Stereoselective synthesis of (6Z,8E)-undeca-6,8,10-trien-3-one (yuzunone) for its characterization in yuzu and various citrus essential oils

Article abstract of DOI:10.1016/j.foodchem.2020.128130

(6Z,8E)-Undeca-6,8,10-trien-3-one (yuzunone) is reported to be one of the main olfactory contributors of the specific fruity-green-balsamic odor of yuzu peel oil. Using an original stereoselective synthesis, we prepared a pure sample of yuzunone, which was used as a reference compound to check its presence by GC–MS and GC–O in 5 commercial samples of yuzu and citrus essential oils. Surprisingly, we could not detect yuzunone by GC–MS in any of our samples. However, it could be detected by a small part of the panelists involved in GC–O/AEDA experiments in a yuzu commercial oil, but its olfactory contribution proved to be very limited.

Full text of DOI:10.1016/j.foodchem.2020.128130

FoodChemistry338(2021)128130
Contents lists available at ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Stereoselective synthesis of (6Z,8E)-undeca-6,8,10-trien-3-one (yuzunone)
for its characterization in yuzu and various citrus essential oils
Ayaka Uehara, Nicolas Baldovini⁎
Institut de Chimie de Nice, CNRS UMR 7272, Université Côte d’Azur, Parc Valrose, F-06108 Nice, France
A R T I C L E I N F O
A B S T R A C T
Keywords:
(6Z,8E)-Undeca-6,8,10-trien-3-one (yuzunone) is reported to be one of the main olfactory contributors of the
specific fruity-green-balsamic odor of yuzu peel oil. Using an original stereoselective synthesis, we prepared a
pure sample of yuzunone, which was used as a reference compound to check its presence by GC–MS and GC–O in
5 commercial samples of yuzu and citrus essential oils. Surprisingly, we could not detect yuzunone by GC–MS in
any of our samples. However, it could be detected by a small part of the panelists involved in GC–O/AEDA
experiments in a yuzu commercial oil, but its olfactory contribution proved to be very limited.
Yuzunone
Yuzu (Citrus junos)
Odor-active constituents
GC–O
1. Introduction
tachibana (Miyazato, 2018), Citrus unshiu, and Citrus kinokuni oils
(Miyazawa, Fujita, & Kubota, 2010). The presence of 7 in yuzu peel oil
Yuzu (Citrus junos) is a sour citrus fruit cultivated in Japan and
South Korea. In Japan, its peel and juice are widely used for cooking as
a natural flavoring agent, and as a bath additive to perfume water. In
Korea, yuzu fruit production is mostly used for the fabrication of “Yuzu
Tea”, a marmalade-like syrup drunk with hot water. As the pleasant
aroma of yuzu is different from those of common citrus fruits such as
lemon, orange, etc., some analytical studies have described the com-
position of yuzu peel oil to better characterize the specificities of this
fruit (Matsuura, Hara, & Abe, 1980; Tajima, Fukuoka, Hasegawa, & Toi,
1992; Tajima, Tanaka, Yamaguchi, & Fujita, 1990; Miyazawa et al.,
was confirmed by the preparation of a sample of reference compounds
for comparison of its chromatographic and spectral properties with
those of the natural constituent of the oil (Miyazawa et al., 2009b).
However, in this synthesis, the Wittig reaction used for the formation of
the C6–C7 double bond of 7 was not stereoselective. Therefore, the
reference sample obtained was indeed a 1:1 mixture of 7 and (6E,8E)-
undeca-6,8,10-trien-3-one 10, the isomers of which could be char-
acterized by GC only on the basis of GC–IR experiments. This mixture
was also used for the formulation of yuzu aroma reconstitutions, that
demonstrated that the addition of 10 ppm of 7/10 in a yuzu aroma
model mixture significantly improved the characteristic yuzu like bal-
samic/sweet/floral notes of the recomposition. Although the authors
reported that 10 was less odorant than 7, it had nevertheless a sig-
nificant olfactory contribution to the odor of the mixture used in these
reformulation experiments. In the course of our analytical investiga-
tions on yuzu oils, our preliminary GC–O experiments seemed to in-
dicate that the olfactory contribution of 7 in our samples was rather low
for most of our panelists. Then, we hypothesized that this compound
might not be a general key odorant for the aroma of yuzu. To bring a
definitive answer, a sample of 7 devoid of 10 was required for use as a
reference compound in GC–MS and GC–O on various columns, and
potentially for static sensorial tests and reformulation assays. Therefore,
we designed a stereoselective synthesis of 7 that could be used in our
analyses of the key odorants of various citrus essential oils.
2009a; Tomita, 2010; Nakanishi
& Watanabe, 2010; Miyazato,
Hashimoto, & Hayashi, 2012, 2013). Although the major constituents of
yuzu oil are classical monoterpenes present in most citrus fruit oils like
limonene and α-pinene (Ohloff, Pickenhagen, & Kraft, 2012), some
characteristic compounds have been reported as important odor-active
constituents of yuzu oil, like lactone 1 (Matsuura et al., 1980), sulfanyl
ketone 2 (Tajima et al., 1992), aldehydes 3 and 4 (Tajima et al., 1990),
5 (Miyazawa et al., 2009a), 6 (Miyazato et al., 2012), and mis-
cellaneous compounds 7, 8 and 9 (Miyazawa et al., 2009a; Miyazato
et al., 2013) (Fig. 1).
Miyazawa et al. (2009a) demonstrated that (6Z,8E)-undeca-6,8,10-
trien-3-one (yuzunone) 7 is one of the key odorants which contributes
to the specific fruity-green-balsamic odor of yuzu. The same authors
showed that it is also an important contributor to the odor of galbanum
(Ferula galbaniflua) (Miyazawa et al., 2009b) and of Japanese Citrus
⁎ Corresponding author.
E-mail addresses: ayaka.uehara@unice.fr (A. Uehara), baldovin@unice.fr (N. Baldovini).
https://doi.org/10.1016/j.foodchem.2020.128130
Received 7 May 2020; Received in revised form 14 September 2020; Accepted 15 September 2020
Availableonline19September2020
0308-8146/©2020ElsevierLtd.Allrightsreserved.

A. Uehara and N. Baldovini
FoodChemistry338(2021)128130
Fig. 1. Specific odor-active constituents of yuzu peel oil.
2. Material and methods
2.4. GC–olfactometry (GC–O)
2.1. Raw materials and chemicals
GC–O analyses were performed on a Shimadzu GC-2010 gas chro-
matograph equipped with an HP-5 capillary column (50 m × 0.32 mm
i.d.; 0.53 μm film thickness), the FID detector and the ATAS olfactory
port OP275 mounted with a glass nose cone. Injection volume was
1.0 μL in splitless mode. Injector temperature was 250 °C; the oven
temperature program ran from 100 °C to 250 °C at 5 °C/min, then
isothermal for 5 min. Carrier gas (nitrogen) flow was 1.50 mL/min;
60% of the flow was directed to the FID and 40% was directed into the
heated sniffing port. The FID temperature was 250 °C. The main odor-
active components were characterized by aroma extract dilution ana-
lysis (AEDA) (Ullrich & Grosch, 1987): At first, a 10% (w/w) solution of
the essential oil (EO) was analyzed (3–5 times) by each panelist, who
was asked to freely describe his/her olfactory perception, and to specify
if the odor was reminiscent of the typical smell of yuzu. These experi-
ments were then followed by the injection of serial successive (1/4)
dilutions until no more odor was detected. Nine evaluators participated
in this study. All of the panelists who were not familiar with the typical
smell of yuzu EO were asked to memorize the odor of the sample stu-
died in this work prior to the GC–O experiments.
2.1.1. Commercial oils
Yuzu cold-pressed oils were purchased from Tree of Life (Tokyo,
Japan) and Kitomura (Tokushima, Japan). Yuzu hydrodistilled oil was
from Ecology Shimanto (Kochi, Japan). Yellow mandarin and Brazil
orange oils were kindly provided by Albert-Vieille (Vallauris, France).
2.1.2. Chemicals
All reagents and solvents were purchased from Sigma-Aldrich (L'lsle
D'Abeau, Chesnes, France) except NaHCO3, MgSO4, Celite®, tBuOK
(VWR, Fontenay-sous-Bois, France) and silica gel (Macherey-Nagel,
Hoerdt, France). 2-(prop-2-yn-1-yloxy)tetrahydro-2H-pyran (Cerutti-
Delasalle et al., 2016), 2-(2-iodoethyl)-1,3-dioxolane (Stowell, King, &
Hauck, 1983) and 2-iodoxybenzoic acid (IBX) (Frigerio, Santagostino, &
Sputore, 1999) were synthesized in our laboratory. Diethyl ether and
THF were dried by distillation over sodium/benzophenone before use.
Dichloromethane and DMSO were dried by distillation over CaH2.
2.2. Sample preparation
2.5. NMR
Each essential oil sample (1 g) was chromatographed on a short pad
silica gel (5 g) with 20 mL of petroleum ether (20 mL) to eliminate the
hydrocarbons, and then with 20 mL of petroleum ether/diethyl ether
(50/50), to give after evaporation a fraction of oxygenated compounds
(20–100 mg) which was analyzed by GC–MS.
NMR experiments were performed on a Bruker NanoBay Avance III
HD (400 MHz) spectrometer (Bruker, FR-Wissembourg) at 25 °C in
CDCl3. The numbering of the positions for the assignment of the 1H and
13C signals of each compound is reported in the supporting information.
2.3. GC–MS
2.6. Syntheses
Analysis was performed on an Agilent 6890 N gas chromatograph
coupled to an Agilent 5973 N mass selective detector working in elec-
tron impact (EI) mode at 70 eV (scanning from m/z 35 to m/z 350 in
scan mode). The gas chromatograph was equipped with a fused silica
capillary column HP-5MS (60 m × 0.2 mm i.d., film thickness: 0.3 µm).
The carrier gas was helium at a flow rate of 0.8 mL/min. The oven
temperature was programmed from 50 to 210 °C at 2 °C/min and held
isothermally for 10 min. The injector (split mode, ratio 1/100) tem-
perature was 240 °C. The temperatures of the ion source and transfer
line were 230 and 250 °C, respectively. In the case of SIM mode, m/z
164 was chosen as a target mass under the same conditions as above.
Retention index (RI) was determined from the retention time of a series
of n-alkanes with linear interpolation. The yuzu oil constituents were
identified by comparison of their mass spectra and RI with those of pure
compounds registered in commercial libraries and literature data, and
with the laboratory-made database built from authentic compounds.
2.6.1. Synthesis of (6Z,8E)-undeca-6,8,10-trien-3-one 7
2.6.1.1. 2-((5-(1,3-Dioxolan-2-yl)pent-2-yn-1-yl)oxy)tetrahydro-2H-
pyran 11. To a solution of 2-(prop-2-yn-1-yloxy)tetrahydro-2H-pyran
(21.0 g, 150 mmol, 1.4 eq.) in dried THF (500 mL) cooled at –70 °C
under argon atmosphere, n-BuLi (1.83 M in hexane, 59 mL, 108 mmol,
1 eq.) was added dropwise, followed by HMPA (28 mL, 161 mmol,
1.5 eq.). The reaction mixture was stirred for 1.5 h, then 2-(2-
iodoethyl)-1,3-dioxolane (27.7 g, 121 mmol, 1.1 eq.) was added
dropwise. The reaction mixture was slowly warmed to room
temperature and stirred overnight, then quenched by addition of
250 mL saturated NaHCO3 solution. After decantation, the aqueous
layer was extracted with 3 × 400 mL of diethyl ether and the combined
organic layers were washed twice with water, then with brine and dried
over MgSO4. After evaporation on a rotatory evaporator, the crude
brown oil thus obtained (32.1 g) was purified by column
chromatography on silica gel with petroleum ether/diethyl ether (80/
2

A. Uehara and N. Baldovini
FoodChemistry338(2021)128130
20) as eluent, to furnish 2-((5-(1,3-dioxolan-2-yl)pent-2-yn-1-yl)oxy)
tetrahydro-2H-pyran 11 (21.1 g, 81% yield) as a slightly yellow oil: 1H
NMR (400 MHz, CDCl3): δ = 4.97 (t, J = 4.7 Hz, 1H11), 4.81 (t,
J = 3.4 Hz, 1H5), 4.24 (qt, J = 17.4, 2.1 Hz, 2H6), 4.00–3.94 (m, 2H12
filtered and the solid was washed with 3 × 100 mL of toluene. The
resulting sticky solid was then dried under vacuum to give (2-oxoethyl)
triphenylphosphonium bromide 14 (17.6 g, 48% yield) as a slightly
brown powder.
13), 3.89–3.84 (m, 2H12 13), 3.88–3.81 (m, 1H1a), 3.56–3.50 (m,
&
1H1b), 2.37 (tt, J = 7.5, 2.1&Hz, 2H9), 1.88 (td, J = 7.6, 4.7 Hz, 2H10),
1.89–1.82 (m, 1H3a), 1.78–1.70 (m, 1H4a), 1.67–1.51 (m,
1H4b + 2H2 + 1H3b). 13C NMR (100 MHz, CDCl3): δ = 103.2 (C11),
96.7 (C5), 85.5 (C8), 76.1 (C7), 65.0 (C12&13), 62.0 (C1), 54.6 (C6), 32.8
(C10), 30.3 (C4), 25.4 (C2), 19.1 (C3), 13.7 (C9). GC–MS (TIC)
area > 99% of the total peak area. MS (EI, 70 eV, m/z %): 39 (17),
41 (17), 45 (24), 55 (40), 67 (19), 73 (100), 83 (14), 84 (29), 85 (26),
95 (14), 139 (24), 155 (7). (No signal for M∙+ = 240).
2.6.1.5. (2E)-8,8-Dimethoxyocten-4-ynal 15. Triethylamine (2.2 mL,
1.61 g, 15.9 mmol, 1.9 eq.) was added dropwise to a suspension of
(2-oxoethyl)triphenylphosphonium bromide 14 (6.4 g, 16.6 mmol,
2 eq.) in toluene (43 mL) at 0 °C under argon atmosphere. This
mixture was stirred for 1 h, and a solution of crude 13 (1.29 g, ca.
8.28 mmol, 1 eq.) in 1 mL of toluene was added dropwise. After stirring
for 1 h at 0 °C, then 48 h at room temperature, the reaction mixture was
fitered through silica gel (58 g) with at first 100 mL of petroleum ether
to remove toluene, and then 300 mL of petroleum ether/diethyl ether
(50/50) to give a filtrate which, after concentration under vacuum gave
(2E)-8,8-dimethoxyocten-4-ynal 15 containing ca. 10% of (2Z)-isomer,
as a yellow oil (969 mg), used in the next step without further
purification. 1H NMR (400 MHz, CDCl3): δ = 9.47 (d, J = 7.8 Hz,
1H1), 6.53 (dt, J = 15.9, 2.0 Hz, 1H3), 6.32 (dd, J = 15.8, 7.8 Hz, 1H2),
4.40 (t, J = 5.6 Hz, 1H8), 3.28 (s, 3H9 + 3H10), 2.44 (td, J = 7.2,
1.5 Hz, 2H6), 1.81 (q, J = 6.7 Hz, 2H7). 13C NMR (100 MHz, CDCl3):
δ = 193.4 (C1), 139.2 (C2), 133.5 (C3), 105.9 (C5), 103.2 (C8), 78.0
(C4), 53.2 (C9&10), 31.1 (C7), 15.51 (C6). GC–MS (TIC) area > 85% of
the total peak area. MS (EI, 70 eV, m/z %): 39 (22), 45 (21), 47 (20), 65
(40), 75 (99), 77 (25), 79 (18), 91 (33), 92 (18), 93 (26), 107 (18), 119
(10), 121 (11), 124 (11), 135 (11), 139 (41), 151 (100), 152 (11), 153
(8), 167 (10), 181 (3), 182 (M∙+, 0.5).
2.6.1.2. 6,6-Dimethoxyhex-2-yn-1-ol 12. A solution of 2-((5-(1,3-
dioxolan-2-yl)pent-2-yn-1-yl)oxy)tetrahydro-2H-pyran 11 (16.1 g,
67.1 mmol, 1 eq.) in methanol (320 mL), containing Amberlyst®15
(3.29 g) was stirred under argon atmosphere at room temperature
overnight. After filtration on a mixture of Celite® and NaHCO3, and
rinsing with 50 mL of diethyl ether and 100 mL of H2O, the filtrate was
evaporated under vacuum. The remaining aqueous mixture was
extracted with 3 × 100 mL of diethyl ether. The combined organic
phases were washed with brine, dried over MgSO4, and concentrated
under vacuum to give 13.0 g of crude 6,6-dimethoxyhex-2-yn-1-ol 12 as
a brown oil which was used in the next step without further
purification. 1H NMR (400 MHz, CDCl3): δ = 4.48 (t, J = 3.4 Hz,
1H7), 4.21 (s, 2H2), 3.32 (s, 3H8 + 3H9), 2.73 (s, 1H1), 2.26 (tt,
J = 7.3, 2.1 Hz, 2H5), 1.79 (q, J = 6.8 Hz, 2H6).13C NMR (100 MHz,
CDCl3): δ = 103.3 (C7), 85.0 (C4), 78.9 (C3), 53.0 (C8&9), 51.0 (C2),
31.3 (C6), 14.2 (C5). GC–MS (TIC) area > 99% of the total peak area.
MS (EI, 70 eV, m/z %): 45 (24), 55 (40), 37 (19), 73 (100), 84 (29), 85
(26), 95 (14), 99 (11), 139 (24), 155 (7), 158 (M∙+, 0.06).
2.6.2. Methyltriphenylphosphonium iodide 16
Iodomethane (18.7 g, 132 mmol, 1 eq.) was added dropwise to a
solution of triphenylphosphine (52.3 g, 200 mmol, 1.5 eq.) in toluene
(100 mL). The reaction mixture was stirred at room temperature for
24 h, then filtered. The resulting solid was washed with 3 × 100 mL of
toluene, then dried under vacuum to give methyltriphenylphosphonium
iodide 16 (53.1 g, 99% yield) as a white powder.
2.6.1.3. 6,6-Dimethoxyhex-2-ynal 13. To a suspension of PCC (36.5 g,
169 mmol, 2.1 eq.) and Celite® (36.2 g) in dichloromethane (450 mL) at
0 °C under argon atmosphere, was added dropwise the crude sample of
12 (12.8 g, ca. 81 mmol, 1 eq.) diluted in 6 mL dichloromethane. After
addition, the solution was stirred at room temperature for 3 h, then
filtered through a pad of silica gel (65.5 g) and Celite®, using 500 mL of
petroleum ether/diethyl ether (80/20) for rinsing, to give after
evaporation 7.95 g of 6,6-dimethoxyhex-2-ynal 13 as a yellow-orange
oil, used in the next step without further purification. 1H NMR
(400 MHz, CDCl3): δ = 9.11 (d, J = 0.76 Hz, 1H1), 4.39 (t,
J = 5.6 Hz, 1H6), 3.28 (s, 3H7 + 3H8), 2.42 (t, J = 7.3 Hz, 2H4),
1.82 (q, J = 6.7 Hz, 2H5). 13C NMR (100 MHz, CDCl3): δ = 177.1 (C1),
103.0 (C6), 98.0 (C3), 81.6 (C2), 53.4 (C7&8), 30.5 (C5), 14.6 (C4).
GC–MS (TIC) area > 99% of the total peak area. MS (EI, 70 eV, m/z %):
45 (11), 47 (12), 65 (11), 75 (100), 81 (4), 82 (6), 95 (4), 97 (3), 125
(41), 126 (3), 156 (M∙+, 0.07).
2.6.3. (E)-9,9-Dimethoxynona-1,3-dien-5-yne 17
Methyltriphenylphosphonium iodide 16 (5.48 g, 13.6 mmol,
2.6 eq.) was added to a suspension of tBuOK (1.45 g, 12.9 mmol,
2.4 eq.) in dried THF (50 mL) under argon atmosphere. The reaction
mixture was stirred for 2 h at room temperature, then overnight at
35 °C. After cooling to room temperature, the sample of crude 15
(969 mg, ca. 5.32 mmol, 1 eq.) in 1 mL of THF was added dropwise to
this mixture. After stirring for 3 h at room temperature, 100 mL of
petroleum ether were added. The reaction mixture was then filtered
through silica gel (59.3 g) with 500 mL of petroleum ether/ diethyl
ether (50/50) for elution. After evaporation of the solvent, the crude
(E)-9,9-dimethoxynona-1,3-dien-5-yne 17 was obtained as a brown oil
(863 mg), used in the next step without further purification. 1H NMR
(400 MHz, CDCl3): δ = 1H NMR (400 MHz, CDCl3) δ 6.44 (dd,
J = 15.5, 10.8 Hz, 1H9), 6.28 (dt, J = 16.8, 10.4 Hz, 1H10), 5.54 (d,
J = 15.5 Hz, 1H8), 5.19 (d, J = 16.8 Hz, 1H11a), 5.07 (d, J = 10.0 Hz,
1H11b), 4.42 (t, J = 5.7 Hz, 1H3), 3.27 (s, 3H1 + 3H2), 2.33 (td,
J = 7.3, 2.2 Hz, 2H5), 1.76 (q, J = 6.8 Hz, 2H4). 13C NMR (100 MHz,
CDCl3): δ = 13C NMR (101 MHz, CDCl3) δ 141.2 (C9), 136.3 (C10),
118.8 (C11), 112.4 (C8), 103.4 (C3), 92.2 (C6), 79.8 (C7), 53.1 (C1&2),
31.6 (C5), 15.1 (C4). GC–MS (TIC) area was > 74% of the total peak
area. MS (EI, 70 eV, m/z %): 65 (24), 75 (29), 91 (100), 105 (14), 115
2.6.1.4. (2-Oxoethyl)triphenylphosphonium bromide 14. Trifluoroacetic
acid (46 mL, 68.5 g, 601 mmol, 6.3 eq.) was added dropwise to a
solution of 2-bromo-1,1-diethoxyethane (18.9 g, 96 mmol, 1 eq.) in
200 mL of dichloromethane under an argon atmosphere. The reaction
mixture was stirred for 24 h, then 200 mL of water were added, and the
mixture was decanted. The aqueous layer was extracted with
3 × 200 mL of dichloromethane, neutralized by the addition of solid
Na2CO3 (22.5 g) then extracted again with 200 mL of dichloromethane.
The combined organic layers were washed with saturated NaHCO3
solution, with brine, and then dried over MgSO4 and concentrated
under vacuum until a volume of ca. 100 mL to give a slightly yellow
solution of 2-bromoacetaldehyde. This solution was then transferred
into a two-neck flask equipped with a reflux condenser, and 100 mL of
toluene and 43.0 g (164 mmol, 1.7 eq.) of triphenylphosphine were
added successively. This mixture was then heated at 50 °C and stirred
overnight. After cooling to room temperature, the reaction mixture was
(17), 117 (22), 122 (36), 135 (43), 149 (64), 150 (7), 165 (6), 180 (M∙+
1).
,
2.6.4. (6E)-Nona-6,8-dien-4-ynal 18
A 3% aqueous solution of sulfuric acid was added to 863 mg (ca.
4.79 mmol, 1 eq.) of the crude sample of 17 obtained as described
above. The reaction mixture was stirred overnight at room temperature,
then quenched by the addition of solid NaHCO3 (10.9 g) at 0 °C and
3

A. Uehara and N. Baldovini
FoodChemistry338(2021)128130
transferred into a separating funnel. The aqueous layer was extracted
with 3 × 150 mL of diethyl ether and the combined organic layers were
washed with brine, dried over MgSO4, and concentrated under vacuum
to afford a brown oil (548 mg) which was then purified by column
chromatography on silica gel using petroleum ether/ diethyl ether (85/
15) as eluent, to furnish eventually (E)-nona-6,8-dien-4-ynal 18
(364 mg, 25% yield for 5 steps) as an orange oil. 1H NMR (400 MHz,
CDCl3): δ = 9.73 (s, 1H1), 6.45 (dd, J = 10.8, 15.5 Hz, 1H7), 6.27 (dt,
J = 16.9, 10.3 Hz, 1H8), 5.52 (d, J = 15.6 Hz, 1H6), 5.20 (d,
J = 16.9 Hz, 1H9a), 5.08 (d, J = 10.0 Hz, 1H9b), 2.66–2.55 (m,
2H2 + 2H3). 13C NMR (100 MHz, CDCl3): δ = 200.4 (C1), 141.7 (C7),
136.1 (C8), 119.2 (C9), 111.9 (C6), 90.7 (C4), 80.4 (C5), 42.6 (C2), 12.9
(C3). GC–MS (TIC) area > 86% of the total peak area. MS (EI, 70 eV, m/
z %): 39 (17), 51 (18), 65 (33), 77 (50), 78 (47), 79 (33), 91 (100), 92
(68), 103 (21), 105 (40), 115 (8), 119 (7), 133 (41), 134 (M∙+, 17).
to rinse the pad. The filtrate was decanted and the aqueous layer was
extracted 3 times with 15 mL of dichloromethane and the combined
organic phases were washed with brine, dried over MgSO4, and con-
centrated under vacuum to give a slightly yellow oil (74 mg) which was
purified by column chromatography on silica gel with petroleum ether/
diethyl ether (80/20) as eluent to furnish (6Z,8E)-undeca-6,8,10-trien-
3-ol 20 (69 mg, 85% yield) as a colorless oil. 1H NMR (400 MHz,
CDCl3): δ = 1H NMR (400 MHz, CDCl3) δ 6.46 (dd, J = 14.8, 11.3 Hz,
1H9), 6.34 (dt, J = 16.9, 10.4 Hz, 1H11), 6.15 (dd, J = 14.9, 10.7 Hz,
1H10), 5.97 (dd, J = 11.0, 11.0 Hz, 1H8), 5.42 (dt, J = 10.6, 7.8 Hz,
1H7), 5.15 (d, J = 16.6 Hz, 1H12a), 5.02 (d, J = 10.1 Hz, 1H12b),
3.53–3.45 (m, J = 12.2, 7.8, 4.7 Hz, 1H3), 2.36–2.17 (m, 2H6),
1.53–1.37 (m, 2H5 + 2H2), 0.87 (t, J = 7.5 Hz, 3H1). 13C NMR
(100 MHz, CDCl3): δ = 137.2 (C11), 133.4 (C10), 132.7 (C7), 128.8 (C8),
128.4 (C9), 117.2 (C12), 72.8 (C3), 36.6 (C5), 30.3 (C2), 24.2 (C6), 9.9
(C1). GC–MS (TIC) area > 85% of the total peak area. MS (EI, 70 eV, m/
z %): 41 (28), 57 (19), 65 (15), 66 (10), 67 (19), 77 (50), 78 (26), 79
(64), 91 (100), 92 (24), 93 (28), 94 (38), 105 (51), 106 (13), 109 (7),
117 (5), 119 (23), 133 (3), 137 (3), 148 (2), 166 (M∙+, 13).
2.6.5. (E)-Undeca-8,10-dien-6-yn-3-ol 19
Bromoethane (615 mg, 5.64 mmol, 2.2 eq.) was added dropwise to a
suspension of magnesium turnings (126 mg, 5.18 mmol, 2.0 eq.) in dry
diethyl ether (12 mL). After generation of the Grignard reagent, the
reaction mixture was stirred for 1.5 h at room temperature, then a so-
lution of (E)-nona-6,8-dien-4-ynal 18 (346 mg, 2.58 mmol, 1 eq.) in
1 mL of diethyl ether was added dropwise. The reaction mixture was
stirred overnight, and then quenched by the addition of saturated
NH4Cl solution (15 mL) and decanted. The aqueous layer was extracted
with 3 × 25 mL of diethyl ether and the combined organic layers were
washed with brine, dried over MgSO4, and concentrated under vacuum
to give a yellow oil (424 mg) which was purified by column chroma-
tography on silica gel with petroleum ether/ diethyl ether (85/15) as
eluent to furnish (E)-undeca-8,10-dien-6-yn-3-ol 19 (254 mg, 60%
yield) as a colorless oil. 1H NMR (400 MHz, CDCl3): δ = 6.44 (dd,
J = 15.5, 10.8 Hz, 1H10), 6.27 (dt, J = 16.8, 10.4 Hz, 1H11), 5.54 (d,
J = 15.5 Hz, 1H9), 5.19 (d, J = 16.8 Hz, 1H12a), 5.07 (d, J = 10.0 Hz,
1H12b), 3.71–3.55 (m, 1H3), 2.41 (td, J = 7.2, 2.2 Hz, 1H6), 1.70–1.51
(m, 2H5), 1.49–1.38 (m, 2H2), 0.89 (t, J = 7.4 Hz, 3H1). 13C NMR
(100 MHz, CDCl3): δ = 141.3 (C10), 136.3 (C11), 118.9 (C12), 112.3
(C9), 92.8 (C7), 80.0 (C8), 72.3 (C3), 35.5 (C5), 30.2 (C2), 16.2 (C6), 9.9
(C1). GC–MS (TIC) area > 88% of the total peak area. MS (EI, 70 eV, m/
z %): 57 (31), 65 (28), 77 (32), 78 (24), 79 (32), 91 (100), 92 (26), 93
(37), 103 (23), 107 (22), 108 (22), 115 (24), 117 (41), 131 (12), 135
(78), 136 (10), 146 (4), 164 (M∙+, 4).
2.6.7. (6Z,8E)-Undeca-6,8,10-trien-3-one 7
To a solution of 20 (40 mg, 0.24 mmol, 1 eq.) in dry DMSO (2 mL)
under argon atmosphere, IBX (118 mg, 0.42 mmol, 1.7 eq.) was added
at room temperature. The reaction mixture was stirred for 24 h at room
temperature, then filtered through a short pad of silica gel using 50 mL
of petroleum ether/diethyl ether (80/20) for rinsing, and concentrated
under vacuum to give a colorless oil (17 mg, 43% yield). 1H NMR
(400 MHz, CDCl3): δ = 6.44 (dd, J = 14.7, 11.5 Hz, 1H8), 6.34 (dt,
J = 16.9, 10.4 Hz, 1H10), 6.14 (dd, J = 14.8, 10.8, 1H9), 5.96 (dd,
J = 10.9, 10.9 Hz, 1H7), 5.35 (dt, J = 10.4, 7.2 Hz, 1H6), 5.15 (d,
J = 16.6 Hz, 1H11a), 5.03 (d, J = 10.0 Hz, 1H11b), 2.47–2.31 (m,
2H4 + 2H5 + 2H2), 0.99 (t, J = 7.3 Hz, 3H1). 7/(6Z,8Z)-isomer = 9/1
13C NMR (100 MHz, CDCl3): δ = 210.7 (C3), 137.1 (C10), 133.7 (C9),
131.0 (C6), 129.3 (C7), 128.1 (C8), 117.4 (C11), 42.0 (C5), 36.0 (C2),
22.3 (C4), 7.8 (C1). GC–MS (TIC) area > 91% of the total peak area. MS
(EI, 70 eV, m/z %): 39 (14), 41 (15), 57 (100), 65 (13), 67 (10), 77 (43),
78 (12), 79 (38), 91 (68), 92 (57), 93 (19), 107 (18), 117 (10), 135 (4),
164 (M∙+, 40).
3. Results and discussion
(6Z,8E)-Yuzunone 7 was synthesized as detailed in Scheme 1 and
used subsequently as a reference compound in our analytical study of
various yuzu and citrus oils. We had to design an original approach for
the synthesis of 7 to ensure the stereoselective construction of the
C6–C7 double bond. Our synthetic plan started by the alkylation of the
tetrahydropyranyl propargyl ether with Stowell iodide (Stowell et al.,
1983) to give 11, which was subsequently deprotected to furnish the
alcohol 12. The corresponding aldehyde 13 could be easily obtained by
oxidation with PCC. Two Wittig reactions were used to build the con-
jugated terminal dienic moiety, by the subsequent reactions on 13 of
2.6.6. (6Z,8E)-Undeca-6,8,10-trien-3-ol 20
Zinc dust (4 g) was ground in a mortar to remove lumps, then stirred
with 15 mL of 2% aqueous HCl in a beaker for a few minutes. The
suspension was filtered and the powder was washed successively twice
with 5 mL of 2% HCl, twice with 10 mL of distilled water, twice with
10 mL of EtOH (95%) and once with 10 mL of dried diethyl ether. It was
then thoroughly dried under vacuum to give a fine powder. This acti-
vated zinc powder (2.0 g, 32 mmol, 65 eq.) was then put in a two-
necked flask and 12 mL of distilled H2O were added under the argon
atmosphere. The resulting suspension was stirred for 20 min, then Cu
(OAc)2 monohydrate (211 mg, 1.06 mmol, 2.2 eq.) was added. The
suspension was stirred for another 15 min, then AgNO3 (213 mg,
1.26 mmol, 2.6 eq.) was added, and the reaction mixture was stirred for
a further 30 min until the end of the exothermic reaction, and was then
filtered. The zinc complex was washed twice with 10 mL of distilled
H2O, twice with 10 mL of MeOH, twice with 10 mL of acetone and
twice with 10 mL of dried diethyl ether successively. The resulting
black paste was immediately transferred into a two-necked flask purged
with argon and 8 mL of MeOH/H2O (50/50) were added. The suspen-
sion was stirred for 30 min at room temperature, then (E)-undeca-8,10-
dien-6-yn-3-ol 19 (80 mg, 0.49 mmol, 1 eq.) in 0.9 mL of MeOH was
added. After stirring overnight, the reaction mixture was filtered
through a short pad of Celite®, and dichloromethane (100 mL) was used
(triphenylphosphoranyliden)acetaldehyde
14
and
methylene-
triphenylphosphorane 16. The first Wittig reaction produced aldehyde
15 accompanied by ca. 10% of its minor 2Z isomer. After deprotection,
the acetal 17 afforded aldehyde 18 which was converted to 19 by the
addition of ethylmagnesium bromide. The stereospecific reduction of
the triple bond to the (Z)-double bond proved to be more difficult than
expected. Indeed, neither the use of P-2 nickel nor Lindlar catalyst were
successful, as in both cases a mixture of isomerized products was ob-
tained. Trienol 20 could be eventually prepared by hydrogenation with
the Zn(Cu/Ag) system described by Boland, Schroer, and Sieler (1987)
and its subsequent oxidation with IBX (Frigerio et al., 1999) furnished
the target compound 7 in good yield. A small amount (ca. 8%) of the 8Z
isomer was still present in the sample, but did not impair the GC–O
experiments since 8Z and 8E isomers could be resolved on the GC–O
4

A. Uehara and N. Baldovini
FoodChemistry338(2021)128130
Scheme 1. Synthesis of (6Z,8E)-undeca-6,8,10-trien-3-one 7. a: nBuLi, HMPA/THF, −70 °C → RT (81%) b: Amberlyst® 15, MeOH, RT c: PCC, DCM, 0 °C → RT, d:
Et3N, toluene, RT e: tBuOK, THF, RT → 35 °C f: H2SO4, H2O, RT 25% for 5 steps) g: Mg, EtBr, Et2O, RT (60%) h: Zn, Cu(OAc)2, AgNO3, MeOH/H2O, RT (85%) i: IBX,
DMSO, RT (43%).
column. Hence, we can report herein the unambiguous chromato-
graphic, spectral and olfactory properties of yuzunone, which were in
good agreement with the data published by Miyazawa et al. (2009b) for
the mixture of 7 and 10. With this sample of pure 7 in hand, we could
investigate by GC–MS its presence in 5 commercial citrus essential oil
samples. These were 2 yuzu cold-pressed oils, a yuzu essential oil ob-
tained by hydrodistillation, and two cold-pressed oils obtained from
other citrus species: a yellow mandarin oil (Citrus reticulata) from Italy,
and a Brazilian orange oil (Citrus sinensis). The olfactory properties of
these last two oils samples were found to be the most similar to yuzu
oils in sensorial tests conducted by a panel of 3 judges on a dozen of
various citrus oils, and we wondered if 7 was not contained in trace
amounts also in these species. Their chemical analysis was motivated by
the fact that we initially suspected that 7 could be present in other
citrus species, and may not necessarily be restricted to yuzu. To in-
crease the detection limit of our analytical procedure, we applied a pre-
concentration step before GC–MS analysis, by a simple fractionation on
a short pad of silica gel to get rid of the hydrocarbon fraction (mostly
limonene) representing usually more than 90% of the samples. The
oxygenated fraction was then analyzed by GC–MS in scan and in SIM
modes (on ion m/z = 164), looking for the presence of 7, the retention
time of which could be unambiguously determined with the help of our
synthetic reference sample. Surprisingly, 7 could not be detected in any
of our samples with this approach, demonstrating clearly that it was
contained in our samples (if actually present) at a much lower con-
centration than the 0.01% reported by Miyazawa et al. (2009a) in their
study. As the very low olfactory detection threshold of yuzunone gave
us hope that it could be more easily detected by GC–O, we conducted a
detailed aroma extract dilution analysis of yuzu cold-pressed oil with 8
panelists. Table 1 shows some of the odorant constituents and their
flavor dilution factor (FD factor) for each panelist. They described a
total of 83 odorant zones, while only 2 zones were detected by all pa-
nelists, which corresponded to nonanal (Entry 4) and cis-4-decenal
(Entry 7). On the other hand, only three panelists detected an odor zone
which could be tentatively attributed to 7 (Entry 13) since it had an
odor quality and a retention time similar to the synthetic sample. Its FD
factor was however moderate (from 0 to 4); therefore we can conclude
that 7 was not a significantly important odor contributor in our sam-
ples, compared to nonanal and cis-4-decenal.
In conclusion, we could prepare for the first time a sample of
(6Z,8E)-undeca-6,8,10-trien-3-one (yuzunone) 7 by an original stereo-
selective synthesis. However, despite the use of this sample as a re-
ference compound, 7 could not be identified by GC–MS in any of our
yuzu peel essential oil samples. Nevertheless, it was characterized as a
Table 1
Aroma extract dilution analysis of yuzu peel oil.
Entry Compounda
RIb
Peak area
Olfactory descriptiond (Panelist No.)
Individual log4(FD)e
AMf
(%)c
P1 P2 P3 P4 P5 P6 P7 P8
1
2
3
4
α-Pinene
Octanal
p-Cresol
Nonanal
928
978
2.04
tr
fresh1, lemon1,5, pine3, citrus4, resinous6
3
3
0
4
3
2
7
7
2
0
3
0
3
lemon-like2,5, citrus4, orange1, menthol3
1
1
1
1054 tr
1080 tr
shoes1, human skin2, citrus3, plastic6, maltol, caramel7,8
orange1,4, floral1,5, yuzu-like2, lemon, mint, fresh, sweet1,8, citrus3,
pleasant2,8, jasmine5, oregano6, almond, burnt7
citrus4, grapefruit, baked apple2, yuzu-like1,5
2
5
1
3
0
3
4
6
4.1
1.8
5
6
7
Linalool
1082 1.35
4
4
1
2
2
1
2
0
0
Unknown
cis-4-Decenal
1133
–
mint, floral1, tangerine2, woody3, plastic5,6, citrus5, grass8
orange, yuzu-like1,6, fatty1, citrus3, aldehyde, choking2, hot plastic5, herb,
lemon6, orange, yuzu, aldehyde, metallic7
4
4
3
4
2
1
0
2
1168 tr
1
8
Decanal
1181 0.03
1268 0.08
1275 tr
woody3, grapefruit5, citrus, herb, yuzu6
0
2
0
0
3
9
Thymol
carvacrol-thymol2, woody1,6, phenolic1, menthol3,8, leaf3, cork6, thymol7
phenol1, carvacrol-thymol2, wood skin5, medicinal8
leaf4, grapefruit5
2
1
1
1
1
1
1
0
10
11
12
Carvacrol
Undecanal
2
0
4
1284 tr
0
2
0
4
(E,E)-2,4-Decadienal 1286 tr
fatty1,2, waxy2, woody3, almond milk5, stink bug4, fruity6, almond, lactonic,
coconut7
3
2
3
0
3
13
14
Yuzunone
1359 tr
1477
lemon/yuzu5, stink bug, bitter, woody7, coriander, herb8
grape, floral1, sweet2,3, coconut5,7, resinous6, spicy7
4
1
2
0
0
Unknown
–
0
0
1
1
a: Linear retention index in GC–O on HP5 column. b: Odorants identified unambiguously by coinjection of reference compounds. Additional constituents identified
unambiguously by coinjection of reference compounds and detected by less than half of the panelists. c: Percentage of each constituent calculated by GC–FID analysis.
tr: trace constituent (< 0.02%). d: Individual olfactory descriptions (corresponding panelist letter in superscript). e: log4(FD) of each individual perceived odor zone.
f: Arithmetic mean of the log4(FD), for each odor zone unanimously perceived.
5

A. Uehara and N. Baldovini
FoodChemistry338(2021)128130
weak odorant in our GC–O experiments, since it could be perceived
with a moderate FD factor by three panelists out of eight, in contrast
with nonanal and cis-4-decenal, detected by all panelists. These results
are somehow contradictory to those reported by Miyazawa et al.
(2009a) who claimed that yuzunone is a key odorant of yuzu peel oil
according to their AEDA experiments. Many factors can explain this
situation, like the differences of methodological approaches and ana-
lytical systems for olfactory analysis, the variations of individual sen-
sitivity among the panelists, the variability of the yuzunone content in
the samples investigated, etc. Besides, such discrepancies are common
among the GC–O studies on yuzu reported in the literature (Miyazawa
et al., 2009a; Tomita, 2010; Miyazato et al., 2012, 2013; Tomiyama
et al., 2012; Song, Sawamura, Ito, Kawashimo, & Ukeda, 1999; Lan-Phi,
Shimamura, Ukeda, & Sawamura, 2009), and nicely illustrate the fact
that the determination of key odorants of natural raw materials still
requires the involvement of large panels of evaluators, and also to take
carefully into consideration the chemical variability inherent to natural
extracts. It should be also reminded that although GC–O is the most
potent technique for the characterization of key odorants, it suffers
from several limitations when used as a single tool (Baldovini &
Chaintreau, 2020). Indeed, all the synergistic effects occurring between
the odorants within a mixture cannot be revealed by a separative
technique like GC–O. Therefore, weakly odorant components can
nevertheless significantly contribute to the odor of the entire mixture.
For this reason, a definitive answer concerning the key odorants of yuzu
and the contribution of yuzunone would require careful recombination
experiments followed by sensorial tests with large panels, and com-
parison with many different samples of yuzu essential oils.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.foodchem.2020.128130.
References
Baldovini N. & Chaintreau A. (2020). Identification of key odorants in complex mixtures
occurring in Nature. Natural Product Reports. In press.
Boland, W., Schroer, N., & Sieler, C. (1987). Stereospecific Syntheses and Spectroscopic
Properties of Isomeric 2,4,6,8-Undecatetraenes. New Hydrocarbons from the Marine
Brown Alga Gijfordia mitchellae. Helvetica Chimica Acta, 70, 1025–1040.
Cerutti-Delasalle, C., Mehiri, M., Cagliero, C., Rubiolo, P., Bicchi, C., Meierhenrich, U. J.,
& Baldovini, N. (2016). The (+)-cis- and (+)-trans-Olibanic Acids: Key Odorants of
Frankincense. Angewandte Chemie International Edition, 55, 13719–13723.
Frigerio, M., Santagostino, M., & Sputore, S. (1999). A User-Friendly Entry to 2-
Iodoxybenzoic Acid (IBX). Journal of Organic Chemistry, 64, 4537–4538.
Lan-Phi, N. T., Shimamura, T., Ukeda, H., & Sawamura, M. (2009). Chemical and aroma
profiles of yuzu (Citrus junos) peel oils of different cultivars. Food Chemistry, 115,
1042–1047.
Matsuura, Y., Hara, G., & Abe, S. (1980). A novel 13-membered macrolide. A flavor
component of Yuzu (Citrus junos) oil. Proceedings of the 8th International Congress of
Essential Oils (pp. 497). .
Miyazato, H., Hashimoto, S., & Hayashi, S. (2012). Identification of the odour-active al-
dehyde trans-4,5-epoxy-(E, Z)-2,7-decadienal in yuzu (Citrus junos Sieb. ex Tanaka).
European Food Research and Technology, 235, 881–891.
Miyazato, H., Hashimoto, S., & Hayashi, S. (2013). First identification of the odour-active
unsaturated aliphatic acid (E)-4-methyl-3-hexenoic acid in yuzu (Citrus junos Sieb. ex
Tanaka). Flavour and Fragrance Journal, 28, 62–69.
Miyazato, H. (2018). Volatile Composition and the Key Aroma Compounds of the Citrus
tachibana (Makino) Tanaka Peel Essential Oil. Journal of Essential Oil-Bearing Plants,
21, 924–937.
Miyazawa, N., Tomita, N., Kurobayashi, Y., Nakanishi, A., Ohkubo, Y., Maeda, T., &
Fujita, A. (2009a). Novel Character Impact Compounds in Yuzu (Citrus junos Sieb. ex
Tanaka) Peel Oil. Journal of Agriculture and Food Chemistry, 57, 1990–1996.
Miyazawa, N., Nakanishi, A., Tomita, N., Ohkubo, Y., Maeda, T., & Fujita, A. (2009b).
Novel Key Aroma Components of Galbanum Oil. Journal of Agriculture and Food
Chemistry, 57, 1433–1439.
CRediT authorship contribution statement
Miyazawa, N., Fujita, A., & Kubota, K. (2010). Aroma Character Impact Compounds in
Kinokuni Mandarin Orange (Citrus kinokuni) Compared with Satsuma Mandarin
Orange (Citrus unshiu). Bioscience, Biotechnology, and Biochemistry, 74, 835–842.
Nakanishi K., & Watanabe H. (2010). JP2010083913A.
Ayaka Uehara: Methodology, Investigation, Writing - original
draft. Nicolas Baldovini: Funding acquisition, Conceptualization,
Writing - review & editing.
Ohloff, G., Pickenhagen, W., & Kraft, P. (2012). Scent and Chemistry: The Molecular World
of Odors. Zurich: Verlag Helvetica Chimica Acta223–225.
Declaration of Competing Interest
Song, H. S., Sawamura, M., Ito, T., Kawashimo, K., & Ukeda, H. (1999). Quantitative
determination and characteristic flavour of Citrus junos (yuzu) peel oil. Flavour and
Fragrance Journal, 15, 245–250.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Stowell, J. C., King, B. T., & Hauck, H. F., Jr. (1983). A Simple Preparation of a β-Iodo
Acetal and a β-Iodo Ketal. The Journal of Organic Chemistry, 48, 5381–5382.
Tajima, K., Tanaka, S., Yamaguchi, T., & Fujita, M. (1990). Analysis of Green and Yellow
Yuzu Peel Oils (Citrus junos Tanaka). Novel Aldehyde Components with Remarkably
Low Odor Thresholds. Journal of Agriculture and Food Chemistry, 38, 1544–1548.
Tajima, K., Fukuoka, K., Hasegawa, Y., & Toi, N. (1992). Research on new key odorants in
yuzu peel oil (Citrus junos TANAKA). Proceedings of the 36th Symposium on the
chemistry of Terpenes, Essential Oils, and Aromatics (pp. 28). .
Acknowledgments
We gratefully thank The Gen Foundation (London, UK) and The
French Ministry of Research for a grant (AU). The authors wish to thank
also Jade Fesancieux, Niko Radulović, Coline Perrin, Marie Rolfo,
Florence Auberon, Lingling Xiong for their kind participation in the
GC–O panel. We also thank Daniel Joulain (SCBZ consulting) and
Kazutoshi Sakurai (Shizuoka Cancer Center Research Institute, Japan)
for their help in the bibliographical research.
Tomita, N. (2010). Identification of yuzunone® and Evaluation of Its Derivatives. Koryo,
248, 69–76.
Tomiyama, K., Aoki, H., Oikawa, T., Sakurai, K., Kasahara, Y., & Kawakami, Y. (2012).
Characteristic volatile components of Japanese sour citrus fruits: Yuzu, Sudachi and
Kabosu. Flavour and Fragrance Journal, 27, 341–355.
Ullrich, F., & Grosch, W. (1987). Identification of the most intense volatile flavor com-
pounds formed during autoxidation of linoleic acid. Zeitschrift fuer Lebensmittel-
Untersuchung und -Forschung, 184, 277–282.
6