic compounds in the crumb of
79
80
81
wheat bread (Frenchꢀtype) ꢀ influence of preꢀferments and studies on the
formation of key odorants during dough processing. Z. Lebensm.ꢀUnters.
Forsch. 1995, 201, 241ꢀ248.
ACS Paragon Plus Environment

Page 29 of 53
Journal of Agricultural and Food Chemistry
29
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
7
7
7
7
7
7
7
82 (27) Pfnür, P. Studies on chocolate aroma (in German). Ph.D. thesis, Technical
83 University of Munich, Munich, Germany, 1998.
84 (28) Blank, I.; Schieberle, P.; Grosch, W. Quantification of the flavour compounds
85
86
87
88
89
3ꢀhydroxyꢀ4,5ꢀdimethylꢀ2(5H)ꢀfuranone and 5ꢀethylꢀ3ꢀhydroxyꢀ4ꢀmethylꢀ2(5H)ꢀ
furanone by a stable isotope dilution assay. In Progress in Flavour Precursor
Studies: Analysis, Generation, Biotechnology, Proceedings of the International
Conference; Schreier, P., Winterhalter, P., Eds.; Allured Publishing
Corporation: Carol Stream, IL, 1993, pp 103ꢀ109.
90 (29) Steinhaus, M.; Fritsch, H. T.; Schieberle, P. Quantitation of (R)ꢀand (S)ꢀlinalool
91
92
in beer using solid phase microextraction (SPME) in combination with a stable
isotope dilution assay (SIDA). J. Agric. Food Chem. 2003, 51, 7100ꢀ7105.
93 (30) Cerny, C.; Grosch, W. Quantification of characterꢀimpact odour compounds of
94 roasted beef. Z. Lebensm.ꢀUnters. Forsch. 1993, 196, 417ꢀ422.
95 (31) Schmitt, R. On the role of ingredients as sources of key aroma compounds in
96
97
Crumb chocolate (in German). Ph.D. thesis, Technical University of Munich,
Munich, Germany, 2005.
98 (32) Granvogl, M.; Beksan, E.; Schieberle, P. New insights into the formation of
99
00
aromaꢀactive Strecker aldehydes from 3ꢀoxazolines as transient intermediates.
J. Agric. Food Chem. 2012, 60, 6312ꢀ6322.
01 (33) Guth, H.; Grosch, W. Identification of the character impact odorants of stewed
02
03
beef juice by instrumental analyses and sensory studies. J. Agric. Food Chem.
1994, 42, 2862ꢀ2866.
04 (34) Fuhrmann, E. Study on the aroma of apples in dependancy on the variety and
05
06
research on structure odor relationships of involved esters (in German). Ph.D.
thesis, Technical University of Munich, Munich, Germany, 1998.
ACS Paragon Plus Environment