Synthesis, olfactory evaluation, and determination of the absolute configuration of the 3,4-didehydroionone stereoisomers

Article abstract of DOI:10.1002/hlca.200690109

The synthesis of 3,4-didehydroionone isomers 4, (+)-6, and (-)-6 and of 3,4-didehydro-7,8-dihydroionone isomers 5, (+)-7, and (-)-7 was accomplished starting from commercially available racemic α-ionone (1). Their preparation of the racemic forms 4-7 was first achieved by mean of a number of chemo- and regioselective reactions (Schemes 1 and 2). The enantio- and diastereoselective lipase-mediated kinetic acetylation of 4-hydroxy-γ- ionone (10a/10b) provided 4-hydroxy-γ-ionone (+)-10a/(±)-10b and (+)-4-(acetyloxy)-γ-ionone ((+)12b) (Scheme 3). The latter compounds were used as starting materials to prepare the 3,4-didehydro-γ-ionones (+)- and (-)-6 and the 3,4-didehydro-7,8-dihydro-γ-ionones (+)- and (-)-7 in enantiomer-enriched form. The absolute configuration of (+)-12b was determine by chemical correlation with (+)-(6S)-γ-ionone ((+)-3) and with (-)-(6S)-α-ionone ((-)-1) therefore allowing to assign the (S)-configuration to (+)-6 and (+)-7. Olfactory evaluation of the above described 3,4-didehydroionone isomers shows a significant difference between the enantiomers and regioisomers both in fragrance feature and in detection threshold (Table).

Full text of DOI:10.1002/hlca.200690109

1110
Helvetica Chimica Acta – Vol. 89 (2006)
Synthesis, Olfactory Evaluation, and Determination of the Absolute
Configuration of the 3,4-Didehydroionone Stereoisomers
by Stefano Serra*, Claudio Fuganti, and Elisabetta Brenna
C.N.R. Istituto di Chimica del Riconoscimento Molecolare, Sezione ‘Adolfo Quilico’, Dipartimento di
Chimica, Materiali ed Ingegneria Chimica ‘Giulio Natta’, Politecnico di Milano, Via Mancinelli 7,
I-20133 Milano
(phone: +390223993076; fax: +390223993080; e-mail: stefano.serra@polimi.it)
The synthesis of 3,4-didehydroionone isomers 4, (+)-6, and (ꢀ)-6 and of 3,4-didehydro-7,8-dihy-
droionone isomers 5, (+)-7, and (ꢀ)-7 was accomplished starting from commercially available racemic
a-ionone (1). Their preparation of the racemic forms 4–7 was first achieved by mean of a number of
chemo- and regioselective reactions (Schemes 1 and 2). The enantio- and diastereoselective lipase-medi-
ated kinetic acetylation of 4-hydroxy-g-ionone (10a/10b) provided 4-hydroxy-g-ionone (+)-10a/(ꢁ)-10b
and (+)-4-(acetyloxy)-g-ionone ((+)12b) (Scheme 3). The latter compounds were used as starting mate-
rials to prepare the 3,4-didehydro-g-ionones (+)- and (ꢀ)-6 and the 3,4-didehydro-7,8-dihydro-g-ionones
(+)- and (ꢀ)-7 in enantiomer-enriched form. The absolute configuration of (+)-12b was determine by
chemical correlation with (+)-(6S)-g-ionone ((+)-3) and with (ꢀ)-(6S)-a-ionone ((ꢀ)-1) therefore allow-
ing to assign the (S)-configuration to (+)-6 and (+)-7. Olfactory evaluation of the above described 3,4-
didehydroionone isomers shows a significant difference between the enantiomers and regioisomers both
in fragrance feature and in detection threshold (Table).
1. Introduction. – The ionone isomers 1–3 are natural1) C13 norterpenoid ketones
[1]. The a- and b-ionones are widely used as flavor ingredients [2] and as starting mate-
rials in several industrial processes, whereas regioisomerically- and enantiomerically
pure a- and g-ionones [3] are suitable building blocks in the synthesis of different nat-
ural products. As part of a programme of synthesis of enantiomer-enriched norterpe-
noid odorants, we have previously reported the preparation and the olfactory evalua-
tion of ionones 1–3 [3][4]. Our studies showed that regioisomer and enantiomer com-
position greatly affected the fragrance properties either in terms of features or as odor
thresholds. For example, (S)-g-ionone was recognized [5] as the most pleasant ionone
isomer with an odor threshold 150 times lower than that of the (R)-enantiomer and 50
times lower than that of the a-isomer. These notable results prompted us to investigate
the preparation of further C13 norterpenoid compounds with a ionone skeleton. In this
context, we focalized our attention on the 3,4-didehydroionone isomers 4–7. The 3,4-
didehydro-b-ionone (4), the corresponding 7,8-dihydro derivative 5, and 3,4-didehy-
dro-g-ionone (6) have been found2), usually as trace components, in several natural
sources [6]. Compounds 4 and 5 are used in a flavor formulation [7] and have been pre-
viously prepared by different methods [7–9]. The acid-catalyzed elimination of 4-hy-
1
)
)
For the distribution of a- and b-ionone in nature, see [1a,b]; for g-ionone see [1c,d].
For the distribution of 4, 5, and 6 in nature, see [6a,b], [6c], and [6d,e], respectively.
2
© 2006 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta – Vol. 89 (2006)
1111
droxy-b-ionone [8c] and the selective reduction of 4 [9] are the most regioselective
paths to 4 and 5, respectively. Moreover, the bromination followed by debromination
of a-ionone (1) [8a] and of dihydro-a-ionone [7], respectively affords the above-men-
tioned compounds contaminated with other regioisomers. For example, 3,4-didehydro-
7,8-dihydro-g-ionone (7) is also produced in the latter process as a by-product. Other-
wise, to the best of our knowledge, no specific preparation of 6 and 7 have been
reported until now. Therefore, the selective preparation of the enantiomeric forms of
6 and 7, as well as the comprehensive olfactory evaluation of all the isomers of com-
pounds 4–7, are still lacking. Based on these observations, it seemed desirable to
develop a new synthetic method to the enantiomerically pure forms of 6 and 7 by
means of a procedure not necessarily involving troublesome isomer separations or
the use of expensive enantiomerically pure starting materials.
2. Results and Discussion. – Accordingly, we prepared compounds 4–7 in racemic
and enantiomer-enriched form starting from commercially available racemic a-ionone
(1). The regioselective base-mediated isomerization of 4,5-epoxy-4,5-dihydro-a-ionone
(8a/8b) followed by elimination of the obtained allylic alcohols were the key steps of
our divergent syntheses. Isomerization of 8a/8b catalyzed by a nucleophilic base [10]
gave access to 4 and 5 (Scheme 1), and the reaction of 8a/8b with a stoichiometric
amount of a strong, non-nucleophilic base furnished 4-hydroxy-g-ionone (10a/10b) as
a key intermediate for the synthesis of 6 and 7 and of their enantiomers (Schemes 2
and 3). The latter were obtained by the lipase-mediated resolution of 10a/10b yielding
enantiomerically and diastereoisomerically pure (+)-4-(acetyloxy)-g-ionone ((+)-12b)
and enantiomer-enriched (+)-10a. The configuration of the stereoisomers of 6 and 7
was assigned unambiguously, by chemical correlation of (+)-12b with (+)-g-ionone
((+)-3) and (ꢀ)-a-ionone ((ꢀ1), both of known absolute configuration [5][11]
(Scheme 3).

1112
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 1
i) MCPBA, CH2Cl2; 8a/8b 5 :1. ii) NaOMe (cat.), MeOH. iii) TsOH (cat.), toluene, reflux. iv) Bu3-
SnH, [PdCl2(PPh3)2] (cat.), NH4Cl, H2O, THF. v) LDA, THF, reflux; 80% of 10a/10b 4 :1, 6% of 11.
ACHTREUNG
vi) 4-N-nitrobenzoyl chloride, pyridine vii) Two crystallizations from MeOH. viii) KOH, MeOH.
2.1. 3,4-Didehydro-b-ionone Isomers 4 and 5. The b-ionone derivatives 4 and 5 were
obtained from a-ionone (1) by a slight modification of a previously reported synthesis.
The bromination/debromination method proved to be efficient but gave regioisomeri-
cally impure products. Thus a two-step process was considered. Epoxidation of 1 with 3-
chloroperbenzoic acid (MCPBA) in CH2Cl2 [12] gave an unseparable 5 :1 mixture of
the two diastereoisomers 8a and 8b in 75% yield (Scheme 1). Treatment of 8a/8b
with a catalytic amount of NaOMe in MeOH [10a,b] yielded crude 4-hydroxy-b-ionone
which was heated in toluene in the presence of a catalytic amount of TsOH [8c] to
afford, after chromatographic separation, pure 4. The regiospecific reduction of the
conjugated C(7)=C(8) bond of 4 was achieved by Pd-catalyzed hydride addition
[5][9a] in THF/H2O/NH4Cl on treatment with an excess of tributylstannane in the pres-
ence of cat. [PdCl2(PPh3)2] (5 mol-%), furnishing isomer-free 5 in 86% yield.
2.2. 3,4-Didehydro-g-ionone Isomers 6 and 7. Although the base-mediated isomer-
ization of epoxides has been extensively used in organic synthesis [13], its application to
the isomerization of 8a/8b is an intriguing synthetic step. Indeed, the necessary use of
harsh condition and the presence of the unsaturated ketone functionality could give rise

Helvetica Chimica Acta – Vol. 89 (2006)
1113
to different regioisomers and/or by-products. However, we found that under carefully
controlled conditions, an efficient isomerization occurs. Thus, the addition of 8a/8b to a
cooled (ꢀ788) solution of lithium diisopropylamide (LDA; 2.2 equiv.) in THF selec-
tively converted the ketone function to the corresponding lithium enolate without
affecting the epoxide group (Scheme 1). To achieve isomerization, the enolate solution
was heated under reflux whereupon the scarcely soluble salt 9 precipitated with con-
comitant disappearance of 8a/8b. After completion of the reaction, the heterogeneous
mixture was added to a cooled diluted HCl solution and rapidly extracted, yielding 80%
of the diasteroisomeric 4-hydroxy-g-ionones 10a/10b as an unseparable 4 :1 mixture,
besides a small amount of the bicyclic ether 11, the latter conceivably deriving from
10a by 1,4-addition of the OH group on the conjugated double bond. Different modi-
fications of the experimental conditions gave inferior results. For example, direct addi-
tion of 8a/8b to a heated LDA solution or basic workup conditions increased the
amount of polymerized by-products or the formation of ether 11, respectively. The
structure of 11 was confirmed by comparison of its NMR data with those reported
by Kaiser and Lamparsky [14] for an identical by-products obtained on Al(iPrO)3 treat-
ment of 8a/8b. Similarly, the diastereoisomers 10a/10b corresponded to the 4-hydroxy-
g-ionone diastereoisomers previously prepared by Griesbeck et al. [15] by photooxyge-
nation of a-ionone (1). Moreover, the transformation of 10a/10b to the corresponding
4-nitrobenzoates followed by crystallization from MeOH afforded diastereoisomeri-
cally pure 12a. All attempts to hydrolyze the ester function of 12a under basic condi-
tions gave mainly ether 11 and only a minor amount of 10a, demonstrating that 4-hy-
droxy-g-ionone is labile on basic treatment; strongly acidic conditions catalyzed both
nonregiospecific elimination and ring closure to 11.
Aimed at finding a selective path to 3,4-didehydro-g-ionone (6) we investigated the
elimination reaction of compound 10a/10b and of its derivatives (Scheme 2). POCl3-
Mediated dehydration of 10a/10b was highly regioselective the didehydroionones
obtained consisting of up to 99% of 6, although not efficient, the yield being only
25–30%, i.e, treatment of 10a/10b in pyridine at 08 with an excess of POCl3 afforded
pure 6 (27% yield) besides a mixture of unseparable chloro derivatives 13, 14a, and
14b (43% yield). Compound 13 was the main product, most likely formed by SN2’ addi-
tion of the Clꢀ ion to the intermediate allyl phosphate, and the diastereoisomers 14a
and 14b [16] arose from the same phosphate by substitution and, under the mild exper-
imental conditions described above, did not undergo elimination.
Subsequently, the stereospecific Pd-mediated elimination of allylic acetate was
tested with the esters of 10a and 10b. According to the Hauser [17] procedure, the ace-
tate 12b or the diastereoisomerically pure 4-nitrobenzoate 12a was heated under reflux
in dioxane in the presence of an excess of CaCO3 and of a catalytic amount of Pd(OAc)2
and PPh3 (Scheme 2). Both substrates gave the same product distribution: 3,4-didehy-
dro-g-ionone (6) was isolated in good yield (78 or 71%, resp.) and was formed regiose-
lectively, the didehydroionones obtained consisting of up to 98% of 6, whereas enol-
ether 15 was the only by-product (18 and 19%, resp.). The latter compound was prob-
ably formed by Pd-catalyzed SN2’ addition of the ketone enolate on the allylic acetate.
This hypothesis was supported by a further experiment in which pure 6 was left unaf-
fected after heating in the presence of CaCO3, Pd(OAc)2, and PPh3. The structure of 15
was corroborated by its spectroscopic data and by chemical derivatization. Indeed,

1114
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 2
i) POCl3, pyridine, CH2Cl2. ii) Ac2O, pyridine. iii) CaCO3, Pd(OAc)2, PPh3, dioxane. iv) Bu3
[PdCl2(PPh3)2] (cat.), NH4Cl, H2O, THF. v) H2, Pd/BaCO3 (cat.), AcOEt.
A
hydrogenation of 15 with Pd/BaCO3 as catalyst afforded saturated ether 16 as a mixture
of four stereoisomers and a small amount of the known tetrahydroionone 17, perhaps
formed from 15 by opening of the seven-membered ring. Finally, the reduction of the
conjugated C(7)=C(8) bond of 6 was accomplished by the same method as that used
for the preparation of 5 from 4, yielding 3,4-didehydro-7,8-dihydro-g-ionone (7)
from 6 in 91% yield.
2.3. Enantiomer-Enriched 3,4-Didehydro-g-ionone Isomers (+)- and (ꢀ)-6 and (+)-
and (ꢀ)-7. Taking advantage of the above-described study, we devised a method for the
preparation of all the enantiomeric forms of 3,4-didehydro-g-ionones 6 and 7. Recently,
we prepared the enantiomers of ionone [3] and irone [18] isomers through the enantio-
selective lipase-mediated resolution of the related ionols and irols. We now extended
this enzymic procedure to the resolution of 4-hydroxy-g-ionone (10a/10b). A prelimi-
nary experiment showed that acetylation of racemic 10a/10b was efficiently catalyzed
by lipase PS (Pseudomonas cepacia) with very high diastereoselectivity and good enan-
tioselectivity. These results are noteworthy since the same enzyme catalyzes the acety-
lation of ionol and irol isomers with complete enantioselectivity and very low (or none)
diastereoselectivity. Therefore, we treated racemic 10a/10b with vinyl acetate in
tBuOMe solution in presence of lipase PS and obtained diastereomerically and enantio-
merically pure (99% ee and de) acetate (+)-12b and a mixture of enantiomer-enriched
(+)-10a and racemic (ꢁ)-10b (Scheme 3). The acetylation was interrupted after ca. 30%
of conversion since a longer reaction time and conversion near to 50% gave (+)-12b
with lower ee although with very high de. The recovered mixture (+)-10a/(ꢁ)-10b
was submitted again to enzymic acetylation to afford (+)-12b with lower ee and a mix-
ture of alcohols which contained a higher amount of (ꢁ)-10b and (+)-10a with

Helvetica Chimica Acta – Vol. 89 (2006)
1115
increased optical purity. Hence, with the methods previously described, we used the lat-
ter set of enantiomerically enriched derivatives to prepare the enantiomers of 6 and 7.
Effectively, Pd-mediated elimination of the acetate group of enantiomerically pure
(+)-12b gave the suitable (+)-6, which by C(7)=C(8) bond reduction, afforded (+)-
7. In the same way, (+)-10a/(ꢁ)-10b was acetylated by Ac2O and pyridine and then sub-
mitted to the above described reaction sequence. Since the elimination step proceeded
without diastereoselectivity, (ꢀ)-6 and (ꢀ)-7 were obtained in low enantiomeric purity
as the result of the relative ratio of racemic (ꢁ)-10b and of the optical purity of (+)-10a.
On the other hand, the POCl3-mediated elimination reaction showed a considerably
diastereoselectivity. Noteworthy, samples of (+)-10a/(ꢁ)-10b with increasing ee and
decreasing de gave samples of (ꢀ)-6 with increasing ee. In that order, we also observed
a change in the relative ratio of the chloro derivative by-products with improving of
compound 13 and 14a. This behavior was probably due to the relative stability of the
allyl phosphate of (ꢁ)-10b. It is reasonable that the latter intermediate will rearrange
to 13 and 14a more easily than the phosphate of (+)-10a, hence rising the optical purity
of (ꢀ)-6. Consequently, we used a sample of (+)-10a/(ꢁ)-10b with good ee and very low
de in the latter elimination reaction to afford (ꢀ)-6 and (ꢀ)-7 in reasonable ee.
2.4. Olfactory Evaluation of 3,4-Didehydroionone Isomers. All the isomers synthe-
sized as described above were submitted to olfactory evaluation (Givaudan Schweiz
AG, Fragrance Research). The results are described in the Table and allow some
remarkable considerations: a) all the isomeric forms show distinct olfactory features
although between enantiomers, the differences are less evident. b) 3,4-didehydro-g-ion-
one isomers are more powerful than 3,4-didehydro-7,8-dihydro-g-ionone isomers. The
(+)-6(S) is more than thousand times more powerful than (+)-7(S), whilst (ꢀ)-6(R) is
about four hundred times more powerful than (ꢀ)-7(R). c) The 3,4-didehydro-7,8-dihy-
dro-g-ionone enantiomers show the same odor threshold, whereas 3,4-didehydro-g-
ionone enantiomers are of different intensity. (+)-6(S) is more than three time more
powerful than (ꢀ)-6(R). Since (ꢀ)-6(R) was obtained with a moderate ee (70%), this
difference may be more pronounced.
2.5. Determination of the Absolute Configuration of the 3,4-Didehydro-g-ionone
Isomers. As mentioned, the absolute configuration of the enantiomeric forms of 6
and 7 are not known. Therefore, we decided to assign these configurations by chemical
correlation. Since we were able to determine accurately the ee of compound (+)-12b,
we converted it to the g-ionone 3 of known absolute configuration [5][11c]. The reduc-
tive removal of the acetyloxy function in the presence of the unsaturated ketone group
was a challenging synthetic steps. We found that the Pd-catalyzed reduction of allylic
acetate [19] by mean of triethylammonium formate as reducing agent was the proce-
dure of choice for this transformation. Accordingly, treatment of enantiomerically
pure (+)-12b with a catalytic quantity of [PdCl2(PPh3)2], and an excess of an equimolar
amount of Et3N and HCOOH under reflux in THF gave enantiomerically pure (+)-3
A
(99% ee) in good yield (87%). GC and chiral GC analysis confirmed that the g-isomer
was formed regioselectively (g/a/b isomer ratio 97:2.5 :0.5) without detectable racemi-
zation. Judging from the measured optical rotation [a]2D0 =+25.2 (c=2, CHCl3) (vs.
[a]2D0 =+36.2 (c=1, CHCl3) [5] for pure (S)-isomer (+)-3), the impurity, i.e., the a-iso-
mer should be optically active ([a]2D0 =ꢀ418 (c=1, CHCl3) [11a] for pure (S)-a-ionone
((ꢀ)-1)). To confirm this assumption, we treated a sample of the g,a,b mixture

1116
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 3
i) Vinyl acetate, t-BuOMe, lipase PS. ii) CaCO3, Pd(OAc)2, PPh3, dioxane. iii) Bu3
[PdCl2(PPh3)2] (cat.), NH4Cl, H2O, THF. iv) Ac2O, pyridine. v) POCl3, pyridine, CH2Cl2. vi) HCO2H,
Et3N, PPh3, [PdCl2(PPh3)2] (cat.), THF, reflux. vii) 85% H3PO4.
SnH,
AHCTREUNG
(97:2.5 :0.5) with H3PO4. By this means we achieved complete isomerization of the g-
isomer (+)-3 to a- and b-ionone. The latter mixture, (ꢀ)-1/2 (a/b isomer ratio 64 :36)
showed an optical rotation ([a]2D0 =ꢀ251.2 (c=2, CHCl3), in good accord to that of
enantiomerically pure a-ionone ((ꢀ)-1). In conclusion, the absolute configurations of
(+)-6 and (+)-7, and of (ꢀ)-6 and (ꢀ)-7 were assign unambiguously as (S) and (R),
respectively.
3. Conclusions. – A new stereospecific approach to the isomeric forms of 3,4-dide-
hydroionone isomers is described. The 3,4-didehydro-b-ionone isomers 4 and 5 and the
3,4-didehydro-g-ionone isomers 6 and 7 were prepared starting from commercially
available a-ionone (1) in a few regioselective steps. Enantiomer-enriched (+)-6 and
(+)-7, and (ꢀ)-6, and (ꢀ)-7 were prepared by means of diastereoselective and enantio-
selective lipase-mediated acetylation of the racemic intermediate 4-hydroxy-g-ionone
10a/10b followed by a number of chemoselective and regioselective reactions. The

Helvetica Chimica Acta – Vol. 89 (2006)
Table. Olfactory Evaluation of 3,4-Didehydro-b- and g-ionone Isomers
Olfactory evaluation
1117
Ionone isomer
Odor
threshold
[ng/l of air]
3,4-Didehydro-b-ionone (4)
ionone-damascone and saffron-like note, with
fruity and slightly leathery aspects
stronger than (ꢀ)-(R)-isomer on blotter,
floral-woody, cetonal-like, slightly fruity,
damasconic, sweet
(S)-3,4-Didehydro-g-ionone ((+)-6)
0.080
0.25
92
(R)-3,4-Didehydro-g-ionone ((ꢀ)-6)
weaker than (+)-(S)-isomer, fruity-floral,
slightly woody
3,4-Didehydro-7,8-dihydro-b-ionone (5) b-ionone-like, leathery, woody-floral
(S)-3,4-Didehydro-7,8-dihydro-g-ionone stronger on blotter than (ꢀ)-(R)-isomer,
((+)-7)
woody-powdery, ionone, slightly fruity-floral,
more pronounced ionone character than
(ꢀ)-(R)-isomer.
(R)-3,4-Didehydro-7,8-dihydro-g-ionone weaker on blotter than (+)-(S)-isomer, fruity, 99.4
((ꢀ)-7)
agrestic, floral, ionone- and damascone-like,
slightly woody, and saffron-like
reduction of acetate (+)-12b to (+)-(S)-g-ionone ((+)-3) and the following isomeriza-
tion to (ꢀ)-(S)-a-ionone ((ꢀ)-1) allowing to assign (S)-configuration to (+)-6 and (+)-
7. The latter process can also be seen as a new enantioselective path to enantiomerically
pure (+)-(S)-g-ionone ((+)-3. Moreover, the resolution step confirms the utility of the
enzymatic approach to the preparation of enantiomer-enriched norterpenoid odorants.
Olfactory evaluation of the above described 3,4-didehydroionone isomers shows a sig-
nificant difference between the enantiomers and regioisomers both in fragrance feature
and in detection threshold.
The authors would like to thank Dr. Philip Kraft, Givaudan Schweiz AG, Fragrance Research,
Dübendorf, Switzerland, for the olfactory descriptions and for the threshold determinations. We thank
COFIN-MURST for partial financial support.
Experimental Part
1. General. All moisture-sensitive reactions were carried out under a static atmosphere of N2. All
reagents were of commercial quality. Lipase PS from Pseudomonas cepacia (Amano Pharmaceuticals
Co., Japan; 30 U mgꢀ1) was employed in this work. TLC: Merk silica gel 60 F254 plates. Column chroma-
tography (CC): silica gel. Gas chromatography (GC): HP-6890 gas chromatograph; HP-5 column
(30 m×0.32 mm; Hewlett Packard) with the following temp. program; 608 (1 min) – 68/min – 1508
(1 min) – 128/min – 2808 (5 min); tR in min. Chiral GC: DANI-HT-86.10 gas chromatograph; enantiomer
excesses determined on a Chirasil-DEX-CB column with the following temp. program; 608 (3 min) – 38/
min – 1808 (5 min); tR in min. M.p.: Reichert apparatus, equipped with a Reichert microscope; uncor-
rected. Optical rotations: Jasco-DIP-181 digital polarimeter. IR Spectra: Perkin-Elmer-2000-FT-IR spec-
trometer; films; n in cmꢀ1. 1H- and 13C-NMR Spectra: CDCl3 solns. at r.t.; Bruker-AC-400 spectrometer at
~
400 MHz; chemical shifts d in ppm rel. to internal SiMe4 (=0 ppm), J values in Hz. Mass spectra: Finni-
gan-Mat-TSQ-70 spectrometer; m/z (rel.%). Microanalyses were determined on an analyzer 1106 from
Carlo Erba.

1118
Helvetica Chimica Acta – Vol. 89 (2006)
2. 3,4-Didehydro-b-ionone Isomers. 2.1. 3,4-Didehydro-b-ionone (4; =(3E)-4-(2,6,6-Trimethylcyclo-
hexa-1,3-dien-1-yl)but-3-en-2-one). Epoxidation of racemic a-ionone (1) with 3-chloroperbenzoic acid
[12a] afforded 4,5-epoxy-4,5-dihydro-a-ionone (8a/8b 5 :1; 75%) (NMR analysis). According to [10a],
8a/8b was treated with NaOMe in dry MeOH. The obtained crude 4-hydroxy-b-ionone was then treated
with TsOH [8c] to afford pure 4 (72%; 96% pure by GC, tR 16.7). IR, 1H-NMR, MS: in accordance with
those reported [9b].
2.2. 3,4-Didehydro-7,8-dihydro-b-ionone (=4-(2,6,6-Trimethylcyclohexa-1,3-dien-1-yl)butan-2-one;
5). Bu
[PdCl2(PPh3)2] (140 mg, 0.2 mmol), NH4Cl (500 mg, 9.4 mmol), and H2O (0.2 ml, 11.2 mmol) in THF
(50 ml) under N2. After 3 h, Et2O (150 ml) was added and the soln. was washed with brine (50 ml) and
A
AHCTREUNG
evaporated. The residue was dissolved in AcOEt (80 ml) and stirred with a sat. NaF soln. (25 ml) for
5 h. The precipitate formed was filtered and the filtrate evaporated. The residue was purified by CC (hex-
ane/AcOEt 95 :5) and bulb-to-bulb distillation: pure 5 (700 mg, 86%; 95% pure by GC, tR 15.1). IR, 1H-
NMR, MS: in accordance with those reported [9b].
3. Racemic 3,4-Didehydro-g-ionone Isomers. 3.1. cis/trans-4-Hydroxy-g-ionone (=(3E)-4-(cis/trans-
5-Hydroxy-2,2-dimethyl-6-methylenecyclohexyl)but-3-en-2-one; 10a/10b). At ꢀ788, 10M BuLi in hexane
i
(55 ml) was added dropwise to Pr2NH (58 g, 573 mol) in dry THF (300 ml) under N2. The mixture was
stirred at ꢀ788 for 30 min, then a soln. of 8a/8b (52 g, 0.25 mol) in dry THF (60 ml) was added dropwise.
The mixture was gradually warmed to r.t. (1 h) and then heated under reflux for 3 h (TLC monitoring).
After cooling to r.t., the mixture was poured into a mixture of crushed ice and 5% HCl soln. (850 ml) and
extracted with Et2O (3×200 ml). The org. phase was successively washed with sat. aq. NH4Cl soln. (100
A
ml) and brine, dried (Na2SO4), and evaporated. The residue was purified by CC (hexane/AcOEt 9 :1 !
hexane/AcOEt 1:1): 11 (3.1 g, 6%) followed by 10a/10b (41.5 g, 80%).
Data of 1-(2,2-Dimethyl-8-methylene-6-oxabicyclo[3.2.1]oct-7-yl)propan-2-one (11): 98% pure by
1
GC, tR 15.4. Colorless oil. IR: 1716, 1366, 1159, 1059, 1013, 965, 892. H-NMR: 4.97 (s, 1 HꢀC=C(8));
4.78 (s, 1 HꢀC=C(8)); 4.58 (t, J=6.7, HꢀC(7)); 4.31 (d, J=4.5, HꢀC(5)); 2.69–2.60 (dd, J=16.1, 6.7,
1 H, CH2
G
COMe); 2.14 (s, MeCO); 1.86 (s, HꢀC(1));
1.81–1.73 (m, 1 H), 1.73–1.62 (m, 1 H), 1.60–1.49 (m, 1 H), 1.20 (dd, J=13.5, 5.5, 1 H) (CH2(3)
CH2(4)); 1.03 (s, 1 MeꢀC(2)); 0.94 (s, 1 MeꢀC(2)). 13C-NMR: 206.9; 149.2; 104.2; 77.7; 74.4; 57.8;
49.8; 35.3; 31.7; 31.4; 30.6; 28.3; 26.2. MS: 208 (1, M+), 193 (5), 175 (2), 165 (6), 157 (1), 150 (30), 147
(10), 135 (20), 122 (10), 119 (9), 107 (36), 95 (20), 91 (18), 81 (18), 79 (19), 69 (13), 67 (11), 55 (17), 53
(12), 43 (100). Anal. calc. for C13H20O2: C 74.96, H 9.68; found: C 75.05, H 9.70.
Data of 10a/10b: 96% pure by GC, tR 20.8 (10a) and tR 20.5 (10b), deduced from the mixture. IR:
1
3428, 1672, 1626, 1366, 1253, 1131, 1069, 1047, 998, 902. H-NMR: cis isomer 10a: 6.93 (dd, J=15.8,
10.3, HꢀC(7)); 6.08 (d, J=15.8, HꢀC(8)); 5.17 (br. s, 1 HꢀC=C(5)); 4.69 (br. s, 1 HꢀC=C(5));
4.10–4.02 (m, HꢀC(4)); 2.56 (d, J=10.3, HꢀC(6)); 2.28 (s, Me(10)); 2.01–1.92 (m, 1 H), 1.64–1.50
(m, 2 H), 1.49–1.39 (m, 1 H) (CH2(2), CH2(3)); 0.89 (s, 1 MeꢀC(1)); 0.88 (s, 1 MeꢀC(1)); trans-isomer
10b: 6.89 (dd, J=15.8, 9.9, HꢀC(7)); 6.14 (dd, J=15.8, 0.6, HꢀC(8)); 5.08–5.06 (m, 1 HꢀC=C(5)); 4.72
(br. s, 1 HꢀC=C(5)); 4.32–4.26 (m, HꢀC(4)); 2.98 (d, J=9.9, HꢀC(6)); 2.26 (s, Me(10)); 1.91–1.81 (m, 1
H), 1.74–1.57 (m, 2 H), 1.49–1.39 (m, 1 H) (CH2(2), CH2(3)); 0.94 (s, 1 MeꢀC(1)); 0.88 (s, 1 MeꢀC(1)).
MS: 208 (3, M+), 193 (8), 190 (4), 175 (12), 165 (100), 157 (3), 150 (39), 147 (78), 135 (29), 122 (43), 109
(40), 107 (69), 105 (43), 95 (58), 91 (59), 81 (39), 79 (41), 69 (41), 67 (32), 55 (36). Anal. calc. for C13H20O2:
C 74.96, H 9.68; found: C 75.10, H 9.65.
cis-4-[(4-Nitrobenzoyl)oxy]-g-ionone (=(3E)-4-{cis-5-[4-Nitrobenzoyl)oxy]-2,2-dimethyl-6-methyl-
enecyclohexyl)but-3-en-2-one; 12a). A sample of 10a/10b (7 g, 33.6 mmol) in CH2Cl2 (30 ml) was treated
with 4-nitrobenzoyl chloride (7.6 g, 41 mmol) and pyridine (20 ml). After workup, the crude product was
purified by CC (hexane/AcOEt 8 :2): 12a (11 g, 92%). Two crystallizations from MeOH afforded crystal-
line 12a (4.9 g, yield of recrystallizations 41%; up to 98% pure by NMR). M.p. 101–1028. IR: 1724, 1671,
1605, 1526, 1357, 1278, 1104, 990, 904, 722. 1H-NMR: 8.34–8.24 (m, 4 arom. H); 6.97 (dd, J=15.6, 10.1,
HꢀC(7)); 6.16 (d, J=15.6, HꢀC(8)); 5.54–5.46 (m, HꢀC(4)); 5.12 (s, 1 HꢀC=C(5)); 4.81 (s, 1 HꢀC=
C(5)); 2.72 (d, J=10.1, HꢀC(6)); 2.23 (s, Me(10)); 2.12–2.01 (m, 1 H), 1.94–1.82 (m, 1 H), 1.79–1.69
(m, 1 H), 1.60–1.49 (m, 1 H) (CH2(2), CH2(3)); 0.96 (s, 2 MeꢀC(1)). 13C-NMR: 197.4; 163.5; 150.6;

Helvetica Chimica Acta – Vol. 89 (2006)
1119
144.7; 144.7, 135.5; 132.6; 130.7; 123.5; 110.2; 75.1; 55.8; 36.1; 35.2; 28.9; 28.5; 27.8; 22.7. MS: 357 (18,
M+), 342 (10), 207 (30), 190 (69), 175 (19), 149 (36), 147 (100), 134 (36), 121 (17), 105 (11), 91 (5), 69
(3), 43 (11). Anal. calc. for C20H23NO5: C 67.21, H 6.49; found: C 67.35, H 6.50.
Alkaline Hydrolysis of 12a. A soln. of 12a (1 g, 2.8 mmol) in MeOH (20 ml) was treated with NaOH
(1 g, 18 mmol) under reflux for 1 h. After workup, the crude product was purified by CC (hexane/AcOEt
9 :1 ! hexane/AcOEt 1:1): 11 (350 mg, 60%) and cis-isomer 10a (150 mg, 26%). IR, 1H-NMR, MS: in
accordance with those reported above.
3.2. Pd-Mediated Elimination of 10a/10b via 12b to 3,4-Didehydro-g-ionone (=(3E)-4-(2,2-Dimethyl-
6-methylenecyclohex-4-en-1-yl)but-3-en-2-one; 6) and 1,6,7,8-Tetrahydro-3,6,6-trimethyl-2-benzoxepine;
15). A sample of 10a/10b (3 g, 14.4 mmol) was converted to the corresponding acetate by treatment
with pyridine (20 ml) and Ac2O (20 ml) at r.t. for 24 h. The crude 12b was dissolved in dioxane (30
ml) and treated with Pd(OAc)2 (50 mg, 0.2 mmol), CaCO3 (2 g, 20 mmol), and PPh3 (0.53 g, 2 mmol).
The resulting heterogeneous mixture was stirred under reflux under N2 for 4 h (TLC monitoring).
Then, the mixture was cooled to r.t., diluted with Et
cessively with sat. aq. NaHCO3 soln. (50 ml) and brine, dried (Na2SO4), and evaporated and the residue
O 8 :2). The first-eluted fractions afforded, after bulb-to-bulb dis-
2O (100 ml), and filtered. The filtrate was washed suc-
purified by CC (hexane ! hexane/Et2
ACHTREUNG
tillation (oven temp. 95–1008/0.05 Torr), pure 15 (0.49 g, 18%; 99% pure by GC, tR 16.4). The last-eluted
fractions gave, after bulb-to-bulb distillation (oven temp. 100–1058/0.05 Torr), 6 (2.14 g, 78%; 98% pure
by GC, tR 15.3).
Data of 15: Colorless oil. IR: 1648, 1626, 1589, 1446, 1394, 1361, 1331, 1214, 1196, 1054, 1029, 943,
848, 810. 1H-NMR: 5.59 (br. d, J=8.6, HꢀC(5)); 5.56 (br. t, J=4.0, HꢀC(9)); 4.93 (dd, J=8.6, 0.7,
HꢀC(4)); 4.41 (d, J=0.7, CH2(1)); 2.25–2.18 (m, CH2(8)); 1.87 (s, MeꢀC(3)); 1.53 (t, J=6.2,
CH2(7)); 1.04 (s, 2 MeꢀC(6)). 13C-NMR: 158.6; 143.8; 133.8; 128.9; 116.3; 101.2; 74.4; 36.8; 34.2; 27.2;
27.2; 23.2; 21.6. MS: 191 (15, [M+1]+), 190 (100, M+), 175 (38), 157 (11), 147 (40), 134 (89), 119 (83),
117 (22), 115 (22), 105 (48), 91 (49), 79 (24), 77 (23), 65 (14), 55 (13), 43 (87). Anal. calc. for C13H18O:
C 82.06, H 9.53; found: C 81.95, H 9.50.
Data of 6: Colorless oil. IR: 1674, 1620, 1426, 1366, 1253, 990, 890, 733. 1H-NMR: 6.70 (dd, J=15.8,
9.2, HꢀC(7)); 6.16 (dt, J=9.9, 1.8, HꢀC(4)); 6.09 (d, J=15.8, HꢀC(8)); 5.74 (dt, J=9.9, 3.6, HꢀC(3));
4.95 (s, 1 HꢀC=C(5)); 4.80 (s, 1 HꢀC=C(5)); 2.72 (d, J=9.2, HꢀC(6)); 2.24 (s, Me(10)); 2.09–1.88 (m,
CH2(2)); 0.93 (s, 2 MeꢀC(1)). 13C-NMR: 198.0; 147.0; 143.2; 132.2; 127.8; 127.5; 114.0; 54.8; 37.6; 33.1;
27.9; 26.9; 25.9. MS: 191 (1, [M+1]+), 190 (8, M+), 175 (9), 157 (8), 147 (32), 131 (20), 119 (19), 105 (48),
91 (42), 79 (16), 77 (21), 65 (15), 55 (15), 43 (100). Anal. calc. for C13H18O: C 82.06, H 9.53; found: C 82.15,
H 9.50.
3.3. Pd-Mediated Elimination of 4-Nitrobenzoate 12a to 6 and 15. As described in Exper. 3.2 with 12a
(1.1 g, 3.1 mmol), Pd(AcO)2 (25 mg, 0.1 mmol), CaCO3 (0.5 g, 5 mmol), and PPh3 (0.26 g, 1 mmol): 15
(0.11 g, 19%; 98% pure by GC) and 6 (0.42 g, 71%; 97% pure by GC).
3.4. POCl3-Mediated Elimination of 10a/10b to 6 and 13/14a/14b. A soln. of POCl3 (2.6 ml, 27.9
mmol) in CH2Cl2 (10 ml) was added dropwise to a cooled (08) soln. of 10a/10b (3 g, 14.4 mmol) in
CH2Cl2 (20 ml) and pyridine (20 ml). The mixture was stirred for 3 h, then diluted with Et2
A
and added to a mixture of crushed ice and sat. aq. NaHCO3 soln. (50 ml). The org. phase was washed
with brine, dried (Na2SO4), and evaporated and the residue purified by CC (hexane ! hexane/Et2
C
Data of 5-methyl-Chloro-a-ionone (=(3E)-4-[2-(Chloromethyl)-6,6-dimethylcyclohex-2-enyl)but-3-
en-2-one; 13): 1H-NMR: 6.61 (dd, J=15.8, 9.6, HꢀC(7)); 6.11 (d, J=15.8, HꢀC(8)); 5.94 (br. t, J=3.7,
HꢀC(4)); 3.95–3.87 (m, CH2Cl); 2.74 (d, J=9.6, HꢀC(6)); 2.25 (s, Me(10)); 2.22–1.87 (m), 1.55–1.22
(m) (CH2(2), CH2(3)); 0.96 (s, 1 MeꢀC(1)); 0.91 (s, 1 MeꢀC(1)).
Data of cis-4-Chloro-g-ionone (=(3E)-4-(cis-5-Chloro-2,2,-dimethyl-6-methylenecyclohexyl)but-3-
1
en-2-one; 14a): H-NMR: 7.03 (dd, J=15.8, 10.3, HꢀC(7)); 6.07 (d, J=15.8, HꢀC(8)); 5.36 (s, 1 Hꢀ

1120
Helvetica Chimica Acta – Vol. 89 (2006)
C=C(5)); 4.85 (s, 1 HꢀC=C(5)); 4.48 (dd, J=8.9, 4.8, HꢀC(4)); 2.64 (d, J=10.3, HꢀC(6)); 2.28 (s,
Me(10)); 2.22–1.87 (m), 1.74 (ddd, J=14.0, 6.5, 4.1, 1 H), 1.55–1.22 (m) (CH2(2), CH2(3)); 0.90 (s, 2
MeꢀC(1)).
Data of trans-4-Chloro-g-ionone (=(3E)-4-(trans-5-Chloro-2,2-dimethyl-6-methylenecyclohexyl)-
1
but-3-en-2-one; 14b): H-NMR: 6.81 (dd, J=15.8, 9.9, HꢀC(7)); 6.20 (d, J=15.8, HꢀC(8)); 5.16 (br. s,
1 HꢀC=C(5)); 4.77 (br. t, J=3.5, HꢀC(4)); 4.71 (br. d, J=1.5, 1 HꢀC=C(5)); 3.15 (d, J=9.9, Hꢀ
C(6)); 2.27 (s, Me(10)); 2.22–1.87, 1.55–1.22 (2m, CH2(2), CH2(3)); 0.95 (s, 1 MeꢀC(1)); 0.86 (s, 1
MeꢀC(1)).
3.5. 3,4-Didehydro-7,8-dihydro-g-ionone (=4-(6,6-Dimethyl-2-methylenecyclohex-3-en-1-yl)butan-2-
one; 7). As described in Exper. 2.2, with 6 (0.9 g, 4.7 mmol), Bu3SnH (2.75 g, 9.4 mmol), [PdCl2(PPh3)2]
N
(140 mg, 0.2 mmol), NH4Cl (600 mg, 11.3 mmol), H2O (0.2 ml, 11.2 mmol), and THF (50 ml). Bulb-to-
bulb distillation (oven temp. 90–958/0.02 Torr) gave 7 (0.82 g, 91%; 98% pure by GC, tR 13.9). IR:
1719, 1638, 1599, 1430, 1364, 1273, 1161, 995, 884, 782. 1H-NMR: 5.99 (dd, J=9.8, 2.0, HꢀC(4));
5.67–5.61 (m, HꢀC(3)); 4.88 (s, 1 HꢀC=C(5)); 4.67 (s, 1 HꢀC=C(5)); 2.46 (ddd, J=17.2, 9.2, 5.4, 1
HꢀC(8)); 2.33 (ddd, J=17.2, 9.2, 6.6, 1 HꢀC(8)); 2.10 (s, Me(10)); 2.09 (br. d, J=17.5, HꢀC(6));
1.89–1.79 (m, 1 HꢀC(7)); 1.77–1.67 (m, CH2(2)); 1.31–1.18 (m, 1 HꢀC(7)); 1.00 (s, 1 MeꢀC(1)); 0.86
(s, 1 MeꢀC(1)). 13C-NMR: 209.1; 145.4; 128.0; 126.9; 112.7; 50.9; 42.1; 36.2; 32.7; 29.9; 28.1; 27.5; 21.9.
MS: 192 (1, M+), 174 (6), 159 (24), 149 (1), 144 (1), 134 (61), 121 (33), 119 (100), 107 (21), 105 (19), 93
(12), 91 (30), 79 (12), 77 (13), 71 (4), 65 (5), 55 (5), 43 (31). Anal. calc. for C13H20O: C 81.20, H 10.48;
found: C 81.15, H 10.50.
3.6. Hydrogenation of 15 to Decahydro-3,6,6-trimethyl-2-benzoxepine (16) and Tetrahydroionone
(=4-(2,2,6-trimethylcyclohexyl)butan-2-one; 17). A soln. of 15 (300 mg, 1.6 mmol) in AcOEt (50 ml)
was treated with 5% Pd/BaCO3 (70 mg) and hydrogenated at 1 atm. for 2 days. The catalyst was then fil-
tered off and the filtrate evaporated. The residue was purified by CC (hexane ! hexane/Et
C
Data of 16: IR: 1466, 1377, 1366, 1271, 1250, 1110. 1H-NMR: 3.96–3.36 (m, 3 H); 1.90–1.20 (m, 12
H); 1.16–0.72 (m, 9 H). MS: 196 (9, M+), 181 (71), 163 (38), 154 (24), 137 (34), 123 (23), 109 (100), 95 (49),
81 (58), 67 (45), 55 (30), 41 (33). Anal. calc. for C13H24O: C 79.53, H 12.32; found: C 79.40, H 12.30.
4. Enantiomer-Enriched 3,4-Didehydroionone Isomers. 4.1. Lipase-Mediated Resolution of cis/trans-
4-Hydroxy-g-ionone (10a/10b). A mixture of racemic 10a/10b (cis/trans 4 :1; 40 g, 192 mmol), lipase PS
(15 g), vinyl acetate (45 ml), and tBuOMe (250 ml) was stirred at r.t. for 5 days. After filtration and evap-
oration of the filtrate, the residue was subjected to CC (hexane/AcOEt 8 :2). The first-eluted fractions
afforded (+)-(4R,6S)-4-(acetyloxy)-g-ionone ((+)-12b; 14.6 g, 30%; 98% pure; 99% de (GC), tR 20.8).
Chiral GC: tR 32.5; ee 99%. [a]2D0 =+22.8 (c=3, CHCl3). IR: 1742, 1698, 1674, 1629, 1367, 1239, 1174,
1
1038, 994, 905. H-NMR: 6.92 (dd, J=15.8, 10.2, HꢀC(7)); 6.10 (d, J=15.6, HꢀC(8)); 5.25–5.14 (m,
HꢀC(4)); 5.02 (s, 1 HꢀC=C(5)); 4.71 (s, 1 HꢀC=C(5)); 2.63 (d, J=10.2, HꢀC(6)); 2.28 (s, Me(10));
2.11 (s, MeCOO); 1.99–1.82 (m, 1 H), 1.78–1.57 (m, 2 H), 1.56–1.38 (m, 1 H) (CH2(2), CH2(3)); 0.91
(s, 1 MeꢀC(1)); 0.89 (s, 1 MeꢀC(1)). 13C-NMR: 197.4; 169.5; 145.1; 145.0; 132.6; 109.1; 73.3; 55.7;
36.4; 35.1; 28.9; 28.5; 27.2; 22.2; 20.9. MS: 250 (1, M+), 235 (5), 208 (26), 190 (33), 175 (19), 165 (31),
157 (6), 149 (51), 147 (100), 134 (32), 121 (31), 105 (34), 91 (34), 79 (14), 69 (14), 55 (12). Anal. calc.
for C15H22O3: C 71.97, H 8.86; found: C 72.10, H 8.85.
The last-eluted fractions gave a mixture of (+)-(4S,6R)-4-hydroxy-g-ionone ((+)-10a) and (4RS,
6RS)-4-hydroxy-g-ionone ((ꢁ)-10b) (26.9 g, 67%; 98% pure; 55% de (GC)). Chiral GC (corresponding
acetate): tR 32.65; ee 70%. [a]2D0 =+4.9 (c=3, CHCl3). IR, 1H-NMR, MS: in accordance with that of rac-
emic 10a/10b.
Repeating the lipase mediated resolution with (+)-10a/(ꢁ)-10b (8 days of reaction) afforded further
(+)-12b (6.6 g; 98% pure; 99% de (GC); ee 76% (chiral GC); [a]2D0 =+16.9 (c=2, CHCl3)) and a mixture
of (+)-10a and (ꢁ)-10b (20.2 g; 98% pure; 35% de (GC); ee 79% (chiral GC (corresponding acetate);
[a]2D0 =+5.8 (c=3, CHCl3).
4.2. (+)-(6S)- and (ꢀ)-(6R)-3,4-Didehydro-g-ionone ((+)- and (ꢀ)-6, resp.) by Pd-Mediated Elimi-
nation. As described in Exper. 3.2, with (+)-12b (0.8 g, 3.2 mmol; 98% pure; 99% de (NMR), 99% ee

Helvetica Chimica Acta – Vol. 89 (2006)
1121
(chiral GC)) Pd(AcO)2 (25 mg, 0.1 mmol), CaCO3 (0.64 g, 6.4 mmol), and PPh3 (0.26 g, 1 mmol): 15 (0.1
g, 16%; 98% pure by GC) and (+)-6 (0.47 g, 77%; 97% pure by GC). (+)-6: [a]2D0 =+11.1 (c=2, CHCl3).
IR, 1H-NMR, MS: in accordance with that of (ꢁ)-6.
The same method was applied to the elimination of (+)-10a/(ꢁ)-10b (0.66 g, 3.2 mmol; 97% pure by
GC; 55% de (NMR); 79% ee (chiral GC)), after conversion to the corresponding acetate: 15 (0.11 g,
18%; 98% pure by GC) and (ꢀ)-6 (0.43 g, 71%; 98% pure by GC). (ꢀ)-6: [a]2D0 =ꢀ3.4 (c=2, CHCl3).
4.3. (ꢀ)-3,4-Didehydro-g-ionone ((ꢀ)-6) by POCl3-Mediated Elimination. As described in Exper. 3.4,
with (+)-10a/(ꢁ)-10b (1.5 g, 7.2 mmol; 96% pure by GC; 55% de (GC); 79% ee (chiral GC)): (ꢀ)-6 (0.33
g, 24%; 99% pure by GC); [a]2D0 =ꢀ5.7 (c=2, CHCl3)) and 13/14a/14b 2.5 :1:1 (0.65 g, 40%; 98% pure by
GC).
The same method was applied to a different sample of (+)-10a/(ꢁ)-10b (0.8 g, 3.8 mmol; 97% pure
by GC); 25% de (NMR); 90% ee (chiral GC)); (ꢀ)-6 (0.17 g, 23%; 98% pure by GC); [a]2D0 =ꢀ7.5 (c=2,
CHCl3)) and 13/14a/14b 3 :2 :1 (0.43 g, 50%; 98% pure by GC).
4.4. (+)-(6S)- and (ꢀ)-(6R)-3,4-Didehydro-7,8-dihydro-g-ionone ((+)- and (ꢀ)-7, resp.). As descri-
bed in Exper. 3.5, with (+)-6 (0.9 g, 4.7 mmol; 96% pure by GC; [a]2D0 =+11.1 (c=2, CHCl3)): (+)-7 (0.8
1
g, 89%; 97% pure by GC). [a]2D0 =+79.5 (c=2, CHCl3). IR, H-NMR, MS: in accordance with that of
(ꢁ)-7.
The same method was applied to (ꢀ)-6 (0.7 g, 3.7 mmol; 97% pure by GC; [a]2D0 =ꢀ7.5 (c=2,
CHCl3)): (ꢀ)-7 (0.62 g, 87%; 97% pure by GC). [a]2D0 =ꢀ55.5 (c=2, CHCl3). IR, 1H-NMR, MS: in
accordance with that of (ꢁ)-7.
5. Conversion of Acetate (+)-12b to (+)-(6S)-g-Ionone ((+)-3) and (ꢀ)-(6S)-a-Ionone ((ꢀ)-1. 5.1.
Reduction of (+)-12b. A soln. of (+)-12b (1.6 g, 6.4 mmol; 98% pure by GC; 99% ee (chiral GC)), formic
acid (0.6 g, 13 mmol), Et3
mmol) in THF (30 ml) was refluxed under N2 for 2 h (TLC monitoring). Then mixture was diluted with
Et2O (100 ml) and washed with H2O (50 ml), 5% HCl soln. (50 ml), sat. aq. NaHCO3 soln. (50 ml), and
brine. The org. phase was dried (Na2SO4) and evaporated and the residue purified by CC (hexane/Et2
N (1.4 g, 13.8 mmol), [PdCl2(PPh3)2] (140 mg, 0.2 mmol), and PPh3 (0.25 g, 0.9
AHCTREUNG
A
9 :1) and bulb-to-bulb distillation: ionone mixture (+)-3/(ꢀ)-1/2 (1.07 g, 87%; 98% pure; ratio g/a/b
97:2.5 :0.5; tR 15.2 (g), 15.1 (a), 16.3 (b)); [a]2D0 =+25.2 (c=2, CHCl3); 99% ee for the g isomer ((+)-
3) (chiral GC [5]). (+)-3: IR, 1H-NMR, MS: in accordance with that reported for enantiomerically
pure (+)-(6S)-g-ionone.
5.2. Isomerization of (+)-(6S)-g-Ionone ((+)-3) to (ꢀ)-(6S)-a-Ionone ((ꢀ)-1). A sample of the above
obtained (+)-3 (0.4 g, 2.1 mmol; containing a- and b-ionone) was stirred in 85% H3PO4 soln. (2 ml) at r.t.
for 2 h (GC monitoring). The mixture was then poured onto crushed ice and extracted with Et
A
ml). The org. phase was washed with sat. aq. NaHCO3 soln. (60 ml) and brine, dried (Na2SO4), and evapo-
rated, and the residue purified by CC (hexane/Et2O 9 :1) and bulb-to-bulb distillation to give a/b-ionone
G
((ꢀ)-1/2; (0.25 g, 62%; 98% pure; a/b 64 :36), [a]2D0 =ꢀ251.2 (c=2, CHCl3).
REFERENCES
[1] a) Y. R. Naues, Rivista Ital. E. P. P. O. S. 1976, 58, 505; b) J. R. Naves, Rivista Ital. E. P. P. O. S. 1977,
59, 495; c) Z. Liu, C. T. Liu, G. Fu, Y. Xu, S. Sun, Z. Chen, A. Li, Youji Huaxue 1982, 443 (Chem.
Abstr.. 1983, 98, 149449j); d) K. Awano, S. Ishizaki, O. Takazawa, T. Kitahara, Flavour Fragr. J. 2005,
20, 18.
[2] K. Bauer, D. Garbe, H. Surburg ‘Common Fragrance and Flavor Materials’ 4th edn., Wiley-VCH
Verlag GmbH, Weinheim, Germany, 2001.
[3] E. Brenna, C. Fuganti, S. Serra, P. Kraft, Eur. J. Org. Chem. 2002, 967.
[4] E. Brenna, C. Fuganti, S. Serra, Tetrahedron: Asymmetry 2003, 14, 1.
[5] C. Fuganti, S. Serra, A. Zenoni, Helv. Chim. Acta 2000, 83, 2761.
[6] a) R. Kaiser, J. Essent. Oil Res. 1991, 3, 129; b) S. Wang, E. L. Ghisalberti, J. Ridsdill-Smith, Phy-
tochemistry 1999, 52, 601; c) F. Peng, L. Sheng, B. Liu, H. Tong, S. Liu, J. Chromatogr., A 2004,
1040, 1; d) I. Flamant, G. Ohloff, ‘Developments in Food Science’, 10 (Prog. Flavour Res.) 1985,
281–300; e) R. ter Heide, H. Schaap, H. J. Wobben, P. J. De Valois, R. Timmer, in ‘The Quality

1122
Helvetica Chimica Acta – Vol. 89 (2006)
of Foods and Beverages: Chemical Technology’, Ed. G. Charalambous and G. Inglett, Academic
Press, New York, 1981, Vol. 1, pp. 183–200.
[7] B. D. M. Matawan, R. W. T. Bricktown, M. H. V. Locust U. S. Patent No 3928645 (Chem. Abstr.
1976, 85, 174491).
[8] a) J. D. Surmatis, R. J. Thommen, J. Org. Chem. 1967, 32, 180; b) W. Skorianetz, G. Ohloff, Helv.
Chim. Acta 1974, 57, 2439; c) K. Karrer, C. H. Eugster, Helv. Chim. Acta 1951, 34, 1400.
[9] a) Y. I. M. Nilsson, A. Aranyos, P. G. Andersson, J. E. Bäckvall, J. L. Parrain, C. Ploteau, J. P. Quin-
tard, J. Org. Chem. 1996, 61, 1825; b) P. G. Andersson, Y. I. M. Nilsson, J. E. Bäckvall, Tetrahedron
1994, 50, 559.
[10] a) A. Haag, W. Eschenmoser, C. H. Eugster, Helv. Chim. Acta 1980, 63, 10; b) M. Leclaire, P. Jean,
R. Lopez, L. Ricard, H. Plessix, J. Y. Lallemand, Tetrahedron 1995, 51, 6983; c) Y. Ohtsuka, T. Oishi,
Chem. Pharm. Bull. 1988, 36, 4722.
[11] a) E. Brenna, C. Fuganti, P. Grasselli, M. Redaelli, S. Serra, J. Chem. Soc., Perkin Trans. 1 1998, 4129;
b) C. Fehr, O. Guntern, Helv. Chim. Acta 1992, 75, 1023; c) T. Oritani, K. Yamashita, Agric. Biol.
Chem. 1987, 51, 1271; d) P. Uebelhart, A. Baumeler, A. Haag, R. Prewo, J. H. Bieri, C. H. Eugster,
Helv. Chim. Acta 1986, 69, 816.
[12] a) J. Aleu, E. Brenna, C. Fuganti, S. Serra, J. Chem. Soc., Perkin Trans. 1 1999, 271; b) C. Fehr,
Angew. Chem., Int. Ed. 1998, 37, 2407.
[13] B. Rickborn, R. P. Thummel, J. Org. Chem. 1969, 34, 3583.
[14] R. Kaiser, D. Lamparsky, Helv. Chim. Acta 1979, 62, 1878.
[15] W. Adam, A. G. Griesbeck, X. Wang, Liebigs Ann. Chem. 1992, 193.
[16] S. G. Hegde, J. Wolinsky, Tetrahedron Lett. 1981, 22, 5019.
[17] F. M. Hauser, R. Tommasi, P. Hewawasam, Y. S. Rho, J. Org. Chem. 1988, 53, 4886.
[18] E. Brenna, C. Fuganti, S. Serra, C. R. Chi. 2003, 6, 529.
[19] J. Tsuji, I. Minami, I. Shimizu, Synthesis 1986, 623.
Received February 9, 2006