Synthesis and olfactory evaluation of optically active β-alkyl substituted γ-lactones and whiskey lactone analogues

Article abstract of DOI:10.1016/j.tet.2020.130984

Optically active β-alkyl substituted γ-lactones and whiskey lactone analogues were synthesized, and the odor properties were evaluated. During the preparation of the chiral intermediates, we found good reaction conditions for the highly enantioselective esterification of 3-arylmethyl-2-methyl-1-propanols to kinetically resolve them. The results of the olfactory evaluations of the synthesized lactones revealed that the alkyl groups on the γ-lactone rings played an important role for the odor profiles.

Full text of DOI:10.1016/j.tet.2020.130984

Journal Pre-proof
Synthesis and olfactory evaluation of optically active β-alkyl substituted γ-lactones
and whiskey lactone analogues
Masashi Kawasaki, Daiki Kato, Takuya Okada, Yuko Morita, Yasuo Tanaka, Naoki
Toyooka
PII:
S0040-4020(20)30079-X
https://doi.org/10.1016/j.tet.2020.130984
TET 130984
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 2 December 2019
Revised Date: 23 January 2020
Accepted Date: 29 January 2020
Please cite this article as: Kawasaki M, Kato D, Okada T, Morita Y, Tanaka Y, Toyooka N, Synthesis
and olfactory evaluation of optically active β-alkyl substituted γ-lactones and whiskey lactone analogues,
Tetrahedron (2020), doi: https://doi.org/10.1016/j.tet.2020.130984.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Synthesis and olfactory evaluation of
optically active ß-alkyl substituted γ-lactones
and whiskey lactone analogues
Leave this area blank for abstract info.
Masashi Kawasakia,*, Daiki Katob, Takuya Okadab, Yuko Moritaa, Yasuo Tanakac, Naoki Toyookab
a Toyama Prefectural University, b University of Toyama, c TAIYO CORPORATION

Synthesis and olfactory evaluation of optically active β-alkyl substituted γ-lactones
and whiskey lactone analogues
b
b
c
d
Masashi Kawasaki a, *, Daiki Kato , Takuya Okada , Yuko Morita , Yasuo Tanaka ,
Naoki Toyooka b
a
Center for Liberal Arts and Sciences, Faculty of Engineering, Toyama Prefectural
University, 5180 Kurokawa, Imizu, Toyama-ken 939-0398, Japan
b
Graduate School of Science and Engineering, University of Toyama, 3190 Gofuku,
Toyama 930-8555, Japan
c
Department of Bioresources Science, College of Technology, Toyama Prefectural
University, 5180 Kurokawa, Imizu, Toyama-ken 939-0398, Japan
d TAIYO CORPORATION, 1-6-27 Higashiawaji, Yodogawa-ku, Osaka 533-0023, Japan
Abstract
Optically active β-alkyl substituted γ-lactones and whiskey lactone analogues were
synthesized, and the odor properties were evaluated. During the preparation of the chiral
intermediates, we found good reaction conditions for the highly enantioselective
esterification of 3-arylmethyl-2-methyl-1-propanols to kinetically resolve them. The
results of the olfactory evaluations of the synthesized lactones revealed that the alkyl
groups on the γ-lactone rings played an important role for the odor profiles.
* Corresponding author.
E-mail address: kawasaki@pu-toyama.ac.jp (M. Kawasaki).
1

Keywords: Asymmetric synthesis; Lactone; Lipase; Odor; Whiskey lactone
1. Introduction
γ-Lactones are compounds containing a five-membered ring [1] and constitute the
structural cores of numerous natural products [2]. They also occur in foods such as dairy
products [3,4], fruit [5-7], and nuts [8,9] and are widely used as flavor and fragrance
chemicals [10]. Representative of γ-lactones are shown in Fig. 1 [1,11]. For example,
γ-decalactone 9f, present in a wide variety of foods, has an odor reminiscent of peaches
and is used in perfumery for flower odors. Whiskey lactone 10, found as a mixture of
cis (4S, 5S)- and trans (4S, 5R)-isomers in aged spirits such as whiskey and oakwood
volatiles [12], has an intense, warm, sweet, and coumarin-like odor and is used in aroma
compositions, for example, for beverages. Sotolone 11, found in coffee and sake, is used
in food flavoring, and its odor changes from a caramel-like one at low concentrations to
a curry-like one at high concentrations.
2

γ-Alkyl substituted γ-lactones 9a-h are especially well-known γ-lactones [13]. They
occur in various kinds of foods and are used for flavor ingredients in (non)alcoholic
beverages, baked goods, gelatins, puddings, and soft candies. Mosandl et al. reported
interesting findings on the odor properties of 9a-h [14]. They synthesized all 16
enantiomers of 9a-h and found that the (R)-enantiomers were responsible for pleasant,
naturally fruity, and aroma notes. The enantiomers of many chemical compounds are
perceived differently as odorants by the human nose [15].
Many flavors and fragrances occur in nature as specific enantiomers, and the aromas
of such specific enantiomers can be distinctive and characteristic [10]. The synthesis of
the enantiomers of fragrance ingredients and the evaluation of the odor properties of the
enantiomers are accordingly of great interest and have been studied [16,17]. Besides
Mosandl’s report [14], there are reports on the evaluation of the odors of the
enantiomers of γ-lactones [17a,k,m]. Shimotori et al. synthesized all four stereoisomers
of whiskey lactone 10. Although all the stereoisomers showed different odors, they had
3

a coconut-like odor commonly [17k]. Dai et al. synthesized all 12 enantiomers of
γ-alkyl substituted α, β-unsaturated γ-lactone and reported that all the lactones more or
less exhibited chiral discrimination in odor quality [17m]. We have been interested in
the differences in the olfactory properties of the enantiomers of fragrance ingredients
and have reported on the synthesis and the olfactory evaluations of the enantiomers of
floral odorants, citrus odorants, and musky odorants [18-22], and are currently
interested in researching differences in the olfactory properties of the enantiomers of
γ-lactones with odors that have not been reported.
Compared with the γ-alkyl substituted γ-lactones 9a-h, little is known about β-alkyl
substituted γ-lactones 1a-h. Lactone 1a is a volatile component of a condensed
fermented corn extractive [23] and contained in the smoke produced during the pyrolytic
process of thyme [24]. To the best of our knowledge, the occurrence of 1b-h in nature
has not been reported. Although the use of racemic 1a as a food flavoring has been
mentioned in several patents [25-29], there have been no similar reports on the
possibility of using 1b-h. Furthermore, the relationship of the absolute stereochemistry
and odor properties of 1a-h have not been reported. We herein report the synthesis of all
16 enantiomers of β-alkyl substituted γ-lactones 1a-h and the odor profiles.
We recently reported that the odors of Phantolide®, a representative synthetic musk
odorant, analogues (±)-12a-d were stronger than those of Phantolide® analogues
(±)-13a-d (Fig. 2) [22]. The positions of the acyl group and that of the methyl group
contributed to the difference in the intensity of the odor. Whiskey lactone 10 has a
methyl group and a n-butyl group on its five-membered ring. We therefore have been
interested in the odor properties of whiskey lactone analogues 2, which differs in the
positions of the methyl group and the n-butyl group from 10. We here also report on the
4

synthesis of all four stereoisomers of whiskey lactone analogue 2 and the evaluation of
their odor profiles.
2. Results and discussion
2.1. Synthesis of optically active β-alkyl substituted γ-lactones (R)- and (S)-1a-h
Ciceri et al. recently reported the synthesis of optically active β-propyl-γ-lactone
(R)-1c [30]. They resolved racemic alcohol (±)-4c shown in Scheme 1 into the
enantiomers by lipase-catalyzed enantioselective transesterification with vinyl acetate
and then transformed the resulting optically active alcohol (R)-4c into (R)-1c. We
followed their method to synthesize all 16 enantiomers of 1a-h. We began with the
synthesis of racemic primary alcohols (±)-4a-h according to the route shown in Scheme
1. The primary alcohol (±)-4a was prepared from α-methylhydrocinnamic acid.
Malonates 3b-d were prepared from the corresponding diethyl alkylmalonates and
benzyl bromide, and malonates 3e-g were prepared from the corresponding alkyl
iodides and diethyl benzylmalonate. Malonates 3b-g were hydrolyzed to the
corresponding dicarboxylic acids. The decarboxylation of the diacids gave the
5

corresponding carboxylic acids, and the reduction of the acids gave the alcohols
(±)-4b-g. Alcohol (±)-4h was prepared from ethyl 2-methylphenyldecanoate (±)-6h
prepared from ethyl decanoate and benzyl bromide.
The most important point of the present syntheses is to prepare all the primary
alcohols 4a-h in highly optically active forms. We achieved that by the optical
resolution of (±)-4a-h via the lipase-catalyzed enantioselective transesterification of
6

(±)-4a-h with vinyl acetate (Scheme 2). Lipase, a kind of hydrolase, has been used as a
catalyst for not only the enantioselective hydrolysis of esters in aqueous media but also
the (trans)esterification of alcohols or carboxylic acids in organic solvents, and many
esters, alcohols, and carboxylic acids have been kinetically resolved [31]. We have also
been using lipase for the kinetic resolution of some alcohols [21,22].
We initially describe the optical resolution of (±)-4a. Santaniello et al. showed that
lipase from Burkholderia cepacia could catalyze the highly enantioselective (E value
[32] >100) transesterification of (±)-4a with vinyl acetate in chloroform [33]. We,
therefore, decided to use the lipase, called Amano PS, as a catalyst and then investigated
the effects of reaction solvents on the enantioselectivity of the lipase to find better
conditions for the enantioselective transesterification (Table 1). Because solvent
hydrophobicity effects the enantioselectivity of enzymatic catalysis [34], we tried five
organic solvents which varies in hydrophobicity. As a result of the screening of the five
organic solvents, we found that toluene was the most suitable solvent (E = 207) for the
transesterification. Next, we tried the small-scale transesterifications of (±)-4b-h with
vinyl acetate in toluene (Table 2). All the transesterifications proceeded highly
7

enantioselectively, and all the (S)-enantiomers were selectively esterified. Neither the
(trans)esterification of primary alcohols catalyzed by lipase from Burkholderia cepacia
nor the hydrolysis of the esters of primary alcohols by the same one proceeds highly
enantioselectively [35,36]. Our findings here are exceptions. Furthermore, Nordin et al.
reported the transesterification of 3-arylmethyl-2-methyl-1-propanols catalyzed by the
lipase proceeded highly enantioselectively [37]. Our and Nordin’s findings suggest that
a combination of Amano PS, vinyl acetate and toluene can be the first choice for the
conditions for the resolution of 3-arylmethyl-2-alkyl-1-propanols.
Table 1
4a
Amano PS-catalyzed transesterification of (±)- with vinyl acetate.
conversiona
(%)
5a
4a
ee of (S )-
ee of (R)-
Ea
solvent
time (h)
(%)
(%)
1,4-dioxane
dichloromethane
ethyl acetate
vinyl acetate
toluene
2.3
5.0
2.3
2.0
2.5
50
49
50
51
50
94
95
95
94
93
90
94
98
97
110
121
139
149
207
96
-1
-1
-1
4a
Conditions: Amano PS (20 mg mL ), (±)- (0.1 mol L ), vinyl acetate (1 mol L ), room
temperature. Refer to Section 4.2.4 for details.
a
5a
4a
Calculated from the ee of (S)- and that of (R)- [32].
8

Table 2
Amano PS-catalyzed transesterification of (±)- - with vinyl acetate in toluene.
4b h
conversiona
(%)
5b h
4b h
-
ee of (S )-
-
ee of (R)-
Ea
alcohol
time (h)
(%)
(%)
4b
4c
4d
4e
(±)-
(±)-
(±)-
(±)-
(±)-
7.5
4.6
6.7
1.8
1.8
2.0
3.1
51
54
52
51
51
51
92
82
86
93
91
92
90
97
96
92
96
93
94
92
101
39
43
108
72
4f
4g
4h
(±)-
(±)-
85
51
62
-1
-1
-1
4b h
Conditions: Amano PS (10-100 mg mL ), (±)- - (0.05-0.1 mol L ), vinyl acetate (0.5-1 mol L ),
room temperature. Refer to Section 4.2.4 for details.
a
5b h
4b h
Calculated from the ees of (S )- - and those of (R)- - [32].
On the basis of the results shown in Table 1, we conducted the large-scale resolution
of (±)-4a in toluene to obtain (S)-5a with 88% ee and (R)-4a with >99% ee, respectively
(Scheme 3). The absolute configuration of (R)-4a was established by comparison of the
value of optical rotation with that reported in the literature (See Section 4.2.5). We could
also obtain (R)-4b-h with ≥99% ee and (S)-5b-h with 73-88% ee in the same manner.
The absolute configurations of (R)-4b-h were established by comparison of the value of
optical rotations with those reported in the literatures (See Supporting information).
9

Ester (S)-5a obtained in the transesterification did not have a sufficiently high ee for
use, and therefore we tried to prepare (S)-5a with a higher ee. Ester (S)-5a was
hydrolyzed to the corresponding alcohol (S)-4a by the use of sodium hydroxide solution,
and then it was subjected to Amano PS-catalyzed transesterification with vinyl acetate
in toluene again to obtain (S)-5a with >99% ee (Scheme 3). We could also obtain
(S)-5b-h with ≥99% ee in the same manner.
The treatment of (R)-4a-h with AcCl afforded the corresponding acetate (R)-5a-h
(Scheme 4). The acetates (R)-5a-h were treated with H5IO6 and RuCl3 to give the
corresponding γ-acetoxycarboxylic acids. The synthesis of (R)-1a-h was accomplished
by the hydrolysis of the ester moiety of the γ-acetoxycarboxylic acids and the
subsequent lactonization of the resultant γ-hydroxycarboxylic acids. We could also
synthesize lactones (S)-1a-h from (S)-5a-h in the same manner.
10

R
-
R
-
R
HO
AcO
R
O
a
b
AcO
c, d
CO2H
O
( )-
( )-
( )-
-
R
AcO
R
O
b, c, d
O
( )-
-
( )-
-
. Preparation of ( )- and ( )- - . (a)AcCl, pyridine, CH2Cl2, 0 ºC
to rt, ( )- : 95%; ( )- : 89%; ( )- : 94%; ( )- : 94%; ( )- : 92%; ( )-
: 94%; ( )- : 90%; ( )- : 86%; (b) RuCl3, H5IO6, H2O, CCl4, CH3CN, 0
ºC to rt, (c) aq NaOH, rt, (d) aq HCl, rt, ( )- : 41% from ( )- ; ( )-
36% from ( )- ; ( )- : 58% from ( )- ; ( )- : 60% from ( )- ; ( )-
57% from ( )- ; ( )- : 55% from ( )- ; ( )- : 46% from ( )- ; ( )-
46% from ( )- ; ( )- : 41% from ( )- ; ( )- : 37% from ( )- ; ( )-
58% from ( )- ; ( )- : 61% from ( )- ; ( )- : 55% from ( )- ; ( )-
:
:
:
:
:
56% from ( )- ; ( )- : 52% from ( )- ; ( )- : 44% from ( )-
.
2.2. Synthesis of all four stereoisomers of whiskey lactone analogues 2
We began with the synthesis of secondary alcohol (3R)-7 according to the route
shown in Scheme 5. The primary alcohol (R)-4d prepared in Section 2.1 was oxidized
with Dess-Martine periodinane to give the corresponding aldehyde, and the addition of
methylmagnesium iodide to the aldehyde gave (3R)-7, which was composed of 59% of
the syn-isomer (2S, 3R)-7 and 41% of the anti-isomer (2R, 3R)-7.
11

Unfortunately, the diastereomers could not be separated with silica gel column
chromatography. However, we could achieve separation via the lipase (Novozym
435)-catalyzed diastereoselective transesterification of (3R)-7 with vinyl acetate,
working as an acyl donor and solvent. We first obtained (1R, 2R)-8 and highly
diastereomerically pure (2S, 3R)-7 with >99% ee (Scheme 5). The ester (1R, 2R)-8 was
not highly diastereomerically pure (syn:anti = 3:97), and therefore we tried to prepare
highly diastereomerically pure (1R, 2R)-8. The ester (1R, 2R)-8 was hydrolyzed to the
corresponding alcohol (2R, 3R)-7 by the use of aqueous NaOH, and then it was
subjected to Novozym 435-catalyzed transesterification with vinyl acetate again to
obtain highly diastereomerically pure (1R, 2R)-8 with >99% ee.
12

The diastereomerically and enantiomerically pure alcohol (2S, 3R)-7 was treated
with AcCl to afford the corresponding acetate (1S, 2R)-8 (Scheme 6). The acetate (1S,
2R)-8 was treated with H5IO6 and RuCl3 to give the corresponding γ-acetoxycarboxylic
acid. The synthesis of (4R, 5S)-2 was accomplished by the hydrolysis of the ester
moiety of the γ-acetoxycarboxylic acid and the subsequent lactonization of the resultant
1
γ-hydroxycarboxylic acid. The H NMR data of the trans isomer of lactone 2, though
racemic, have been reported [38], and the data were identical to those of (4R, 5R)-2.
Therefore, the absolute configuration of C5 of the lactone ring of (4R, 5R)-2 is R. We
could also synthesize (4R, 5R)-2 from (1R, 2R)-8 in the same manner.
We could also synthesize (4S, 5R)-2 and (4S, 5S)-2 from (S)-4d in the same manner
as (4R, 5S)-2 and (4R, 5R)-2 (Scheme 7).
13

2.3. Olfactory evaluation of lactones 1a-h, 2
The odor profiles of the chiral lactones and racemic β-alkyl substituted γ-lactones
(±)-1a-h (See the experimental section and supporting information for the preparation)
were evaluated by seven skilled perfumers.
The results of the olfactory evaluation of all optically active β-alkyl substituted
γ-lactones (R)- and (S)-1a-h are shown in Table 3. Mosandl et al. used 17 keywords to
express the odor profiles of optically active γ-alkyl substituted γ-lactones (R)- and
(S)-9a-h [14]. We also evaluated the odor properties of (R)- and (S)-1a-h by the use of
the keywords here and showed the five keywords that showed the interesting results of
evaluations. The numerical values in Table 3 express how many among the seven
perfumers perceived the odor properties described with the keywords. All the results of
14

the olfactory evaluation of (R)- and (S)-1a-h are shown in Supporting information.
Table 3
1a h
Olfactory evaluation of (R)- and (S )- - .
1a
1b
1c
1d
1e
1f
1g
1h
(R)-
(R)-
(R)-
(R)-
(R)-
(R)-
(R)-
(R)-
fatty
2
2
1
4
4
3
2
3
4
4
6
3
3
6
6
1
6
6
4
1
3
5
4
1
4
5
4
1
4
5
4
1
aldehydic
green
sweet
coconut
5
(S )-
0
3
(S )-
1
4
(S )-
3
1
(S )-
5
3
(S )-
4
1
(S )-
7
0
(S )-
6
0
(S )-
7
1a
1b
1c
1d
1e
1f
1g
1h
fatty
aldehydic
green
1
2
1
2
6
5
6
4
4
3
5
4
7
4
6
2
sweet
4
3
2
4
1
2
1
1
coconut
6
4
2
1
0
0
1
2
More than half of the perfumers perceived fatty odors in (R)-1b, c, e, g, h, and
(S)-1d-h, and interestingly, the number of perfumers who perceived fatty odors in
(S)-1a-h roughly increased with the length of the alkyl groups of (S)-1a-h. Mosandl et
al. reported that fatty odors were perceived in (R)-9b, e-g and (S)-9c-h [14]. Fatty odors
were roughly found to be ones common to the β- and γ-alkyl substituted γ-lactones.
More than half of the perfumers perceived aldehydic odors in (R)- and (S)-1c-h and
green odors in (R)-1c-h and (S)-1c, d, f, g. Mosandl et al. reported that aldehydic odors
were perceived in (R)-9h and (S)-9g and a green odor was perceived in (R)-9d [14].
These results led to the conclusion that the aldehydic odors and the green odors could be
attributed to the existence of the alkyl groups in the β-positions of γ-lactone rings.
The number of perfumers who perceived sweet odors in (R)- and (S)-1a-h roughly
decreased with the length of the alkyl groups of (R)- and (S)-1a-h, and only three
15

lactones (R)-1a, (S)-1a, and (S)-1d had sweet odors which more than half of the
perfumers perceived. On the other hand, 11 chiral lactones among the 18 chiral γ-alkyl
substituted γ-lactones (R)- and (S)-9a-h had sweet odors [14]. The number of perfumers
who perceived coconut odors in (R)- and (S)-1a-h also roughly decreased with the
length of the alkyl groups of (R)- and (S)-1a-h, and only four lactones (R)-1a, (R)-1c,
(S)-1a, and (S)-1b had coconut odors which more than half of the perfumers perceived.
On the other hand, eight chiral lactones among the 18 chiral γ-alkyl substituted
γ-lactones (R)- and (S)-9a-h had coconut odors [14]. These results led to the conclusion
that the coconut and the sweet odors could be attributed to the existence of the alkyl
groups in the γ-positions of γ-lactone rings.
We have reported interesting interactions, such as masking, mutual negation and
combination effects, between the odors of the enantiomers of the Phantolide® analogues
[22] and therefore evaluated not only the odor profiles of the optically active β-alkyl
substituted γ-lactones (R)- and (S)-1a-h but also those of the corresponding racemates
(±)-1a-h. The results of the olfactory evaluation of (±)-1a-h were shown in Table 4.
Although more than half of the perfumers perceived that both (R)- and (S)-1a had
coconut odors, only one perfumer perceived that (±)-1a had a coconut odor. The odors
of (R)- and (S)-1a negated each other’s coconut odors, and therefore (±)-1a did not
show a clearly perceptible coconut odor. The mutual negation effect was also observed
in the green odor. Only a small number of perfumers perceived green odors in (±)-1a-g
compared to in (R)- and (S)-1a-g.
16

Table 4
Olfactory evaluation of (±)- - .
1a h
1a
1b
1c
1d
1e
1f
1g
1h
(±)-
2
(±)-
2
(±)-
2
(±)-
3
(±)-
2
(±)-
4
(±)-
6
(±)-
6
fatty
aldehydic
green
2
0
0
1
4
4
4
1
2
2
3
3
4
3
6
3
sweet
6
5
3
2
1
1
1
2
coconut
1
4
3
4
3
0
1
0
More than half of the perfumers perceived that lactone (±)-1d had a coconut odor,
but only one perfumer perceived that both (R)- and (S)-1d had a coconut odor. The
combination of the odors of (R)- and (S)-1d produced a clearly perceptible coconut
odor.
The results of the olfactory evaluation of whiskey lactone analogue 2 are shown in
Table 5. A difference in the odor profiles was found between the enantiomers. (4R, 5S)-
and (4R, 5R)-2 had coconut odors although weak. However, neither (4S, 5R)- nor (4S,
5S)-2 had coconut odors.
Shimotori et al. reported that all four stereoisomers of whiskey lactone 10 had
17

coconut odors [17k]. However, (4R, 5S)- and (4R, 5R)-2 had slightly coconut-like odors.
Furthermore, no perfumers perceived coconut odors in (4S, 5R)- or (4S, 5S)-2. These
results led to the conclusion that the positions of the methyl groups and the n-butyl
groups contributed to the presence and the intensity of the coconut odors. Interestingly,
(4S, 5R)-2 did not even have a lactone-like odor.
3. Conclusions
We synthesized optically active β-alkyl substituted γ-lactones 1a-h. During the
preparations of the intermediates of the chiral lactones, we found good reaction
conditions for the lipase-catalyzed enantioselective transesterification of
3-arylmethyl-2-alkyl-1-propanols to kinetically resolve: Amano PS as lipase, vinyl
acetate as an acyl donor, and toluene as solvent. We believe that our findings will reduce
the time taken to screen for the optimal conditions for the resolutions of
3-arylmethyl-2-alkyl-1-propanols.
We compared the odor profiles of 1a-h with the odor profiles of γ-alkyl substituted
γ-lactones 9a-h, which Mosandl et al. originally reported [14]. The existence of the
alkyl groups in the β-positions of γ-lactone rings were attributed to aldehydic odors and
green odors and the existence of the alkyl groups in the γ-positions of γ-lactone rings
were attributed to coconut odors and sweet odors.
We also found the mutual negation effect of the enantiomers of some β-alkyl
substituted γ-lactones by evaluating not only the odors of the optically active isomers
but also those of the racemates. This finding shows the importance of the olfactory
evaluation of the enantiomers of odorous compounds, and even racemic compounds
18

establishing a firm position as odorous compounds.
We could synthesize all four stereoisomers of whiskey lactone analogues 2 by the
use of the chiral intermediates of (R)- and (S)-1d. Although all the stereoisomers of
whiskey lactones 10 showed coconut odors [17k], all the isomers of 2 showed weak or
no coconut odor.
4. Experimental
4.1. General
All commercially available reagent chemicals were obtained from Aldrich, Kanto
Kagaku, Nacalai Tesque, Tokyo Kasei, and Wako Chemicals and used without further
purification. Ruthenium chloride hydrate (RuCl3·nH2O) was used as an anhydrous salt
(n = 0). Lipase Amano PS was purchased from Amano Enzyme and dried over P2O5 for
three days at room temperature. Lipase Novozym 435 was donated by Novozymes
Japan. HMPA was distillated (70.1-72.1 ºC/0.5 mmHg) from CaH2. Dry Et2O for the
Grignard reaction was distilled from Na/benzophenone under an argon atmosphere.
Water refers to deionized water. NMR spectra were recorded on a Bruker Advance II
13
400 spectrometer (1H 400 MHz and C 100 MHz). Chemical shifts (δ) are reported in
parts per million (ppm) and referenced to TMS (δ = 0.00 ppm). IR spectra were
recorded on a Jasco FT/IR-410 spectrometer. MS was performed on a Jeol MStation
JMS-700 instrument. Optical rotations were recorded on a Jasco P-1300 polarimeter.
Gas chromatography (GC) analyses were performed on a Shimadzu GC-2014 gas
chromatograph equipped with InertCap 1 capillary column (GL Sciences, 0.25 mm ϕ ×
19

30 m), GAMMA DEXTM 120 capillary column (SUPELCO, 0.25 mm ϕ × 30 m) and
CP-Cyclodextrin-β-2,3,6-M-19 capillary column (Agilent, 0.25 mm ϕ × 25 m). High
performance liquid chromatography (HPLC) analyses were performed on a Shimadzu
LC-20AD intelligent pump with a Shimadzu SPD-20A UV detector using a
CHIRALCEL OD-H column, CHIRALCEL OZ-H column, CHIRALPAK AS-H
column (Daicel, 4.6 mm ϕ × 250 mm).
4.2. Synthesis of racemic and optically active dihydro-4-methyl-2(3H)-furanone 1a
4.2.1. (±)-2-Methyl-3-phenyl-1-propanol (±)-4a
α-Methylhydrocinnamic acid (5.100 g, 31.06 mmol) in dry THF (19 mL) was added
dropwise to a suspension of LiAlH4 (1.532 g, 40.37 mmol) in dry THF (20 mL) at 0 ºC
under an argon atmosphere. The reaction mixture was stirred for 26 h at room
temperature. The reaction was quenched with water (30 mL) and aqueous HCl (6 mol
L-1, 30 mL) at 0 ºC. The mixture was extracted with Et2O (3 × 50 mL). The combined
organic layers were washed with water (30 mL), aqueous NaOH (6 mol L-1, 30 mL),
and brine (30 mL). The organic layers were dried over anhydrous Na2SO4 and filtered.
After removal of the solvent, the residue was purified by column chromatography on
1
silica gel (hexane/acetone = 3/2) to afford (±)-4a (4.387 g, 94%) as a colorless oil. H
NMR data were identical to those reported in the literature [39].
4.2.2. (±)-2-Methyl-3-phenylpropyl acetate (±)-5a
Acetyl chloride (1.816 g, 23.14 mmol) was added slowly to a solution of (±)-4a
(1.728 g, 11.50 mmol) and dry pyridine (2.757 g, 34.85 mmol) in dry CH2Cl2 (30 ml) at
20

0 ºC. The reaction mixture was stirred overnight at room temperature. The reaction was
quenched with water (30 mL) at 0 ºC. The mixture was extracted with Et2O (50 mL).
The organic layer was washed with brine (30 mL), dried over anhydrous Na2SO4 and
filtered. After removal of the solvent, the residue was purified by column
chromatography on silica gel (hexane/EtOAc = 5/1) to afford (±)-5a (2.079 g, 94%) as a
colorless oil. 1H NMR data were identical to those reported in the literature [40].
4.2.3. (±)-Dihydro-4-methyl-2(3H)-furanone (±)-1a
A solution of H5IO6 (48.570 g, 21.308 mmol) in water (60 mL) was added to a
solution of (±)-5a (2.079 g, 10.81 mmol) in CCl4 (40 mL) and CH3CN (40 mL) at 0 ºC.
RuCl3·nH2O (0.465 g, 2.24 mmol) was added slowly to the mixture at 0 ºC, and the
resultant mixture was vigorously stirred for 20 h at room temperature. The reaction was
quenched with Et2O (50 mL) at 0 ºC, and the mixture was stirred for 30 min. The
mixture was extracted with Et2O (3 × 50 mL). The combined organic layers were
washed with brine (30 mL), dried over anhydrous Na2SO4 and filtered. After removal of
the solvent, a deep yellow oil (2.125 g) was obtained. The oil was dissolved in aqueous
NaOH (3 mol L-1, 40 mL), and the resulting solution was stirred overnight at room
temperature. The solution was washed with Et2O (30 mL) and acidified with aqueous
HCl (6 mol L-1, 40 mL) at 0 ºC. The mixture was stirred overnight at room temperature.
The aqueous layer was saturated with NaCl; however, the saturation process was not
applied to the synthesis of lactones 1b-h, 2. The mixture was extracted with Et2O (5 ×
40 mL). The combined organic layers were washed with saturated aqueous NaHCO3 (30
mL), brine (30 mL), dried over anhydrous Na2SO4 and filtered. After removal of the
solvent, the residue was purified by column chromatography on silica gel
21

(hexane/acetone = 2/1) and subsequent distillation (119-138 ºC/13.5 mmHg) to afford
1
(±)-1a (0.380 g, 35%) as a colorless oil. H NMR data were identical to those reported
in the literature [41].
4.2.4. Amano PS-catalyzed transesterification of (±)-2-methyl-3-phenyl-1-propanol
(±)-4a with vinyl acetate (screening experiments)
In a typical experiment, 1 mL of solution of (±)-4a (15.0 mg, 0.0999 mmol) and
vinyl acetate (86.1 mg, 1.00 mmol) in an organic solvent was added to a vial in which
Amano PS (20.0 mg) was placed. The suspension was stirred at room temperature. The
reaction was monitored by GC (InertCap 1 capillary column, 170 °C) and terminated at
approximately 50% conversion by the filtration of the lipase. The extent of the
conversion was calculated from the ratio of the peak area of the produced ester 5a to the
total of the peak areas of 5a and the unreacted alcohol 4a. The peak area of 5a was
corrected on the basis of the number of the carbon atoms in the molecules of 4a and 5a
[42]. The filtrate was concentrated, and the residue was purified by column
chromatography on silica gel (hexane/EtOAc = 10/1 to 1/1) packed in a short column (9
mm ϕ × 60 mm) to afford 5a and 4a. The enantiomeric excess (ee) of 4a was
determined by HPLC using a CHIRALCEL OD-H column (hexane/2-propanol = 91/9).
The ee of 5a was determined by measuring the ee of the corresponding alcohol 4a
prepared by hydrolysis. The ester 5a obtained above was dissolved in EtOH (0.5 mL),
and ten drops of aqueous NaOH (6 mol L-1) were added to the solution. The mixture
was stirred overnight at room temperature. The ethanol was removed under reduced
pressure, and water was added to the residue. The mixture was extracted with Et2O (15
mL). The organic layer was washed with brine (10 mL), dried over anhydrous Na2SO4
22

and filtered. After removal of the solvent, the residue was purified by column
chromatography on silica gel (hexane/EtOAc = 10/1 to 1/1) packed in a short column (9
mm ϕ × 60 mm) to afford 4a. The ee of 4a was determined by HPLC using a
CHIRALCEL OD-H column (hexane/2-propanol = 91/9).
4.2.5. Preparative resolution of (±)-2-methyl-3-phenyl-1-propanol (±)-4a
Amano PS (1.450 g) was added to a solution of (±)-4a (4.377 g, 29.14 mmol) and
vinyl acetate (25.089 g, 29.143 mmol) in dry toluene (260 mL), and the resulting
suspension was stirred at room temperature. The reaction was monitored by GC
(InertCap 1 capillary column, 185 °C) and terminated at approximately 52% conversion
(approximately 4.2 h, see the first paragraph of Section 4.2.4 for determining the extent
of the conversions) by the suction filtration of the lipase with Celite®. After removal of
the solvent, the residue was purified by column chromatography on silica gel
(hexane/EtOAc = 5/1 to 1/1) to afford (S)-5a (2.971 g, 53%) as a colorless oil and
(R)-4a (1.985 g, 45%) as a colorless oil.
(R)-4a: 1H NMR data were identical to those of (±)-4a. ee: >99% (HPLC, CHIRALCEL
22
OD-H column, hexane/2-propanol = 91/9); [α]D = +12 (c 0.96, CHCl3), {Lit [39].
[α]D25 = +11.62 (c 1.0, CHCl3, R)}.
Aqueous NaOH (6 mol L-1, 5 mL) was added to a solution of (S)-5a (2.971 g, 15.45
mmol) obtained above in EtOH (25 mL), and the mixture was stirred overnight at room
temperature. The ethanol was removed under reduced pressure, and water was added to
the residue. The mixture was extracted with Et2O (3 × 50 mL). The combined organic
layers were washed with brine (30 mL), dried over Na2SO4 and filtered. After removal
of the solvent, the residue was purified by chromatography on silica gel (hexane/EtOAc
23

= 1/1) to afford (S)-4a (2.192 g, 94%) as a colorless oil.
1
(S)-4a: H NMR data were identical to those of (±)-4a. ee: 88% (HPLC, CHIRALCEL
OD-H column, hexane/2-propanol = 91/9).
4.2.6. (R)-2-Methyl-3-phenylpropyl acetate (R)-5a
This was prepared from (R)-4a (1.984 g, 13.21 mmol, >99% ee), AcCl (2.083 g,
26.54 mmol), and dry pyridine (3.181 g, 40.21 mmol) in dry CH2Cl2 (30 mL) in 95%
1
yield (2.418 g, colorless oil) according to the procedure shown in Section 4.2.2. H
22
NMR data were identical to those of (±)-5a. [α]D = -14 (c 0.87, benzene), {Lit [43].
[α]D = +14 (c 1.15, benzene, S, 98% ee)}.
4.2.7. (R)-Dihydro-4-methyl-2(3H)-furanone (R)-1a
This was prepared from (R)-5a (2.418 g, 12.58 mmol) as the starting material in
41% yield (0.522 g, colorless oil) according to the procedure shown in Section 4.2.3.
The reagents and reaction solvents for the preparation are as below: CCl4 (50 mL),
CH3CN (50 mL), H5IO6 (57.231 g, 251.08 mmol), water (75 mL), RuCl3·nH2O (0.522 g,
2.52 mmol), aqueous NaOH (3 mol L-1, 40 mL), aqueous HCl (6 mol L-1, 30 mL). Ethyl
1
acetate was used for the extraction of (R)-1a. H NMR data were identical to those of
21
(±)-1a. Bp: 129-165 ºC/16.5 mmHg; [α]D = +26 (c 0.73, MeOH), {Lit [44]. [α]D =
+21.6 (c 0.5, MeOH, R, 98% ee)}.
4.2.8. (S)-2-Methyl-3-phenylpropyl acetate (S)-5a
Amano PS (0.363 g) was added to the solution of (S)-4a (2.191 g, 14.59 mmol, 88%
ee) and vinyl acetate (12.613 g, 14.651 mmol) in dry toluene 130 mL, and the resulting
24

suspension was stirred at room temperature. The reaction was monitored by GC
(InertCap 1 capillary column, 185 °C) and terminated at approximately 86% conversion
(approximately 4.3 h, see the first paragraph of Section 4.2.4 for determining the extent
of the conversions) by the suction filtration of the lipase with Celite®. After removal of
the solvent, the residue was purified by column chromatography on silica gel
1
(hexane/EtOAc = 5/1) to afford (S)-5a (2.410 g, 86%) as colorless oil. H NMR data
were identical to those of (±)-5a. The ee of (S)-5a was determined by measuring the ee
of the corresponding alcohol (S)-4a prepared by hydrolysis; a drop of (S)-5a was
hydrolyzed to (S)-4a according to the procedure shown in the second paragraph of
Section 4.2.4, and the ee of (S)-4a was determined by HPLC (CHIRALCEL OD-H
21
column, hexane/2-propanol = 91/9) to be >99%. [α]D = +15 (c 0.79, benzene), {Lit
[43]. [α]D = +14 (c 1.15, benzene, S, 98% ee)}.
4.2.9. (S)-Dihydro-4-methyl-2(3H)-furanone (S)-1a
This was prepared from (S)-5a (2.399 g, 12.48 mmol) as the starting material in
41% yield (0.517 g, colorless oil) according to the procedure shown in Section 4.2.3.
The reagents and reaction solvents for the preparation are as below: CCl4 (50 mL),
CH3CN (50 mL), H5IO6 (56.072 g, 245.99 mmol), water (75 mL), RuCl3·nH2O (0.516 g,
2.49 mmol), aqueous NaOH (4 mol L-1, 30 mL), aqueous HCl (6 mol L-1, 30 mL). Ethyl
acetate was used for the extraction of (S)-1a. 1H NMR data were identical to (±)-1a. Bp:
22
126-146 ºC/16 mmHg; [α]D = - 26 (c 0.69, MeOH), {Lit [44]. [α]D = +21.6 (c 0.5,
MeOH, R, 98% ee)}.
4.3. Synthesis of two stereoisomers of whiskey lactone analogue {(4R, 5S)- and (4R,
25

5R)-4-butyldihydro-5-methyl-2(3H)-furanone} (4R, 5S)- and (4R, 5R)-2
4.3.1. (3R)-3-Benzyl-2-heptanol (3R)-7
A solution of (R)-4d (0.905 g, 4.71 mmol, >99% ee) prepared according to the
procedure for the preparative resolution of (±)-4d shown in Supporting information in
dry CH2Cl2 (3 mL) was added dropwise to a suspension of 95% Dess-Martin
periodinane (2.395 g, 5.4 mmol) in dry CH2Cl2 (9 mL) at room temperature under an
argon atmosphere, and the reaction mixture was stirred for 30 min. Et2O (30 mL),
saturated aqueous Na2S2O3 (10 mL), and saturated aqueous NaHCO3 (10 mL) were
successively added to the mixture. The resulting mixture was extracted with Et2O. The
organic layer was washed with brine, dried over anhydrous Na2SO4, and filtered. After
removal the solvent, the residue was purified by column chromatography on silica gel
(hexane/acetone = 10/1) to afford the corresponding aldehyde (0.735 g, 82%) as a
colorless oil. The product was immediately used without characterization.
Mg shavings (0.378 g, 15.5 mmol) were placed in a three-necked flask. The
shavings were covered with about 1 mL of a solution of distilled (44.5 °C) iodomethane
(2.051 g, 14.45 mmol) in dry Et2O (12 mL) under an argon atmosphere. When the
reaction started, the remainder of the solution was slowly added. The mixture was
gently refluxed with a mantle heater for 30 min to give methylmagnesium iodide. After
cooling to room temperature, a solution of the aldehyde obtained above (0.735 g, 3.86
mmol) in dry Et2O (10 mL) was slowly added to the solution of the methylmagnesium
iodide at room temperature. The reaction mixture was stirred overnight under an argon
atmosphere. After cooling the mixture to 0 ºC, saturated aqueous NH4Cl and water were
successively added. Unreacted Mg shavings were removed by suction filtration with
26

Celite®. The filtrate was extracted three times with Et2O. The combined organic layers
were washed with water, aqueous NaOH (1 mol L-1), and brine. The organic layers were
dried over anhydrous Na2SO4 and filtered. After removal of the solvent, the residue was
purified by column chromatography on silica gel (hexane/EtOAc = 3/1) to afford (3R)-7
{0.612 g, 63% from (R)-4d} as a diastereomeric mixture {syn:anti = 59:41 (GC,
1
GAMMA DEXTM 120 capillary column, 160 °C)} and a colorless oil. Refer to the H
NMR data of (2S, 3R)-7 shown in Section 4.3.2 and those of (2S, 3S)-7 shown in
Supporting information.
4.3.2. Separation of diastereomers of (3R)-3-benzyl-2-heptanol (3R)-7
Novozym 435 (2.672 g) was added to a solution of (3R)-7 (0.544 g, 2.64 mmol,
syn:anti = 59:41) in vinyl acetate (53 mL), and the resulting suspension was stirred at
room temperature. The reaction was monitored by GC (GAMMA DEXTM 120 capillary
column, 160 °C) and terminated at approximately 42% conversion (approximately 22 h,
see the first paragraph of Section 4.2.4 for determining the conversion) by the suction
filtration of the lipase with Celite®. After removal of the solvent, the residue was
purified by column chromatography on silica gel (hexane/acetone = 10/1 to 1/1) to
afford (1R, 2R)-8 (0.248 g, 38%) as a colorless oil and (2S, 3R)-7 as a yellow oil.
Alcohol (2S, 3R)-7 was purified again by column chromatography on silica gel
(hexane/acetone = 3/1) to become a colorless oil (0.315 g, 58%).
(2S, 3R)-7: IR (neat): 3358 cm-1; 1H NMR (400 MHz, CDCl3): δ 0.86 (3H, t, J = 7.2 Hz),
1.20 (3H, d, J = 6.4), 1.17-1.37 (5H, m), 1.39-1.48 (1H, m), 1.65-1.73 (1H, m), 2.59
(1H, dd, J = 7.6, J = 14), 2.67 (1H, dd, J = 6.8, J = 14), 3.82-3.87 (1H, m), 7.17-7.34
(5H, m); 13C NMR (100 MHz, CDCl3): δ 14.04, 19.78, 23.02, 28.84, 29.49, 36.13,
27

46.91, 68.96, 125.75, 128.27, 129.09, 141.35; MS (EI): m/z 206 (M+); HRMS: Calcd for
C14H22O 206.1671, Found 206.1672; syn:anti >99:<1 (GC,
=
CP-Cyclodextrin-β-2,3,6-M-19 capillary column, 120 °C); ee: >99% (GC,
22
CP-Cyclodextrin-β-2,3,6-M-19 capillary column, 120 °C); [α]D = +8.3 (c 0.73,
CHCl3).
Ester (1R, 2R)-8 (0.248 g, 0.999 mmol) obtained above was hydrolyzed to the
corresponding alcohol (2R, 3R)-7 with aqueous NaOH (6 mol L-1, 0.5 mL) in EtOH (5
mL) according to the procedure shown in the second paragraph of Section 4.2.5. The
crude product was purified by chromatography on silica gel (hexane/EtOAc = 3/1) to
afford (2R, 3R)-7 (0.173 g, 84%) as a colorless oil.
1
(2R, 3R)-7: H NMR data were almost identical to those of (2S, 3S)-7 shown in
Supporting information. syn:anti = 3:97 (GC, CP-Cyclodextrin-β-2,3,6-M-19 capillary
column, 120 °C); ee: >99% (GC, CP-Cyclodextrin-β-2,3,6-M-19 capillary column,
120 °C).
4.3.3. (1S, 2R)-2-Benzyl-1-methylhexyl acetate (1S, 2R)-8
This was prepared from (2S, 3R)-7 (0.306 g, 1.48 mmol) and AcCl (0.214 g, 2.73
mmol) with dry pyridine (0.503 g, 6.36 mmol) in dry CH2Cl2 (5 mL) according to the
procedure Section 4.2.2. The crude product was purified by chromatography on silica
gel (hexane/acetone = 10/1) to afford (1S, 2R)-8 (0.347 g, 94%) as a colorless oil. IR
(neat): 1245, 1737 cm-1; 1H NMR (400 MHz, CDCl3): δ 0.84 (3H, t, J = 7.0), 1.22 (3H,
d, J = 6.4), 1.25-1.39 (6H, m), 1.82-1.90 (1H, m), 2.00 (3H, s), 2.54 (1H, dd, J = 8.4, J
= 14), 2.69 (1H, dd, J = 6.4, J = 14), 4.94 (1H, qd, J = 4.8, J = 6.4), 7.15-7.30 (5H, m);
13C NMR (100 MHz, CDCl3): δ 13.95, 16.67, 21.34, 22.87, 28.96, 28.99, 36.48, 44.11,
28