Enzyme-mediated preparation of (+)- and (-)-cis-α-irone and (+)- and (- )-trans-α-irone

Article abstract of DOI:10.1002/(SICI)1522-2675(19991215)82:12<2246::AID-HLCA2246>3.0.CO;2-3

The preparation of (-)- and (+)-trans-α-irone (1a and 1b, resp.) and of (+)- and (-)-cis-α-irone (1c and 1d, resp.) from commercially available Irone alpha is reported. The relevant step in the synthetic sequence is the initial chromatographic separation

Full text of DOI:10.1002/(SICI)1522-2675(19991215)82:12<2246::AID-HLCA2246>3.0.CO;2-3

2246
Helvetica Chimica Acta ± Vol. 82 (1999)
Enzyme-Mediated Preparation of (+)- and (±)-cis-a-Irone
and (+)- and (±)-trans-a-Irone
by Elisabetta Brenna, Claudio Fuganti*, Giovanni Fronza, Luciana Malpezzi, Annalisa Righetti, and Stefano Serra
Dipartimento di Chimica del Politecnico, Centro CNR per lo Studio delle Sostanze Organiche Naturali, Via
Mancinelli 7, I-20131 Milano
The preparation of ( )- and (‡)-trans-a-irone (1a and 1b, resp.) and of (‡)- and ( )-cis-a-irone (1c and 1d,
resp.) from commercially available Irone alpha ꢀ is reported. The relevant step in the synthetic sequence is the
initial chromatographic separation of crystalline (Æ)-4,5-epoxy-4,5-dihydro-cis-a-irone ((Æ)-5) from oily (Æ)-
4,5-epoxy-4,5-dihydro-trans-a-irone ((Æ)-4). The latter was subsequently converted, after NaBH4 reduction,
into the crystalline 3,5-dinitrobenzoate ester (Æ)-8, thus allowing a complete separation of the two corresponding
diastereoisomeric alcohol derivatives. Suitable enantiomerically pure precursors of the desired products 1a ± d
were obtained by kinetic resolution of the racemic allylic alcohols derived from (Æ)-5 and (Æ)-8, mediated by
lipase PS (Amano). The last steps consisted of MnO2 oxidation and removal of the epoxy moiety with Me3SiCl/
NaI in MeCN. External panel olfactory evaluation showed that ( )-cis-a-irone (1d) has the finest and most
distinct ꢁorris butterꢂ character.
Introduction. ± Irones 1 ± 3 are the olfactory active components of Iris essential oil,
one of the most precious raw materials in fine perfumery [1a]. These ketones are not
present in fresh rhizomes, but they are slowly produced from two isomeric C31
precursors, after the harvesting and during ageing, through a still obscure oxidative
mechanism. The manufacture of the essential oil begins with the rhizomes of three-
years-old plants, which, once peeled, are kept in a dry and aerated environment for two
to three years. The powdered material is then incubated with diluted sulfuric acid and
steam-distilled to provide the ꢁorris butterꢂ, containing up to 90% myristic acid. The
desired essential oil is finally recovered from the butter through delicate operations [2].
More recently, Belcour et al. reported on a new microbial process allowing the
expression of up to 1 g of irones per kilogram of three-year-old rhizomes, instead of the
0.4 g provided by the traditional procedure [3].
Irones were first isolated from Iris rhizomes by Tiemann more than one century ago
[4], but the correct structural formula was assigned in 1947 by Ruzicka and Naves at the
end of an extended degradative work [5][6]. The materials extracted from Iris pallida
Lemarck, and used by Ruzicka in the structure elucidation, were shown to possess
(2R)-configuration through chemical correlation with ( )-camphor by Rautenstrauch
and Ohloff in 1971 [7]. Not surprisingly, the set of chemically identical products
isolated from Iris germanica Linnaeus possess the opposite configuration [8]. Indeed,
the configurational assignment explains, in terms of olfactory enantiodifferentiation,
the defect in radiant power, long known to professional perfumers, of the Iris germanica
extracts in comparison to those of Iris pallida and Iris florentina Linnaeus. Similarly,
the odor response of synthetic racemic a-irone was described to be different from that
of the natural extract [1b].

Helvetica Chimica Acta ± Vol. 82 (1999)
2247
Detailed comparative olfactory descriptions of all the stereoisomers of a-, b-, and g-
irones 1 ± 3 are rather limited, since these materials are not easily accessible in
enantiomerically enriched forms. Indeed, the number of syntheses of optically active
irones is rather small [9], especially when compared to the number of the racemic
modifications [10]. Petrzilka and co-workers reported the olfactory evaluation of the
single enantiomers of irones 1 ± 3 by chiral GC-sniff analyses of the corresponding
racemic materials on modified cyclodextrins as stationary phases [11]. According to
this study, with the exception of trans-a-irone, all the other pairs of enantiomers could
be distinguished by the human nose. In particular, of cis-a-irone and cis-g-irone, which
constitute up to 80% of the essential oils of different origin in ca. equal ratio, only the
(2R)-enantiomers appeared to be odor active, showing also the finest and most intense
iris-like notes. Independently, Ohloff considered (2R)-trans-a-irone and (2R)-b-irone
to possess the highest odor quality amongst the olfactorily active compounds known at
the time [1b].
At the light of these observations, it seemed desirable to have access to substantial
quantities of each stereoisomer of 1 ± 3 in enantiomerically enriched forms through a
general procedure from inexpensive readily available starting materials. We report now
on the preparation of the (‡)- and ( )-trans- and (‡)- and ( )-cis-a-irones (1a ± d)
from Irone alphaꢀ, a commercially available 55 :45 mixture of racemic trans- and cis-a-
irone, with 5% of b-isomer.
Results and Discussion. ± To prepare the enantiomerically pure substrates 1a ± d, we
first considered exploitation of the same procedure we had successfully developed for
the resolution of racemic a-ionone [12a]. (R)- and (S)-a-Ionone were obtained by
MnO2 oxidation of enantiomerically pure a-ionols produced by lipase-mediated
acetylation of the racemic materials and subsequent saponification. However, in the
case of a-irone, the presence of one more Me group at C(2) gave rise to eight
stereoisomers upon NaBH4 reduction of commercial a-irone: eight peaks of nearly
equal intensity were indeed observed by chiral GC analysis of the corresponding ironol
acetates (see Exper. Part). Thus, a chemical separation of the two racemic
diastereoisomers 1a,b and 1c,d was desirable before attempting enzymic resolution.
Fortunately, a series of favorable circumstances greatly facilitated the achievement of
this goal.

2248
Helvetica Chimica Acta ± Vol. 82 (1999)
On the basis of an earlier experience with a- and b-ionone epoxides [12b], we found
that key derivatives for the separation of trans- and cis-a-irone were the corresponding
epoxides. The 3-chloroperbenzoic acid oxidation of Irone alphaꢀ in CH2Cl2 afforded an
oily mixture, showing two main well-resolved spots by TLC. Column chromatography
(SiO2, 10 ± 15% AcOEt/hexane) gave, in order of elution, (Æ)-4 and (Æ)-5. The latter
was obtained in pure form since it crystallized nicely from hexane; X-ray single-
crystal analysis (by L.M.) (Fig. 1) allowed the assignment of the structural formula
of (Æ)-5. This crystalline material was found to be the diastereoisomerically pure
epoxide derivative of cis-a-irone (1c,d) bearing the oxirane ring in ꢀantiꢂ position
to the side chain at C(6), consistent with the following measured torsional
angles: C(4) C(5) (6) C(7) 150.0(1)8, O(1) C(5) C(6) C(7) 143.6(1)8, and
C(11) C(5) C(6) C(7) 7.7(2)8.
Fig. 1. Molecular structure of (Æ)-5 with the atom-numbering scheme. Displacement ellipsoids correspond to
30% probability; H-atoms are shown as spheres of arbitrary radius.
The first eluted fraction of the epoxidation mixture was composed mainly of (Æ)-
epoxy-trans-a-irone (Æ)-4, with 20% of the other diastereoisomer of epoxy-cis-a-irone,
i.e. the one obtained by reaction of the peracid on the same side of the substituent at
C(6). As a matter of fact, the treatment of this chromatographic fraction with
chlorotrimethylsilane and NaI in MeCN [13] afforded a 80 : 20 mixture of trans- and cis-
a-irone. The configuration of the epoxy moiety of derivative (Æ)-4 was determined by
1H-NMR analysis of a suitable derivative.

Helvetica Chimica Acta ± Vol. 82 (1999)
2249
It is worth noting that the steric course of the epoxidation of cis-a-irone was
completely different from that we observed for trans-a-irone and the one Eugster
reported for a-ionone [14], in complete accordance with the known ꢁantiꢂ-epoxidation
of cis-a-sub-skeletons under the same reaction conditions [15].
Substrate (Æ)-4, contaminated with the cis-derivative, was reduced with NaBH4 in
CH2Cl2/MeOH 2 : 1 to provide, as major components, the two racemic diastereoiso-
meric alcohols (Æ)-6 and (Æ)-7 (1 : 1 ratio), which could be separated by column
chromatography (SiO2) (Scheme 1). The first-eluted alcohol (Æ)-6, containing over
90% of a single component, was obtained in pure form by crystallization of the
corresponding 3,5-dinitrobenzoate ester (Æ)-8 from EtOH. This material was assigned
the shown relative configuration at the cyclic moiety by NMR data.
Scheme 1
i) NaBH4, CH2Cl2/MeOH 2 :1, column chromatography. ii) 3,5-(NO2)2C6H3COCl, pyridine, CH2Cl2. iii)
Mitsunobuꢂs esterification. iv) Crystallization from EtOH. v) KOH, MeOH.
A strong NOE was observed between H C(7) and H C(2) of (Æ)-8, thus confirming the trans-
configuration of Me C(2) and the side chain. The oxirane ring was found to be on the same side of the
substituent at C(6) on the basis of the clear NOEs between the axial Me C(1) and H C(4) and Me C(5).
Moreover, the values of the vicinal coupling constants J(H C(4),Heq C(3)) and J(H C(4),Hax C(3)) were
rather low (1.7 and 2.4 Hz, resp.), in agreement with an equatorial location of H C(4).
The relative configuration of the OH-bearing C-atom in the side chain was
tentatively attributed by analogy to the first-eluted diastereoisomeric epoxy-a-ionol,
showing ꢁantiꢂ relationship between the side chain and the OH group [12b]. This
assignment was confirmed by the outcome of the synthetic sequence described below.
The 3,5-dinitrobenzoate of (Æ)-7 was an oil. Crude (Æ)-7 was then esterified with
3,5-dinitrobenzoic acid under Mitsunobu conditions [16] to provide (Æ)-8. The

2250
Helvetica Chimica Acta ± Vol. 82 (1999)
crystalline ester (Æ)-8, pooled from the two reactions, was crystallized from EtOH and
hydrolyzed under basic conditions to afford (Æ)-6. The homogeneity of the sample was
verified by chiral GC analysis of the corresponding acetate (Fig. 2,a). The allylic
alcohol (Æ)-6 was straightforwardly converted to the ( )- and (‡)-trans-a-irones (1a
and 1b, resp.) by the chemo-enzymatic process outlined in Scheme 2 [12b]. Treatment
of (Æ)-6 with vinyl acetate in t-butyl methyl ether in the presence of lipase PS (Amano)
afforded enantiomerically pure acetate (‡)-9 (see Fig. 2,c) and unreacted alcohol ( )-6
showing an ee of 98% (GC of the corresponding acetate derivative, see Fig. 2,b;
enantiomer ratio ˆ 461 and conversion ˆ 0.5, calculated according to [17]). The acetate
(‡)-9 was then hydrolyzed under basic conditions, and the derived alcohol was oxidized
with MnO2 in CH2Cl2 to give epoxy ketone (‡)-4 (ee 98% by GC). Similarly, alcohol
( )-6 afforded enantiomerically pure (GC) epoxy ketone ( )-4. Finally, ( )- and (‡)-4
Scheme 2
i) Lipase PS (Amano), vinyl acetate, tBuOMe, column chromatography. ii) MnO2, CH2Cl2. iii) KOH, MeOH.
iv) Me3SiCl, NaI, MeCN.

Helvetica Chimica Acta ± Vol. 82 (1999)
2251
Fig. 2. Chiral GC analyses a) of the acetate derivatives of (Æ)-6, b) of the acetate derivative of ( )-6, c) of (‡)-9,
d) of ( )-4, and e) of (‡)-4
were deoxygenated separately in good yields to ( )-trans-a-irone (1a; chemical purity
88% and ee 98% by GC) and (‡)-trans-a-irone (1b; chemical purity 96% and ee 98%
by GC). This deoxygenation process caused the formation of impurities which could
not be easily separated from trans-a-irone (see Fig. 3,d and e).
Lipase PS (Amano) usually reacts with (R)-alcohols [12]; thus, the conversion of
( )-6 and (‡)-6 into 1a (( )-trans) and 1b ((‡)-trans), respectively, supported the
configurational assignment of (Æ)-6 and (Æ)-7. Moreover, the conversion of (Æ)-7 to
(Æ)-6 by Mitsunobuꢂs esterification and hydrolysis allowed us to resolve (Æ)-4 in 33%
overall yield without loss of material.
More direct was the access to the enantiomers 1c and 1d of cis-a-irone from the
racemic trans/cis mixture thanks to the crystallinity of epoxy-cis-a-irone (Æ)-5
(Scheme 3). This latter was reduced with NaBH4 to afford a 1 : 1 mixture of the two
diastereoisomeric alcohols (Æ)-10 and (Æ)-11, which could be separated by column
chromatography. The second-eluted alcohol (Æ)-11 was then purified by crystallization

2252
Helvetica Chimica Acta ± Vol. 82 (1999)
Fig. 3. Chiral GC analyses a) of commercial a-irone, b) of (‡)-cis-a-irone (1c), c) of ( )-cis-a-irone (1d), d) of
)-trans-a-irone (1a), and e) of (‡)-trans-a-irone (1b)
(
from hexane. The configurational assignment of (Æ)-10 and (Æ)-11 was based on the
same arguments given for (Æ)-6 and similarly confirmed by the synthetic sequence. In
separate experiments, (Æ)-10 and (Æ)-11 were submitted to lipase PS mediated
acetylation in tBuOMe solution in the presence of vinyl acetate to provide, after column
chromatography, acetates (‡)-12 (ee 98%, by GC, see Fig. 4,e) and (‡)-13 (ee 98%) by
GC see Fig. 4,f), respectively, and unreacted alcohols (‡)-10 (ee 98% by GC of the
corresponding acetate derivative, see Fig. 4,g) and ( )-11 (ee 98% by GC of the
corresponding acetate derivative, see Fig. 4,h), respectively (enantiomer ratio ˆ 461,

Helvetica Chimica Acta ± Vol. 82 (1999)
2253
Scheme 3
t
i) NaBH4, CH2Cl2/MeOH 2 :1. ii) Lipase PS (Amano), vinyl acetate, BuOMe, column chromatography. iii)
KOH, MeOH. iv) MnO2, CH2Cl2. v) Me3SiCl, NaI, MeCN.
conversion ˆ 0.5 for both the kinetic resolutions; see [17]). Acetate (‡)-12 gave, upon
hydrolysis and oxidation with MnO2 the same enantiomeric epoxy-irone ( )-5 as
obtained by direct oxidation of ( )-11. The two samples of ( )-5 were combined (ee
98% by GC, see Fig. 4,i) and submitted to deoxygenation with chlorotrimethylsilane
and NaI in MeCN, to give (‡)-cis-a-irone (1c, chemical purity 81% and ee ˆ 98% by
GC, see Fig. 3,b). Derivative (‡)-5 (ee 98% by GC, see Fig. 4,j), prepared both from
acetate (‡)-13 and from alcohol (‡)-10, was transformed according to the same
deoxygenation procedure into ( )-cis-a-irone (1d; chemical purity 85% and ee 98% by
GC, see Fig. 3,c). Thus, enantiospecific acetylation of the two diastereoisomeric

2254
Helvetica Chimica Acta ± Vol. 82 (1999)
Fig. 4. Chiral GC analyses a) of (Æ)-5, b) of the acetate derivatives of (Æ)-10/(Æ)-11 1 : 1, c) of the acetate
derivatives of (Æ)-10, d) of the acetate derivatives of (Æ)-11, e) of (‡)-12, f) of (‡)-13, g) of the acetate derivative
of (‡)-10, h) of the acetate derivative of ( )-11, i) of ( )-5, and j) of (‡)-5

Helvetica Chimica Acta ± Vol. 82 (1999)
2255
alcohols (Æ)-10 and (Æ)-11 allowed us to convert both acetates and unreacted alcohols
to ( )- and (‡)-5, precursors of the valuable (‡)- and ( )-cis-a-irones, allowing a
formal resolution of epoxy-cis-a-irone (Æ)-5 in 49% overall yield. The weak point of
the sequence was again the deoxygenation process, in which some impurities not easily
separable were produced, thus lowering the chemical purity of the samples.
Olfactory Evaluation. ± Olfactory evaluation of irones 1a ± d was performed by
twelve trained evaluators (Robertet S.A., Grasse, France) using 0.1% solutions in
EtOH. The following results were obtained: a) The trans enantiomers 1a and 1b did not
possess the characteristic ꢁorrisꢂ odor. The ( )-enantiomer 1a showed a weak violet/
wood/red berry character, while the (‡)-enantiomer 1b was defined as the weakest
isomer by unanimous decision. b) (‡)-cis-a-Irone (1c) was violet-like, with woody,
methylionone undertones. ( )-cis-a-Irone (1d) was slightly stronger with a distinct
ꢁorris-butterꢂ character (iris-like).
The olfactory detection threshold (triangular test) was ca. 10 ppm for 1d and ca.
100 ppm for 1c (in H2O). As for derivatives 1c and 1d, remarkable differences were
observed with respect to the olfactory evaluation of cis-a-irone enantiomers reported
by Petrzilka and co-workers [11]. In contrast with our results, by GC-sniff analysis of
racemic cis-a-irone, the ( )-enantiomer was found to be odorless, and the (‡)-
enantiomer was described as floral, fruity, with a woody ionone and sweet irone odor.
Conclusions. ± The preparation of enantiomerically pure irones 1a ± d from the
commercially available mixture of trans- and cis-a-irone 1 (Irone alphaꢀ) might be of
some utility. The enzyme-mediated kinetic resolution of allylic alcohols (Æ)-6, (Æ)-10,
and (Æ)-11 was an extension of some previous findings [12]. However, the success of
these bio-catalyzed reactions was based on the simple, and still unknown, separation of
racemic trans-a-irone from racemic cis-a-irone through their epoxide derivatives. The
crystallinity of epoxide (Æ)-5 was of great help in this separation process. Moreover, the
conversion of both (Æ)-6 and (Æ)-7 to (Æ)-8, which was easily purified by
crystallization, was another key step in the synthetic sequence.
Thus, the unexceptional reactions described in this work might be of some
significance in the chemistry of irones, a class of compounds that have been known for
more than one century, but which are still subject to new chemical investigations.
The partial financial support of MURST is acknowledged. The authors would like to thank Dr. Daniel
Joulain, Robertet S.A., Grasse, France, for his kind co-operation and the twelve evaluators of Robertet S.A. for
the olfactory evaluation.
Experimental Part
1. General. Irone alphaꢀ was purchased from IES, Allauch, France. Lipase PS Pseudomonas cepacia
(Amano Pharmaceuticals Co., Japan) was employed in this work. TLC: Merck silica gel 60 F254 plates. Column
chromatography (CC): silica gel. Chiral GC analysis: DANI-HT-86.10 gas chromatograph; enantiomer and
diastereoisomer excesses determined on a Chirasil DEX CB column (25 m  0.25 mm; Chrompack) temp.
program 708 (3 min) 3.58/min 1408 88/min 1808 (1 min); analysis of a-ironol acetates on a Hydrodex-b-
PM column (50 m  0.25 mm; Superchrom), temp. program 708 (3 min) 18/min 1508 (10 min) 18/min
1608 (1 min) 108/min 2008 (1 min): tR 84.10, 84.61, 84.96, 85.24, 90.32, 90.54, 91.07, and 91.57; mass-
detection limit for anal. GC ca. 10 ng for an injected volume of 1 ml; tR in min. Optical rotations: Jasco-DIP-181
digital polarimeter. 1H-NMR Spectra: CDCl3 solns. at r.t., unless otherwise stated; Bruker-AC-250 spectrometer

2256
Helvetica Chimica Acta ± Vol. 82 (1999)
at 250 MHz (1H); chemical shifts in ppm rel. to internal SiMe4 (ˆ 0 ppm), J values in Hz. Microanalyses were
determined on an analyzer 1106 from Carlo Erba.
2. X-Ray Crystal-Structure Determination of (Æ)-5. A colorless crystal, suitable for X-ray analysis, with the
shape of a hexagonal prism, was obtained by crystallization from hexane. For crystal data and measurement
conditions, see Table. Diffraction data were collected on
a Siemens-P4 diffractometer with graphite
monochromated Cu-Ka radiation (l 1.54179 Š) by the q/2q scan technique. Unit-cell parameters were
determined using 48 reflections in the range 188 ꢀ 2q ꢀ 808. Empirical adsorption correction was applied by
means of the y-scan methods; no decay correction was deemed necessary. The structure was solved by direct
methods with SIR97 [18] and refined with SHELXL97 [19] programs. All H-atoms were located from a
difference electron density map, and their positions were allowed to refine together with individual isotropic
temp. factors. The six-membered ring has a slightly distorted envelope form, with the C(1) atom at 0.690(2) Š
out of the best plane of the other five-ring C-atoms; the O(1) atom lies at 1.283 Š on the opposite side of the
plane. Crystallographic data for (Æ)-5 have been deposited at the Cambridge Crystallographic Data Centre as
supplementary publication No. CCDC 132368.
Table. Crystal Data and Structure Refinement of (Æ)-5
Empirical formula
Mr
C14H22O2
222.32
q Range [8]
Index ranges
5.27 to 67.96
10 ꢀ h ꢁ 10, 15 ꢀ k ꢁ 16,
12 ꢀ l ꢁ 12
Crystal system
Space group
Unit cell dimension a[Š]
monoclinic
P21/c
9.131(5)
Reflect. collected
Independent reflect.
Completeness to
4552
2309 (R(int) ˆ 0.081)
2q ˆ 67.96
91.0%
b[Š]
c[Š]
b[8]
14.912(5)
10.734(5)
111.71(1)
1343.0(11)
4
Refinement method
Data/restraints/parameters
Goodness of fit on F2
Final R indices (I > 2s(I))
R indices (all data)
full-matrix least-squares on F2
2309/0/234
1.068
Volume [Š3]
Z
R1 ˆ 0.0492, wR2 ˆ 0.1250
R1 ˆ 0.0541, wR2 ˆ 0.1296
3
Density calc. [Mg m
]
1.100
Largest difference peaks
and hole [e ´ Š
3
]
0.158 and 0.144
1
Adsorption coeff. [mm
F(000)
]
0.560
488
Crystal size [mm]
0.8 Â 0.6 Â 0.3
3. (Æ)-(2RS,4RS,5SR,6SR)-4,5-Epoxy-4,5-dihydro-a-irone (ˆ(E)-4-[(1RS,2RS,4SR,6SR)-1,3,3,4-Tetra-
methyl-7-oxabicyclo[4.1.0]hept-2-yl]but-3-en-2-one; (Æ)-4) and (Æ)-(2RS,4RS,5SR,6RS)-4,5-Epoxy-4,5-di-
hydro-a-irone (ˆ(E)-4-[(1RS,2SR,4SR,6SR)-1,3,3,4-Tetramethyl-7-oxabicyclo[4.1.0]hept-2-yl]but-3-en-2-one;
(Æ)-5). To a soln. of racemic a-irone (1; ca. 55 : 45 trans/cis mixture; 50 g, 0.24 mol) in CH2Cl2 (450 ml) at 08, 3-
chloroperbenzoic acid (62 g, 0.36mol) was added. The mixture was stirred at 08 for 2 h, then poured into H2O.
The org. phase was washed with sat. NaHCO3 soln., dried (Na2SO4) and evaporated. The residue was submitted
to CC (hexane/AcOEt 9 : 1): (Æ)-4 (18.6 g, 35%) and then pure (Æ)-5 (10.1 g, 19%; after crystallization).
Data of (Æ)-4 (containing 20% (GC, 1H-NMR ) of isomer (Æ)-(2RS,4SR,5RS,6RS)-4,5-epoxy-4,5-dihydro-
a-irone): Chiral GC: tR 24.58, 24.96; isomer (20%) at tR 25.41, 25.78. 1H-NMR: 6.76 (dd, J ˆ 15.7, 10.0, H C(7));
6.10 (d, J ˆ 15.7, H C(8)); 3.11 (m, H C(4)); 2.30 (s, Me(10)); 2.11 (d, J ˆ 10, H C(6)); 2.01 (m, 1 H); 1.59
(m, 2 H); 1.27 (s, Me C(5)); 0.83 (s, 1 Me C(1)); 0.81 (d, J ˆ 6, Me C(2)); 0.70 (3 H, s, 1 Me C(1)); isomer
(20%) at 3.07 (d, J ˆ 6, H C(4)). Anal. calc. for C14H22O2: C 75.63, H 9.97; found: C 75.69, H 9.91.
Data of (Æ)-5: Crystallization from hexane. M.p. 658. Chiral GC: tR 25.41, 25.78. 1H-NMR 6.71 (dd, J ˆ 11.2,
15.5, H C(7)); 6.16 (d, J ˆ 15.5, H C(8)); 3.00 (m, H C(4)); 2.30 (s ‡ d, J ˆ 11.2, MeCO, H C(6)); 2.01
(ddd, J ˆ 14.7, 3.9, 1.2, Heq C(3)); 1.59 (td, J ˆ 14.7, 1.9, Hax C(3)); 1.45 (m, H C(2)); 1.22 (s, Me C(5));
0.82 (d, J ˆ 6.5, Me C(2)); 0.74 (s, 2Me C(1)). Anal. calc. for C14H22O2: C 75.63, H 9.97; found: C 75.59, H 9.93.
4. (Æ)-(2RS,4RS,5SR,6SR,9SR)-4,5-Epoxy-4,5-dihydro-a-ironol (ˆ(2RS,E)-4-[(1RS,2RS,4SR,6SR)-
1,3,3,4-Tetramethyl-7-oxabicyclo[4.1.0]hept-2-yl]but-3-en-2-ol; (Æ)-6) and (Æ)-(2RS,4RS,5SR,6SR,9RS)-4,5-
Epoxy-4,5-dihydro-a-ironol (ˆ(2RS,E)-4-[(1SR,2SR,4RS,6RS)-1,3,3,4-Tetramethyl-7-oxabicyclo[4.1.0]hept-2-

Helvetica Chimica Acta ± Vol. 82 (1999)
2257
yl]but-3-en-2-ol; (Æ)-7). Reduction of (Æ)-4 (18.0 g, 0.081 mol) with NaBH4 (3.7 g, 0.097 mol) in CH2Cl2/
MeOH 2 : 1 (100 ml) gave, after the usual workup, (Æ)-6/(Æ)-7 1 : 1 (GC) which was chromatographed (hexane/
AcOEt 8 : 2): (Æ)-6 (7.4 g, 41%) and (Æ)-7 (6.9 g, 38%).
Data of (Æ)-6 (containing 5% (GC) of isomer (Æ)-(2RS,4SR,5RS,6RS,9RS)-4,5-epoxy-4,5-dihydro-a-
ironol): Chiral GC (corresponding acetate): tR 25.41, 25.80, isomer (5%) at tR 26.15, 26.51. 1H-NMR: 5.63
(m, 2 olef. H); 4.34 (quint., J ˆ 6, H C(9)); 3.07 (m, H C(4)); 1.94 (m, 2 H); 1.55 (m, 2 H); 1.29 (d, J ˆ 6,
Me(10)); 1.25 (s, Me C(5)); 0.78 (s, Me C(1)); 0.76 (d, J ˆ 6, Me C(2)). Anal. calc. for C14H24O2: C 74.95,
H 10.78; found: C 74.99, H 10.71.
Data of (Æ)-7 (containing 6% (1H-NMR) of isomer (Æ)-(2RS,4SR,5RS,6RS,9SR)-4,5-epoxy-4,5-dihydro-a-
1
ironol): Chiral GC (corresponding acetate): tR 25.80, 26.01. H-NMR: 5.63 (m, 2 olef. H); 4.34 (m, H C(9));
3.08 (m, H C(4)); 1.91 (d, J ˆ 9, H C(6)); 1.2 ± 1.7 (m, 3 H); 1.29 (d, J ˆ 6, Me(10)); 1.25 (s, Me C(5)); 0.78
(s, Me C(1)); 0.76 (d, J ˆ 6, Me C(2)); 0.69 (s, Me C(1)); isomer (6%) at 3.01 (d, J ˆ 6, H C(4)). Anal.
calc. for C14H24O2: C 74.95, H 10.78; found: C 74.91, H 10.82.
5. (Æ)-(2RS,4RS,5SR,6SR,9SR)-4,5-Epoxy-4,5-dihydro-a-ironol 3,5-Dinitrobenzoate (Æ-8). From (Æ)-6:
A soln. of (Æ)-6 (7.0 g, 0.031 mol) in CH2Cl2 (70 ml) was treated with 3,5-dinitrobenzoyl chloride (8.5 g,
0.037 mol) in the presence of pyridine (10 ml). After the usual workup, purification of the residue by CC
(hexane/AcOEt 9 : 1) gave (Æ)-8 which was crystallized from EtOH: diastereoisomerically pure (GC) (Æ)-8
(9.2 g, 71%). M.p. 898. 1H-NMR (C6D6): 8.69 (d, J ˆ 2.1, 2 arom. H); 8.42 (t, J ˆ 2.1, 1 arom. H); 6.05
(ddd, J(7,6) ˆ 10.4, J(7,8) ˆ 15.7, J(7,9) ˆ 0.95, H C(7)); 5.54 (qd, J(9,Me) ˆ 6.2, J(9,8) ˆ 7.1, H C(9)); 5.42
(dd, J(8,7) ˆ 15.7, J(8,9) ˆ 7.1, H C(8)); 2.75 (t, J(4,3ax) ˆ J(4,3eq) ˆ 1.9, H C(4)); 1.79 (ddd, J (3eq,3ax) ˆ
15.1, J(3eq,2) ˆ 4.7, J(3eq,4) ˆ 1.6, Heq C(3)); 1.76 (d, J(6,7) ˆ 10.4, H C(6)); 1.57 (dqd, J(2,Me) ˆ 6.9,
J(2,3ax) ˆ 11.2, J(2,3eq) ˆ 4.7,
H
C(2)); 1.23 (ddd, J(3ax,3eq) ˆ 15.1, J(3ax,2) ˆ 11.2, J(3ax,4) ˆ 2.4,
Hax C(3)); 1.23 (d, J ˆ 6.4, Me(10)); 1.16 (s, Me C(5)); 0.70 (s, Meeq C(1)); 0.63 (d, J ˆ 6.9, Me C(2));
0.60 (s, Meax C(1)). Anal. calc. for C21H26N2O7: C 60.28, H 6.26; found: C 60.21, H 6.32.
From (Æ)-7 by Mitsunobuꢂs Esterification: a soln. of (Æ)-7 (6.0 g, 0.027 mmol) and PPh3 (8.49 g, 0.032 mol)
in THF (30 ml) was added dropwise into a soln. of diisopropyl azodicarboxylate (6.54 g, 0.032 mmol) and 3,5-
dinitrobenzoic acid (6.8 g, 0.032 mol) in THF (30 ml). The mixture was stirred at r.t. for 4 h and then
evaporated, the residue poured into H2O and extracted with AcOEt, the org. layer dried (Na2SO4) and
evaporated, the residue chromatographed (hexane/AcOEt 9 : 1) and the product crystallized from EtOH:
diastereoisomerically pure (Æ)-8 (6.54 g, 58%). Anal. and spectral data: in accordance with those of the sample
obtained from (Æ)-6.
6. (Æ)-(2RS,4RS,5SR,6SR,9SR)-4,5-Epoxy-4,5-dihydro-a-ironol ((Æ)-6). Diastereoisomerically pure
(GC) (Æ)-6 (6.11 g, 88%) was obtained by hydrolysis of (Æ)-8 (13 g, 0.031 mol) with 85% KOH (2.45 g,
0.037 mol) in MeOH soln. (50 ml): Chiral GC (corresponding acetate): tR 25.41, 25.80. 1H-NMR: in accordance
with that described above.
7. ( )-(2R,4R,5S,6S,9S)-4,5-Epoxy-4,5-dihydro-a-ironol (( )-6) and (‡)-(2S,4S,5R,6R,9R)-4,5-Epoxy-4,5-
dihydro-a-ironol Acetate ((‡)-9). A mixture of (Æ)-6 (de > 99%; 6 g, 0.026 mol), lipase PS (Pseudomonas
cepacia; 6 g), and vinyl acetate (10 ml) in tBuOMe (80 ml) was stirred at r.t. for 24 h. The residue obtained upon
evaporation of the filtered mixture was chromatographed (hexane/AcOEt 7 : 3): (‡)-9 (3.11 g, 45%) and ( )-6
(2.44 g, 42%).
Data of (‡)-9: Chiral GC: tR 25.80; enantiomerically pure. [a]2D0 ˆ ‡218.8 (c ˆ 0.95, CH2Cl2). 1H-NMR:
5.72 (dd, J ˆ 15, 10,
H
C(7)); 5.51 (dd, J ˆ 15, 6.2,
H
C(8)); 5.35 (quint., J ˆ 6.2,
H C(9)); 3.05
(m,H C(4)); 2.03 (s, MeCOO); 1.95 (dd, J ˆ 10, 1, 1 H); 1.88 (d, J ˆ 10, H C(6)); 1.53 (m, 2 H); 1.32
(d, J ˆ 6.2, Me(10)); 1.24 (s, Me C(5)); 0.77 (s, Me C(1)); 0.76 (d, J ˆ 6.2, Me C(2)); 0.66 (s, Me C(1)).
Anal. calc. for C16H26O2: C 72.14, H 9.84; found: C 72.09, H 9.92.
Data of ( )-6: Chiral GC (corresponding acetate): tR 25.41; ee 98%. [a]D20
1H-NMR: consistent with that of the racemic starting material.
ˆ
280 (c ˆ 0.45, CH2Cl2).
8. ( )-(2R,4R,5S,6S)-4,5-Epoxy-4,5-dihydro-a-irone (( )-4). At r.t. ( )-6 (ee 98% by GC; 2.40 g,
0.011 mol) was oxidized with MrO2 (0.017 mol, 1.48 g) in CH2Cl2 (30 ml). The crude material was purified by
CC (hexane/AcOEt 9 :1): ( )-4 (1.97 g, 83%). Chiral GC: tR 24.58; ee 97%. [a]D20
1H-NMR: in accordance with that of (Æ)-4.
ˆ
258 (c ˆ 1.0 CH2Cl2).
9. (‡)-(2S,4S,5R,6R,9R)-4,5-Epoxy-4,5-dihydro-a-ironol ((‡)-6). Enantiomerically pure acetate (‡)-9
(3.0 g, 0.011 mol) was hydrolyzed with 85% KOH (1.09 g, 0.017 mol) in MeOH (30 ml), to afford, after CC,
enantiomerically pure (‡)-6 (2.24 g, 91%). [a]2D0 ˆ ‡270 (c ˆ 0.7 CH2Cl2). 1H-NMR: in accordance with that
of (Æ)-6.

2258
Helvetica Chimica Acta ± Vol. 82 (1999)
10. (‡)-(2S,4S,5R,6R)-4,5-Epoxy-4,5-dihydro-a-irone ((‡)-4). According to Exper. 8, with (‡)-6 (2.20 g,
0.010 mol), MnO2 (0.015 mol, 1.30 g), and CH2Cl2 (30 ml): (‡)-4 (2.02 g, 91%). Chiral GC: tR 24.96; ee 98%.
1
[a]2D0 ˆ ‡260 (c ˆ 1.15, CH2Cl2). H-NMR: in accordance with that of (Æ)-4.
11. (Æ)-(2RS,4RS,5SR,6RS,9RS))-4,5-Epoxy-4,5-dihydro-a-ironol ((Æ)-10) and (Æ)-(2RS,4RS,5SR,
6RS,9SR)-4,5-Epoxy-4,5-dihydro-a-irone ((‡)-11). According to Exper. 4, with (Æ)-5 (15.0 g, 0.068 mol),
NaBH4 (3.1 g, 0.081 mol), and CH2Cl2/MeOH 2 : 1 (100 ml). CC (hexane/AcOEt 8 : 2) of (Æ)-10/(Æ)-11 1 : 1
(GC) gave (Æ)-10 (6.4 g, 42%) and (Æ)-11 (5.6 g, 37%; after crystallization from hexane).
Data of (Æ)-10: Chiral GC (corresponding acetate): tR 27.43, 27.65. 1H-NMR: 5.56 (m, 2 olef. H); 4.36
(m, H C(9)); 2.97 (m, H C(4)); 2.14 (d, J ˆ 10, H C(6)); 1.97 (ddd, J ˆ 14.6, 3.9, 1.2, Heq C(3)); 1.52
(td, J ˆ 14.6, 1.9, Hax C(3)); 1.40 (m, H C(2)); 1.30 (d, J ˆ 6, Me(10)); 1.16 (s, Me C(5)); 0.80 (d, J ˆ 7,
Me C(2)); 0.76 (s, Me C(1)); 0.67 (s, Me C(1)). Anal. calc. for C14H24O2: C 74.95, H 10.78; found: C 74.88,
H 10.71.
Data of (Æ)-11: M.p. 648. Chiral GC (corresponding acetate): tR 27.43, 27.83. 1H-NMR: 5.57 (m, 2 olef. H);
4.35 (m, H C(9)); 2.97 (m, H C(4)); 2.14 (d, J ˆ 9.6, H C(6)); 1.97 (ddd, J ˆ 14.7, 4.2, 1.6, Heq C(3)); 1.52
(td, J ˆ 14.7, 1.9, Hax C(3)); 1.40 (m, H C(2)); 1.30 (d, J ˆ 6.5, Me(10)); 1.19 (s, Me C(5)); 0.80 (d, J ˆ 7,
Me C(2)); 0.73 (s, Me C(1)); 0.66 (s, Me C(1)). Anal. calc. for C14H24O2: C 74.95, H 10.78; found: C 75.01,
H 10.75.
12. (‡)-(2R,4R,5S,6R,9R)-4,5-Epoxy-4,5-dihydro-a-ironol Acetate ((‡)-12) and (‡)-(2S,4S,5R,6S,9S)-4,5-
epoxy-4,5-dihydro-a-ironol ((‡)-10). According to Exper. 7, with (Æ)-10 (6 g, 0.026 mol), lipase PS (Pseudo-
monas cepacia; 6 g), vinyl acetate (10 ml), and tBuOMe (80 ml). (‡)-12 (3.22 g, 46%) and (‡)-10 (2.39 g, 41%).
Data of (‡)-12: Chiral GC: tR 27.43; enantiomerically pure. [a]2D0 ˆ ‡58 (c ˆ 1.8, CH2Cl2). 1H-NMR: 5.57
(m, 2olef. H); 5.36 (quint., J ˆ 6.5, H C(9)); 2.96 (m, H C(4)); 2.12 (d, J ˆ 10, 1 H); 2.04 (s, MeCOO); 1.96
(ddd, J ˆ 14.5, 4.2, 1.2, Heq C(3)); 1.52 (td, J ˆ 14.5, 2 Hax C(3)); 1.40 (m, 1 H); 1.35 (d, J ˆ 6.5, Me(10)); 1.15
(s, Me C(5)); 0.80 (d, J ˆ 6.5, Me C(2)); 0.71 (s, Me C(1)); 0.66 (s, Me C(1)). Anal. calc. for C16H26O2:
C 72.14, H 9.84; found: C 72.08, H 9.81.
Data of (‡)-10: Chiral GC (corresponding acetate): tR 27.65; ee 98%. [a]2D0 ˆ ‡16 (c ˆ 1.0, CH2Cl2).
1H-NMR: consistent with that of the racemic starting material.
13. ( )-(2R,4R,5S,6R,9R)-4,5-Epoxy-4,5-dihydro-a-ironol (( )-10). (‡)-12 (3.2 g, 0.012 mol) was hydro-
lyzed with 85% KOH (1.18 g, 0.018 mol) in MeOH (25 ml), to afford, after CC (hexane/AcOEt 7:3),
1
enantiomerically pure ( )-10 (2.39 g, 89%), [a]D20
(Æ)-10.
ˆ
14 (c ˆ 1.05, CH2Cl2). H-NMR: consistent with that of
14. (‡)-(2S,4S,5R,6S,9R)-4,5-Epoxy-4,5-dihydro-a-ironol Acetate ((‡)-13) and ( )-(2R,4R,5S,6R,9S)-4,5-
Epoxy-4,5-dihydro-a-ironol (( )-11). According to Exper. 7, with (Æ)-11 (5.5 g, 0.025 mol), lipase PS
t
(Pseudomonas cepacia; 6 g), vinyl acetate (10 ml), and BuOMe (80 ml): (‡)-13 (2.72 g, 41%) and ( )-11
(2.13 g, 38%).
Data of (‡)-13: Chiral GC: tR 27.83; enantiomerically pure. [a]2D0 ˆ 89 (c ˆ 1.55, CH2Cl2). H-NMR: 5.58
1
(m, 2 olef. H); 5.37 (quint., J ˆ 6.5, H C(9)); 2.97 (m, H C(4)); 2.11 (d, J ˆ 10, 1 H); 2.04 (s, MeCOO); 1.96
(ddd, J ˆ 14.4, 4.1, 1.1, Heq C(3)); 1.51 (td, J ˆ 14.4, 2.1, Hax C(3)); 1.40 (m, H C(2)); 1.35 (d, J ˆ 6.5,
Me(10)); 1.15 (s, Me C(5)); 0.80 (d, J ˆ 7, Me C(2)); 0.73 (s, Me C(1)); 0.66 (s, Me C(1)). Anal. calc. for
C16H26O2: C 72.14, H 9.84; found: C 72.19, H 9.93.
Data of ( )-11: Chiral GC (corresponding acetate): tR 27.43; ee 98%. [a]D20
ˆ
38 (c ˆ 0.95, CH2Cl2).
1H-NMR: consistent with that of the racemic starting material.
15. (‡)-(2S,4S,5R,6S,9R)-4,5-Epoxy-4,5-dihydro-a-ironol ((‡)-11). Acetate (‡)-13 (2.7 g, 0.010 mol) was
hydrolyzed with 85% KOH (1.0 g, 0.015 mol) in MeOH (25 ml), to afford, after CC (hexane/AcOEt 7:3),
enantiomerically pure (‡)-11 (1.93 g, 86%). [a]D20 ˆ ‡32 (cˆ 1.2, CH2Cl2): 1H-NMR: consistent with that of (Æ)-11.
16. ( )-(2R,4R,5S,6R)-4,5-Epoxy-4,5-dihydro-a-irone (( )-5). In separate experiments, ( )-10 (2.3 g,
0.010 mol) and ( )-11 (2.10 g, 0.0094 mol) were oxidized with MnO2 (0.015 mol, 1.30 g) in CH2Cl2 (30 ml), to
afford the same enantiomer ( )-5. The two samples of ( )-5 were combined (3.77 g, 0.017 mol), thus providing
an overall yield from (Æ)-5 of 25%. ( )-5: Chiral GC: tR 25.41; ee 98%. [a]D20
ˆ
18 (c ˆ 1.45, CH2Cl2).
17. (‡)-(2S,4S,5R,6S)-4,5-Epoxy-4,5-dihydro-a-irone ((‡)-5). In separate experiments, (‡)-10 (2.3 g,
0.010 mol) and (‡)-11 (1.90 g, 0.0085 mol) were oxidized with MnO2 (0.015 mol, 1.30 g) in CH2Cl2 (30 ml), to
afford the same enantiomer (‡)-5. The two samples of (‡)-5 were combined (3.55 g, 0.016 mol), thus providing
an overall yield from (Æ)-5 of 24%. (‡)-5: Chiral GC: tR 25.78; ee 98%. [a]2D0 ˆ ‡21 (c ˆ 1.15, CH2Cl2).
18. General Procedure for the Conversion of Epoxy-a-irones to a-Irones. In separate experiments, epoxy-a-
irone ( )-4, (‡)-4, ( )-5, or (‡)-5 was deoxygenated according to the following procedure to afford a-irone 1a
(( )-trans)), 1b ((‡)-trans), 1c ((‡)-cis), or 1d, (( )-cis), respectively. Chlorotrimethylsilane (1.63 g, 0.015 mol)

Helvetica Chimica Acta ± Vol. 82 (1999)
2259
was added dropwise to a soln. of NaI (4.5 g, 0.03 mol) in dry MeCN (10 ml) under N2. After a few min, a soln. of
epoxy-a-irone (2.22 g, 0.01 mol) in MeCN (5 ml) was added. After stirring at r.t. for 30 min, the mixture was
poured into 4n sodium thiosulfate and extracted with AcOEt. The org. layer was dried (Na2SO4) and
evaporated, and the residue chromatographed (hexane): 1a ± d.
Data of ( )-trans-a-Irone (ˆ(E)-4-[(1R,5R)-2,5,6,6-Tetramethylcyclohex-2-en-1-yl]but-3-en-2-one; 1a):
Yield 60%. After bulb-to-bulb distillation, 88% chemical purity (GC). Chiral GC: tR 21.28; ee 98%. [a]D20
ˆ
1
400 (c ˆ 1.05, CH2Cl2) [7a]: [a]D20
ˆ
420 (c ˆ 0.98, CH2Cl2)). H-NMR: 6.67 (dd, J ˆ 15, 10, H C(7)); 6.01
(d, J ˆ 15, H C(8)); 5.47 (m, H C(4)); 2.30 (m, 1 H); 2.26 (s, MeCO); 2.05 (m, 1H); 1.71 (m, 1H), 1.64
(m, 1 H); 1.55 (m, Me C(5)); 0.88 (s, Me C(1)); 0.85 (d, J ˆ 7, Me C(2)); 0.82 (s, Me C(1)).
Data of (‡)-trans-a-Irone (ˆ(E)-4-[(1S,5S)-2,5,6,6-Tetramethylcyclohex-2-en-1-yl]but-3-en-2-one; 1b):
Yield 57%. After bulb-to-bulb distillation, 96% chemical purity (GC). Chiral GC: tR 21.72; ee 98%. [a]D20
ˆ
‡427 (c ˆ 0.95, CH2Cl2) ([9a]: [a]2D0 ˆ ‡432 (c ˆ 2.85, CH2Cl2; 86% ee)). 1H-NMR: in accordance with that of 1a.
Data of (‡)-cis-a-Irone (ˆ(E)-4-[(1S,5R)-2,5,6,6-Tetramethylcyclohex-2-en-1-yl]but-3-en-2-one; 1c):
Yield 63%. After bulb-to-bulb distillation, 81% chemical purity (GC). Chiral GC: tR 22.69; ee 98%. [a]D20
ˆ
‡117 (c ˆ 1.5, CH2Cl2) ([7a]: [a]2D0 ˆ ‡111 (c ˆ 0.92, CH2Cl2)). 1H-NMR: 6.65 (dd, J ˆ 15.7, 11, H C(7)); 6.12
(d, J ˆ 15.7,
H C(8)); 5.52 (m, H C(5)); 2.55 (m, 1 H); 2.28 (s, MeCO); 1.5 ± 1.8 (m, 2 H); 1.53
(m, Me C(5)); 1.46 (m, 1 H); 0.88 (d, J ˆ 6, Me C(2)); 0.86 (s, Me C(1)); 0.71 (s, Me C(1)).
Data of
(
)-cis-a-Irone (ˆ(E)-4-[(1R,5S)-2,5,6,6-Tetramethylcyclohex-2-en-1-yl]but-3-en-2-one; 1d):
Yield 55%. After bulb-to-bulb distillation, 85% chemical purity (GC). Chiral GC: tR 22.77; ee 98%. [a]D20
ˆ
130 (c ˆ 1.55, CH2Cl2) ([9a]: [a]D20
ˆ
103.7 (c ˆ 0.65, CHCl3; 86% ee)). 1H-NMR: in accordance with that of 1c.
REFERENCES
[1] a) G. Ohloff, ꢁScent and Fragrancesꢂ, Springer-Verlag, Berlin-Heidelberg, 1994, p. 164; b) p. 166.
[2] E. Guenther, in ꢁThe Essential Oilsꢂ, Vol. IV, Van Nostrand, New York, 1952, p. 79.
[3] B. Belcour, D. Courton, C. Ehret, V. Petiard, Phytochemistry 1993, 34, 1313.
[4] F. Tiemann, P. Kruger, Chem. Ber. 1893, 26, 2675; ibid. 1898, 31, 809, 870, 874, 1737.
[5] L. Ruzicka, C. F. Seidel, H. Scinz, M. Pfeiffer, Helv. Chim. Acta 1947, 30, 1807; H. Schinz, L. Ruzicka, C. F.
Seidel, C. Tavel, ibid. 1947, 30, 1810; ibid. 1948, 31, 257.
[6] Y. R. Naves, Helv. Chim. Acta 1948, 31, 151; ibid. 1948, 31, 1871.
[7] a) V. Rautenstrauch, G. Ohloff, Helv. Chim. Acta 1971, 54, 1776; b) V. Rautenstrauch, B. Willhalm, W.
Thommen, G. Ohloff, ibid. 1984, 67, 325.
[8] F. J. Marner, L. Jaenicke, Helv. Chim. Acta 1989, 72, 287 and ref. cit. therein.
Â
[9] a) C. Chapuis, R. Brauchli, Helv. Chim. Acta 1993, 76, 2070 and ref. 9 therein; b) D. Helmlinger, G. Frater,
ibid. 1989, 72, 1515.
[10] K. H. Schulte-Elte, H. Pamingle, A. P. Uijttewaal, R. Snowden, Helv. Chim. Acta 1992, 75, 759 and ref. 6
therein.
Â
[11] A. Galfre, P. Martin, M. Petrzilka, J. Essent. Oil Res. 1993, 5, 265.
[12] a) E. Brenna, C. Fuganti, P. Grasselli, M. Redaelli, S. Serra, J. Chem. Soc., Perkin Trans. 1 1998, 4129; b) J.
Aleu, E. Brenna, C. Fuganti, S. Serra, ibid. 1999, 271.
[13] R. Caputo, L. Mangoni, O. Neri, G. Palumbo, Tetrahedron Lett. 1981, 22, 3551.
[14] P. Uebelhart, A. Baumeler, H. Haag, R. Prewo, J. H. Bieri, C. H. Eugster, Helv. Chim. Acta 1986, 69, 816.
[15] C. Chapuis, K. H. Schulte-Elte, Helv. Chim. Acta 1995, 78, 165; C. Fehr, Angew. Chem., Int. Ed. 1998, 37,
2407.
[16] O. Mitsunobu, Synthesis 1981, 1.
[17] C. S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih, J. Am. Chem. Soc. 1982, 104, 7294.
[18] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, A. G. G. Moliteni,
G. P. Polidori, R. Spagna, ꢁSIR97: A New Tool for Crystal Structure Determination and Refinementꢂ, J.
Appl. Crystallogr. 1999, 32, 115.
[19] G. M. Sheldrick, ꢁSHELXL97. Release 97-2. Program for the Refinement of Crystal Structures, 1999ꢂ,
University of Göttingen, Germany.
Received July 23, 1999