Challenging the Molecular Parameters of Vetiver: Can 4,5-Dimethyl-3-(3′-methylbut-1′-en-2′-yl)-4-phenylcyclopent-2-en-1-one Mimic Zizaenones in Structure and Odor?

Article abstract of DOI:10.1002/ejoc.201900163

In order to investigate the molecular parameters of vetiver odorants and scrutinize the significance of ziza-6(13)-en-3-ones as structural templates in the design of new vetiver odorants, 4,5-dimethyl-3-(3′-methylbut-1′-en-2′-yl)-4-phenylcyclopent-2-en-1-one (10) was synthesized in 6 steps from cyclopentane-1,3-dione (20) following a Stork–Danheiser strategy after synthetic approaches by Ni-catalyzed [3+2] annulation and enyne metathesis had failed. Addition of EtOH, α-methylation of the resulting vinylogous ester 18, Pd-catalyzed α-arylation, and CeCl₃·2LiCl mediated alkylation with (3-methylbut-1-en-2-yl)lithium, generated by Shapiro reaction from the corresponding hydrazone 23, furnished after a concluding α-methylation the target compound that possessed only a weak woody odor devoid of typical vetiver characteristics. While the design approach utilizing London dispersion forces could not reveal the secret of vetiver, it showed that the molecular parameters are far more complex than initially expected.

Full text of DOI:10.1002/ejoc.201900163

A Journal of
Accepted Article
Title: Challenging the Molecular Parameters of Vetiver:
Can 4,5-Dimethyl-3-(3’-methylbut-1’-en-2’-yl)-4-
phenylcyclopent-2-en-1-one Mimic Zizaenones in Structure and
Odor?
Authors: Patrick Pfaff, Halima Mouhib, and Philip Kraft
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201900163
Link to VoR: http://dx.doi.org/10.1002/ejoc.201900163
Supported by

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
Challenging the Molecular Parameters of Vetiver:
Can 4,5-Dimethyl-3-(3’-methylbut-1’-en-2’-yl)-4-phenylcyclopent-2-
en-1-one Mimic Zizaenones in Structure and Odor?
[a]
[b]
*[a]
Patrick Pfaff , Halima Mouhib , and Philip Kraft
Dedicated to Cornelius Nussbaumer on the occasion of his retirement
Abstract: In order to investigate the molecular parameters of vetiver
odorants and scrutinize the significance of ziza-6(13)-en-3-ones as
structural templates in the design of new vetiver odorants, 4,5-
dimethyl-3-(3’-methylbut-1’-en-2’-yl)-4-phenylcyclopent-2-en-1-one
Today, the worldwide production of vetiver oil is estimated at
300–400 t/year due to its popularity in modern perfumery.
[
3, 4]
This popularity is based on its distinct and characteristic suave
and sweet transparent woody-ambery odor with characteristic
earthy, green, grapefruit and rhubarb-type facets. Vetiver oil
connects a fresh hesperidic-citrusy top with a dark woody base
note forming a universal skeleton that can be interpreted either
in a clean and soft or a rough and dirty way. An example for the
latter is ‘Vetiver Extraordinaire’ (Frederic Malle, 2002) by
Dominique Ropion with some 18% of vetiver oil, the grapefruit
top note of which is enhanced by pink pepper, while cloves,
myrrh and oak moss fortify its smoky, woody-balsamic base
note. In contrast, Olivier Pescheux emphasized the clean and
soft transparent woody character the in ‘Vetyverio’ (Diptyque,
2017) by combining 22% of vetiver oil with a Turkish rose–
patchouli harmony in the heart and a warm mandarin note in the
top of the fragrance to tame the harsher aspects of grapefruit.
As compiled in Fig. 1, this hesperidic grapefruit tonality is due
mainly to α- (1) and β-vetivone (2) as well as nootkatone (3),
(10) was synthesized in 6 steps from cyclopentane-1,3-dione (20)
following a Stork–Danheiser strategy after synthetic approaches by
Ni-catalyzed [3+2] annulation and enyne metathesis had failed.
Addition of EtOH, α-methylation of the resulting vinylogous ester 18,
3
Pd-catalyzed α-arylation, and CeCl ∙ 2 LiCl mediated alkylation with
(3-methylbut-1-en-2-yl)lithium, generated by Shapiro reaction from
the corresponding hydrazone 23, furnished after a concluding α-
methylation the target compound that possessed only a weak woody
odor devoid of typical vetiver characteristics. While the design
approach utilizing London dispersion forces could not reveal the
secret of vetiver, it showed that the molecular parameters are far
more complex than initially expected.
“
If you see me in the vetiver
[
2, 4]
Wildflowers in my hair
while earthy aspects are due to geosmine.
Concerning the
Never thinking like the regular
So we better be prepared”
woody character of vetiver oil, there is an important creamy
sandalwood contribution from (E)-isovalencenol (4) that also
displays rougher woody aspects in cedarwood direction and is
Made in Heights,
[
4]
[
1]
the most abundant alcohol in Haitian vetiver oil. More important
for the typical vetiver character though are transparent woody-
ambery odorants that share the characteristic suave and sweet
vetiver tonality. However, there has been no consensus about
the constituent(s) that are responsible for this typical note. The
only component which had unanimously been agreed upon to
‘
Wildflowers (Exhale Efreet)’
Introduction
Native to India and Indonesia, vetiver grass [Chrysopogon
zizanioides (L.) Roberty, formerly Vetiveria zizanioides (L.) Nash
Vetiveria odorata Virey = Anatherum muricatum (Retz.) P.
Beauv.] is a tufted perennial plant that grows in tropical and
subtropical countries up to 3 m tall with roots up to 4 m in
These dense root clumps are the source of an
essential oil, which is hydrodistilled at elevated temperatures
after cleaning, drying, chopping and soaking the roots in water.
[
5]
=
possess such a typical vetiver odor was khusimone (5), which
[
6]
is contained in up to 2% in the essential oil. Together with
[
5]
synthetic spiro[4.5]decan-2-ones, eremophiladienal and 1,7-
[
2,3]
[
7]
depth.
cyclogerma-1(10),4-dienal it thus served as the basis of a
[
8]
vetiver rule. According to this rule, a vetiver odorant should
possess an α-branched carbonyl osmophore at 5.0±0.5 Å
distance to a bulky group, e.g. a quaternary carbon atom.
However,
a spirocyclic seco-structure of khusimone (5),
[
[
a]
b]
P. Pfaff, Dr. P. Kraft
designed according to this rule, had a disappointingly high odor
–1
Givaudan Fragrances S+T, Ingredients Research
Kemptpark 50, CH-8310 Kemptthal, Switzerland,
E-mail: philip.kraft@givaudan.com
Homepage: www.givaudan.com
threshold of 42.0 ng L air and no woody vetiver but a floral,
[9]
rosy, green geranium odor. While this disproved the simplistic
rule, it also put into question the relevance of khusimone (5) as a
lead structure, all the more since it was found to possess a
Dr. H. Moubib
–1
[9]
rather high odor threshold of 4.7 ng L air.
Albeit Belhassen et al. confirmed in a recent in-depth analysis
Université Paris-Est, Laboratoire Modélisation et Simulation Multi
Echelle, MSME UMR 8208 CNRS, 5 bd Descartes, 77454 Marne-
la-Vallée, France
[
4]
a typical woody-ambery vetiver note with green-rooty facets for
khusimone (5), they however stated that “it is not the main
contributor of vetiver scent.” Instead, they reported ziza-
Supporting information and ORCID(s) from the author(s) for this
article are available on the WWW under
http://dx.doi.org/10.1002/ejoc...
6(13)en-3-one (6) and 2-epi-ziza-6(13)en-3-one (7) to display a
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
[
4]
far higher odor intensity and constitute the main odor vectors.
We could confirm the clean, fresh, transparent woody-ambery
odor of 6, and a strong vetiver odor of woody-ambery tonality for
work of Belhassen et al. are still somewhat shrouded in
secrecy.
[
10]
7,
more intense than 6; however, without being able to
Design of the Target Compound 10
measure odor thresholds or determine odor values. Besides, the
corresponding (2R,3R)-configured alcohol to 6, the ziza-6(13)-
en-3α-ol (8), was ascribed by Belhassen et al. a typical vetiver
note with cedarwood facets, and it was also held responsible for
the desired suave and sweet vetiver tonality, but again without
As outlined in Fig. 2, we derived our simplified target (TGT)
structure 10 by first dissecting all bonds to C-8 in the tricyclic
framework of ziza-4(5),6(13)-dien-3-one (9). That way,
a
[
4]
cyclopent-2-en-1-one with one 3’-methylbut-1’-en-2’-yl
quantitative threshold data reported.
substituent at C-3 and one methyl group at C-4 was derived.
While this latter 4-Me substituent should sterically imitate the
methano bridge, the bulk of the bicyclo[3.2.1]octane
substructure on the Western face was projected to be imitated
by a simple phenyl ring at C-4. The resulting (4S,5S)-configured
4,5-dimethyl-3-(3’-methylbut-1’-en-2’-yl)-4-phenylcyclopent-2-en-
1-one [(4S,5S)-10, Fig. 2, in black] superimposes well on the
strong vetiver odorant 2-epi-ziza-6(13)-en-3-one (7, in white) as
calculated with the Molecular Operation Environment (MOE)
[
12]
software
employing an Amber10:EHT force field. Especially,
both anticipated Eastern binding motifs overlay very well on one
another as is apparent from Fig. 2.
Figure 1. The important character-impact odorants of vetiver oil grouped into a
hesperidic grapefruit family comprising the vetivones 1 and 2 as well as
nootkatone (3),
a significant sandal-/cedarwood contribution from (E)-
isovalencenol (4) and a typical vetiver-like, woody-ambery block made up from
khusimone (5) and the transparent woody-ambery ziza-6(13)en-3-yl
derivatrives 6–8.
Figure 2. Derivation of the target (TGT) compound 10 by cutting C-8 out of the
tricyclic framework of the racemic ziza-4(5),6(13)-dien-3-one (9) and replacing
the Western hydrophobic part by a phenyl ring. The biflexible superposition
analysis of the resulting (4S,5S)-configured enantiomer of 10 (in black) on the
strong vetiver odorant 2-epi-ziza-6(13)-en-3-one (7, in white) with the MOE
None of woody-ambery ziza-6(13)en-3-yl derivatives 6–8 in Fig.
1
is easily accessible synthetically, especially on an industrial
[12]
software package (Amber10:EHT) indicated a good overlay of the structural
scale. Therefore, these can only serve as structural templates
for the design of new vetiver odorants. Upon structural
modifications of the lead structures 6 and 7, the racemic ziza-
features, especially concerning the anticipated binding motif. The feature
overlap (F), which describes the configurational similarity as the negative
value of the probability density overlap function, has a value of –118.0. The
grand alignment score (S) of the probability-density overlap is –81.8.
4
(5),6(13)-dien-3-one (9, Fig. 2) had been found to possess a
[
11]
woody-ambery vetiver-type odor as well.
Since its odor
intensity was found comparable to that of the ziza-6(13)-en-3-
one (6), it appeared as if the docking of 9 to the respective
olfactory receptors as H-bond acceptor would primarily involve
the eastern part of the zizaenones in this case featuring a
trienone motif, while the sterically less accessible western 1,1-
spiroannulated cyclopentyl face would interact with a less
specific hydrophobic binding pocket. If these assumptions were
true, then it should be possible to construct a modified seco-
structure that would preserve these two binding features on a
simple cyclopentenone core. Of course, getting rid of the tricyclic
skeleton should significantly facilitate the synthetic access. With
some caution, such a target structure could even serve to
assess the significance of the structural features and potentially
even the importance of the zizaenones 6 and 7 for the odor of
vetiver oil, the principal odorants of which, despite the recent
The reason behind selecting a phenyl ring to mimic the western
steric bulk of the ziza-6(13)-en-3-ones 6, 7 and 8 was the idea
[13]
that London dispersion forces of the aromatic system at C-4
with the 3-(3’-methylbut-1’-en-2’-yl) substituent would bring both
substituents in rather close proximity, so that the resulting
densely packed moiety would in effect be similar in its
hydrophobicity to the Western part of the zizaenones 6–9.
Thereby, it should be possible to compensate the steric
repulsion of the C-3 and C-4 substituents as apparent for
[
13]
branched alkyl groups.
To support this hypothesis, the conformational space of TGT 10
was characterized by quantum chemical calculations at different
levels of theory. The precise quantification of weak inter- and
intramolecular London dispersion forces at quantum chemical
level is not trivial, and its proper description is still a topic of
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
[
14–18]
active research.
The effect becomes all the more important
systematic overview of stabilizing effects arising from dispersion
interaction between these structural subunits, we expanded the
calculations and optimized our starting geometries with
additional functionals, now explicitly including Grimmes’ D3
when large molecular systems are treated, which makes the
right choice of quantum chemical method crucial to obtain
realistic structural insight into the system investigated. Starting
with a conformational sampling at molecular mechanics level,
we subsequently optimized several resulting geometries using
the density functional theory exchange-correlation functional
B3LYP and the ab initio Møller-Plesset second order
perturbation theory method (MP2). We used the Avogadro and
[
23]
dispersion correction. The relative energy differences between
the two conformers are given in Table 1. All methods that
explicitly include the dispersion interaction during the
optimization process favor conformer I over conformer II, which
…
exhibits an intramolecular CH π interaction between the
[
19,20]
GAUSSIAN09 software packages,
to carry out the classical
hydrogen atom of the isopropyl-group and the aromatic ring,
while the other methods are in favor of conformer II. By
comparison of different density functionals, the importance of the
dispersion interaction becomes important. This is the major
and quantum chemical calculations, respectively. Our
preliminary results yielded five different conformers of TGT 10,
of which the two lowest energy conformers differ mainly in the
orientation of the hydrogen atom of the isopropyl group and the
hydrogen atom of the C=C double bond towards the aromatic
ring. The optimized geometries are depicted in Fig. 3 and
display two distinct orientations of the 3-(3’-methylbut-1’-en-2’-yl)
moiety, though interestingly the distances of the center of the
aromatic ring to the respective hydrogen atoms are almost
identical around 3.1Å. The lowest energy conformer 10 conf. I
optimized at the MP2 level of theory (Fig. 3) is indeed very close
to the (4S,5S)-configured conformer of 10 of the superposition
analysis in Fig. 2, which nicely confirmed our initial assumptions
in the derivation of the target (TGT) compound 10 from the
tricyclic ziza-4(5),6(13)-dien-3-one (9). Due to the possible
stabilization by π–π stacking interactions between the exocyclic
double bond and the aromatic ring, one would have intuitively
supposed the s-trans conformer 10 conf. II to be the lowest one
in energy. However, steric interactions, intramolecular strains
and geometric constraints make the target structure 10
incomparable with unrestrained CH/π hydrocarbon–benzene
…
[24]
source of attraction in the CH π interaction,
and the reason
why 10 conf. I is lowest in energy. In the optimized structure of
conformer 10 conf. II, the orientation of the benzene ring and
the methylene group does not allow for a classical stabilization
by π–π stacking. These results demonstrate that benchmark
calculations are still required to properly describe the structure
and stabilizing interactions in large molecular systems.
Table 1. Relative electronic energies of the two lowest energy
conformer of TGT 10 depicted in Fig, 3 as obtained from quantum
chemical geometry optimizations at different levels of theory. Unless
otherwise stated, the triple-zeta basis set 6-311++G(d,p) set and
the electronic zero-point energy correction (ZPE) is included in the
data A more detailed Table including higher energy conformers is
provided in the Supporting Information.
a
conf.
HF
MP2
PBE0
PBE0-D3
I
II
0.00
2.20
0.00
6.61
2.75
0.00
0.00
0.85
b
b
Conf.
B3LYP
B3LYP-D3
B3LYP
B3LYP-D3
[
21, 22]
systems.
I
II
2.37
0.00
0.00
1.59
2.77
0.00
0.00
1.22
Conf.
B3LYP-D3BJ
M052X
M052X-D3
M062X
I
0.00
0.95
0.00
1.39
0.00
2.65
0.00
2.66
II
a
The MP2 values are not corrected by the electronic zero-point
energy (ZPE).
b
Def2TZVP basis set was used.
Figure 3. Geometries of the two lowest energy conformers I and II of the like-
configured target structure 10 as optimized at the MP2/6-311++G(d,p) level of
theory. The relative energies are given without the electronic zero-point energy
correction. It should be noted that due to the weak dispersion interactions
between the phenyl moiety and the 3-(3’-methylbut-1’-en-2’-yl) moiety, the
lowest energy structure strongly depends on the applied quantum chemical
method (cf. Table 1 for further details).
Results and Discussion
The retrosynthetic analysis of our target compound is detailed in
Scheme 1. As an alternative to the sterically sensitive and often
low-yielding
Pauson–Khand
cyclopentenone
synthesis,
reductive Ni-catalyzed [3+2] annulations had independently
been reported by the groups of Montgomery and Ogoshi as
[
25]
[26]
alternatives. Recently, even an enantioselective strategy had
It has been previously reported that the MP2 and the B3LYP
method tend to over- and underestimate intramolecular London
dispersion interactions, respectively. Therefore, to get a more
[
27]
been worked out by the group of Cramer.
While the
intermolecular reductive cycloaddition of cinnamic aldehyde (11)
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
with alkynes such as 12 in the presence of a Ni(cod)
2
-derived
as
B), and an
the Grela catalyst 14 forms the ruthenium carbene 15, which
then closes to a 6-membered ring with the vinyl carbinol
substituent. While endo-products quite frequently occur as by
products, only in rare cases are they obtained as main products
of enyne metatheses, for instance with 12-membered or larger
complex (cod = cycloocta-1,5-diene) with for instance PBu
ligand, an organoborane reducing agent (e.g. Et
3
3
alcohol as co-solvent (e.g. MeOH in THF), should furnish the
corresponding cyclopent-2-enon-1-ols, enolates of cinnamates
should provide the corresponding cyclopent-2-enones.
However, in our hands the reductive Ni-catalyzed [3+2]
cycloaddition (Scheme 1, route A) proved sterically too sensitive
for both types of substrates, and was therefore abandoned.
[25]
[35]
rings, or when sterically demanding substituents are present
[
36]
in the alkyne moiety.
Based on these findings, the classical Stork–Danheiser
[37]
strategy
was finally opted for. As delineated in Scheme 1
(route C), this strategy to target structure 10 comprised the α-
arylation of a vinylogous ester 18 with a halobenzene 17,
followed by the addition of a metalated alkenylspecies 19. The
vinylogous ethyl ester 18 was prepared in 89% isolated yield
from cyclopenta-1,3-dione 20 as starting material by addition of
EtOH in the presence of p-toluenesulfonic acid (pTsOH, Scheme
2), and displayed a creamy, fruity, sweet odor recalling red
berries. This vinylogous ester 18 was then α-methylated by
treatment with lithium diisopropyl amide (LDA) and subsequent
reaction of the resulting enolate with iodomethane at –78 °C to
afford in 52% isolated yield the intermediate 21 with a green,
fatty, balsamic odor reminiscent of coumarin with additional
[
38]
resinous aspects.
arylation, recently reported by Lautens and coworkers,
transform the 3-ethoxy-5-methylcyclopent-2-en-1-one (21) into
the corresponding 5-methyl-5-phenyl-substituted ethoxy
Next, we implemented a Pd-catalyzed α-
[38]
Scheme 1. Retrosynthetic analysis of the target compound 10 by reductive Ni-
catalyzed [3+2] cycloaddition (dissection A), an enyne metathetic approach
to
(dissection B) employing the highly-active nitro-Grela catalyst 14, and on the
classical Stork–Danheiser route (dissection C) featuring a Woods–Grignard
reaction with a vinyl metal species 19 on a vinylogous ester constructed from
cyclopentenone 22. At –20 °C, the lithium enolate of 21 was
formed by deprotonation with 1.2 eq. of lithium
hexamethyldisilazane (LiHMDS) and reacted with 1.2 eq. of
bromobenzene in the presence of 4.0 mol% chloro[(tri-tert-
3-ethoxycyclopent-2-enone (18) by α-methylation and subsequent reaction
with a phenyl halide 17.
3
butylphosphine)-2-(2-aminobiphenyl)] palladium(II)) (tBu PPd–
In general, cyclopentenol systems are also accessible by an
enyne metathesis as is as well illustrated in Scheme 1
G2), a commercially available second generation (G2) 2-
phenylaniline-based Buchwald palladacycle complex used as
[
39]
precatalyst for cross-coupling reactions.
chromatographic purification the 5-methyl-5-phenylcyclopenten-
-one, 22 was isolated in 69% as off-white solid. Noteworthy,
After work-up and
(dissection B). Precedence for such enyne metathetic ring
closures had for instance been established by Ramhartzer and
1
Mulzer in their synthesis of valerenic acid employing Grubbs 1
[
28]
only decomposition products were obtained upon submitting 18
devoid of the α-methyl substituent to the same reaction
conditions.
catalyst under ethylene atmosphere (“Mori conditions”),
and
[
29]
by Grubbs and co-workers themselves in the construction of
bicyclic ring systems, for instance the triethylsilyl (TES)
protected 2-isopropylbicyclo[4.3.0]nona-2,9-dien-6-ol. It was
thus reasoned that a (3-methylbut-1-yn-1-yl)-substituent should
present a surmountable obstacle for such an enyne-ring closing
metathesis. However, when we subjected the tert-butyl dimethyl
silyl (TBS) protected 8-methyl-5-phenylnon-1-en-6-yn-3-ol 13,
prepared by conjugate addition of 3-methylbut-1-yn-1-yl lithium
to cinnamic aldehyde (11) in the presence of stoichiometric
[
30]
amounts of aluminum tris(2,6-diphenylphenoxide),
and
Scheme 2. Synthesis of the intermediate 22 on the Stork–Danheiser route.
subsequent vinyl magnesium Grignard reaction with TBS
protection, no metathesis product was observed under the “Mori
With compound 22 at hand, the addition of alkenyl nucleophiles
to the carbonyl group was investigated, followed by acid-
catalyzed transposition of the vinylogous ester moiety in Stork–
[
31]
conditions”,
even when utilizing the highly reactive nitro-
[
32]
Grela
metathesis catalyst 14 instead. Subjecting 13 to 12.5
mol% nitro-Grela catalyst in refluxing toluene for 27 h provided
in 53% yield the TBS-protected 5-phenyl-3-isopropyl-4-
methylenecyclohex-2-en-1-ol 16. This endo-metathesis product
3
Danheiser fashion. Gratifyingly, the CeCl ∙ 2 LiCl mediated
[
40]
activation of the enone moiety enabled nucleophilic addition of
the corresponding alkenyl lithium species, which was
regioselectively generated from the 2,4,6-triisopropylbenzene-
[
33, 34]
16 results from an “yne-then-ene” pathway,
as delineated
[41]
sulfyl hydrazone 23 under Shapiro conditions, making use of
in Scheme 1. Instead of first reacting with the vinyl group of 13,
the improved procedure as reported by F. T. Bond and
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
[
42]
coworkers (Scheme 3).
Interestingly, the addition of LiCl was
found to be crucial for activating the vinylogous ester 22; in the
absence of LiCl, no addition occurred. In the presence of 1 eq.
of anhyd. CeCl
ethoxy-5-methyl-5-phenylcyclopent-2-en-1-one (22) with 2 eq. of
3-methylbut-1-en-2-yl)lithium, generated by Shapiro reaction
3
and 2 eq. of anhyd. LiCl, the alkylation of 3-
(
from hydrazone 23, afforded the 3-(3’-methylbut-1’-en-2’-yl)-
substituted cyclopentenone 24 in 76% yield after
chromatographic purification (Scheme 3).
In the last step of the synthetic sequence, the obtained Stork–
Danheiser product 24 was finally α-methylated by deprotonation
with LDA and reaction with methyl iodide to furnish the projected
target compound 10 as a diastereomeric mixture (52:48, E/Z) in
Scheme 4. Alternative approach to intermediate 24 via the respective alkenyl
lithium species from an isomeric mixture of alkenyl iodides 28a and 28b.
Olfactory Properties and Conclusions
5
0% yield based on recovered starting material (brsm). Besides,
the bis-methylated analogue 25 was isolated in 15% yield brsm
Scheme 3), which in terms of structure–odor relationship was
The two unintended 4-methyl-4-phenylcyclopent-2-en-1-ones 29
and 30 were also related to the target compound 10 and
therefore evaluated for their olfactory properties as well. Yet,
both 29 and 30 were very weak in odor strength and displayed
only a faintly woody character with odor detection thresholds (th)
(
highly welcome as an additional data point.
–1
–1
of 400 ng L air (29) and 500 ng L air (30), respectively. The
related 3-(3’-methylbut-1’-en-2’-yl) intermediate 24 en route to
the target structure 10 was also weak, woody with slightly
creamy, caramel-type facets, but had a lower odor threshold of
–1
1
40 ng L air (Scheme 3).
Scheme 3. Synthesis of the target compound 10. An alkenyl lithium species is
regioselectively generated from sulfonohydrazone 23 via a Shapiro reaction,
3
and then trapped by addition to 22, activated by CeCl · 2 LiCl. Subsequent
methylation with LDA/MeI afforded 10 as a diastereomeric mixture, together
with the bis-methylated analogue 25.
Figure 4. Rigid-body alignment of the lowest energy conformer I of the target
structure 10 (in black) at the MP2/6-311++G(d,p) level of theory (cf. Fig. 2) on
the strong vetiver odorant 2-epi-ziza-6(13)-en-3-one (7, in white) with the MOE
In an earlier attempt to generate the required alkenyl lithium
species in situ, we prepared the alkenyl iodides 28a and 28b via
a recently reported Ni-catalyzed regioselective hydroalumination
[
12]
[
43]
software package (Amber10:EHT) . The feature overlap (F), which describes
the configurational similarity as the negative value of the probability density
overlap function, has a value of –104.1. The grand alignment score (S) of the
probability-density overlap is –62.2.
of terminal alkynes.
The yields of the addition were, however,
only moderate and required distillation of the volatile
regioisomeric mixture 28a/b which after purification was
obtained as a solution in pentane. Remarkably, in an initial
attempt, when the halogen–metal exchange of the undistilled
alkenyl iodide mixture 28a/b was incomplete, even the addition
of tBuLi to the vinylogous ester moiety of 22 was observed,
which demonstrates the pronounced electrophilic activation of
The target structure 10 possessed a woody odor, though it was
rather weak, and no typical vetiver odor characteristics were
present. The odor threshold of the diastereomeric (E/Z)-mixture
–1
was determined at 230 ng L air (10) but was only due to the
major (52:48) (E)-configured isomer (Fig. 4), for which a
the enone system in the presence of CeCl
The resulting 3-tert-butyl-4-methyl-4-phenylcyclopent-2-en-1-one
30) as well as the also unintended 3-(isopent-1’-en-1’-yl)-
derivative 29 were isolated in 32% and 5% yield, respectively.
3
· 2 LiCl (Scheme 4).
–1
threshold value of 120 ng L air was measured. The relative
stereochemistry of the potent diastereomer (E)-10 would at least
correspond to the configuration in the superposition analysis in
Fig 2. The 5,5-bis-methyl analogue 25 was even a bit weaker
than the diastereomeric mixture 10 with an odor threshold of 250
(
–1
ng
L
air. This may indicate that the additional methyl
substituent would rather hinder receptor interaction and
therefore only exhibit a slightly woody odor.
But no matter how one looks at it, either the 2-epi-ziza-6(13)en-
3-one (7) and ziza-4(5),6(13)-dien-3-one (9) are no potent lead
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
[
38]
structures, or the design of the target (TGT) compound 10 was
to be too simplistic. If one aligns the lowest energy conformer I
of the target structure 10 (Fig. 4 in black) as obtained at the
MP2/6-311++G(d,p) level of theory (cf. Fig. 2) on 2-epi-ziza-
3-Ethoxycyclopent-2-en-1-one (18) : At room temp., EtOH (48 mL)
was added to a suspension of cyclopentane-1,3-dione (20; 10.0 g, 102
mmol, 1.00 eq.) in MePh (170 mL), followed by p-toluenesulfonic acid
monohydrate (368 mg, 1.94 mmol, 1.90 mol%). The resulting mixture
was heated to 80°C for 17 h, and then concentrated under reduced
pressure. The resulting residue was dissolved in EtOAc and
subsequently neutralized with 2 M aq. NaOH. The organic layer was
separated, and the aqueous one extracted with EtOAc (2×). The
6(13)en-3-one (7, Fig. 4 in white) in a rigid-body manner, one
easily recognizes that despite the London dispersion forces
between the phenyl ring and the branched 3-methylbut-1-en-2-yl
substituent of 10, the phenyl ring protrudes quite a bit from the
tricyclic framework of the lead structure 7. If the Western side of
the zizaenones 6–9 does not tolerate additional bulk, it would
explain why our design concept to focus on the eastern part in
the construction of a simplified, easier accessible synthetic
vetiver odorant did not succeed. At this stage however, any
attempt to explain the structural parameters of vetiver odorants
at a biological level remains rather speculative, as it is not even
known which odorant receptors are involved in the perception of
vetiver. Sufficient quantities of 7 from a partial or total synthesis
would thus be highly in demand if one wants to uncover the
secret of the molecular parameters of vetiver odorant – the
design approach we followed in this paper could not reveal the
secrets of vetiver; yet, it clearly showed that there is more
behind the eastern part of 2-epi-ziza-6(13)en-3-one (7) and the
rough molecular dimensions. For the current work, we assumed
that 2-epi-ziza-6(13)en-3-one (7) is the main principal vetiver
4
combined organic extracts were washed with brine, dried (MgSO ), and
concentrated in vacuo. The resulting crude orange oil was purified by
filtration through a pad of silica (5 cm) with EtOAc as eluent to furnish the
target compound 18 (11.4 g, 89%) as a yellowish oil.
IR (ATR): ṽ = 2984 / 2929 (w, νC–H), 1699 / 1676 (m, νC=O), 1581 (s,
νC=C–C=O), 1475 / 1439 / 1414 (w, δCH
2
, CH
3
), 1373 / 1337 (m, δCH
3
),
H NMR (400 MHz,
): δ = 5.29 (t, J = 1.0 Hz, 1 H, 2-H), 4.06 (q, J = 7.0 Hz, 2 H, 1’-H ),
–1
1
1
287 / 1246 / 1220 / 1180 / 1024 (m, νC–O) cm
.
CDCl
3
2
2.64–2.59 (m, 2 H, 5-H ), 2.46–2.42 (m, 2 H, 4-H ), 1.42 (t, J = 7.0 Hz, 3
) ppm. C NMR (101 MHz, CDCl
C-3), 104.6 (d, C-2), 67.7 (t, C-1’), 34.0 (t, C-5), 28.5 (t, C-4), 14.1 (q, C-
2
2
1
3
H, 2’-H
3
3
) δ = 206.0 (s, C-1), 190.1 (s,
+
+
+
2
’) ppm. MS (EI): m/z = 126 [M] , 98 [M – CO] , 97 [M – C
H
2 5
] , 82 [M –
– CO] . These spectroscopic data correspond
as far as detailed. Odor description (10% DPG,
blotter): creamy, fruity, sweet, red berries.
+
+
3 2 5
CH CHO] , 69 [M – C H
to those in ref.
[44]
[38]
3
-Ethoxy-5-methylcyclopent-2-en-1-one (21) : At –78 °C under a N
2
atmosphere, a 2.5 M solution of nBuLi in hexanes (17.4 mL, 43.4 mmol)
was slowly added to a solution of diisopropylamine (6.33 mL, 44.0 mmol)
in THF (20 mL). A solution of the β-ethoxy cyclopentenone 18 (4.00 g,
1.7 mmol) in THF (20 mL) was then added dropwise at the same temp.
After 15 min. of stirring, a solution of methyl iodide (2.97 mL, 47.6 mmol)
[
4]
odorant Belhassen et al. reported. In future, a more accurate
verification of the threshold of 2-epi-ziza-6(13)en-3-one (7) and
its olfactory properties assessed by perfumers will be necessary
to address the enigma of this exciting molecule.
3
and N,N’-dimethylpropyleneurea (DMPU, 7.65 mL, 63.4 mmol) in THF
20 ml) was added slowly to the reaction mixture at –78 °C. The mixture
(
was allowed to warm to 0 °C prior to the careful addition of water (50
mL). The phases were separated, and the aqueous layer was extracted
with EtOAc. The combined organic extracts were washed with water and
Experimental Section
brine, dried (NaSO
crude product was purified by automated silica-gel column
chromatography (10% to 80% EtOAc/pentane, 0.34 in 50%
EtOAc/pentane) furnishing the title compound 21 (2.20 g, 52%) as a
yellowish oil.
4
), and concentrated under reduced pressure. The
All reactions involving chemicals sensitive to water and/or O
2
were
carried out under dry N atmosphere. Unless otherwise stated,
a
2
f
R =
reagents and solvents (puriss. or purum) were purchased from SAFC or
Acros and used without further purification. Purification of the reaction
products was carried out on a Biotage Isolera One Spektra System with
an Accelerated Chromatographic Isolation (ACI) using SNAP Ultra (Ultra-
Fast Flash Chromatography) Cartridges Biotage packed with HP-Sphere
spherical silica, 25 µm. TLC analyses were performed on Merck
Kieselgel 60 F254 (particle size 5–20 µm, layer thickness 250 µm on
glass, 10 cm × 10 cm) employing phosphomolybdic acid spray (Merck
IR (ATR): ṽ = 2985 / 2929 (w, νC–H), 1690 (m, νC=O), 1586 (s, νC=C–
C=O), 1475 / 1457 / 1432 (w, δCH
2
, CH
022 (m, νC–O) cm . H NMR (400 MHz, CDCl
H, 2-H), 4.05 (q, J = 7.0 Hz, 2 H, 4’-H ), 2.83 (ddd, J = 17.5, 7.5, 1.0
3
), 1373 / 1338 (m, δCH
3
), 1173 /
–1
1
1
1
3
): δ = 5.23 (t, J = 1.0 Hz,
2
Hz, 1 H, 4-H), 2.50 (quintd, J = 7.5, 3.0 Hz, 1 H, 5-H), 2.21 (ddd, J =
1
7
7.5, 3.0, 1.0 Hz, 1 H, 4-H), 1.41 (t, J = 7.0 Hz, 3 H, 2’-H
.5 Hz, 3 H, 5-Me). C NMR (101 MHz, CDCl
3
), 1.21 (d, J =
3
) δ = 208.9 (s, C-1), 188.5
1
.00480.0100) as visualization reagent. Melting points (Mp) were
13
determined on a Büchi melting point apparatus B 545 (uncorr.). IR
spectroscopic data were recorded on a Bruker Alpha Platinum instrument
with integrated diamond ATR. NMR experiments were performed on
Bruker Avance III HD 400 MHz, a Bruker Avance I 500 MHz, or a Bruker
(
5
CH
CH
s, C-3), 103.3 (d, C-2), 67.5 (t, C-1’), 39.7 (d, C-5), 37.1 (t, C-4), 16.7 (q,
+
-Me or C-1’), 14.1 (q, 5-Me or C-1’). MS (EI): m/z = 140 [M] , 125 [M –
+
+
+
+
3
] , 112 [M – CO] , 97 [M – CH
3
– CO] , 84 [M – CO – C
2 4
H ] , 69 [M –
+
3
– CO – C H ] . Odor description (10% DPG, blotter): green, fatty,
2 4
Avance II 600 MHz / TCI 1.7 mircocryoprobe spectrometer in CDCl
3
or
balsamic, coumarin (2H-chromen-2-one), resinous.
1
C
0
6 6
D , as indicated. H NMR chemical shifts are given relative to TMS (δ =
13
ppm). C NMR chemical shifts are given relative to CDCl
3
(δ = 77 ppm)
[38]
3
-Ethoxy-5-methyl-5-phenylcyclopent-2-en-1-one (22) : At –20 °C, a
solution of 3-ethoxy-5-methylcyclopent-2-enone (21, 1.50 g, 10.7 mmol,
.00 equiv) in toluene (20 mL) was added dropwise over the course of 10
6 6
or C D
(δ = 128 ppm). The multiplicities were assigned by DEPT-
experiments. MS spectra were recorded with an Agilent 7890A GC
equipped with an Agilent MSD5975C. Liquid chromatography high
resolution mass spectrometry (LC–HRMS) was performed using
Thermo Scientific Q Exactive Orbitrap instrument with following settings:
ESI in positive and/or negative mode (ESI+/–), full scan from 50–750 m/z
with high mass resolution of 140,000 FWHM (<5 ppm mass accuracy).
Elemental analyses were performed by Eidgenössische Technische
Hochschule Zürich, Laboratorium für Organische Chemie, ETH-
Hönggerberg, HCI E 304, CH-8093 Zürich, Switzerland.
1
min. to 1 M solution of LiHMDS in MePh (12.8 mL, 12.8 mmol, 1.20 eq.)
The mixture was left to stir at this temp. for 30 min. In the meantime,
a
bromobenzene (1.35 mL, 12.8 mmol, 1.20 eq.) was added to
suspension of P(tBu) Pd–G2 (220 mg, 0.43 mmol, 4.0 mol%) in MePh
10 ml). At –20 °C, the previously prepared enolate suspension was
a
3
(
quickly transferred via syringe to this 2-(2-aminobiphenyl)]Pd(II) catalyst
suspension in one dash, and the reaction mixture was stirred at this
temp. for 20 min. Then, the cooling bath was removed, and 10 drops of
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
–
water were added with stirring at room temp. The resulting mixture was
CH , νC=C), 920 (w, δC=CH ), 770 / 699 (m, δoopC=CH=C, aromat.) cm
3
2
1
1
filtered through a pad of Celite, covered with a layer of MgSO
4
. The
3
. H NMR (400 MHz, CDCl ): δ = 7.34–7.19 (m, 5 H, Ph–H), 6.33 (s, 1 H,
filtrate was concentrated under reduced pressure and further purified by
automated silica-gel column chromatography (12% to 100%
2-H), 5.21 (d, J = 1.0 Hz, 1 H, 1’-H), 4.88 (s, 1 H, 1’-H), 2.63 (s, 2 H, 5-
H ), 2.58 (heptd, J = 7.0 Hz, 1.0 Hz, 1 H, 3’-H), 1.76 (s, 3 H, 4-Me), 1.09
2
13
Et
2
O/pentane, R
f
= 0.28 in 50% Et
2
O/pentane) to furnish the title
3 3
(d, J = 7.0 Hz, 3 H, 4’-H ), 1.04 (d, J = 7.0 Hz, 3 H, 4’-H ). C NMR (101
compound 22 (1.60 g, 69%) as an off-white solid.
MHz, CDCl ) δ = 208.3 (s, C-1), 179.9 (s, C-3), 147.5/146.3 (s, C-2’, -1’’),
3
IR (ATR): ṽ = 3074 (w, νC=C–H), 2980 / 2956 (w, νC–H), 1689 (m,
129.3 (d, C-2), 128.8 (2 d, C-3’’), 126.5 (d, C-4’’), 125.2 (2 d, C-2’’), 119.5
νC=O), 1588 (s, νC=C–C=O), 1498 / 1473 / 1445 (w, δCH
ν(C=C)], 1376 (m, δCH ), 1235 / 1192 / 1023 (m, νC–O), 869 (m,
δC=CH–R), 702 (m, δC=CH=C, aromat.) cm
CDCl ): δ = 7.34–7.28 (m, 4 H, 2’’-, 3’’-H), 7.23–7.18 (m, 1 H, 4’’-H), 5.33
t, J = 1.0 Hz, 1 H, 2-H), 4.11 (q, J = 7.0 Hz, 2 H, 1’-H ), 2.99 (dd, J =
8.0, 1.0 Hz, 1 H, 4-H), 2.77 (dd, J = 18.0, 1.0 Hz, 1 H, 4-H), 1.59 (s, 3 H,
2
,CH
3
),
(t, C-1’), 57.0 (t, C-5), 49.7 (s, C-4), 31.6 (d, C-3’), 25.4 (q, 4-Me),
+
+
2 3
22.5/21.9 (2 q, 3’-Me ). MS (EI): m/z = 240 [M] , 225 [M – CH
] , 212 [M
O] . Odor description (10%
DPG, blotter): weak, woody, slightly creamy, caramel. Odor threshold
3
–1
1
+
+
+
.
H NMR (400 MHz,
3 7 3 7 2
– CO] , 197 [M – C H ] , 179 [M – C H – H
3
–1
(
2
(GC): 140 ng L air.
1
5
2
13
-Me), 1.44 (t, J = 7.0 Hz, 3 H, 2’-H
08.2 (s, C-1), 187.8 (s, C-3), 143.9 (s, C-1’’), 128.5 (2 d, C-2’’), 126.6 (2
3
). C NMR (101 MHz, CDCl
3
) δ =
4,5-Dimethyl-3-(3’-methylbut-1’-en-2’-yl)-4-phenylcyclopent-2-en-1-
one (10a and 10b) and 4,5,5-trimethyl-3-(3’-methylbut-1’-en-2’-yl)-4-
d, C-3’’), 125.9 (d, C-4’’), 102.7 (d, C-2), 67.8 (d, C-1’), 51.2 (s, C-5), 46.7
phenylcyclopent-2-en-1-one (25): At –78 °C under an N
a 2.5 M solution of nBuLi in hexanes (0.18 mL, 0.45 mmol, 1.08 eq.) was
slowly added with stirring to a solution of iPr NH (65 µL, 0.46 mmol, 1.1
2
atmosphere,
+
(
t, C-4), 24.5 (q, 5-Me), 14.2 (q, C-2’). MS (EI): m/z = 216 [M] , 201 [M –
+
+
+
CH
3
] , 187 [M – C
2
H
5
] , 173 [M – CH
3
– CO] .
2
eq.) in THF (1.5 mL). 4-Methyl-3-(3’-methylbut-1’-en-2’-yl)-4-
phenylcyclopent-2-en-1-one (24) was dissolved in THF (1.5 mL) and
added dropwise with stirring to the reaction mixture at –78 °C. The
resulting mixture was stirred for 15 min. at this temperature, and then a
solution of MeI (26 µl, 0.42 mmol, 1.0 eq) and DMPU (0.10 mL, 0.83
mmol, 2.0 eq.) in THF (1.5 mL) was added dropwise to the reaction
mixture. After further 15 min. of stirring, the mixture was allowed to warm
(
2’E)-2,4,6-Triisopropyl-N'-(3’-methylbutan-2’-ylidene)benzenesulfo-
nohydrazide (23): At room temp., 2,4,6-triisopropylbenzenesulfonohy-
drazide (2.50 g, 8.38 mmol, 1.00 eq.) was dissolved in EtOH (10 mL) and
5
drops of aq. conc. HCl. 3-Methylbutan-2-one (900 µl, 8.38 mmol, 1.00
eq.) was added, and the resulting suspension stirred for 20 min. After
evaporation of the solvent in vacuo, the solid residue was dissolved in
CH
silica-gel column chromatography (5% Et
Et O/pentane, dry loading on silica, R = 0.27 in 20% Et
afford the title compound 23 as a colorless crystalline solid (2.34 g, 76%).
2
Cl
2
and dried (MgSO
4
). The crude product was purified by automated
O/pentane to 35%
O/pentane) to
4
to room temp., and quenched by addition of sat. aq. NH Cl, prior to being
diluted with water and MTBE. The phases were separated, and the
aqueous phase was extracted with MTBE (2×). The combined organic
2
2
f
2
3
extracts were washed with 2 M aq. HCl, aq. NaHCO and brine. After
1
H NMR (400 MHz, CDCl
3
): 7.28–7.23 (m, 2 H, Ph–H), 4.22 (hept, J =
), 2.90 (hept, J = 7.0, 1 H, 4’- CHMe ), 2.32 (hept, J
7.0, 1 H, 3’-H), 1.73 (s, 3 H, 1’-H ), 1.28–1.24 (m, 18 H, 2-, 4-, 6-
), 0.96 (d, J = 7.0 Hz, 6 H, 3’-Me
160.2 (s, C-2’), 153.0 (s, C-4), 151.2 (2 s, C-2, -6), 131.5 (s, C-1),
23.6 (2 d, C-3, -5), 36.8 (d, C-3’), 34.2 (d, 4-CMe ), 29.9 (2 d, 2-, 6-
), 24.8 (4 q, 2-, 6-CMe ) , 23.6 (2 q, 4-CMe ), 19.6 (q, 3’-Me ), 13.2
q, C-1’). MS (EI): m/z = 366 [M] , 351 [M – CH
drying (MgSO ), the crude product was further purified by silica-gel
4
7
=
.0, 2 H, 2-, 6-CHMe
2
2
column chromatography (3% to6% EtOAc/pentane) to furnish the
compounds 10 as a mixture (dr 52:48, E/Z) of diastereomers (33 mg,
3
13
CHMe
=
1
CMe
(
2
2
). C NMR (101 MHz, CDCl
3
) δ
50% brsm, yellowish oil, R = 0.24 (5% EtOAc/pentane)) and 25 (10 mg,
f
15% brsm, yellowish oil, R = 0.32 (5% EtOAc/pentane)).
f
2
Diastereomeric mixture 10: IR (ATR): ṽ = 2966 (w, ν CH ), 2922 (w,
s
3
2
2
2
+
2
2 as 3
νCH ), 2865 (w, ν CH ), 1693 (s, νC=O), 1563 (m, νC=C–C=O), 1495 /
+
+
3
] , 323 [M – C
H
3 7
] .
1461 / 1447 / 1379 (w, δCH , CH , νC=C), 911 / 896 (m, δC=CH ), 762 /
2
3
2
–
1
+
7
50 / 725 (m, δoopC=CH=C, aromat.) cm . MS (EI): m/z = 254 [M] , 239
+
+
+
[
M – CH
3
] , 226 [M – CO] , 211 [M – CH
3
– C
2 4 3 2 4
H ] , 193 [M – CH – C H
4
-Methyl-3-(3’-methylbut-1’-en-2’-yl)-4-phenylcyclopent-2-en-1-one
+
–
H O] . Odor description (10% DPG, blotter): weak, woody. Odor
2
(
(
24): At room temp., 3-ethoxy-5-methyl-5-phenylcyclopent-2-en-1-one
22, 431 mg, 1.91 mmol, 1.00 eq.) was dissolved in a 0.5 M solution of
–1
threshold (GC): 230 ng L air.
Assignment of the two diastereomers by 2D NMR analysis in CDCl
3
anhyd. LiCl in THF (7.64 mL, 3.82 mmol, 2.00 eq.) and then transferred
to a flask containing vacuum-dried, anhydrous CeCl (471 mg, 1.91
1
(
NOESY, HMBC). Major trans-isomer (E-10a): H NMR (500 MHz,
3
CDCl3) δ = 7.34–7.10 (m, 5 H, Ph–H), 6.35 (s, 1 H, 2-H), 5.21 (s, 1 H, 1’-
H), 4.90 (s, 1 H, 1’-H), 2.68 – 2.61 (m, 1 H, 3’-H), 2.56 (q, J = 7.5 Hz, 1H,
mmol, 1.00 eq.). The resulting suspension was stirred vigorously at room
temp. for 3 h to furnish an off-white slurry. In the meantime, a 2.5 M
solution of nBuLi in hexanes (3.21 mL, 8.02 mmol, 4.20 eq.) was added
dropwise with stirring at –78 °C to a solution of (E)-2,4,6-triisopropyl-N'-
5
-H), 1.57 (s, 3 H, 4-Me), 1.10 (d, J = 7.0 Hz, 3 H, 3’-Me), 1.06 (d, J = 7.0
13
Hz, 3 H, 3’-Me), 1.07 (d, J = 7.5 Hz, 3 H, 5-Me). C NMR (126 MHz,
CDCl ) δ = 210.0 (s, C-1), 178.7 (s, C-3), 147.2 (s, C-2’), 146.7 (s, C-1’’),
28.8 (2 d, C-3’’/-5’’), 128.2 (d, C-2), 126.4 (d, C-4’’), 125.9 (2 d, C-2’’/-
’’), 120.0 (t, C-1’), 59.5 (d, C-5), 53.2 (s, C-4), 31.3 (d, C-3’), 22.4 (2 q,
3
(
3-methylbutan-2-ylidene)benzenesulfonohydrazide (23, 1.40 g, 3.82
1
6
3
mmol, 2.00 eq.) in THF (10 ml). The mixture was left to stir at this temp.
for 15 min. and was then allowed to warm to 0 °C and left to stir until the
–
1
2
’-Me ), 21.9 (q, 4-Me), 9.0 (q, 5-Me). Odor threshold (GC): 120 ng L
N
2
evolution had ceased completely (1.5 h). This yellow suspension was
1
air. Minor cis-isomer (Z-10b): H NMR (500 MHz, CDCl
7
1
3
) δ = 7.34–
.10 (m, 5 H, Ph–H), 6.40 (s, 1 H, 2-H), 5.17 (s, 1 H, 1’-H), 4.85 (s, 1 H,
’-H), 2.63 – 2.57 (m, 1 H, 3’-H), 2.44 (q, J = 7.5 Hz, 1 H, 5-H), 1.72 (s, 3
added dropwise to the slurry of 22 with CeCl and LiCl, and the resulting
3
mixture was allowed to warm to room temp. prior to being diluted with
MTBE (20 mL) and 5 M aq. HCl (5 mL). After further dilution with water,
the solution was left to stir for 10 min., upon which the mixture parted into
two clear phases. These were separated, and the aqueous layer was
extracted with MTBE (2×). The combined organic extracts were washed
H, 4-Me), 1.11 (d, J = 7.0 Hz, 3 H, 3’-Me), 1.04 (d, J = 7.0 Hz, 3 H, 3’-
13
Me), 0.57 (d, J = 7.5 Hz, 3 H, 5-Me). C NMR (126 MHz, CDCl
10.3 (s, C-1), 177.9 (s, C-3), 147.9 (s, C-2’), 142.6 (s, C-1’’), 128.8 (d,
C-2), 128.5 (2 d, C-3’’’/-5’’), 126.8 (d, C-4’’), 125.9 (2 d, C-2’’/-6’’), 118.6
t, C-1’), 56.6 (d, C-5), 53.6 (s, C-4), 32.0 (d, C-3’), 24.6 (q, 4-Me),
2.9/21.8 (2 q, 3’-Me ), 10.6 (q, 5-Me).
By-Product 25: H NMR (400 MHz, CDCl
H), 6.34 (s, 1 H, 2-H), 5.16 (s, 1 H, 1’-H ), 4.83 (s, 1 H, 1’-H
m, 1 H, 3’-H), 1.61 (s, 3 H, 4-Me), 1.14 (s, 3 H, 5-Me ), 1.13 (d, J = 7.0
Hz, 3 H, 3’-Me), 1.04 (d, J = 7.0 Hz, 3 H, 3’-Me), 0.48 (s, 3 H, 5-Me
NMR (101 MHz, CDCl ): δ = 213.8 (s, C-1), 176.8 (s, C-3), 148.1 (s, C-
3
) δ =
2
with brine and sat. aq. NaHCO
product was further purified by automated silica-gel column
chromatography (5% to 40% Et O/pentane, = 0.41 (20% Et
pentane)) to furnish the title compound 24 (350 mg, 76%) as a yellowish
3 4
, and dried over MgSO . The crude
(
2
2
2
R
f
2
O
1
3
): δ = 7.36–7.15 (m, 5 H, Ph–
), 2.67–2.57
/
a
b
oil.
(
a
s 3 2 3
IR (ATR): ṽ = 2963 (w, ν CH ), 2932 (w, νCH ), 2873 (w, νasCH ), 1693
13
b
).
C
(
2
s, νC=O), 1563 (m, νC=C–C=O), 1496, 1461, 1446, 1408 (w, δCH ,
3
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
2
1
3
’), 143.9 (s, C-1’’), 128.3 (2 d, C-2’’/-6’’), 127.4 (2 d, C-3’’/-5’’),
26.6/126.0 (2 d, C-2, -4’’), 118.0 (t, C-1’), 56.7 (s, C-5), 54.8 (s, C-4),
2.2 (d, C-3’), 24.6/22.9/22.2/21.8/21.1 (5 q, 4-Me, 5-, 3’-Me ). Odor
2
3-tert-Butyl-4-methyl-4-phenylcyclopent-2-en-1-one (30): Following
the identical procedure for the synthesis of 10 and 26, an undistilled,
diluted mixture of alkenyl iodides 28a and 28b in pentane was utilized.
The Li–halogen exchange was unsuccessful, and the residual t-BuLi was
added to the activated ketone of 10 (60 mg, 0.28 mmol). After column
description (10% DPG, blotter): weak, slightly woody. Odor threshold
–1
(
GC): 250 ng L air.
chromatography (5% to 40% Et
2 f 2
O/pentane) 30 (R (30% Et O/pentane) =
0
.48; 20 mg, 32%) was obtained as a yellowish oil.
2
(
-Iodo-3-methylbut-1-ene (28a) and
1-Iodo-3-methylbut-1-ene
28b): At room temp., a 1 M solution of DIBAL in hexane, 16.2 mL, 1.10
eq.) was added dropwise with stirring to a suspension of Ni(dppp)Cl
160 mg, 0.29 mmol, 2.0 mol%) in N -purged THF (15 mL). The
[43]
s 3 2 as 3
IR (ATR): ṽ = 2967 (w, ν CH ), 2930 (w, νCH ), 2872 (w, ν CH ), 1693
(
s, νC=O), 1586 (m, νC=C–C=O), 1496 / 1459 / 1446 / 1377 / 1366 (w,
δCH , CH , νC=C), 918 (w, δC=C), 770 / 741 / 700 (m, δoopC=CH=C),
aromat.) cm . H NMR (400 MHz, CDCl
2
(
2
2
3
–1 1
) δ = 7.35–7.19 (m, 6 H, Ph–H),
.24 (s, 1 H, 2-H), 2.65/2.55 (2 d, J = 19.0 Hz, 1 H, 4-H ), 1.85 (s, 3 H, 4-
) δ = 208.8 (s, C-1),
suspension was cooled to 0 °C, and 3-methylbut-1-yne (1.50 mL, 14.7
mmol, 1.00 eq.) was added dropwise. After stirring at room temp. for 2 h,
a solution of N-iodosuccinimide (3.31 g, 14.7 mmol, 1.00 eq.) in THF (10
ml) was added slowly to at 0 °C. The reaction mixture was then stirred at
room temp. for 17 h, followed by addition of sat. aq. Rochelle salt at 0 °C.
The resulting mixture was diluted with water and Et
separated, and the aqueous phase was extracted with Et
combined organic extracts were washed with brine. After drying (MgSO
3
6
2
13
3 3
Me), 1.03 (s, 9 H, 1’-Me ). C NMR (101 MHz, CDCl
1
94.8 (s. C-3), 145.6 (s, C-1’), 130.5 (d, C-2), 128.6 (2 d, C-3’’), 126.5 (d,
C-4’’), 125.6 (2 d, C-2’’), 57.3 (t, C-5), 51.3 (s, C-4), 36.9 (s, C-1’), 31.4 (3
q, 1’-Me ) 25.3 (q, 4-Me). Odor description (10% DPG, blotter): very
2
O, the phases were
O (2×). . The
),
3
–1
weak, faintly woody. Odor threshold (GC): 500 ng L air.
2
4
column chromatography (100% pentane) provided ca. 70%-solution of 2-
iodo-3-methylbut-1-ene (major) and 1-iodo-3-methylbut-1-ene (minor
constituent) in pentane (due to the volatility of the constituents the exact
conc. could not be determined). Distillation in vacuo of this solution (32.5
mbar, 32.5 °C head temp.) afforded the regioisomeric alkenyl iodides (2.0
g, 9:1 mixture, 70%) as used in the subsequent steps.
Acknowledgements
We thank Dr. Veronika Zelenay and Dr. Gerhard Brunner for
NMR and IR experiments, Benjamin Spengler for 2D NMR
assignments, Dr. Susanne Kern for the mass spectrometric data,
Sandro Dossenbach, Daniel Bieri, Heinz Koch and their
panelists for the determination of odor thresholds, and
Dominique Lelievre for the olfactory evaluation of the
compounds. For the photography of Indonesian vetiver grass,
Vetiveria zizanioides (L.) Nash, we are grateful to Prof. Dr. Topul
Rali, University of Papua New Guinea. Fruitful discussions with
Prof. Dr. Amir H. Hoveyda, Boston College, as well as with Prof.
Dr. Erick M. Carreira, ETH Zürich, are acknowledged with
gratitude as well.
(
1’E)-4-Methyl-3-(3’-methylbut-1’-en-1’-yl)-4-phenylcyclopent-2-en-1-
one (26): At room temp., 3-ethoxy-5-methyl-5-phenylcyclopent-2-en-1-
one (22, 100 mg, 0.462 mmol, 1.00 eq.) was dissolved in a 0.5 M solution
of anhyd. LiCl in THF (1.85 mL, 0.925 mmol, 2.00 eq.) and then
transferred to a flask containing vacuum-dried, anhyd. CeCl
.463 mmol, 1.00 eq.). The resulting suspension was stirred vigorously at
room temp. for 3 h to furnish an off-white slurry. At –78 °C under N
atmosphere, the freshly distilled mixture of the iodides 28a and 28b (270
mg, 1.39 mmol, 3.00 eq.) was dissolved in dry Et O (1.0 ml) and added
dropwise with stirring to a 1.7 M solution of t-BuLi in hexane (1.63 mL,
.00 eq.) in dry Et O (0.3 mL). The mixture was stirred for 15 min. at this
temp., and then at 0 °C transferred via syringe to the suspension of 22,
CeCl and LiCl in THF. The resulting suspension was allowed to warm to
room temp., prior to being diluted with Et O (3 mL) and 5 M aq. HCl (3
3
(114 mg,
0
2
2
6
2
Keywords: fragrances • London dispersion forces • structure–
odor correlation • Stork–Danheiser strategy • vetiver
3
2
mL). After further dilution with water, the solution was left to stir for 10
min., upon which the mixture parted into two clear phases, which were
[1]
[2]
[3]
Made In Hights (K. B. Bulkin, A. S. Mohajerjasbi), Wildflowers (Exhale
Efreet) on: Made In Hights, HEIGHTS, 2014, B00GHSTS2K, track #4;
http://madeinheights.com/.
separated. The aqueous layer was extracted with Et
combined organic extracts were washed with brine and sat. aq. NaHCO
After drying (MgSO ), the crude mixture was purified by column
chromatography (5% to 40% Et O/pentane) to furnish both 24 (R (20%
Et O/pentane) = 0.41; 40 mg, 36%; spectroscopic data vide infra) and 29
(20% Et O/pentane) = 0.28; 5 mg, 5%) as yellowish oils.
2
O (2×), the
3
.
G. Ohloff, W. Pickenhagen, P. Kraft, Scent and Chemistry – The
Molecular World of Odors, Verlag Helvetica Chimica Acta, Zurich, 2011,
p. 104 and pp. 291–296.
4
2
f
2
(
R
f
2
M. Maffei (Ed.), Vetiveria: The Genus Vetiveria, Taylor & Francis,
London and New York, 2002.
[
[
4]
5]
E. Belhassen, N. Baldovini, H. Brevard, U. J. Meierhenrich, J.-J. Filippi,
Chem. Biodivers. 2014, 11, 1821–1842.
Minor product 29: IR (ATR): ṽ = 2961 (w, ν
w, νasCH ), 1690 (s, νC=O), 1636/1563 (m, ν(C=C-C=C–C=O), 1497 /
461 / 1445 / 1378 (w, δCH , CH , νC=C), 1029 / 974 (w, δC=C), 768 /
99 (m, δoopC=CH=C, aromat.) cm . H NMR (400 MHz, CDCl
.36–7.27 (m, 2 H, Ph–H), 7.28–7.18 (m, 3 H, Ph–H), 6.22 (s, 1 H, 2-H),
.11 (dd, J = 16.0, 7.0 Hz, 1 H, 2’-H), 5.87 (dd, J = 16.0, 1.5 Hz, 1 H, 1’-
s 3 2
CH ), 2930 (w, νCH ), 2865
(
3
P. Kraft, R. Cadalbert, Synthesis 2002, 2243–2253.
P. Weyerstahl, H. Marschall, U. Splittgerber, D. Wolf, H. Surburg,
Flavour Fragrance J. 2000, 15, 395–412.
1
6
7
6
2
3
–1 1
[6]
3
) δ =
[
[
7]
8]
P. Weyerstahl, H. Marschall, U. Splittgerber, D. Wolf, Flavour
Fragrance J. 2000, 15, 61–83.
2
H), 2.66/2.59 (2 d, J = 18.5 Hz, 1H, 5-H ), 2.30 (doct, J = 7.0, 1.5 Hz, 1
P. Kraft, Chem. Biodiv. 2008, 5, 970–999
H, 3’-H), 1.70 (s, 3 H, 4-Me), 0.92 (d, J = 7.0 Hz, 3 H, 3’-Me), 0.91 (d, J =
13
[9]
P. Kraft, N. Denizot, Eur. J. Org. Chem. 2013, 49–58.
7
.0 Hz, 3 H, 3’-Me). C NMR (101 MHz, CDCl
3
) δ = 208.2 (s, C-1), 180.0
[
[
[
10] R. Clery, M. Ambühl, private communication, November 2013.
11] M. Dao, P. Kraft, to be published.
(
s, C-3), 150.6 (d, C-2’), 145.2 (s, C-1’’), 128.8/126.0 (4 d, C-2’’/C-3’’),
1
3
26.7/126.7 (2 d, C-2/C-4’’), 120.4 (d, C-1’), 55.6 (t, C-5), 49.0 (s, C-4),
2.2 (d, C-3’), 25.2 (q, 4-Me), 21.8/21.7 (2 q, 3’-Me ). Odor description
12] Molecular Operating Environment (MOE), release 2018.01.01,
Chemical Computing Group, Montreal, Quebec, Canada H3A 2R7,
2014. For more information, see http://www.chemcomp.com/.
13] J. P. Wagner, P. R. Schreiner, Angew. Chem. Int. Ed. 2015, 54, 12274–
2
(
10% DPG, blotter): very weak, faintly woody. Odor threshold (GC): 400
–
1
ng L air.
[
12296.
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
[
14] C. H. Gottschalk Hannes, A. Poblotzki, M. A. Suhm, M. M. Al-Mogren,
J. Antony, A. Auer, L. Baptista, D. Benoit, G. Bistoni, F. Bohle, R.
Dahmani, D. Firaha, S. Grimme, A. Hansen, M. Harding, M. Hochlaf, C.
Holzer, G. Jansen, W. Klopper, W.A Kopp, L. C. Kröger, K. Leonhard,
H. Mouhib, F. Neese, M. N. Pereira, I. S. Ulusoy, A. Wuttke, R. Mata,
J. Chem. Phys. 2018, 148, 014301.
[27] J. S. E. Ahlin, P. A. Donets, N. Cramer, Angew. Chem. Int. Ed. 2014,
53, 13229–13233.
[28] J. Ramharter, J. Mulzer, Eur. J. Org. Chem. 2012, 2041–2053.
[29] S.-H. Kim, W. J. Zuercher, N. B. Bowden, R. H. Grubbs, J. Org. Chem.
1996, 61, 1073–1081.
[30] K. Maruoka, H. Imoto, S. Saito, H. Yamamoto, J. Am. Chem. Soc.
1994, 116, 4131–4132.
[
15] S. Grimme, S. Ehrlich, L. Goerigk, J. Comp. Chem. 2011, 32(7), 1456-
1
465.
[31] A. Kinoshita, N. Sakakibara, M. Mori, J. Am. Chem. Soc. 1997, 119,
12388–12389.
[
[
16] S. Grimme, J. Comput. Chem. 2004, 25, 1463–1473
17] E. G. Hohenstein, S. T. Chill, C. D. Sherrill, J. Chem. Theory Comput.
[32] K. Grela, S. Harutyunya, A. Michrowska, Angew. Chem. Int. Ed. 2014,
41, 4038–4040.
2
008, 4, 1996–2000.
18] M. Biczysko, P. Paneka, V. Barone, Chem. Phys. Let. 2009, 475(1),
05-110.
[
[
[
[
[
[
[33] E. Vedrenne, F. Royer, J. Oble, L. El Kaïm, L. Grimaud, Synlett 2005,
2379–2381.
1
19] M. D. Hanwell, D. E. Curtis, D. C. Lonie, T. Vandermeersch, E. Zurek,
G. R. Hutchison, J. Cheminform. 2012, 4, 1-17.
[34] B. R. Galan, A. J. Giessert, J. B. Keister, S. T. Diver, J. Am. Chem.
Soc. 2005, 127, 5762–5763.
20] M. J. Frisch et al., Gaussian 09, Revision A.02, Gaussian, Inc.,
Wallingford CT, 2009.
[35] E. C. Hansen, D. Lee, J. Am. Chem. Soc. 2003, 125, 9582–9583.
[36] H. Villar, M. Frings, C. Bolm, Chem. Soc. Rev. 2007, 36, 55–66
[37] G. Stork, R. L. Danheiser, J. Org. Chem. 1973, 38, 1775–1776.
[38] T. Johnson, F. Pultar, F. Menke, M. Lautens, Org. Lett. 2016, 18,
21] S. Tsuzuki, K. Honda, T. Uchimaru, M. Mikami, K. Tanabe, J. Am.
Chem. Soc. 2000, 122, 15, 3746–3753
22] P. Tarakeshwar, H. S. Choi, K. S. Kim, J. Am. Chem. Soc. 2001, 123,
6488−6491.
1
4, 3323–3331.
23] S. Grimme, J. Antony, S. Ehrlich, H. A. Krieg, J. Chem. Phys. 2010,
32, 154104.
[39] H.-H. Zhang, J. Dong, Q.-S. Hu, Eur. J. Org. Chem. 2014, 1327–1332.
[
40] A. Krasovskiy, F. Kopp, P. Knochel, Angew. Chem. Int. Ed. Eng. 2006,
45, 497–500.
1
[
[
41] R. H. Shapiro, M. J. Heath, J. Am. Chem. Soc. 1967, 89, 5734–5735.
42] A. R. Chamberlin, J. E. Stemke, F. T. Bond, J. Org. Chem. 1978, 43,
[
[
24] S. Tsuzuki, A. Fujii, Phys. Chem. Chem. Phys. 2008, 10, 2584–94.
25] A. D. Jenkins, A. Herath, M. Song, J. Montgomery, J. Am. Chem.
Soc. 2011, 133, 14460–14466.
147–154.
[43] F. Gao, A. H. Hoveyda, J. Am. Chem. Soc. 2010, 132, 10961–10963.
[44] H.-S. Wang, H.-J. Li, J.-L. Wang, Y.-C. Wu, Green Chem. 2017, 19,
2140–2144.
[
26] M. Ogoshi, T. Taniguchi, S. Ogoshi, J. Am. Chem. Soc. 2011, 133,
1
4900–14903.
This article is protected by copyright. All rights reserved.

European Journal of Organic Chemistry
10.1002/ejoc.201900163
FULL PAPER
Entry for the Table of Contents
Layout 1:
FULL PAPER
Unfortunately Not!
– While the
Fragrance Chemistry
synthesis was finally successful on a
classical Stork–Danheiser route, the
simplified target structure, designed
to be confined by London dispersion
forces to mimic zizaenons, did only
possess a weak woody odor but no
typical vetiver character. So, the
question of the ultimate vetiver lead
structure and the race for an easily
accessible synthetic vetiver odorant
remain open.
P. Pfaff, H. Mouhib, P. Kraft*
Page No. – Page No.
Challenging
the
Molecular
Parameters of Vetiver: Can 4,5-
Dimethyl-3-(3’-methylbut-1’-en-2’-yl)-
4-phenylcyclopent-2-en-1-one Mimic
Zizaenones in Structure and Odor?
This article is protected by copyright. All rights reserved.