Mammalian blood odorant and chirality: synthesis and sensory evaluation by humans and mice of the racemate and enantiomers of trans-4,5-epoxy-(E)-2-decenal

Article abstract of DOI:10.1016/j.tetasy.2015.06.011

Abstract The racemate and enantiomers of trans-4,5-epoxy-(E)-2-decenal 1, an odorant with a metallic blood smell, were synthesized. The detection threshold for humans was 0.019 ppb for (2E,4S,5S)-(-)-1 and 0.62 ppb for (2E,4R,5R)-(+)-1, respectively. The

Full text of DOI:10.1016/j.tetasy.2015.06.011

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Tetrahedron: Asymmetry 26 (2015) 861–867
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Mammalian blood odorant and chirality: synthesis and sensory
evaluation by humans and mice of the racemate and enantiomers
of trans-4,5-epoxy-(E)-2-decenal
b
c
Kenji Mori a, , Kazumi Osada , Masayasu Amaike
⇑
a Photosensitive Materials Research Center, Toyo Gosei Co., Ltd, 4-2-1 Wakahagi, Inzai-shi, Chiba 270-1609, Japan
b Department of Oral Physiology, School of Dentistry, Health Science University of Hokkaido, Ishikari-tobetsu, Hokkaido 061-0293, Japan
c R&D Center, T. Hasegawa Co., Ltd, 29-7 Kariyado, Nakahara-ku, Kawasaki-shi, Kanagawa 211-0022, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The racemate and enantiomers of trans-4,5-epoxy-(E)-2-decenal 1, an odorant with a metallic blood
smell, were synthesized. The detection threshold for humans was 0.019 ppb for (2E,4S,5S)-(À)-1 and
0.62 ppb for (2E,4R,5R)-(+)-1, respectively. The racemate ( )-1 made mice scared stiff at a dosage of
approximately 1 mg, while the enantiomers were not significantly bioactive.
Received 19 May 2015
Accepted 10 June 2015
Available online 1 July 2015
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction
2-deoxy-D-ribose as an intermediate in the course of their leuko-
triene synthesis, although they did not describe its physical data
including the specific rotation.8 In the patent literature, Daniher
et al. described the synthesis of both (+)- and (À)-1 by Jacobsen’s
asymmetric epoxidation, and assigned the absolute configuration
of (À)-1 as (4S,5S) by converting it into a compound reported by
Zamboni.9
We became interested in synthesizing the racemate as well as
the enantiomers of 1 with known absolute configuration and
enantiomeric purity so as to compare their odorant properties.
In 2014, Nilsson et al. reported that trans-4,5-epoxy-(E)-2-decenal
1 (Fig. 1) could be as efficient in eliciting behavioral responses among
dogs and tigers as the odor of mammalian blood.1 The fact that the
metallic blood smell of 1 attracts predators is of interest in its ecolog-
ical context to study the chemistry of how carnivores find their
prey.2 Nilsson et al., however, examined only racemic 1, which had
been known since 1991 as one of the primary odorants in wheat
bread crumb3 and popcorn4 by Schieberle et al. Aldehyde 1 was
found in many other food systems5 such as black tea,6 and its
biogenetic pathway was speculated.7
Herein we report
a scalable and efficient synthesis of ( )-,
(2E,4R,5R)-(+)- and (2E,4S,5S)-(À)-1, together with the preliminary
results of their sensory evaluation by humans and mice. It was
shown that to the human nose (2E,4S,5S)-(À)-1 smelled about
twenty to thirty times more intensive than its enantiomer. Mice
were scared stiff when approximately 1 mg of ( )-1 was released.
There are two reports, which describe the preparation of
optically active 1. Zamboni et al. prepared (2E,4S,5S)-1 from
O
CHO
n-C5H11
2. Results and discussion
trans-4,5-epoxy-( E)-2-decenal
1
2.1. Synthesis of racemic trans-4,5-epoxy-(E)-2-decenal 1
O
O
CHO
CHO
n-C5H11
(2E,4R,5R)-(+)-1
n-C5H11
(2E,4S,5S)-(–)-1
Prior to the synthesis of the enantiomers of 1, that of the race-
mate was examined as shown in Scheme 1 so as to find a scalable
route, which was needed, because a multi-gram amount of 1 was
necessary to carry out mammalian sensory evaluation of 1 against
mice and deer. Schieberle’s route was first examined with regard to
its efficiency. This protocol involved epoxidation of commercially
available (2E,4E)-2,4-decadienal 2.3 In our hands, however, epoxi-
dation of 2 with m-chloroperbenzoic acid in CH2Cl2 gave impure
Figure 1. Structure of trans-4,5-epoxy-(E)-2-decenal 1.
⇑
Corresponding author. Tel./fax: +81 3 3816 6889.
E-mail address: kjk-mori@arion.ocn.ne.jp (K. Mori).
http://dx.doi.org/10.1016/j.tetasy.2015.06.011
0957-4166/Ó 2015 Elsevier Ltd. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
862
K. Mori et al. / Tetrahedron: Asymmetry 26 (2015) 861–867
O
2.2. Synthesis of the enantiomers of trans-4,5-epoxy-(E)-2-
decenal 1
a
CHO
CHO
n-C5H11
n-C5H11
n-C5H11
n-C5H11
2
(2E,4E)-
(2E,4R*,5R*)-( )-1
2.2.1. Attempted asymmetric acetylation of ( )-trans-2,3-epoxy-
1-octanol 4
Since epoxy alcohol ( )-4 was readily available, we first
attempted its asymmetric acetylation with vinyl acetate in the
presence of lipase PS, which has been employed widely to prepare
enantiopure building blocks.17 Lipase PS was a remarkably suitable
enzyme to give (2S,3R)-4-t-butyldimethylsilyloxy-2,3-epoxybutyl
acetate (98% ee) from the corresponding racemic alcohol by
enzymatic asymmetric acetylation with vinyl acetate.18 In the
present case with ( )-4 however, lipase PS coated with an ionic
O
a
b or c or d
OH
n-C5H11
OH
3
(E)-
(2R*,3R*)-( )-4
O
O
e
CHO
CHO
n-C5H11
(2E,4R*,5R*)-( )-1
liquid [IL 1:a
-cetylpolyoxyethylene(19)ether sulfate]19 could not
(2R*,3S*)-( )-5
discriminate the enantiomers of 4, and after only 4 h at room
temperature, ( )-4 furnished the corresponding racemic acetate
( )-6 in 95% yield (Scheme 2). The enzymatic route was therefore
abandoned.
(OAc)3
I
OH
I(OAc)2
N
O
N
O
O
AZADOL®
DAIB
O
O
IL 1-coated lipase PS
Dess-Martin periodinane
TEMPO
n-C5H11
(2R*,3R*)-( )-4
OH
n-C5H11
(2R*,3R*)-( )-6
OAc
CH2=CHOAc, (i-Pr)2O
Scheme 1. Synthesis of ( )-1. Reagents and conditions: (a) m-chloroperbenzoic
acid, CH2Cl2 [3% for ( )-1; 94% for ( )-4]: (b) Dess–Martin periodinane, CH2Cl2
(62%); (c) TEMPO, DAIB, CH2Cl2 (80%); (d) AZADOL,Ò DAIB, CH2Cl2 (78%);
(e) Ph3P@CHCHO, C6H6 (87%).
(92%)
OH
1) HBr, AcOH
2) MeOH
AD-mix-α
CO2Me
CO2Me
n-C9H19
n-C9H19
( )-1 in a very low yield of 3% after chromatographic purification
on SiO2.
OH
A
B
(mp 43.5-44.5 ºC)
The next attempt was to convert commercially available
(E)-2-octen-1-ol 3 into ( )-1, hoping to improve the efficiency
OH
of the previous results.5,7,10,11 Epoxidation of
3
with
O
1) K2CO3, MeOH
2) DIBAL-H
CO2Me
CHO
m-chloroperbenzoic acid in CH2Cl2 gave epoxy alcohol ( )-4 in
a satisfactory yield of 94% after distillation. It should be noted
that this step can be modified to give either (+)- or (À)-4 by
employing an asymmetric epoxidation developed by Katsuki
and Sharpless.12,13
n-C9H19
n-C9H19
Br
(mp 44.0-44.5 ºC)
D
(ref. 21)
C
OH
Although Swern oxidation of ( )-4 failed to give aldehyde ( )-5
under standard conditions [DMSO/(COCl)2 and Et3N], Dess–Martin
periodinane14 in CH2Cl2 cleanly converted ( )-4 into ( )-5
AD-mix-α
AD-mix-β
CO2Me
n-C5H11
OH
8
(2R,3S)-(–)- (oil)
in 62% yield. Since the oxidation of the enantiomers of
4
CO2Me
n-C5H11
with 2,2,6,6-tetramethyl-1-piperidinyloxyl (TEMPO) and
(diacetoxyiodo)benzene (DAIB) in CH2Cl2 was recently published
by Rizzo et al.,15 the TEMPO/DAIB oxidation of ( )-4 was carried
out to give ( )-5 as a slightly pink-colored oil in 80% yield after
chromatographic purification and distillation. Iwabuchi’s highly
active 2-azaadamantan-N-oxyl (AZADO) catalyst16 was then used
for the oxidation. Only 1.6 mol % of AZADOLÒ (2-hydroxyo-2-aza-
adamantane) catalyzed the oxidation of ( )-4 into ( )-5 within
1 h to give ( )-5 as a colorless oil in 78% yield, while oxidation of
( )-4 in the presence of 17 mol% of TEMPO required 3.5 h for
completion. AZADOLÒ/DAIB oxidation was therefore adopted for
the multi-gram scale preparation of ( )-5.
Two-carbon elongation of ( )-5 to give ( )-1 was carried out
by a Wittig reaction between ( )-5 and commercially available
(triphyenylphosphoranylidene)acetaldehyde in C6H6. After
chromatographic purification and distillation, the desired
( )-trans-4,5-epoxy-(E)-2-decenal 1 was obtained in 87% yield
as a colorless and odoriferous oil. Its 1H NMR, 13C NMR and MS
spectra were in good accord with those previously reported.3,5
The overall yield of ( )-1 was 65% based on 3 (three steps). The
procedure reported in Section 4 is highly reproducible, and over
10 g of ( )-1 could be secured for sensory evaluation against mice
and deer. The present method is the most efficient one for
the preparation of ( )-1.
OH
7
(E)-
CO2Me
n-C5H11
OH
(2S,3R)-(+)-8 (oil)
Scheme 2. Attempted routes for the synthesis of optically active 1.
2.2.2. Attempted AD (asymmetric dihydroxylation)-mediated
route
Asymmetric dihydroxylation (AD) as developed by Sharpless
et al. is one of the most versatile enantioselective methods in
organic synthesis.20 In 2001 we applied AD in our pheromone syn-
thesis, converting
a,b-unsaturated ester A into enantiopure alde-
hyde D.21 This was possible because both dihydroxy ester B and
bromohydrin ester C were nicely crystalline. Their purification
was possible by recrystallization.
The AD of methyl (E)-2-octenoate 7 with either AD-mix-aÒ or
AD-mix-bÒ gave (2R,3S)- or (2S,3R)-8, respectively, both as oils.
Since recrystallization of 8 was impossible, we gave up this
approach, and came back to the more traditional endeavor to use
the Sharpless asymmetric epoxidation for the preparation of the
enantiomers of 4.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
K. Mori et al. / Tetrahedron: Asymmetry 26 (2015) 861–867
863
2.2.3. Asymmetric epoxidation-mediated synthesis of the
enantiomers of 1
Subsequent Dess–Martin periodinane oxidation of (2R,3R)-(+)-4
gave aldehyde (2S,3R)-(À)-5, [
a
]
23 = À93.4 (c 1.21, pentane) {Ref.
D
In 1980, Katsuki and Sharpless invented asymmetric epoxida-
tion.12 Since then asymmetric epoxidation has been widely used
in organic synthesis.13 Indeed, the asymmetric epoxidation of
(E)-2-octen-1-ol 3 (Scheme 3) was repeatedly reported to give
the enantiomers of 4.13,22,23 With regard to the experimental
procedure for the conversion of 3 into (+)-4 or (À)-4, we adopted
Taber’s protocol for the asymmetric epoxidation of geraniol,
avoiding chromatographic purification.24
15. [a
]
D
25.7 = À98.6 (c 0.35, CHCl3)}. Similarly, (2S,3S)-(À)-4 was
oxidized with Dess–Martin periodinane to give (2R,3S)-(+)-5,
[a] [a]
21 = +93.2 (c 1.40, pentane){Ref. 15 24.7 = +98.3 (c 0.48,
D
D
CHCl3)}.
Finally, two-carbon elongation of the enantiomers of 5 was car-
ried out with Ph3P@CHCHO in C6H6. Accordingly, (2S,3R)-(À)-4 was
subjected to the Wittig reaction to give (4R,5R)-(+)-1, [a]
24 = +26.2
D
(c 2.24, hexane), while (2R,3S)-(+)-4 furnished (4S,5S)-(À)-1,
[a
]
22 = À26.0 (c 2.33, hexane). Their IR, 1H and13C NMR, and MS
D
spectra were identical with those of ( )-1. Their enantiomeric puri-
ties were determined by enantioselective GC on a ChiramixÒ
column,26 and found to be 97.2% ee for (4R,5R)-(+)-1 and 96.2%
ee for (4S,5S)-(À)-1. The overall yield of (+)-1 was 26% and that
of (À)-1 was 24% based on 3 (five steps including the enrichment
steps). Both enantiomers of trans-4,5-epoxy-(E)-2-decenal 1 were
secured in quantities sufficient for their sensory evaluation.
O
a
d
n-C5H11
OH
n-C5H11
OR
3
(E)-
4
(2R,3R)-(+)- R = H
O
c
b
NO2
(2R,3R)-(+)-9 R =
NO2
2.3. Sensory evaluation of the racemate and enantiomers of
trans-4,5-epoxy-(E)-2-decenal 1
O
CHO
O
e
n-C5H11
CHO
n-C5H11
(2E,4R,5R)-(+)-1
2.3.1. By humans
5
(2S,3R)-(–)-
[α]D24 = +26.2 (c 2.24, hexane)
The odors of the synthetic (2E,4R,5R)-(+)-1, ( )-1 and (2E,4S,5S)-
(À)-1 were examined by dissolving trace amounts of each of the
samples into water, and their recognition threshold as well as
detection threshold were determined by a group (22–26 persons)
of flavor scientists at T. Hasegawa Co. The results are shown in
Table 1.
Similarly
O
f
n-C5H11
OH
n-C5H11
OH
Table 1
(E)-3
(2S,3S)-(–)-4
Recognition and detection thresholds of (+)-, ( )- and (À)-1 as determined by a panel
of flavor scientists
O
Sample
Recognition
threshold (ppb)
Detection
threshold (ppb)
Number of
panelists
CHO
O
n-C5H11
CHO
n-C5H11
(2E,4R,5R)-(+)-1
( )-1
(2E,4S,5S)-(–)-1
1.1
0.20
0.047
0.62
0.073
0.019
22
24
26
(2E,4S,5S)-(–)-1
[α]D22 = –26.0 (c 2.33, hexane)
(2R,3S)-(+)-5
Scheme 3. Synthesis of the enantiomers of 1. Reagents and conditions: (a) t-BuOOH,
Ti(Oi-Pr)4, (À)-DET, MS 4 Å, CH2Cl2, -20 °C, 6 h (88%); (b) 3,5-dinitrobenzoyl chloride,
DMAP, C5H5N, Et2O; recryst’n (58%); (c) K2CO3, MeOH, THF, 0–5 °C, 1 h (95%); (d)
Dess–Martin periodinane, CH2Cl2 (66%); (e) Ph3P@CHCHO, C6H6 (86%); (f) t-BuOOH,
Ti(Oi-Pr)4, (+)-DET, MS 4 Å, CH2Cl2, À20 °C, 6 h (81%).
It should be noted that (À)-1 was twenty to thirty times more
intensive odorant than (+)-1. People at Givaudan Co. also noticed
that the odor of (À)-1 was at least ten times stronger than (+)-1.9
Their enantiomers of 1 were prepared by Jacobsen’s asymmetric
epoxidation.27
Impressions of the odors of (+)- and (À)-1 as 0.01% solution in
triacetin (triacetylglycerol) were examined by flavorists at T.
Hasegawa Co. Their responses were as follows:—(+)-1: weak,
metallic, green and oily, and (À)-1: strong, metallic, oily, fatty
and peely. Both enantiomers of 1 showed similar odor quality,
although they were different in odor intensity.
The asymmetric epoxidation of 3 with t-BuOOH, Ti(Oi-Pr)4 and
diethyl (2S,3S)-(À)-tartrate [(À)-DET] in the presence of powdered
MS 4 Å in CH2Cl2 at À20 °C for 6 h afforded (2R,3R)-(+)-4 in 88%
yield after work-up and distillation. Its specific rotation,
[a]
18 = +34.5 (c 5.56, CHCl3), was consistent with the reported
D
value of [a]
27 = +38.6 (c 1.0, CHCl3).22 Similarly, by employing
D
diethyl (2R,3R)-(+)-tartrate [(+)-DET] instead of (À)-DET, (2S,3S)-
22 = À36.1 (c 4.39, CHCl3)
2.3.2. By mice
(À)-4 was obtained in 81% yield, [
a
]
D
22 = À34.6 (c 5.7, CHCl3)}.
The effects of (+)-, (À)- and ( )-trans-4,5-epoxy-(E)-2-decenal 1
on mice were examined by behavioral observation at Health
Science University of Hokkaido. Fear induction among mice caused
by blood odor seems to be the result of an instinctive memory that
conveys information about predation and injury. To assess the
behavioral responses of mice to the enantiomers and racemate of
1, a test for freezing (i.e., immobilization) was performed with
the synthetic samples released for 5 min.28
{Ref. 23. [
a
]
D
In order to increase the enantiomeric purities of the
enantiomers of 4, they were converted into the corresponding
3,5-dinitrobenzoates (3,5-DNB esters), which were crystalline.cf.25
Two recrystallizations afforded (2R,3R)-(+)-9 {mp 50.5–51.0 °C;
[
[
a
]
22 = +34.5 (c 3.38, CHCl3)} and (2S,3S)-(À)-9{mp 50 °C;
D
a]
21 = À34.9 (c 3.80, CHCl3)}. These enantiomers of 9 were shown
D
to be of ca. 95% ee by enantioselective HPLC analysis on ChiralpakÒ
AD-H (see Section 4). Methanolysis of the crystalline enantiomers
Three-month-old female C57BL/6j mice29 (n = 24) were used in
the present experiment. One
tended to increase the freezing duration, but not so significant.
On the other hand, 1 L (=ca. 1 mg) of ( )-1 evoked significantly
23 = +33.4
lL (=ca. 1 mg) of either (+)- or (À)-1
of 9 with K2CO3/MeOH gave back (2R,3R)-(+)-4 {[a]
D
(c 4.59, CHCl3); 96.2% ee as determined by enantioselective GC
analysis on heptakis (2,3-di-O-acetyl-6-O-t-butyldimethylsilyl)-
l
21 = À34.5
longer freezing duration: 53.2 8.5 s, P <0.01 vs. control
b-cyclodextrin (see Section 4)} and (2S,3S)-(À)-4{[
a]
D
16.2 3.4 s. ANOVA followed by Dunnett’s post-hoc test.30
(c 4.50, CHCl3); 95.8% ee}, respectively.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
864
K. Mori et al. / Tetrahedron: Asymmetry 26 (2015) 861–867
These results clearly indicate that blood must produce a mix-
amount of Na2S2O3 and brine, dried (MgSO4), and concentrated in
vacuo. The residue (10.2 g) was distilled to give 9.02 g (94%) of
(2R*,3R*)-( )-4 as a colorless oil, bp 75–77 °C/2 Torr; n1D8 = 1.4412;
mmax (film): 3420 (br m), 2957 (s), 2930 (s), 2860 (s), 1467 (m),
1073 (m), 1031 (m), 888 (m); dH (CDCl3): 0.90 (3H, t, J 6.8),
1.25–1.40 (4H, m), 1.40–1.50 (2H,m), 1.55–1.59 (2H, m), 2.17
(1H, br), 2.90–2.98 (2H, m), 3.61 (1H, d, J 12), 3.91 (1H, d, J 12);
GC–MS [column: HP-5MS, 5% phenylmethylsiloxane, 0.25 mm
i.d. Â 30 m; carrier gas, He; press 60.7 kPa; temp 70–230 °C
(+10 °C/min)]: tR 7.88 min (97.2%); MS (70 eV, EI): m/z: 101 (9),
83 (92), 55 (100), 41 (51). HRMS calcd for C8H17O2
[(C8H16O2+H)+]: 145.1229, found: 145.1229.
ture of the enantiomers of 1, if mice are to communicate the fear
situation with each other. Synergistic response to enantiomers
was first discovered in 1976 in the case of an insect aggregation
pheromone sulcatol(6-methyl-5-hepten-2-ol).31 Further studies
are currently in progress to clarify the enantiomeric composition
of 1 derived from mammalian blood, and the results will be pub-
lished in due course by Osada et al. It should be added that people
at T. Hasegawa Co. already found that 1, generated by heating a
mixture of beef oil and water, was an almost racemic mixture of
(+)-1 (55.9%) and (À)-1 (44.1%) as determined by GC–MS analysis
on a ChiramixÒ column.32
4.4. (2R*,3S*)-( )-2,3-Epoxyoctanal 5
3. Conclusion
4.4.1. By Dess–Martin periodinane oxidation of ( )-4
An efficient and scalable synthesis of ( )-trans-4,5-epoxy-(E)-2-
decenal 1 was developed. The enantiomers of 1 were synthesized
by employing Sharpless asymmetric epoxidation as the source of
chirality. (2E,4S,5S)-(À)-1 was a twenty to thirty times more
intense odorant for humans than (2E,4R,5R)-(+)-1, although their
odor quality as metallic and oily was similar to each other. The
odor of ( )-1 made mice scared stiff.
Dess–Martin periodinane (6.00 g, 14 mmol) was added portion-
wise to
a stirred and ice-cooled solution of ( )-4 (1.50 g,
10.4 mmol) in CH2Cl2 (75 mL) containing five drops of water. The
mixture was stirred for 20 min at 0–5 °C and then for 1 h at room
temperature. Colorless precipitates of o-iodobenzoic acid were
generated. It was then ice-cooled, and the reaction was quenched
by adding NaHCO3 solution. After stirring for 10 min, the mixture
was extracted with Et2O. The extract was washed with brine, dried
(MgSO4), and concentrated in vacuo. The residue was diluted with
hexane, and transferred onto a column of SiO2 (15 g) in hexane.
Elution with hexane/EtOAc (20:1) gave 918 mg (62%) of
(2R⁄,3S⁄)-( )-5, nD20 = 1.4358; mmax (film): 2958 (s), 2932 (s), 2860
(s), 2734 (w), 1731 (s), 1467 (m), 1437 (m), 1380 (w), 1241 (w),
1151 (w), 1060 (w), 981 (w), 852 (m), 727 (w); dH(CDCl3): 0.90
(3H, t, J 7.2), 1.30-1.40 (4H, m), 1.42-1.54 (2H, m), 1.60-1.72 (2H,
m), 3.14 (1H, dd, J 2, 6.4), 3.22 (1H, dt, J 2, 6), 9.01 (1H, d, J 6); dC
(CDCl3): 13.87, 22.42, 25.40, 31.12, 31.32, 56.75, 59.11, 198.51;
GC–MS (same conditions as those for 4): tR 6.58 min (93.5%); MS
(70 eV, EI): m/z: 141 (<1) [(MÀH)+], 95 (7), 71 (100), 57 (12), 55
(14), 41 (25). HRMS calcd for C8H14O2: 142.0994, found: 142.1002.
4. Experimental33
4.1. General
Melting points are uncorrected values. Refractive indices were
measured on an Atago DMT-1 refractometer. Optical rotations
were measured on a Jasco P-1020 polarimeter. IR spectra were
measured on a Jasco FT/IR-410 spectrometer. 1H NMR spectra
(400 MHz, TMS at d = 0.00 as the internal standard) and 13C NMR
spectra (100 MHz, CDCl3 at d = 77.0 as the internal standard) were
recorded on a Jeol JNM-AL 400 or JNM-ECZ 400S/L1 spectrometer.
GC-MS were measured on Agilent Technologies 5975 inert XL.
HRMS were recorded on Jeol JMS-SX 102A, Waters Synapt G2
HDMS, or Varian 901-MS Ion Spec QFT-7. Column chromatography
was carried out on Merck Kieselgel 60 Art 1.07734.
4.4.2. By TEMPO/DAIB oxidation of ( )-4
At first, DAIB (20.6 g, 64 mmol) was added to a stirred solution
of ( )-4 (8.40 g, 58 mmol) and TEMPO (1.56 g, 10 mmol) in dry
CH2Cl2 (125 mL). The red solution was stirred for 3.5 h at room
temperature. The reaction was slightly exothermic. It was then
diluted with CH2Cl2 (100 mL) and stirred with 5% Na2S2O3 solution
(100 mL). The organic layer was separated and the aqueous layer
was extracted with CH2Cl2. The combined CH2Cl2 solution was
washed successively with NaHCO3 solution and brine, dried
(MgSO4), and concentrated in vacuo. The residue was chro-
matographed over SiO2 (100 g). Elution with hexane gave 12.9 g
(99% recovery) of C6H5I. Further elution with hexane/EtOAc
(20:1) gave 8.94 g of crude ( )-5. This was purified by distillation
to give (2R⁄,3S⁄)-( )-5 (6.64 g, 80%) as a slightly pink-colored oil,
bp 57–58 °C/3 Torr; n1D8 = 1.4375; GC–MS (same conditions as
those for 4); tR 6.42 min (88.5%). Its IR, 1H NMR and MS spectra
were consistent with those of ( )-5 prepared by Dess–Martin
periodinane oxidation. This was employed in the next step without
further purification.
4.2. (2E,4R*,5R*)-( )-4,5-Epoxy-2-decenal 1 by epoxidation of
(2E,4E)-2,4-decadienal 2
m-Chloroperbenzoic acid (70% purity, 4.16 g, 170 mmol) was
added portionwise to an ice-cooled and stirred solution of 2
(2.28 g, 150 mmol) in CH2Cl2 (35 mL) at 0–5 °C. The mixture was
stirred for 3 h at 0–5 °C, after which it was diluted with hexane,
and filtered to remove MCBA. The filter cake was washed with hex-
ane. The combined organic solution was washed with Na2CO3 solu-
tion containing a small amount of Na2S2O3 and brine, dried
(MgSO4), and concentrated in vacuo. The residual oil (2.53 g) was
chromatographed over SiO2 (30 g). Elution with hexane/EtOAc
(50:1) gave 0.25 g (11%) of recovered 2. Further elution with
hexane/EtOAc (30:1) gave 80 mg (3%) of (2E,4R*,5R*)-( )-1. Its IR,
1H NMR and MS spectra were identical with those described
for ( )-1 in Section 4.5. Further elution with hexane/EtOAc
(10:1–4:1) gave 1.09 g of unidentified polar materials.
4.4.3. By AZADOLÒ/DAIB oxidation of ( )-4
4.3. (2R*,3R*)-( )-2,3-Epoxy-1-octanol 4
At first, DAIB (22.5 g, 70 mmol) was added to a stirred solution
of ( )-4 (9.2 g, 63.9 mmol) and AZADOLÒ (153 mg, 1 mmol) in dry
CH2Cl2 (125 mL). The reaction was slightly exothermic, and the
temperature of the pale orange-colored solution reached 22 °C.
After stirring for 1 h, the mixture was transferred into a separatory
funnel, washed successively with NaHCO3 solution containing a
small amount of Na2S2O3 and brine, dried (MgSO4), and concen-
trated in vacuo. The residual colorless oil (26.9 g) was
m-Chloroperbenzoic acid (70% purity, 18.3 g, 75 mmol) was
added portionwise to a stirred and ice-cooled solution of 3 (8.5 g,
66.4 mmol) in CH2Cl2 (100 mL) at 0–10 °C. The mixture was stirred
for 2 h at 0–5 °C, then diluted with hexane, and filtered. The filter
cake was washed with hexane. The combined organic solution
was washed successively with Na2CO3 solution containing a small