Enantioselective divergent synthesis of (-)-cis-a- And (-)-cis-g-irone by using wilkinson's catalyst

Article abstract of DOI:10.1002/chem.201404642

A simple, efficient synthesis is reported for (-)-cis-a-and (-)-cis-g-irone, two precious constituents of iris oils, in ≥99% diastereomeric and enantioselective ratios. The two routes diverge from a common intermediate prepared from (-)-epoxygeraniol. Of general interest in this approach is the installation of the enone moiety of irones through a NHC-Au^I-catalyzed Meyer-Schuster-like rearrangement of a propargylic benzoate and the use of Wilkinson's catalyst for the stereoselective hydrogenation of a prostereogenic exocyclic double bond to secure the critical cis stereochemistry of the alkyl groups at C2 and C6 of the irones. The stereochemical aspects of this reaction are rationally supported by DFT calculation of the conformers of the substrates undergoing the hydrogenation and by a modeling study of the geometry of the rhodium ν² complexes involved in the diastereodifferentiation of the double bond faces. Thus, computational investigation of the ν² intermediates formed in the catalytic cycle of prostereogenic alkene hydrogenation by using Wilkinson's catalyst could be highly predictive of the stereochemistry of the products.

Full text of DOI:10.1002/chem.201404642

DOI: 10.1002/chem.201404642
Full Paper
&
Total Synthesis
Enantioselective Divergent Synthesis of (ꢀ)-cis-a- and (ꢀ)-cis-g-
Irone by Using Wilkinson’s Catalyst
Serena Bugoni, Debora Boccato, Alessio Porta,* Giuseppe Zanoni, and Giovanni Vidari*[a]
Dedicated to Professor Paul A. Grieco on the occasion of his 70th anniversary
Abstract: A simple, efficient synthesis is reported for (ꢀ)-cis-
a- and (ꢀ)-cis-g-irone, two precious constituents of iris oils,
in ꢁ99% diastereomeric and enantioselective ratios. The
two routes diverge from a common intermediate prepared
from (ꢀ)-epoxygeraniol. Of general interest in this approach
is the installation of the enone moiety of irones through
a NHCꢀAuI-catalyzed Meyer–Schuster-like rearrangement of
a propargylic benzoate and the use of Wilkinson’s catalyst
for the stereoselective hydrogenation of a prostereogenic
exocyclic double bond to secure the critical cis stereochem-
istry of the alkyl groups at C2 and C6 of the irones. The ste-
reochemical aspects of this reaction are rationally supported
by DFT calculation of the conformers of the substrates un-
dergoing the hydrogenation and by a modeling study of the
geometry of the rhodium h2 complexes involved in the dia-
stereodifferentiation of the double bond faces. Thus, compu-
tational investigation of the h2 intermediates formed in the
catalytic cycle of prostereogenic alkene hydrogenation by
using Wilkinson’s catalyst could be highly predictive of the
stereochemistry of the products.
Introduction
Irones are C14 nor-terpenoids, which impart a powerful and
pleasant violet-like scent to the essential oil of Iris rhizomes.[1]
They have a long history in the chemistry of natural products
and odorants. Indeed, from the 16th century, with the inven-
tion of the fragrance “Queen’s water” at the French royal court
of Catherine de’ Medici, the scent of Iris extracts, obtained
from orris roots through expensive and years-long processes,
have continued to impart a delicate violet tone to many ex-
pensive fragrances, perfumes, and cosmetics, even nowadays.[2]
For example, the oil of the Italian Iris pallida donates the boisꢀ
note to Armani Privꢁ Nuances, a high-priced unisex fragrance
launched by Giorgio Armani in 2013.
Figure 1. Components of Iris oil.
Three double-bond regioisomers a, b, and g (1–3; Figure 1)
are present in different proportions in most iris oils, and iris
oils of different origin show different enantiomeric composi-
tion of the three regioisomers.[3] cis-a-irone and cis-g-irone are
the major components of the oils, while trans-a-irone has
been found in minor amounts and trans-g-irone is even rar-
er.[3b,f] In addition, Iris plants of different geographical origin
afford oils with different enantiomeric composition of the
three regioisomers. For example, the dextrorotatory irones
characterize the oil of Italian Iris pallida varieties, whereas the
levorotatory enantiomers that obtained from Moroccan I. ger-
manica.[3d] Notably, the human nose is able to distinguish the
odor potency and characters of each of the ten irone iso-
mers.[3g] Thus, (ꢀ)-(2S,6R)-cis-a-irone 4 and (ꢀ)-(2S,6R)-cis-g-
irone 5 (Figure 1) have been determined to be stronger odor-
ants than their enantiomers and their trans-stereoisomers, and
(ꢀ)-cis-a-irone (4) shows the finest and strongest “orris-butter”
character.[4] Therefore, the synthesis of the single enantiomers
4 and 5 has long been a target for organic chemists. Fuganti
and co-workers were able to prepare the single stereoisomers
of the three regioisomeric irones, including 4 and 5, by using
the commercial product Irone Alpha. The process was, howev-
er, rather long and tedious, requiring multiple lipase-mediated
resolutions of racemic mixtures and separation of diastereo-
meric products.[4,5] Ex novo enantioselective synthesis of (ꢀ)-4
and (ꢀ)-5 was even more challenging, since the severe 1,3-al-
[a] S. Bugoni, D. Boccato, Dr. A. Porta, Prof. G. Zanoni, Prof. G. Vidari
Dipartimento di Chimica, Sezione Chimica Organica,
Universitꢂ degli Studi di Pavia, Via Taramelli 12, 27100 Pavia (Italy)
E-mail: alessio.porta@unipv.it
vidari@unipv.it
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201404642.
Chem. Eur. J. 2014, 20, 1 – 10
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
lylic interaction[6] between the substituents at C5 and C6 of the
irone skeleton required the invention of ingenious strategies
for installing the cis stereochemistry between H2 and H6.
Moreover, enantioenriched optimal starting materials were
seldom easily available. These difficulties account for the very
few syntheses of 4 and 5 published to date.[7]
Results and Discussion
We initially envisioned
a novel and simple approach
(Scheme 1) to install the critical cis stereochemistry of 4 and 5
Figure 2. Structure of model diene 9 and 3D representations of its calculated
boat (S1a–c) and twist-chair conformations (S2a–c). The relative energy
(kcalmolꢀ1) of each conformer is shown within parentheses.
prising three rotamers a–c resulting from free rotation around
the C6ꢀC7 bond (Figure 2).
However, on the basis of the relative energies, it can be con-
cluded that only conformers S2a–c were significantly populat-
ed at room temperature.
Scheme 1. Retrosynthesis of irones 4 and 5.
in the regioselective and stereoselective hydrogenation of the
exocyclic double bond of diene 6. An added value of this strat-
egy was the ready availability and excellent enantiomeric
purity of the starting material. In fact, diol (ꢀ)-7 can readily be
obtained at ꢁ99% ee by electrophilic cyclization of (ꢀ)-epoxy-
geraniol 8, followed by crystallization.[8]
Our assumption that hydrogen addition to the exocyclic
double bond of a diene of general formula 6 would deliver the
resulting methyl group at C2 in cis fashion to the substituent
at C6 was predicted by exploration of the conformational
space of model diene 9. In fact, we envisioned that, in the ab-
sence of significant electronic effects displayed by the silyloxy
substituent, the different steric hindrance of the two faces of
the exo double bond would be the predominant factor con-
trolling the reaction stereoselectivity. Indeed, calculations (see
below) suggested that the face syn to H6 was more accessible
than the anti face of the exo double bond, thus leading to the
desired stereochemistry.
Inspection of the most representative geometry of diene 9
clearly showed two sterically well-differentiated faces of the
exo double bond, with the lower (si,si’) face severely obstruct-
ed by the pseudo-axial silyloxymethyl group, while the upper
(re,re’) face was mainly hindered by the pseudoaxial methyl
group at C1. However, due to the bent shape of the molecule,
the endo approach to the double bond appeared to be more
difficult than the exo approach, indicating that the cis product
would be favored in addition reactions to the exocyclic double
bond of 9.
For synthetic purposes, a protecting group of the primary al-
cohol more robust than TMS was used. Conversion of diol 7 to
substrate 12 was carried out straightforwardly in three stan-
dard steps (Scheme 2). Regioselective protection of 7 with tert-
Modeling studies
Scheme 2. Conversion of diol 7 to exo olefin 12: a) TBDPSCl, Im, DMAP,
DCM, room temperature, 2 h, 100% yield; b) TPAP, NMO, 4 ꢁ molecular
sieves, DCM, room temperature, 3 h, 78% yield; c) Ph3PCH3Br, NaN(SiMe3)2,
THF, 708C, 20 h, 82% yield.
The conformational space of compound 9 was explored by
using the B3LYP functional. The symmetrical Me3Si group was
selected as the alcohol protecting group to facilitate the mod-
eling study. All atoms were described with the 6-31+G(d,p)
basis set,[9] with the polarizable continuum model.[10] Vibration-
al frequencies were computed at the same level of theory to
verify that the optimized structures were minima. Calculations
showed that diene 9 exists in two sets of low-energy confor-
mations, a boat-like S1 and a twist-chair-like S2, separated by
a gap of about 6 kcalmolꢀ1 (1 kcal=4.184 kJ) with each com-
butyldiphenylsilyl chloride (TBDPSCl) yielded silyl ether 10
quantitatively, which smoothly afforded ketone 11 in 78%
yield upon exposure to N-methylmorpholine N-oxide (NMO)
and catalytic tetra-n-propylammonium perruthenate (TPAP) in
dichloromethane (DCM).[11] Other oxidants such as pyridinium
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
chlorochromate (PCC) in DCM,[12] Py·SO3 in DMSO,[13] or stabi-
lized 2-iodoxybenzoic acid (SIBX) in DMSO,[14] gave lower
yields. Finally, Wittig reaction of 11 with the ylide generated
from methyltriphenylphosphonium bromide and NaN(SiMe3)2
in THF, slowly produced the exo olefin 12, isolated in 82%
yield. An analogous route, by substituting TBDPSCl with tert-
butyldimethylsilyl chloride (TBSCl) in the first step, afforded the
corresponding TBS ether 13 in similar overall yields.
DFT modeling studies (see below), we anticipated that com-
plexation of substrate 12 with the rhodium center to give the
active complex would be sterically demanding, resulting in
a highly preferred coordination of the more accessible exo face
(re,re’) of the less-hindered double bond of the diene, suggest-
ing a highly regioselective reaction. Moreover, we expected
the cis product 15 to be largely predominant, considering that,
in the mechanism of the Wilkinson’s hydrogenation,[15] hydro-
gen ligands are transferred in syn fashion from the rhodium
center to the coordinated double bond.
With dienes 12 and 13 in hand, we examined different pro-
tocols for adding hydrogen to the exo double bond in a regio-
selective and stereoselective fashion (Table 1). Diimide reduc-
In the event, this hypothesis was confirmed by the isolation,
in almost quantitative yield, of a mixture of 15 and 18, in
a ratio of 9.5:1 (GC; Table 1, entry 6), whereas no product from
hydrogen addition to the trisubstituted olefin of 12 was de-
tected. Wilkinson’s hydrogenation showed a slightly higher dia-
stereoselectivity in dichloromethane (Table 1, entry 7), afford-
ing 15 and 18 in a ratio of 10.8:1 (GC). In contrast, diastereose-
lectivity decreased for the hydrogenation of free alcohol 14 in
DCM (Table 1, entry 8), confirming the importance of the bulky
silyl group for sheltering the concave face of the disubstituted
double bond.
Table 1. Results of hydrogen addition to dienes 12–14.
Entry Substrate Reagents and cata- Solvent T [8C] Products Yield
lysts
(ratio)[a]
[%][b]
1
12
excess NH=NH from MeOH/ reflux 15/18 50
NH2NH2/H2O2
H2, 5% Rh/Al2O3
NiCl2·NaBH4
DCM
EtOAc
MeOH
(4:1)
[c]
2
3
13
13
22
0
95
16/19 90
(3:2)
Modeling study of epoxide 22 and synthesis of epoxides
23–25
4
5
6
7
8
12
12
12
12
14
NiCl2·NaBH4
MeOH
EtOAc
0
15/18[c] 90
(3:2)
Although quite satisfactory, we considered the results of the
hydrogenation of dienes 12–14 merely a confirmation of our
hypothesis that steric factors encoded in the bent molecular
shape and the pseudo-axial orientation of the silyoxymethyl
substituent strongly determined the stereochemistry of prod-
ucts, mainly resulting from hydrogen addition to the disubsti-
tuted double bond of the substrate. By this reasoning, we ex-
pected that the cis diastereoselectivity of the hydrogenation
would be improved significantly by increasing the steric hin-
drance around the endo (si,si’) face of the double bond. The
4,5-epoxides 24 and 25 were identified as the best candidates
to test this hypothesis. In fact, the epoxy ring could be instal-
led with high a-diastereoselectivity and regioselectivity on the
trisubstituted double bond of diene 14.[16] Moreover, after the
key hydrogenation step, the epoxy ring could be removed to
restore the original double bond or serve as a useful synthetic
linchpin (see below).
H2, 10% Pd/C
H2, [(Ph3P)3RhCl]
H2, [(Ph3P)3RhCl]
H2, [(Ph3P)3RhCl]
22
15/18[e] 95
(3.7:1)
benzene 22
15/18 97
(9.5:1)
15/18 96
(10.8:1)
17/20 96
(7.3:1)
DCM
DCM
22
22
[a] Product ratio was determined by GC; [b] yield of isolated product;
[c] mixture of tetrahydrostereoisomers; [d] accompanied by minor
amounts of over-reduced products; [e] accompanied by 20% 21.
tion of 12 (Table 1, entry 1) was completely regioselective, re-
sulting in the production of the cis (15) and trans (18) products
in the ratio of 4:1; however, the conversion was largely incom-
plete in spite of using excess reagent. Hydrogenation of 13
over 5% Rh on Al2O3 (Table 1, entry 2) was apparently highly
cis-stereoselective (GC); however, at the end of the reaction,
mainly the fully saturated product was formed, as an undeter-
mined mixture of diastereomers. Exposure of 13 to nickel
boride in MeOH (Table 1, entry 3) afforded a 60:40 mixture of
cis and trans dihydroderivatives, 16 and 19, while reduction of
12 with the same reagent (Table 1, entry 4) gave, in addition to
15 and 18, a minor amount of unidentified over-reduced prod-
ucts. Hydrogenation of 12 in EtOAc over 10% Pd on carbon
(Table 1, entry 5) afforded a 3.7:1 mixture of cis and trans dihy-
droderivatives, accompanied by 20% fully hydrogenated
products 21.
To gain some insight into the stereochemical features of ep-
oxides 24 and 25, we carried out a modeling study on the sim-
pler homologue 22. Calculations showed that 22 existed, in
analogy with parent diene 9, in two sets of low-energy confor-
mations, a boat-like S3 and a twist-chair-like S4, with an
energy gap higher than 5 kcalmolꢀ1, and each comprising
three rotamers a--c resulting from the free rotation around the
C6ꢀC7 bond (Figure 3). However, the relative energies clearly
indicated that only conformers S4a–b significantly represented
the geometry of epoxide 22 at room temperature. We consid-
ered that the preference for this geometry should be largely
independent of the silyl protecting group of the primary alco-
hol and, for synthetic purposes, we continued our work on
TBDPS- or TBS-protected epoxides, 24 and 25, respectively.
Finally, diene 12 was hydrogenated at room temperature in
the presence of Wilkinson’s catalyst [(Ph3P)3RhCl] (0.3 equiv) in
carefully degassed benzene (Table 1, entry 6). On the basis of
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 2. Results of hydrogenation of epoxides 23–25.
Entry Substrate Reagents and Solvent
catalysts
T
t
Products
Yield
[%][b]
[8C] [h] (ratio)[a]
1
2
3
4
5
25
24
24
24
23
H2, 5% Rh/
Al2O3
H2, 10% Pd/C
EtOAc
EtOAc
22 24
22 24
28:31
(3.5:1)
27:30
(4.3:1)
27:30
95
N.D.[c]
94
H2,
benzene, 22
22
DCM
7
7
7
[(Ph3P)3RhCl]
H2,
[(Ph3P)3RhCl]
H2,
(ꢁ99:1)
22
27:30
95
(ꢁ99:1)
26:29
(91:9)
DCM
22
96
[(Ph3P)3RhCl]
[a] Product ratio was determined by GC; [b] yield of isolated product;
[c] yield was not determined. Expected products were accompanied by
considerable amounts of compounds in which the epoxy ring was
cleaved.
Figure 3. Structure of model epoxide 22 and 3D representations of its mod-
eled boat (S3a–c) and twist-chair conformations (S4a–c). The relative
energy (kcalmolꢀ1) of each conformer is shown between parentheses.
In the folded shape of compounds 24 and 25 the exocyclic
methylene group was thus projected inside the molecular
cavity even more deeply than in parent dienes 12 and 13, re-
spectively. This geometry and the additional steric hindrance
created by the a-epoxy ring, were expected to hamper hydro-
gen addition to the si,si’ face of the double bond of com-
pounds 24 and 25 more efficiently than for dienes 12 and 13.
This would result in a higher proportion of 2,6-cis products. On
the basis of this assumption, diene 12 was converted to 24
and 25 in a few simple steps (Scheme 3); however, to secure
entry 1) and epoxide 24 with 10% Pd/C catalyst (Table 2,
entry 2) proceeded with higher diastereoselectivity than with
dienes 13 and 12 (Table 1, entries 2 and 5), respectively, as ex-
pected; however, the diastereomeric purities of cis diastereo-
mers 28 and 27 were still unacceptably low.
In striking contrast, hydrogenation of 24 by using Wilkin-
son’s catalyst (0.3 equiv) afforded the desired cis product 27 in
an excellent ꢁ99:1 diastereomeric ratio (GC), either in benzene
(Table 2, entry 3) or in DCM (Table 2, entry 4). The relative ste-
reochemistry of epoxide 27 was fully confirmed by the NOESY
spectrum (see the Supporting Information) and the subse-
quent conversion to known alcohols 17 and 43 (see below). In
contrast, as previously observed with diene 14, Wilkinson’s hy-
drogenation of the free alcohol 23 occurred with high but sig-
nificantly lower diastereoselectivity (Table 2, entry 5) than with
the bulky silyl ether 24.
Modeling study of the Wilkinson’s hydrogenation
Scheme 3. Conversion of diene 12 to epoxides 23–25: a) TBAF, THF, room
temperature, overnight, 100% yield. b) m-CPBA, DCM, 08C, 3 h, 74% yield.
c) TBDPSCl, Im, DMAP, CH2Cl2, room temperature, 2 h, 100% yield. d) TBSCl,
Im, CH2Cl2, room temperature, 2 h, 100% yield.
To shed further light on the stereochemical aspects of the Wil-
kinson’s hydrogenation of epoxide 24 and diene 12, at first we
carried out DFT modeling studies on the complexes formed by
the rhodium catalyst with the model epoxide 22. Calculations
were carried out by DFT at the B3LYP[9] and M06[17] levels with
a differentiated basis set in benzene as a solvent with polariza-
ble continuum model.[10] C, H, and O atoms were described by
using the 6-31G(d) basis set, whereas Cl, Si, and P atoms were
described with the 6-31+G(d,p) basis set and Rh atom with
the LANL2DZ[18] relativistic effective core potential basis set, by
using the Gaussian 09 software suite.[19]
syn-epoxidation of the D4 double bond to give the a-epoxide
23, 12 had to be deprotected to the free alcohol 14. As antici-
pated,[16] epoxidation of 14 was completely regio- and diaste-
reoselective, and configuration of the a-epoxide was con-
firmed by the NOESY spectrum of subsequent compound 27.
Hydrogenation of epoxides 23–25
Several different reaction mechanisms have been proposed
for the Wilkinson’s hydrogenation catalytic cycle, of which two
“hydride route” mechanisms are believed to be the predomi-
nant pathways.[15] However, the absence of a solvent effect on
Epoxides 23–25 were then submitted to the key hydrogena-
tion reaction of the exocyclic double bond (Table 2). Both the
reaction of epoxide 25 with 5% Rh/Al2O3 catalyst (Table 2,
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 4. Hydride route mechanism proposed for the Wilkinson’s hydroge-
nation of alkenes.[15]
the reaction stereoselectivity (see above) and the steric hin-
drance of ligands tend to support the mechanism in which no
solvent molecule is involved and the two phosphine ligands
are cis oriented in the key h2 intermediate 32 formed by coor-
dination of a terminal olefin to rhodium (Scheme 4).[15] More-
over, computational studies on the h2 coordination complex of
a model epoxide mimicking 22 (see the Supporting Informa-
tion) indicated that the chloro ligand is spatially oriented far
from the bulky geminal dimethyl group in the octahedral dihy-
dride alkene complex. We assumed that such geometry is pre-
served also in the complexes with epoxides 22 and 24 (see
below). According to the proposed mechanism (Scheme 4), the
first four steps leading to 32 are considered to be fast equili-
bria, whereas the rate-determining alkene insertion (32!33)
and the fast alkyl reductive elimination (33!34) are essentially
irreversible steps.[15] In the case of a prostereogenic olefin, such
as 22, the stereochemistry of hydrogenated products is deter-
mined by the olefin recognition step, namely by the face of
the double bond that coordinates to rhodium in the key 32-
type h2 intermediate. In fact, in both the alkene insertion
(32!33) and the reductive elimination steps (33!34), the
two hydrogens add in syn fashion to the face of the double
bond that is initially coordinated to the rhodium center
(Scheme 4).
Figure 4. Computed exo and endo h2 complexes of epoxide 22 in the twist-
chair (35a) and boat conformations (35b and c).
conformation of 22 (Figure 3); 35b (Figure 4), resulting from
complex formation on the re,re’ face of the boat conformation
of 22 (Figure 3); 35c (Figure 4), resulting from complex forma-
tion on the si,si’ face of the boat conformation of 22 (Figure 3).
Therefore, both 35a and 35b would give rise to the cis-hydro-
genated product 36 at the end of the catalytic cycle, while the
trans-hydrogenated product 37 would originate from the com-
plex 35c.[20] Then, it was reasonable to assume that the steric
and conformational effects present in the h2 intermediates of
type 32, likely occurred also in the transition states of the
olefin insertion step (32!33). Indeed, the calculated transition
states for the Wilkinson’s hydrogenation of epoxide 22 were of
the early type and showed geometries very similar to the start-
ing complexes (data not included). As a consequence, the dif-
ference in energy between the competitive transition states of
olefin insertion was comparable with the relative energy of the
h2 intermediates. In this scenario, according to the Curtin–
Hammett principle,[21] the diastereoselectivity of the hydroge-
nation of the model epoxide 22 by using Wilkinson’s catalyst
could confidently be predicted on the basis of the relative sta-
bility of the h2 complexes.
Considering the two faces of the exo olefin and the two
types of the most-populated conformations of epoxide 22
(Figure 3), four families of h2 complexes could in principle be
formed, each family comprising three different rotamers. How-
ever, a finite value of the energy could be calculated for only
three types of h2 complexes. In fact, rhodium coordination to
the si,si’ (endo) face of the twist-chair conformation of epoxide
22 was predicted not to form a stable complex. All of the
stable complexes were modeled (see the Supporting Informa-
tion); however, in Figure 4 we depict, as an example, only the
geometries of the complexes between rhodium and the most
stable rotamer of the two conformers:[20] 35a (Figure 4), result-
ing from complex formation on the re,re’ face of the twist-chair
The energies of the different complexes 35a–c in the gas
phase and in benzene solution were thus determined at the
B3LYP[9] and M06[17] levels (Table 3).[20]
The calculations clearly indicated that 35a would be, to
a great extent, the most representative geometry of the h2 in-
termediate in the Wilkinson’s hydrogenation of epoxide 22,
whereas 35b or 35c would play an insignificant role or no role
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 3. Energies E (aU) and relative energies DE (kcalmolꢀ1) of complexes
35a–c calculated in gas phase and in benzene as a solvent.[a]
Complex E[b] (DE)
E[c] (DE)
E[d] (DE)
E[e] (DE)
35a
35b
35c
ꢀ3632.66472 ꢀ3632.674516 ꢀ3630.953748 ꢀ3630.963414
(0) (0) (0) (0)
ꢀ3632.653793 ꢀ3632.664076 ꢀ3630.936633 ꢀ3630.946957
(+6.8) (+6.5) (+10) (+10)
ꢀ3632.654990 ꢀ3632.665330 ꢀ3630.924848 ꢀ3630.935827
(+6.1) (+5.8) (+18) (+17)
Figure 6. Computed exo and endo h2 complexes, 39a and b, respectively, of
diene 12 in the twist-chair conformation with Wilkinson’s catalyst.
[a] Cartesian Coordinates and complete Boltzmann distribution analyses of
all structures are included in the Supporting Information; [b] calculated in
gas phase at RB3LYP level; [c] calculated in benzene at RB3LYP level;
[d] calculated in gas phase at M06 level; [e] calculated in benzene at M06
level.
twist-chair conformation of substrate 12 was considered. Car-
tesian coordinates and energies (aU) are reported in the Sup-
porting Information.
The calculated energy difference between the exo and endo
complexes 39a and 39b was 1.3 kcalmolꢀ1, which correspond-
ed to a product ratio of 9:1 (15/18), in good agreement with
the experimental data obtained for the Wilkinson’s hydrogena-
tion of diene 12 (Table 1, entries 6 and 7).
at all in the catalytic cycle. Therefore, the computational inves-
tigation on h2 complexes of model epoxide 22 was highly indi-
cative of the factors determining the almost complete cis dia-
stereoselectivity for the homogeneous hydrogenation of epox-
ide 24. To further support the experimental data, we then ex-
tended the modeling study to epoxide 24 itself, starting from
the geometries found for the model 22.
Final steps of the syntheses of (ꢀ)-a-irone 4 and (ꢀ)-g-irone
5
Only the h2 intermediates 38a and 38b (Figure 5), resulting
from complex formation on the exo and endo faces, respective-
With epoxide 27 securely in hand, the synthesis of irones 4
and 5 was completed through two straight divergent routes
(Scheme 5). Isomerization of epoxide 27 to the corresponding
allylic alcohol 40 was realized in excellent yield upon exposure
to excess Al(OiPr)3 in refluxing xylene.[16b] After acetylation of
40 to acetate 41, the acetoxy group was removed by Pd-medi-
ated hydride reduction to afford protected alcohol 42 in high
yield.[22] Deprotection of 42 under standard conditions yielded
free alcohol 43, identical with the literature,[7b,d] in ꢁ99% dia-
stereomeric and enantiomeric purity by chiral HPLC analysis.
Conversion of compound 43 to (ꢀ)-g-irone 5 was then accom-
plished, with conservation of the stereoisomeric purity, by fol-
lowing a well-established procedure.[7d,e]
Figure 5. Computed exo and endo h2 complexes, 38a and b, respectively, of
epoxide 24 in the twist-chair conformation with Wilkinson’s catalyst.
To prepare an advanced intermediate for the synthesis of
(ꢀ)-a-irone 4, we first attempted direct isomerization of the
exo olefin 42 to the thermodynamically more stable trisubsti-
tuted endo olefin 15. However, upon treating 42 with I2 in
DCM or with H3PO4 in toluene at reflux, the desired compound
was produced in unacceptable yields. In contrast, de-epoxida-
tion of 27 by use of Zn powder and NaI,[23] followed by cleav-
age of the silyl protecting group by using a standard protocol,
delivered free alcohol 17, identical with the literature,[7b] in
76% overall yield, and ꢁ99% e.r. and d.r. (GC).
ly, of the double bond in the largely most populated twist-
chair conformer of compound 24, were considered. Rather un-
expectedly, contrary to the results found for the model epox-
ide 22, the endo complex 38b also converged to a minimum
(imaginary frequencies=0) and its energy could be estimated.
Indeed, the change of the TMS with the TBDPS group was ac-
companied by a significant change in the geometric parame-
ters of the endo complex (see Supporting Information), so that
severe repulsive interactions between the phosphine ligands
and the substituents at silicon were partially alleviated. Howev-
er the calculated energy difference between 38a and 38b re-
mained very large (4.36 kcalmolꢀ1), in full agreement with the
ratio ꢁ99:1 of cis and trans hydrogenated products, 27 and
30, formed from epoxide 24 (Table 2, entries 3 and 4).
It is well known that construction of an enone side chain,
such as that present in a-irone, from a 17-like alcohol precur-
sor is not a trivial process. In fact, Wittig-like olefinations of
corresponding 44-type aldehydes, under non-racemizing con-
ditions, proceeded with unacceptably low yields.[8] Moreover,
aldol homologation of aldehyde 44 with Me2CO and EtONa oc-
curred with considerable erosion of the cis stereochemistry.[7b]
To circumvent this problem, a few years ago we resorted to an
alternative methodology based upon a Julia–Lythgoe olefina-
tion, although the synthetic sequence was significantly length-
For comparison, the two h2 complexes 39a and 39b
(Figure 6), resulting from coordination of the rhodium center
to the re,re’ and si,si’ faces, respectively, of the exocyclic
double bond of diene 12, were also submitted to modeling
studies. To simplify the calculations, only the most populated
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
yield, a chromatographically inseparable mixture of (E)-(ꢀ)-a-
irone 4 and (E)-(ꢀ)-b-irone 2, in the ratio of 7:1 by NMR analy-
sis. GC analysis of chemically pure (E)-(ꢀ)-a-irone 4 indicated
no loss of stereochemistry from starting alcohol 17.
Conclusion
In conclusion, the extremely precious components of iris oils,
(ꢀ)-(2S, 6R)-cis-a-irone (4) and (ꢀ)-(2S, 6R)-cis-g-irone (5) have
been obtained in enantiomeric and diastereomeric ratios of
ꢁ99:1 by following two simple synthetic pathways diverging
from common epoxide 27. This intermediate was prepared
straightforwardly from (ꢀ)-epoxygeraniol 8 through the cyclic
diol (ꢀ)-7. Since the enantiomer (+)-7 is obtainable from
(+)-epoxygeraniol,[8] this synthetic pathway formally consti-
tutes also the synthesis of the antipodes (+)-(2R, 6S)-cis-a-
irone and (ꢀ)-(2R, 6S)-cis-g-irone. Of general interest in our ap-
proach are the installation of the enone moiety of irones
through an NHCꢀAuI-catalyzed Meyer–Schuster-like rearrange-
ment[24] of a propargylic benzoate and the use of Wilkinson’s
catalyst for the stereoselective hydrogenation of a prostereo-
genic exocyclic double bond to secure the critical cis stereo-
chemistry of the alkyl groups at C2 and C6 of irones. The ste-
reochemical aspects of this reaction were rationally supported
by DFT calculation of the conformers of the substrates under-
going the hydrogenation and by modeling study of the ge-
ometry of the rhodium h2 complexes involved in the diastereo-
differentiation of the double bond faces. The almost complete
diastereoselectivity predicted by the calculations was con-
firmed well by the experimental data. They indicated that com-
putational investigation of the h2 intermediates formed in the
catalytic cycle of alkene hydrogenation by using Wilkinson’s
catalyst could be highly predictive of the stereochemistry of
products. In addition, our results support the hydride route
mechanism with isomerization, proposed for homogeneous
hydrogenation of alkenes by using Wilkinson’s catalyst.[15]
Scheme 5. Synthesis of (ꢀ)-(2S,6R)-cis-a-irone 4 and (ꢀ)-(2S,6R)-cis-g-irone 5
from epoxide 27: a) Al(OiPr)3, xylene at reflux, 12 h, 95%; b) Ac2O, pyridine,
DMAP, DCM, room temperature, 2 h, 98%. c) [Pd(PPh3)4], Et3N, HCO2H, THF
at reflux, 24 h, 83% yield; d) TBAF, THF, room temperature, overnight, 100%
yield; e) Zn, NaOAc, NaI, AcOH, DCM, 408C, 24 h, 76% yield; f) TBAF, THF,
room temperature, overnight, 100% yield; g) SIBX, DMSO, room tempera-
ture, 2 h; h) 1-propynil-magnesium bromide, THF, ꢀ788C!08C, 2 h, fol-
lowed by BzCl, DMAP, 08C!258C, 2 h, 63% overall yield from 17;
i) [(IprAu)2(m-OH)]BF4, butan-2-one - H2O, 100:1, 608C, 12 h, 68% total yield
of 4+46, 70:15.
ened.[8] More recently, we introduced an innovative methodol-
ogy involving an NHCꢀAuI (NHC=N-heterocyclic carbene) cat-
alyzed Meyer–Schuster-like rearrangement of propargylic
esters,[24] which was readily employed for the synthesis of the
enone moiety of a-ionone.[25] To extend this procedure also to
the synthesis of a-irone, we oxidized alcohol 17 with SIBX[14] to
the highly epimerizable aldehyde 44, which, without purifica-
tion, was immediately submitted to the addition of propynyl-
magnesium bromide in THF. Quenching the resulting alkoxy
derivative with benzoyl chloride (BzCl), in the presence of a cat-
alytic amount of 4-dimethylaminopyridine (DMAP), produced
propargyl benzoate 45 in 63% yield (see the Supporting Infor-
mation for the detailed procedure). The ester 45 was then ex-
posed to the Nolan dinuclear NHCꢀAuI catalyst [(IprAu)2(m-
The novel synthetic pathway to enantioenriched cis-a-irone
and cis-g-irone disclosed herein competes favorably with exist-
ing synthetic routes in terms of simplicity and stereoselectivity.
Experimental Section
Wilkinson’s hydrogenation of epoxide 24 to give tert-butyl-
diphenyl(((1S,2R,4S,6R)-1,3,3,4-tetramethyl-7-oxabicy-
clo[4.1.0]heptan-2-yl)methoxy)silane (27).
[(PPh3)3RhCl] (206 mg, 0.2 mmol, 0.3 equiv) was added to a solution
of epoxide 24 (313 mg, 0.7 mmol, 1.0 equiv) in degassed dry di-
chloromethane (7.4 mL). The mixture was exposed to H2
(101325 Pa) under magnetic stirring at room temperature for 7 h.
After removal of volatiles under reduced pressure, the residue was
separated by column chromatography (150 mmꢂ30 mm) on silica
gel (15 g; eluent=2% Et2O/hexane), affording epoxide 27 (296 mg,
yield=95%) as a colorless viscous oil. Rf =0.31 (2% Et2O/hexane);
½aꢂ2D2 =ꢀ12.3 cm3 gꢀ1 dmꢀ1 (c=1.3 in CH2Cl2); 1H NMR (300 MHz,
CD2Cl2): d=7.75–7.55 (m, 4H), 7.50–7.35 (m, 6H), 4.12–3.78 (m,
2H), 2.92 (d, J=2.9 Hz, 1H), 1.93–1.84 (m, 1H), 1.73–1.68 (m, 1H),
1.59–1.53 (m, 1H), 1.45 (s, 3H), 1.35–1.25 (m, 1H), 1.09 (s, 9H), 0.82
[26]
OH)]BF4 in a 100:1 butan-2-one/H2O solvent mixture, at 608C
for 12 h. A 4.7:1 mixture of (E)-(ꢀ)-a-irone 4 and (Z)-(ꢀ)-a-irone
46[7a] was delivered in 68% combined yield. The two diastereo-
mers were cleanly separated by chromatography on silica gel,
and the (Z)-isomer 46 was then treated with I2 in DCM to
induce Z!E isomerization of the enone unit.[27] The reaction
was completed in 2 h; however, it produced, in quantitative
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
(s, 3H), 0.78 (d, J=5.2 Hz, 3H), 0.65 ppm (s, 3H); 13C NMR (75 MHz,
CD2Cl2): d=136.5 (d), 136.4 (d), 134.5 (s), 130.4 (d), 130.3 (d), 128.4
(d), 64.2 (t), 62.2 (d), 60.5 (s), 52.1 (d), 38.8 (d), 35.2 (s), 30.7 (t), 29.2
(q), 27.4 (q), 25.1 (q), 19.7 (s), 16.7 (q), 16.2 ppm (q); IR (oil film on
Acknowledgements
We thank the Italian MIUR (funds PRIN) for financial support,
and Professor Mariella Mella for the NMR spectra measure-
ment.
~
NaCl disks): n=2962, 1472, 1428, 1390, 1112, 856, 824, 740,
701 cmꢀ1; HRMS (EI): m/z calcd for C27H38O2Si: 422.2641; found
422.2645.
The diastereomeric ratio (d.r.) ꢁ99:1 of compound 27 was deter-
mined by GC analysis on a capillary HP5 column (30 m, 0.25 mm
i.d., 0.25 mm f.t.); carrier gas=He, flow=1 mLminꢀ1; injector tem-
perature=2508C; detector: MS; temperature program: 808C
(1 min), then 108Cminꢀ1 to 2808C (5 min); tR of cis compound 27=
21.44 min; tR of trans compound 30=21.04 min.
Keywords: fragrances · hydrogenation · terpenoids · total
synthesis · Wilkinson’s catalyst
[1] a) A. Tschirsch in Handbuch der Pharmakognosie, C. H. Tachnitz Verlag,
Leipzig, 1917, p. 1143; b) G. Ohloff, Scent and Fragrances; Springer-
Verlag, Berlin, Heidelberg, 1994, p. 164; c) L. Jaenicke, F. J. Marner, Pure
Appl. Chem. 1990, 62, 1365–1368.
[2] a) D. H. Pybus in The Chemistry of Fragrances (Eds.: D. H. Pybus, C. S.
Sell), Royal Society of Chemistry, Cambridge, 1999, p. 14; b) G. Ohloff,
W. Pickenhagen, P. Kraft, Scent and Chemistry: The Molecular World of
Odors, Wiley-VCH, Weinheim, 2012, p. 214.
Meyer–Schuster rearrangement[24] of propargylic benzoate
45.
[3] a) W. Krick, F. J. Marner, L. Jaenicke, Helv. Chim. Acta 1984, 67, 318–323;
b) V. Rautenstrauch, B. Willhalm, W. Thommen, G. Ohloff, Helv. Chim.
Acta 1984, 67, 325–331; c) F. J. Marner, L. Jaenicke, Helv. Chim. Acta
1989, 72, 287–294; d) F. J. Marner, T. Runge, W. A. Kçnig, Helv. Chim.
Acta 1990, 73, 2165–2170; e) A. Galfrꢃ, P. Martin, M. Petrizlka, J. Essent.
Oil Res. 1993, 5, 265–277; f) E. Brenna, C. Fuganti, S. Ronzani, S. Serra,
Helv. Chim. Acta 2001, 84, 3650–3666; g) E. Brenna, C. Fuganti, S. Serra,
Tetrahedron: Asymmetry 2003, 14, 1–42.
[(IprAu)2OH]BF4 (2.7 mg, 0.002 mmol, 0.02 equiv) was added to a so-
lution of ester 45 (33 mg, 0.1 mmol, 1.0 equiv) in a 100:1 butan-2-
one/H2O solvent mixture (1.1 mL/0.011 mL). The mixture was stirred
and heated at 608C for 12 h. The solvent was then removed under
reduced pressure (P>10.66 kPa). The residue, composed by a mix-
ture of E and Z irones, was separated by column chromatography
(150 mmꢂ15 mm) on silica gel (5 g; eluent=1% Et2O/pentane), af-
fording (Z,2S,6R)-(ꢀ)-cis-a-irone 46 (2.6 mg, 12%) as a colorless oil,
followed by the desired (E,2S,6R)-(ꢀ)-cis-a-irone 4 (12.3 mg, 56%)
as a colorless oil.
[4] E. Brenna, C. Fuganti, S. Serra, C. R. Chim. 2003, 6, 529–546.
[5] E. Brenna, C. Fuganti, S. Serra, Chem. Soc. Rev. 2008, 37, 2443–2451,
and references therein.
[6] R. W. Hoffmann, Chem. Rev. 1989, 89, 1841–1860.
[7] (ꢀ)-cis-a-Irone: a) Y. Ohtsuka, F. Itoh, T. Oishi, Chem. Pharm. Bull. 1991,
39, 2540–2544; b) C. Chapuis, R. Brauchli, Helv. Chim. Acta 1993, 76,
2070–2088; c) T. Inoue, H. Kiyota, T. Oritani, Tetrahedron: Asymmetry
2000, 11, 3807–3818; (ꢀ)-cis-g-Irone: references [7b] and [7c]; d) S. Bes-
zant, E. Giannini, G. Zanoni, G. Vidari, Eur. J. Org. Chem. 2003, 3958–
3968; Formal synthesis: e) G. Laval, G. Audran, J.-M. Galano, H. Monti, J.
Org. Chem. 2000, 65, 3551–3554.
[8] M. Bovolenta, F. Castronovo, A. Vadalꢄ, G. Zanoni, G. Vidari, J. Org.
Chem. 2004, 69, 8959–8962.
[9] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. Lee, W. Yang,
R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[10] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999–3093.
[11] W. P. Griffith, S. V. Ley, G. P. Whitcombe, A. D. White, J. Chem. Soc. Chem.
Commun. 1987, 1625–1627.
[12] E. J. Corey, J. W. Suggs, Tetrahedron Lett. 1975, 16, 2647–2650.
[13] J. R. Parikh, W. von E Doering, J. Am. Chem. Soc. 1967, 89, 5505–5507.
[14] A. Ozanne, L. Pouysꢃgu, D. Depernet, B. Francois, S. Quideau, Org. Lett.
2003, 5, 2903–2906.
[15] D. J. Nelson, R. Li, C. Brammer, J. Org. Chem. 2005, 70, 761–767.
[16] a) C. Fehr, Angew. Chem. Int. Ed. 1998, 37, 2407–2409; Angew. Chem.
1998, 110, 2509–2512. b) M. Luparia, P. Boschetti, F. Piccinini, A. Porta,
G. Zanoni, G. Vidari, Chem. Biodiversity 2008, 5, 1045–1057.
(E,2S,6R)-(ꢀ)-cis-a-irone (4): The enantiomeric (e.r.) and diastereo-
meric ratio (d.r.) ꢁ99:1 of compound 4 were determined by GC
analysis on a capillary Mega DEX column (20 m, 0.25 mm i.d.,
0.18 mm f.t.); dacnPentilSilBETACDX as stationary phase; carrier
gas=He, split 1:30, flow=1 mLminꢀ1
; injector temperature=
2508C; detector: FID; temperature program: 808C (1 min), then
108Cminꢀ1 to 2808C (5 min); tR =10.36 min. Rf =0.28 (pentane-
Et2O, 99:1). ½aꢂ2D2 =ꢀ109.9 cm3 gꢀ1 dmꢀ1 (c=0.42 in CH2Cl2);[7a]
½aꢂ2D1 =ꢀ119.2 cm3 gꢀ1 dmꢀ1 (c=0.25 in CH2Cl2);[7b] ½aꢂ2D0 =ꢀ103.7
(c=0.65 in CHCl3; 86% ee);[4] ½aꢂD20 =ꢀ130 cm3 gꢀ1 dmꢀ1 (c=1.55 in
CH2Cl2; chemical purity=85%, 98% ee)]; 1H NMR (300 MHz,
CD2Cl2): d=6.65 (dd, J=15.9 and 10.8 Hz, 1H), 6.12 (d, J=15.8 Hz,
1H), 5.56 (m, 1H), 2.60 (br d, J=10.0 Hz, 1H), 2.25 (s, 3H), 2.05–
1.95 (m, 1H), 1.87–1.70 (m, 1H), 1.53 (d, J=1.6 Hz, 3H), 1.55–1.45
(m, 1H), 0.88 (d, J=6.5 Hz, 3H), 0.87 (s, 3H), 0.71 ppm (s, 3H);
13C NMR (75 MHz, CD2Cl2): d=198.8 (s), 150.1 (d), 135.5 (d), 133.2
(s), 123.9 (d), 57.0 (d), 39.0 (d), 36.7 (s), 32.8 (t), 27.9 (q), 27.7 (q),
~
23.7 (q), 16.2 (q), 15.8 ppm (q); IR (oil film on NaCl disks):n=2965,
1675, 1618, 1364, 1254, 966 cmꢀ1; HRMS (EI): m/z calcd for C14H22O
1
206.1671; found 206.1674. The H NMR and 13C NMR spectroscopic
[17] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215–241.
[18] a) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270–283; b) P. J. Hay,
W. R. Wadt, J. Chem. Phys. 1985, 82, 299–310; c) W. R. Wadt, P. J. Hay, J.
Chem. Phys. 1985, 82, 284–298.
data of compound 4 are in accordance with those reported in the
literature.[7a,b]
22
(Z,2S,6R)-(ꢀ)-cis-a-irone (46): Rf =0.32 (1% Et2O/pentane); ½aꢂD
=
ꢀ39 cm3 gꢀ1 dmꢀ1 (c=1.3 in CH2Cl2) ;[7a] ½aꢂ2D1 =ꢀ37.7 cm3 gꢀ1 dmꢀ1
[19] Gaussian 09, Revision B.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith,
R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dan-
1
(c=0.83 in CH2Cl2)]; H NMR (300 MHz, CD2Cl2): d=6.37 (d, J=11.8,
Hz, 1H), 5.96 (t, J=11.8 Hz, 1H), 5.51 (m, 1H), 3.98 (br d, J=11.7,
Hz, 1H), 2.23 (s, 3H), 1.94–1.86 (m, 1H), 1.75–1.67 (m, 1H), 1.55 (d,
J=1.5 Hz, 3H), 1.35–1.25 (m, 1H), 0.89 (d, J=6.8 Hz, 3H), 0.87 (s,
3H), 0.71 ppm (s, 3H); 13C NMR (75 MHz, CD2Cl2): d=199.9 (s),
148.5 (d), 133.7 (s), 130.4 (d), 122.8 (d), 50.2 (d), 38.8 (d), 36.7 (s),
32.7 (t), 32.5 (q), 26.9 (q), 23.0 (q), 15.0 (q), 14.6 ppm (q); HRMS (EI):
1
m/z calcd for C14H22O 206.1671; found 206.1675. The H NMR spec-
trum of compound 46 is in complete accordance with the
literature.[7a]
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
nenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2010.
[20] From a qualitative point of view, the diastereoselectivity of the Wilkin-
son’s hydrogenation predicted by our modeling studies did not change
significantly upon taking into consideration also the complexes of rho-
dium with minor rotamers of epoxide 22.
S. P. Nolan, Tetrahedron 2009, 65, 1767–1773; c) D. A. Engel, G. B.
Dudley, Org. Biomol. Chem. 2009, 7, 4149–4158.
[25] V. Merlini, S. Gaillard, A. Porta, G. Zanoni, G. Vidari, S. P. Nolan, Tetrahe-
dron Lett. 2011, 52, 1124–1127.
[26] S. Gaillard, J. Bosson, R. S. Ramon, P. Nun, A. M. Z. Slawin, S. P. Nolan,
Chem. Eur. J. 2010, 16, 13729–13740.
[21] E. L. Eliel, S. H. Wilen, L. N. Mander, Stereochemistry of Organic Com-
pounds, Wiley, Hoboken, 1994, p. 647.
[27] S. W. Hwang, M. Adiyaman, S. P. Khanapure, J. Rokach, Tetrahedron Lett.
1996, 37, 779–782.
[22] S. Serra, C. Fuganti, E. Brenna, Helv. Chim. Acta 2006, 89, 1110–1122.
[23] D. E. Cane, G. Yang, R. M. Coates, H. J. Pyun, T. M. Hohn, J. Org. Chem.
1992, 57, 3454–3462.
Received: July 29, 2014
[24] a) N. Marion, P. Carlqvist, R. Gealageas, P. de Frꢃmont, F. Maseras, S. P.
Nolan, Chem. Eur. J. 2007, 13, 6437–6451; b) R. S. Ramꢅn, N. Marion,
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
9
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Total Synthesis
S. Bugoni, D. Boccato, A. Porta,*
G. Zanoni, G. Vidari*
&& – &&
Science about scents: The well-known
(ꢀ)-cis-a- and (ꢀ)-cis-g-irone, the finest
and strongest smell components of iris
oils, have been synthesized in ꢁ99%
diastereomeric and enantioselective
ratios. The diastereoselectivity of the
key Wilkinson’s hydrogenation is sup-
ported by a detailed DFT modeling
study of the geometry of the rhodium
h2 complexes involved in the diastereo-
differentiation of the prostereogenic
double bond faces.
Enantioselective Divergent Synthesis
of (ꢀ)-cis-a- and (ꢀ)-cis-g-Irone by
Using Wilkinson’s Catalyst
&
&
Chem. Eur. J. 2014, 20, 1 – 10
www.chemeurj.org
10
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!