Atropisomerism about aryl-csp3 bonds: The electronic and steric influence of ortho -substituents on conformational exchange in cannabidiol and linderatin derivatives

Article abstract of DOI:10.1021/jo5006069

Terpenylation reactions of substituted phenols were used to prepare cannabidiol and linderatin derivatives, and their structure and conformational behavior in solution were investigated by NMR and, for some representative examples, by DFT. VT-NMR spectra and DFT calculations were used to determine the activation energies of the conformational change arising from restricted rotation about the aryl-Csp³ bond that lead to two unequally populated rotameric epimers. The NBO calculation was applied to explain the electronic stabilization of one conformer over another by donor-acceptor charge transfer interactions. Conformational control arises from a combination of stereoelectronic and steric effects between substituents in close contact with each other on the two rings of the endocyclic epoxide atropisomers. This study represents the first exploration of the stereoelectronic origins of atropisomerism around C(sp²)-C(sp³) single bonds through theoretical calculations.

Full text of DOI:10.1021/jo5006069

Article
pubs.acs.org/joc
Atropisomerism about Aryl−Csp3 Bonds: The Electronic and Steric
Influence of ortho-Substituents on Conformational Exchange in
Cannabidiol and Linderatin Derivatives
Hatice Berber,†,* Pedro Lameiras,‡ Clement Denhez,†,⊥ Cyril Antheaume,§ and Jonathan Clayden∥
́
†ICMR, CNRS UMR 7312, Faculte
Cedex, France
́
de Pharmacie, Universite
́
de Reims Champagne-Ardenne, 51 rue Cognacq-Jay, F-51096 Reims
iment 18, BP 1039, F-51687 Reims
́
de Reims Champagne-Ardenne, UFR Sciences Exactes et Naturelles, F-51687
‡ICMR, CNRS UMR 7312, Universite
Cedex 2, France
́
de Reims Champagne-Ardenne, Moulin de la Housse, Bat
̂
⊥Multiscale Molecular Modeling Platform, Universite
Reims Cedex 2, France
§Faculte
́
de Pharmacie, Service Commun d’Analyse, 74 route du Rhin, BP 60024, F-67401 Illkirch Cedex, France
∥School of Chemistry, University of Manchester, Oxford Road, Manchester M13 9PL, U.K.
S
* Supporting Information
ABSTRACT: Terpenylation reactions of substituted phenols were used
to prepare cannabidiol and linderatin derivatives, and their structure and
conformational behavior in solution were investigated by NMR and, for
some representative examples, by DFT. VT-NMR spectra and DFT
calculations were used to determine the activation energies of the
conformational change arising from restricted rotation about the aryl−
Csp3 bond that lead to two unequally populated rotameric epimers. The
NBO calculation was applied to explain the electronic stabilization of one
conformer over another by donor−acceptor charge transfer interactions.
Conformational control arises from a combination of stereoelectronic and
steric effects between substituents in close contact with each other on the
two rings of the endocyclic epoxide atropisomers. This study represents
the first exploration of the stereoelectronic origins of atropisomerism around C(sp2)−C(sp3) single bonds through theoretical
calculations.
substituents), atropisomers have been isolated.1a In the course
of the total synthesis of (−)-linderol A, we prepared endocyclic
INTRODUCTION
■
Atropisomerism is a stereochemical phenomenon in which
epoxide intermediates which also display restricted rotation
hindered single-bond rotation leads to isolable stereoisomers.
around the aryl−Csp3 bond leading to two unequally populated
Ortho-substituted biaryls are the most well-known examples
where the restricted rotation about C(sp2)−C(sp2) single
rotameric epimers with a high rotational barrier between them,
of around 80 kJ mol−1.6
bonds can generate conformationally separable enantiomers.1
New methods for asymmetric (atrop-selective) preparation of
this compound class still continue to develop.1c−h Recently, the
importance of axial chirality has been recognized in drug
discovery.2 Indeed, biological activity at target molecules such
as receptors and enzymes depends on the ability to recognize
the active conformer of a conformationally flexible molecule.
Remarkable results have been reported in this field3 with a
predominant emergence of nonbiaryl atropisomeric com-
pounds such as aryl amides4 or aromatic carbamates5 in
which restricted rotation occurs about an aryl C−N bond.
On the other hand, barriers to rotation about C(sp2)−C(sp3)
single bonds are, in general, so small that isomers arising from
restricted rotation cannot be separated at room temperature.
Nonetheless, in highly hindered systems (when the sp2-
hybridized carbon atom is part of an aryl ring with ortho
These compounds happened to be derivatives in the
(+)-enantiomeric series of natural products with pharmaceut-
ical or potential therapeutic interests, such as cannabidiol7−11
((−)-CBD) and ( )-linderatins12,13 (Figure 1).
Internal rotation of the bond connecting the two rings of
10b
cannabidiol was studied by Tamir10a and Osawa
in the
̆
1980s. A brief description10a of an investigation of the
1
conformational properties of (−)-CBD by dynamic H NMR
and PCILO calculations suggesting a low rotational barrier for
the aryl−Csp3 bond of 28.4 kJ mol−1 at 288 K in CDCl3 was
followed by more detailed study in which the conformational
behavior of (−)-CBD and its mono- or bis-ortho-methoxy
Received: March 14, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
generating atropisomeric diastereoisomers (Scheme 1). DFT
calculations carried out for three representative compounds (1c
and 2c,k) are also described.
a
Scheme 1. Structure of the Synthesized Compounds
Figure 1. Structure of natural products.
derivatives was investigated by dynamic NMR supported by
molecular mechanics at the MM2 level.10b Two major points
arise from this study: first, the previously reported low
rotational barrier was revised (61.4 kJ mol−1 at 295 K in
CDCl3) to be in close agreement with the recalculated value by
Mechoulam (62.7 kJ mol−1 at 288 K in CDCl3). Second,
calculations of model structures showed that phenylcyclohex-
ane derivatives appropriately substituted at β-carbons of the
cyclohexane ring (with two methyl groups) and at the ortho
positions of the phenyl moiety (with two hydroxyl groups)
should exhibit atropisomerism.10b
Regarding epoxide derivatives previously reported in the
literature, restricted rotation around the aryl−Csp3 bond had
never been mentioned, probably due to their symmetrically
substituted aromatic ring.7,11
High rotational barriers, resulting in half-lives of at least 1000
s, are required for separation of atropisomers at room
temperature. Among the different contributions to the
rotational barriers, the bulkiness of the ortho-substituents at
the axis of chirality plays a dominant role.1a,b
However, there are also results indicating that (stereo)-
electronic effects can influence these barriers around C(sp2)−
C(sp2) single bonds. Indeed, when an electron-withdrawing
group is placed at the aryl moiety excluding ortho positions in
aromatic carbamates,5 significant influence on the rotational
barriers induced by the electronic properties of this group has
been observed. Conformational analysis in peptoids4c,d has also
shown stereoelectronic effects such as intramolecular hydrogen-
bonding pattern playing important role in the cis/trans
equilibrium process. Likewise, conformational analysis in
indomethacin derivatives has highlighted stereoelectronic
interactions between neighboring substituents at the indole
unit and the benzoyl moiety.4e A recent Wilcox molecular
torsion balance study has also shown CH−π interactions with
rigid side arm which enables the separation of the two
atropisomers.14a
a
Reagents and conditions: (a) PivCl, py, 80 °C; (b) Ac2O, py, rt or 80
°C; (c) Me2SO4, K2CO3, Me2CO, reflux; (d) (1) Tf2O, py, CH2Cl2, 0
°C; (2) Et3Al, Pd(PPh3)4, THF, reflux; (e) m-CPBA, CH2Cl2, rt.
RESULTS AND DISCUSSION
■
Synthesis of Cannabidiol and Linderatin Derivatives.
A useful approach for the synthesis of this class of molecules
has been recently reported for 1a−c and 2c.6 All the
investigated compounds displayed in Scheme 1 were prepared
starting from a terpenylation reaction between commercially
available (−)-α-phellandrene and conveniently substituted
phenols. This reaction also provided the monoterpenylated
derivatives 1e,h from the corresponding phenols: 2′,6′-
dihydroxyacetophenone and 3,5-dimethylphenol, respectively
(see the Experimental Section). Then, protection of phenols by
esterification conducted to pivalates (or acetates) 1d from 1a,
1f,g from 1e and 1i,j from 1h. The monoterpenylated phenol
1h was also converted into its methyl ether 1k. Alkene 1l with
ortho alkyl substituents (ethyl and methyl) was also obtained
from 1h in two steps.
Stereoelectronic effects can even allow the conformation of
atropisomeric molecules to be controlled in diaryl ethers14b and
in substituted or ring-fused 2-phenylpyridyls.1c The rotational
control in these compounds is possible by exploitation of
dipolar interactions using sulfoxide substituent at ortho position.
In a particular case of C(sp2)−C(sp3) single bonds, π−π
interaction influencing the conformation of aromatic rings in
isolable atropisomers of 2-arylindoline derivatives has been
demonstrated.14c Nevertheless, stereoelectronic effects about
C(sp2)−C(sp3) single bonds and specially in cannabidiol
remain unexplored due probably to the fact that restricted
rotation around such bonds is very rare.
In this paper, we report studies of the influence of steric and/
or electronic factors on conformational exchange in some new
epoxide derivatives of CBD and comparison with the parent
phenylcyclohexene precursors using NMR spectroscopy, with
the aim of shedding light on their fluxional properties and thus
Epoxidation of the resulting phenylcyclohexenes (1d,f,g,i−l)
with m-CPBA gave the corresponding epoxide derivatives
2d,f,g,i−l.
Two major parameters seem to have an influence on the
rotation of the aryl−Csp3 bond: the nature (and in particular
the bulkiness) of ortho substituents, and rigidity of the
endocyclic epoxide ring. According to the nature of the ortho
substituents attached to the phenyl ring, these derivatives were
classified into three categories in order to study steric and
electronic effects on the conformational exchange process
(Scheme 1).
The first group of ortho di-O-substituted compounds (with
hydroxyl, ether or ester groups) includes alkenes 1a−d and
epoxides 2c,d. In this type of compounds, steric and electronic
factors are associated on both ortho positions bearing similar
substituents to probe their both effects on the conformational
equilibria. Their structures are also interesting because they are
B
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
the closest to those of the natural products. However, they
differ from the structure of CBD by the asymmetrically
substituted aromatic ring which can lead to two conformers.
The second group of ortho mono-O-substituted compounds
(with hydroxyl or ester groups) includes alkenes 1e−g and
epoxides 2f,g. This group differs only from the precedent by the
nature of one ortho substituent. Indeed, one O-substituted
group is replaced by a smaller hydrogen atom in order to study
the influence of the endocyclic epoxide ring on restricted bond
rotation versus ortho substituents.
In the last case, we decided to introduce two ortho
substituents, with at least one sufficiently large to generate
observable atropo-diastereoisomeric conformers at room
temperature. And specifically, we chose to put a hindered
group on one ortho position and an O-substituted group on the
other ortho position to gauge steric effects on one side and
electronic effects on the other side. The methyl group seemed
to us a sound choice insofar as it is known to have a high steric
bulk (van der Waals radius of 2 Å). The other ortho substituent
was a hydroxyl, ester, ether or alkyl group in alkenes 1h−l and
in epoxides 2i−l. We also included in this group compounds 1l
and 2l with alkyl ortho substituents as comparison references.
NMR Studies of Conformational Exchange. Eight
alkene (1a−d,i−l) and eight epoxide (2c,d,f,g,i−l) derivatives
were chosen for an intensive NMR study (Scheme 1).
Conformational Analysis. In epoxide derivatives 2 (Figure
2), two slowly intervonverting conformers (P) and (M)15 are
1
detectable at room temperature by H NMR in DMSO-d6.
(P,M)-2c,d were obtained in a ratio of 68:32 and 67:33,
respectively. The structure of rotamers (P)-2c,d and (M)-2c,d
could be identified from the NOESY experiments in DMSO-d6.
In the case of rotamers (P)-2d and (M)-2d, assignment of the
two ortho pivaloyl signals was possible from the six resolved
signals arising from all the t-butyl groups of each rotamer.
1
Moreover, in the H NMR spectra in DMSO-d6, the C3−H
proton is observed at 2.76 and 2.81 ppm in major conformers
(P)-2c,d, respectively and at 3.34 and 2.96 ppm in minor
conformers (M)-2c,d, respectively. This downfield shift, also
observed in other solvents (see Table S1 in the Supporting
Information), is caused by a deshielding effect of the oxygen
lone-pair electrons of the ortho O-substituent, meaning that
important electronic interactions exist between C3−H and the
ortho group in front. It is also noteworthy that the deshielding
effect observed between conformers of 2c (Δδ ∼ 0.5 ppm) with
OAc/OMe as ortho groups is stronger than that observed for
compounds 2d (Δδ = 0.15 ppm) with diester groups (see also
Table S1 in the Supporting Information). This could be
explained by the decreased electron density on the oxygen atom
when the oxygen lone pairs are engaged in delocalization into
the carbonyl group, inducing less interaction with C3−H.
Furthermore, these observations are in agreement with the
structures deduced by NOESY experiments for each conformer
of 2c,d.
Figure 2. Conformational analysis based on the 1H NMR spectra and
on significant NOESY interactions of epoxides 2c,d,i−l in DMSO-d6.
both (M)-2i,j and at 2.90 ppm in (M)-2k, and at 3.05, 3.00, and
3.61 ppm in (P)-2i−k respectively, which are in agreement with
NOESY results. Likewise, the deshielding effect is in the same
range as seen in the “first group” of epoxides 2c,d with the
strongest effect between 2k conformers (Δδ ∼ 0.7 ppm).
For the conformers of 2l (at the bottom in Figure 2), the
C3−H proton chemical shift remains unchanged for the two
conformers (δ ∼ 3.1 ppm), which is not surprising in view of
the similar nature of the substituents (ethyl and methyl
1
groups). In the H NMR spectrum, the ethyl group of the
The observed stereoselectivity for the formation of the major
rotamer (P) seems interesting insofar as (P)-2c is the more
polar conformer identified previously by TLC at low temper-
ature.6 In accordance with this observation, the dipole moments
calculated for 2c are 9.95 D for (P)-2c and 6.63 D for (M)-2c
(Table 6 in the computational section).
In the “third group” of compounds, a mixture of two
conformers (M)-2i−l and (P)-2i−l was also obtained in ratios
of 57:43, 24:76, 72:28 and 45:55 in DMSO-d6, respectively. In
the 1H NMR spectra, C3−H proton is observed at 2.96 ppm in
major conformer (P)-1l is an ABX3 system, while the equivalent
signal in the minor conformer (M)-1l is a simple A2X3 (quartet
of triplets). As the environment of the ethyl group in
interaction with C3−H is chiral, this observation allows us to
assign with certainty the stereochemistry of the major
conformer, which was also confirmed by NOESY experiments.
Furthermore, 2i−k appear as two separate spots in each case
by TLC at room temperature with different Rf values (see the
Experimental Section), suggesting that the two conformers
interconvert slowly enough to be considered as atropisomers.
C
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
The identity of the more polar epimers in 2i,j was established
by analytical HPLC at room temperature (see the Experimental
Section and for chromatograms in the Supporting Informa-
tion): ratios of 67:33 (M)-2i:(P)-2i and 27:73 (M)-2j:(P)-2j
were obtained in 10% EtOAc/hexane. Thus, the more polar
epimers turn out to be (P)-2i,j, which is contradictory in view
of the obtained ratios. Indeed the less polar epimer (M)-2i
prevails as expected, while the formation of the major more
polar epimer (P)-2j remains intriguing. The ratios of the peaks
on the chromatograms differ elsewhere substantially from those
1
obtained by H NMR, which could probably be explained by
the high degree of overlapping peaks associated with a solvent
effect.
The identity of the more polar conformer of 2k was
established from dipole moments (see Table 6 in the
computational part) calculated for each conformer: 3.62 D
for (M)-2k (the more polar epimer) and 2.63 D for (P)-2k (the
less polar epimer). Here again, interestingly major rotamer
(M)-2k turns out to be the more polar.
Thus, the stereoselectivity in the “third group” of epoxides
2i−k seems to be governed by electronic effects as seen for the
“first group” of compounds (2c,d). This aspect is discussed in
detail in the next part with the modeled conformers of 2c,k.
Epoxidation of mono-O-substituted arylalkenes 1f,g (“second
group”) gave, to our surprise, epoxides 2f,g (Scheme 1)
showing a single set of NMR signals in each case with broad
signals for the benzylic (C3−H) and the neighboring (C2−H)
protons. This means that restriction decreased considerably
around the C(sp2)−C(sp3) bond rotation even with the
endocyclic epoxide, and the presence of two ortho substituents
is crucial for observation of a mixture of rotamers. VT-NMR
experiments were conducted at low temperature and two
rotamers appear in a ratio of 76:24 and 95:5 in THF-d8 for 2f
and 2g, respectively (see entries 3, 4 in Table 2). Unfortunately
NOESY experiments at low temperature were unable to
discriminate each epimeric rotamer due to the fact that the
C3−H and ortho-aromatic proton signals were not duplicated.
Alkene derivatives 1 were obtained as single compounds in
each case with a very broad signal for the benzylic C3−H
proton except for 1l. The rotation at the CAr−C3 bond is
restricted but not enough to observe conformers at room
1
temperature by H NMR. At lower temperature, analysis of
proton spectra of alkenes 1a,c,d,i−k provided interesting results
(Figure 3).
Alkene 1a exists as a 56:44 equilibrium mixture of the
rotamers (P) and (M) at 250 K in MeOH-d3. The individual
rotamers (P)-1a and (M)-1a could be identified by NOESY
experiments. Although NOESY experiments failed at low
temperature, rotamers (P)-1c,d and (M)-1c,d could be
identified with the aid of the chemical shift difference for
C3−H protons in the 1H NMR spectra as observed in epoxide
conformers of 2c,d (see spectra and Tables S1, S2 in the
Supporting Information). Indeed, the C3−H proton is
observed at 3.82 and 3.45 ppm for major rotamers (M)-1c
and (P)-1d, respectively, and at 3.28 and 3.55 ppm for minor
rotamers (P)-1c and (M)-1d. As for epoxide 2c, the stronger
deshielding effect is also observed between rotamers of 1c (Δδ
∼ 0.5 ppm). Furthermore, interconversion between di-O-
substituted conformers (P)-1c and (M)-1c led to a ratio of 7:93
at 213 K in THF-d8, which is surprisingly diastereoselective.
The observed deshielding effect between (P)-1d and (M)-1d
(Δδ = 0.10 ppm) seems consistent with the similar nature of
the ortho-O-substituents (two pivaloyl groups).
Figure 3. Conformational analysis based on significant NOESY
correlations in alkenes 1a,i−l and on the H NMR spectra of alkenes
1c,d,i−l.
1
In this “first group” of ortho di-O-substituted alkenes 1c,d, it
seems that the origin of the stereoselectivity is more electronic
than steric. These stereoelectronic features are discussed in
detail in the Computational Studies section with 1c as model.
For the “third group” of compounds, alkenes 1i−k showed a
single set of NMR signals in each case while 1l was observed as
a mixture of two epimeric conformers (M, P) on the NMR time
scale at room temperature (Figure 3). VT-NMR experiments
were conducted at low temperature for 1i−k and two rotamers
(M:P) become evident in a ratio of 75:25, 51:49 and 20:80,
respectively below 233 K in THF-d8. They could be identified
by NOESY correlations and by the chemical shift variations of
1
C3−H proton in H NMR spectra. Here again, the stronger
deshielding effect is observed between ether conformers of 1k
D
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 1. VT-NMR Studies of Alkenes 1a (in MeOH-d3), 1b,c,d,i−k (in THF-d8) and 1l (in DMSO-d6) at 500 MHz
a
b
c
d
e
f
‡
‡
entry
1
compd
coalescing signals
OH (R1)
Δν (Hz)
Tc (K)
K
kc (s−1) kA kB
ΔG‡ (kJ mol−1) ΔGA ΔGB
t1/2 (s) maj (A) min (B)
1a
312
285
151
16
9
283
283
273
248
243
253
283
300
283
283
298
293
288
283
343
333
1.3
1.1
13
1.4
3
388
494
354
451
174
189
18
55.2
54.6
55.4
54.8
54.9
54.8
54.3
54.2
58.3
53.1
58.6
53.1
60.0
59.2
59.5
58.7
62.7
60.1
62.6
60.0
63.2
63.1
63.0
62.9
62.6
59.2
63.0
59.6
74.6
74.1
74.4
73.9
8 × 10−4
6 × 10−4
8.6 × 10−4
7 × 10−4
5 × 10−4
4.5 × 10−4
6.5 × 10−4
6 × 10−4
3.4 × 10−3
4 × 10−4
3.6 × 10−3
4 × 10−4
5.5 × 10−3
4 × 10−3
4 × 10−3
3 × 10−3
2 × 10−2
6 × 10−3
2 × 10−2
6 × 10−3
1.9 × 10−2
1.8 × 10−2
1.8 × 10−2
1.7 × 10−2
2 × 10−2
4 × 10−3
2 × 10−2
5 × 10−3
1.7
OH (R2)
OH
2
3
4
5
6
7
8
1b
1c
1d
1i
Har
20
Har
1
19
CH3CO
Har
26
39
210
23
24
46
32
53
28
25
12
4
56
50
69
CHa
270
373
16
Har (R3)
Har
47
16
49
1j
Har
1.0
4.0
1.2
52
54
C3−H
C2−H
Har (R3)
C2−H
36
38
1k
1l
28
109
15
58
31
37
1.4
Har (R3)
14.7
18
1.6
1.3
a
b
c
Chemical shift separation. Coalescence temperature ( 2−5 K). Equilibrium constant between the two rotamers: A (major) ⇌ B (minor), where
d
e
K = PA/PB. Rate constants at coalescence: kA for A → B and kB for B → A. Free energies of activation ( 0.3−1.1 kJ mol−1) for bond rotation at Tc:
f
‡
‡
ΔGA for A → B and ΔGB for B → A. Half-lives for interconversion of rotamers calculated at 298 K assuming ΔS‡ ≈ 0.
(Δδ = 0.72 ppm) compared to ester conformers of 1i,j (Δδ ∼
0.1 ppm) as found for 1,2c and 1,2d (see also in Tables S1 and
S2 in the Supporting Information). It is noteworthy that in the
1H NMR spectrum of 1k two broad signals for C3−H are
observed at 3.43 and 4.14 ppm in CDCl3 at room temperature,
with a ratio of 25:75.
The conformer ratios obtained for acetate and ether rotamers
of 1i,k are surprisingly high and opposite to those of pivalate
rotamers of 1j even though they all bear a methyl group at ortho
position. Therefore, a steric effect is not sufficient to explain the
high diastereoselectivity between conformers in 1i,k.
Interconversion between (M)-1l and (P)-1l with ortho alkyl
substituents is slow enough on the NMR time scale to allow the
detection of two conformers at room temperature in a ratio of
45:55 in DMSO-d6 or CDCl3 which is interesting because until
now no similar alkene derivative has shown this property
(Figure 3, bottom). Their structures were identified in a similar
manner as the conformers of epoxide 2l (Figure 2, bottom).
As shown for epoxides 2f,g, this result confirms that the
presence of the epoxide ring is not crucial for generating
rotamers on the NMR time scale, unlike the ortho substituents
which seem to be essential.
Determination of Kinetic and Activation Parameters by
VT-NMR Experiments. Dynamic NMR studies were conducted
on eight alkene (Table 1) and eight epoxide (Table 2)
derivatives at low and high temperatures to help understand the
structural features that dictate the dynamics of Ar−Csp3 bond
rotation barriers. We applied the Gutowski−Holm equations16a
and/or the graphical method of Shanan−Atidi and Bar−Eli16b
(see details in the Supporting Information) to all coalescing
signals observed to obtain crude estimates of the rate of Ar−
Csp3 bond rotation in these rotamers. Tables 1 and 2
summarize the results of these investigations and give the
chemical shift differences between the proton (or carbon)
signals which coalesce in the epimeric rotamers (P, M), their
coalescence temperatures and corresponding rate constants, the
E
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Table 2. VT-NMR Studies of Endocyclic Epoxides 2c,d,i−l (in DMSO-d6) and 2f,g (in THF-d8) at 500 MHz
a
b
c
d
e
f
‡
‡
entry
1
compd
2c6
coalescing signals
Har
Δν (Hz)
Tc (K)
K
kc (s−1) kA kB
ΔG‡ (kJ mol−1) ΔGA ΔGB
t1/2 (s) maj (A) min (B)
13
10
19
40
11
42
8
370
363
378
170
253
275
398
395
390
408
403
313
2.1
2
20
42
15
31
28
57
26
82
7
82.1
79.8
81.3
79.2
82.8
80.6
58.7
56.0
57.6
55.0
63.6
56.8
90.8
89.8
92.3
88.4
91.1
87.2
92.3
89.0
92.6
89.4
73.4
72.9
32
13
2
2d
2f
CH3CO
C2−H
Har
24
10
42
17
3
3.2
3 × 10−3
1 × 10−3
2 × 10−3
0.9 × 10−3
2.5 × 10−2
2 × 10−3
337
CH3CO
Har
23
5
4
5
6
2g
2i
2j
19
92
10
13
5
Har
1.3
3.2
234
Har (R3)
CH3ortho
Har CH3O
8
1814
390
17
5.5
17
13.5
33
8.7
21.2
3.7
4.5
8
1127
247
7
8
2k
2l
16.5
10.6
3
2.6
1700
492
1951
574
Har (R3)
1.2
1.1
0.9
a
b
c
Chemical shift separation. Coalescence temperature ( 2−5 K). Equilibrium constant between the two rotamers: A (major) ⇌ B (minor) where
d
e
K = PA/PB. Rate constants at coalescence: kA for A → B and kB for B → A at 298 K. Free energies of activation ( 0.5−1.3 kJ mol−1) for bond
f
‡
‡
rotation at Tc: ΔGA for A → B and ΔGB for B → A. Half-lives for interconversion of rotamers calculated at 298 K by assuming ΔS‡ ≈ 0.
equilibrium constants between the two unequally populated
rotamers, the barriers to Ar−Csp3 bond rotation, and estimates
of half-lives for Ar−Csp3 bond rotation at 298 K for both
rotamers.
combination of OMe/Me as ortho substituents in conformers
of 1k shows the highest diastereoselectivity and ΔG‡ values
(best correlation).
In epoxide series (Table 2), ΔG‡ values are in the range of 55
to 92 kJ mol−1. The highest bond rotation barriers above 90 kJ
mol−1 with half-lives sufficiently higher than those required for
atropisomers (or near-atropisomers) are obtained for 2i−k
(entries 5,6,7) bearing a bulky methyl and an ester or ether O-
substituent. Interconversion between atropisomers of 2j,k has
both the highest stereoselectivities and barriers, meaning that
the combination of Me/OPiv or Me/OMe as ortho substituents
lead to atropisomers (M)-2j,k and (P)-2j,k favoring one of the
two epimers. Barriers around 80 kJ mol−1 were found in
rotamers of 2c,d (entries 1 and 2) bearing two O-substituents.
The increase of about 10 kJ mol−1 from 2c,d to 2i−k is
therefore due to the hindrance of the methyl group larger than
the others. Surprisingly, barriers found between epoxide
conformers of 2l (entry 8) and those of 1l (Table 1, entry 8)
are in the same range, which confirms that in the case of alkyl
groups in the ortho positions the presence of the epoxide ring
does not influence the rotation about the indicated C−C bond.
It is also noteworthy that the barriers increased (about 23−29
kJ mol−1) from the parent alkenes 1c,d,i−k to the
corresponding endocyclic epoxides 2c,d,i−k. This means that
for these compounds with at least one ortho O-substituent the
presence of the epoxy moiety influences the barriers
In alkene series (Table 1), ΔG‡ values are in the range of 54
to 75 kJ mol−1. Among the different data reported in the
literature for cannabidiol (Figure 1) and its monomethyl ether
derivative, rotational barriers calculated in acetone-d6 (53.6 and
54.4 kJ mol−1, respectively)10b are the closest values to those
obtained for the rotamers of 1a,b in MeOH-d3 and THF-d8,
respectively (entries 1, 2; approximately 55 and 54 kJ mol−1)
which possess quite similar structures. Higher bond rotation
barriers above 60 kJ mol−1 are obtained for the conformers of
1i−l (entries 5−8) all bearing one ortho methyl substituent.
The conformers of 1d (entry 4) bearing two ortho pivaloyl
groups also achieve barriers around 60 kJ mol−1. VT-NMR
HSQC experiments were also conducted on 1d (see spectra in
the Supporting Information). From the coalescing aromatic
carbon signals, we also deduced kinetic parameters for the
rotamers of 1d which are in good agreement with those
obtained from coalescing proton signals. Barriers above 70 kJ
mol−1 are reached for conformers of 1l (entry 8) bearing two
ortho alkyl groups which are known to be sterically hindered
substituents. On the other hand, higher diastereoselectivities
with equilibrium constants of 13, 4, and 3 were obtained for 1c,
k, and i respectively (entries 3,7 and 5). Therefore, a
F
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
significantly. Lower barriers of around 60 kJ mol−1 were
obtained for mono-O-substituted endocyclic epoxides 2f,g
(entries 3 and 4). Indeed, barriers decreased considerably over
other epoxides confirming the importance of the two ortho
substituents. Comparison between 2g and 2d, which differ only
in the nature of one ortho substituent R1 (H and OPiv,
respectively), shows a significant decrease of barriers (around
20 kJ mol−1). Moreover, the equilibrium constant for 2g is
substantially higher (K = 19) than others, which is probably due
to the bulkiness of the ortho pivalate group versus H atom.
The ΔG‡ values were also determined for (M)-1c, (P)-2c
and (M)-2j from the Eyring plots of the corresponding rates of
interconversion k obtained by line-shape simulation (Table 3),
Mechoulam10b) and epoxide conformers (P)-2c and (M)-2j
(ΔS‡ < 0) is opposite, indicative of highly organized transition
state in endocyclic epoxides unlike more flexible alkene
derivatives. The discussion is further developed in the next
part with the computed models.
Computational Studies. Alkene 1c and epoxide 2c,k
derivatives were selected as representative structures for
theoretical investigations. The first step involved molecular
mechanics exploration of their conformational space, which for
each compound provided initial geometries of the local minima
using Macromodel software implemented in Schrodinger suite
̈
2013. The conformations were then subjected to a further full
geometry optimization in solution in THF for 1c and in DMSO
for 2c,k using DFT methods implemented in Gaussian09
program at the IEFPCM/M06/6-31G** level of theory.18 All
stationary points (minima, transition states) were properly
characterized by frequency calculations using the same DFT
methodology. Transition states were identified using the
synchronous transit-guided quasi-Newton method (STQN)
and optimized using the Berny algorithm at the IEFPCM/
M06/6-31G** level of theory.19 In the second step, accurate
estimations of energies of minima and transition states were
obtained through single point MP2 calculations at the
IEFPCM/MP2/6-31G**//M06/6-31G** level (see the Sup-
porting Information for full structural parameters). The
structures of the lowest-energy conformers (minima) and
transition-state structures detected for 1c and 2c are shown in
Figure 4, including energetic barriers (ΔE‡ and ΔE° at the MP2
level). In alkene conformers of 1c, the computed barriers agree
well with those determined by VT-NMR experiments (Table 1,
entry 3 and Table 3, entry 1). In epoxide conformers of 2c, the
computed values, although slightly higher than those obtained
experimentally, remain close to them (Table 2, entry 1 and
Table 3, entry 2).
Table 3. Activation Parameters of Alkene 1c and Epoxide
2c,j Conformers by Line Shape Simulation at 298 K
ΔH‡ (kJ
ΔS‡ (J K−1
ΔG‡ (kJ
entry
1
compd
k (s−1
)
mol−1
60.6
)
mol−1
4.5
)
mol−1
59.3
)
t1/2 (s)
a
(M)-1c
241
4 ×
10−3
a
2
3
(P)-2c
(M)-2j
4 ×
78.7
88.7
− 13.8
− 3.8
82.8
89.8
25
10−2
b
3 ×
333
10−3
a
From Eyring plot and rate constants kA for A (major) → B (minor).
From Eyring plot and rate constants kB for B (minor) → A (major).
b
including the activation enthalpies (ΔH‡) and entropies
(ΔS‡).17 We found that this method gives ΔG‡ values that
match those from the coalescence temperature measurements
described above in Tables 1 and 2. Table 3 shows also that ΔS‡
values are near zero, in the range of 4.5 to −13.8 J mol−1 K−1.
The dominating activation-enthalpy terms clearly point to a
free-energy barrier arising primarily from potential energy
constraints for achieving the transition state. It is also
noteworthy that the sign of entropy of activation ΔS‡ calculated
for alkene conformer (M)-1c (ΔS‡ > 0, which also is in
agreement with the previously recalculated value for CBD by
Table 4 reports the complete computed energetic parameters
of 1c and 2c,k conformers. In terms of relative energy between
conformers, (M)-1c was calculated to be the major conformer
at 298 K with a ΔG° of 11.2 kJ mol−1 (6.4 kJ mol−1 at the MP2
Figure 4. Structures and activated energies (kJ mol−1) associated with the epimerization pathway between (P)-1c and (M)-1c through TS-1c (A)
and between (P)-2c and (M)-2c through TS-2c (B) calculated at the IEFPCM/MP2/6-31G**//M06/6-31G** level of theory.
G
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
likewise for their corresponding TS structures. Indeed, the two
rings linked by this bond are almost perpendicular (the aryl ring
is slightly oblique) in each conformer, as observed in biaryl
systems. Similarly, at the saddle point, the two rings are almost
coplanar with Φ in the range of 143−155° in TS-1c and TS-
2c,k. Regarding dihedral angles Θ responsible for the
conformational changes in the monoterpene ring, the half-
chair conformation adopted by each conformer in both alkene
and endocyclic epoxide derivatives 1c and 2c,k is confirmed.10b
On the other hand, a huge difference was observed in Θ angles
in TS-1c versus TS-2c,k. Indeed, a dihedral angle Θ of −66.6°
was calculated for the alkene transition state structure (TS-1c),
suggesting a ring inversion from diequatorial half-chair form in
(P)-1c or (M)-1c to diaxial half-chair form in TS-1c. Therefore,
the ΔS‡ contribution influenced by this deviation should be
positive by favoring more flexible conformational changes, as
obtained experimentally in Table 3 for (M)-1c. However, the
negative computed value in Table 4 is in contradiction with the
experimental value, which is probably due to the difference
(∼10 kJ mol−1) found between its ΔG‡exp (59.3 kJ mol−1) and
Table 4. Computed Energetic Parameters of 1c and 2c,k
Conformers at 298 K
ab
,
ab
,
c
ab
ad
,
ad
,
,
compd
ΔG‡
ΔH‡
ΔS‡
ΔG°
11.2
ΔE‡
ΔE°
(P)-1c
(M)-1c
(P)-2c
(M)-2c
(P)-2k
(M)-2k
57.2
68.3
94.8
93.3
101.8
113.2
56.3
64.3
87.1
87.7
102.1
−2.32
−13.1
−25.8
−18.9
0.07
49.7
56.1
89.8
87.3
93.3
101.2
6.4
0
0
0
0
1.45
11.4
0
2.5
7.9
0
104.8
b
−2.5
a
Energies in kJ mol−1. Energies determined using single point
calculations at the IEFPCM/M06/6-31G** level of theory.
c
Activation entropy in J K−1 mol−1 derived from ΔG = ΔH − TΔS.
d
Energies determined using single point calculations at the IEFPCM/
MP2/6-31G** level of theory.
level) versus (P)-1c. Surprisingly, the minor conformer for 2c is
featured by (M)-2c with a ΔG° of 1.45 kJ mol−1 (2.5 kJ mol−1
at the MP2 level) versus (P)-2c. This inversion in the major
conformers between 1c and 2c seems to come from
stereoelectronic effects arising from the epoxy moiety. In
conformers of 2c, (P)-2k was found to have higher energies
(ΔG° = 11.4 kJ mol−1 and ΔE° = 7.9 kJ mol−1) than the more
stable (M)-2k. The calculated energetic differences between
these conformers are entirely in accordance with ratios
ΔG‡
(68.3 kJ mol−1) values. For TS-2c and TS-2k, a
calc
calculated dihedral angle Θ of +37.4° and +18.5°, respectively,
shows that the endocyclic epoxide undergoes a “small”
contraction to adopt a diaxial twist chair conformation,
meaning that epoxide conformers are less flexible to conforma-
tional change. The expected negative ΔS‡ value calculated for
(P)-2c (Table 4) is in total accordance with the experimental
value in Table 3.
1
determined by H NMR. Likewise, the computed values for
ΔG‡ compare well with those obtained by VT-NMR
experiments reported in Tables 3−5. Barriers are nonetheless
Moreover, the MP2 calculated dipole moments for epoxide
derivatives 2c,k show significant differences between their
conformers versus values found between alkene conformers
(P)-1c and (M)-1c (Table 6). This fact implied that it is the
Table 5. Computed Dihedral Angles of 1c and 2c,k
Calculated at 298 K using IEFPCM/M06/6-31G** and
IEFPCM/MP2/6-31G**//M06/6-31G**
Table 6. Computed Dipole Moments μ of Conformers of 1c
and 2c,k at the IEFPCM/MP2/6-31G**//M06/6-31G**
Level of Theory
compd
1c
2c
2k
a
b
Epimer
(P)-1c
(M)-1c
(P)-2c
(M)-2c
(P)-2k
(M)-2k
compd
Φ
Θ
μ (D)
7.40
7.17
9.95
6.63
2.63
3.62
(P)-1c
TS-1c
117.2
63.7
−66.6
61.7
65.8
37.4
62.2
63.7
18.5
62.2
−154.9
−55.2
121.6
(M)-1c
(P)-2c
TS-2c
presence of the polar epoxy moiety that intrinsically modifies
the polarity of the saturated ring. Interestingly, major
conformers (P)-2c and (M)-2k turn out to be the more
polar. In these conformers, dipoles resulting from ortho
substituents on the aryl ring and from the endocyclic epoxide
appear to be aligned in a rather more parallel fashion than in
the less polar conformers, in which the dipoles are able to
oppose one another and would make them more stable. Thus,
other stereoelectronic effects associated with or compensating
for the polarity of (P)-2c and (M)-2k should make them more
stable than the less polar conformers.
To account for this, the second-order perturbation energy
E(2) of donor−acceptor interactions in the NBO (natural bond
orbital)21 basis was calculated for each conformer of 1c and
2c,k at the IEFPCM/M06/6-31G** level of theory, as reported
in Figure 5 and Table 7. Indeed, deletion energies obtained by
NBO analysis can identify stereoelectronic interactions. In
these computed models, the interactions occur mainly between
O-substituted groups at the ortho positions of the aryl ring and
bonds on the C2 (C2C or C2−O), C3 (C3−H) and C4
(C4−H) carbon atoms in close contact with each other (see
O−H, O−C2 interatomic distances in Table S3 in the
−151.6
−55.0
131.8
(M)-2c
(P)-2k
TS-2k
(M)-2k
−143.4
−56.9
a
b
C2−C3−C6′-C1′ dihedral angle in degree. C3−C4−C5-C6
dihedral angle in degree.
slightly higher (∼10 kJ mol−1) than experimental values, due to
the fact that accurate transition state energy determination by
DFT methods requires meticulous calculations and still remains
a challenge.20 A more precise estimation of these energies was
achieved by single point calculations using second-order
Møller−Plesset perturbation (IEFPCM/MP2/6-31G**/ meth-
odology).
Also, included in Table 5 are the dihedral angles Φ, featuring
the Ar−Csp3 bond rotation limit, and Θ, representing the shape
of the monoterpene ring in conformers of 1c and 2c,k.
Calculated dihedral angles Φ related to the restricted rotation
are found to be similar in both conformers of 1c and 2c,k
H
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 5. Representative interactions by charge transfer between donor−acceptor in the NBO basis for major conformers (M)-1c, (P)-2c and (M)-
2k at the IEFPCM/M06 level of theory.
Table 7. NBO Values of E(2) (kJ mol−1) for 1c and 2c,k at the IEFPCM/M06 Level of Theory
Olp → σ*C2−O
Olp → σ*C3−H
OAc
Olp → σ*C4−H
OMe
compd
OAc
OMe
OMe
5.36
OAc
πC2C → π*CO
total
(P)-1c
(M)-1c
(P)-2c
(M)-2c
(P)-2k
(M)-2k
1.93, 3.48
3.22
8.62
12.85
7.66
6.99
5.28
5.95
2.72
4.77
2.68
3.18
1.55
1.63, 1.80
2.14
4.86
5.28
<1.05
1.63, 1.80
Supporting Information). It is noteworthy that the interactions
between O atoms at the ortho positions and the C3−H proton
observed by 1H NMR (Figures 2 and 3, Tables S1 and S2 in the
Supporting Information) correlate well with these calculations.
And in particular, lone pairs of the OMe goup lead to the
strongest interactions compared to those of the OAc group.22
In the case of alkene conformers of 1c, two sets of close
intramolecular contacts were highlighted: (1) between oxygen
lone pairs at ortho positions of OMe and OAc groups (donors)
and the σ* antibonding orbital of C3−H and C4−H bonds
(acceptors), (2) between the π bonding orbital of C2C bond
(donor) and the π* antibonding orbital of the CO bond of
the OAc group (acceptor). The sum of the amounts of the
charge transfer between donor and acceptor is in favor of
conformer (M)-1c, making it more stable. Note that E(2)
quantities for interactions Olp → σ*C3−H and πC2C → π*CO
are the highest in this conformer versus (P)-1c.
In the conformers of epoxide 2c, Olp → σ*C2−O interaction
occurs in addition to the interactions between ortho oxygen
lone pairs and the σ* antibonding orbital of C3−H and C4−H
bonds. (P)-2c conformer is thereby more stabilized than (M)-
2c by total electron transfers from filled to vacant orbitals (7.66
against 6.99 kJ mol−1). Likewise, for conformers of 2k, two
interactions (Olp → σ*C2−O and Olp → σ*C4−H) occur in favor
of (M)-2k whereas only one (Olp → σ*C3−H) for (P)-2k.
These interactions could reasonably be extended to explain
the stereoselectivity observed in the rotamers of alkenes 1a,i,k,l
(Figure 3) and in rotamers of epoxides 2i,l (Figure 2). Indeed,
for compounds bearing similar ortho substituents, such as the
computed models (1i,k and 2i), the formation of major
conformers (M)-1i, (P)-1k and (M)-2i could easily be
explained by similar charge transfer interactions. Thus, by
comparison of 1c (model) and 1i conformers, (M)-1i appears
σ*C3−H). In case of 1k, (P)-1k is the more stabilized conformer
due to Olp → σ*C3−H stronger interaction than Olp → σ*C4−H
interaction in (M)-1k. In conformers of 2i compared to those
of 2k (model), two interactions (Olp → σ*C2−O and Olp
→
σ*C4−H) are in favor of (M)-2i, whereas only one for (P)-2i
(Olp → σ*C3−H). For conformers in 1a,l and 2l, the poor
stereoselectivity could also be explained by the fact that their
ortho-substituents induce either equivalent interactions (1a) or
no interaction at all (1l and 2l). However, in the more
complicated cases with OPiv group as ortho-substituent(s), as in
the conformers of 1d,j (Figure 3) and 2d,j (Figure 2), it is more
difficult to rationalize the formation of the major conformers,
which requires a more detailed computational analysis.
CONCLUSION
■
Some new cannabidiol and linderatin derivatives were
synthesized by a terpenylation reaction. Spectroscopic NMR
and computational studies of their conformational exchanges
were carried out on both alkene and epoxide derivatives to
identify clearly the different parameters that influence the
restriction to aryl−Csp3 bond rotation. The structures of each
conformer in alkenes 1a,c,d,i−l and epoxides 2c,d,i−l were
1
determined by means of H NMR and NOESY spectra, and
those for 1c and 2c,k were also confirmed by DFT
computation, which also identified transition structures. VT-
NMR spectra provided the barriers and activation parameters
for the conformational processes, which were reasonably in line
with the computed data. Comparison of the results from
differently substituted compounds highlighted unusual stereo-
electronic effects influencing the formation of the major and the
minor conformers.
From these results, we have demonstrated that the potential
for atropisomerism with high diastereoselectivity in these
natural product derivatives depends mainly on the nature of the
two ortho substituents in association with the epoxy moiety.
more stabilized than (P)-1i by two interactions (πC2C
→
→
π*CO and Olp → σ*C4−H) against one for (P)-1i (Olp
I
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
%) 556 (M+, 7), 471 (31), 387 (100), 303 (15); HRMS (EI-TOF) m/
z calcd for C33H48O7 556.3400, found 556.3387.
The combination of a substituted oxygen and a hindered
methyl group at ortho positions in epoxide 2i−k conformers
can exhibit atropisomerism.
2-Acetyl-4-((1R,6R)-6-Isopropyl-3-methylcyclohex-2-enyl)-
1,3-phenylene bis(2,2-dimethylpropanoate) (1g). 60%, liquid:
[α]D 21 + 76.9 (c 1.5, CHCl3); 1H NMR (300 MHz, CDCl3) δ 7.27 (d,
J = 8.7 Hz, 1H), 6.88 (d, J = 8.7 Hz, 1H), 5.11 (s, 1H), 3.30 (br m,
1H), 2.42 (s, 3H), 2.20−1.94 (m, 2H), 1.84−1.43 (m, 7H), 1.41−1.24
(m, 18H), 0.82 (d, J = 6.8 Hz, 3H), 0.74 (d, J = 6.8 Hz, 3H); 13C
NMR (75 MHz, CDCl3) δ 199.5 (C), 176.5 (C), 176.2 (C), 145.5
(C), 145.3 (C), 136.5 (C), 134.3 (C), 130.2 (CH), 128.3 (C), 124.8
(CH), 120.1 (CH), 45.7 (CH), 39.2 (CH), 39.1 (C), 39.0 (C), 31.6
(CH3), 30.6 (CH2), 27.4 (CH), 27.1 (3CH3), 27.0 (3CH3), 23.5
(CH3), 22.0 (CH2), 21.6 (CH3), 16.3 (CH3); MS (EI) m/z (rel
intensity, %) 456 (M+, 1.5), 371 (91), 287 (100), 203 (31), 165 (27),
85 (40); HRMS (EI-TOF) m/z calcd for C28H40O5 456.2876, found
456.2873.
EXPERIMENTAL SECTION
■
1-(2,4-Dihydroxy-3-((1R,6R)-6-isopropyl-3-methylcyclohex-
2-enyl)phenyl)ethanone (1e). A solution of 2′,6′-dihydroxyaceto-
phenone (152 mg, 1 mmol) in dry CH3CN (5 mL) was added
dropwise to a suspension of (−)-α-phellandrene (0.4 mL, 2.5 mmol)
and TsOH (273 mg, 1.5 mmol) in dry toluene (10 mL) at 0 °C under
a N2 atmosphere. The mixture was stirred for 3 h at 0 °C. After
neutralization with saturated NaHCO3, the mixture was extracted with
EtOAc. The combined extracts were washed, dried, and evaporated.
The residue was chromatographed (3:7 CH2Cl2/ petroleum ether) to
give 1e (128 mg, 44%) as a crystalline yellow solid: mp 94−96 °C;
2-((1R,6R)-6-Isopropyl-3-methylcyclohex-2-enyl)-3,5-dime-
21
1
[α]D + 80.7 (c 1.9, CHCl3); H NMR (300 MHz, CDCl3) δ 10.73
(br s, 1H), 9.08 (br s, 1H), 7.09 (d, J = 8.4 Hz, 1H), 6.38 (d, J = 8.4
Hz, 1H), 5.40 (s, 1H), 3.39 (br dt, J = 9.1 Hz, J = 2.4 Hz, 1H), 2.74 (s,
3H), 2.10 (m, 2H), 1.80 (s, 3H), 1.79−1.30 (m, 4H), 0.89 (d, J = 6.8
Hz, 3H), 0.82 (d, J = 6.8 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ
205.6 (C), 160.6 (C), 158.9 (C), 139.0 (C), 137.3 (CH), 124.1 (CH),
121.9 (C), 110.7 (C), 107.7 (CH), 44.3 (CH), 41.4 (CH), 33.7
(CH3), 30.3 (CH2), 27.3 (CH), 23.7 (CH3), 21.6 (CH2), 21.57
(CH3), 16.4 (CH3); IR (NaCl) 3333, 2951, 2925, 1620, 1597, 1429,
1370, 1241, 1225, 1042, 807, 758 cm−1; MS (EI) m/z (rel intensity,
%) 288 (M+, 44), 245 (23), 218 (58), 203 (100), 165 (34), 149 (70);
HRMS (EI-TOF) m/z calcd for C18H24O3 288.1725, found 288.1734.
Anal. Calcd for C18H24O3: C, 74.97; H, 8.39. Found: C, 75.22; H, 8.59.
2-((1R,6R)-6-Isopropyl-3-methylcyclohex-2-enyl)-3,5-dime-
thylbenzene-1-ol (1h). A solution of 3,5-dimethylphenol (934 mg,
7.6 mmol) in dry toluene (10 mL) was added dropwise to a
suspension of (−)-α-phellandrene (3 mL, 18.6 mmol) and TsOH
(2.21 g, 11.6 mmol) in dry toluene (70 mL) at rt under a N2
atmosphere. The mixture was stirred for 3 h at rt. After neutralization
with saturated NaHCO3, the mixture was extracted with EtOAc. The
combined extracts were washed, dried, and evaporated. The residue
was chromatographed (98:2 petroleum ether/EtOAc) to give 1h (1.06
24
thylphenyl pivalate (1j). 56%, liquid: [α]D
+ 69.1 (c 0.44,
CHCl3); 1H NMR (300 MHz, CDCl3) δ 6.82 (s, 1H), 6.51 (br s, 1H),
5.21 (br s, 1H), 3.52 (br s, 1H), 2.32 (s, 3H), 2.26 (s, 3H), 2.12−1.72
(m, 4H), 1.62 (br s, 3H), 1.48−1.08 (m, 11H), 0.81 (d, J = 6.8 Hz,
3H), 0.77 (d, J = 6.8 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 174.1
(C), 150.5 (C), 138.5 (C), 136.3 (C), 132.1 (C), 130.8 and 128.7
(CH), 126.1 and 125.1 (CH), 122.0 and 120.1 (CH), 42.5 (CH),
122.4 (CH), 40.3 (CH), 40.1 (C), 29.8 (CH2), 27.9 (CH), 27.4
(CH3), 27.1 (CH3), 26.6 (CH3), 23.4 (CH3), 23.1 (CH2), 21.9 (CH3),
20.8 (2 CH3), 16.7 (CH3); IR (NaCl) 2956, 2925, 1749, 1476, 1458,
1269, 1117, 1034, 845, 758 cm−1; MS (EI) m/z (rel intensity, %) 342
(M+, 34), 257 (57), 241 (31), 173 (100), 171 (67), 135 (31); HRMS
(EI-TOF) m/z calcd for C23H34O2 342.2559, found 342.2555.
General Procedure for the Synthesis of Acetates 1f,i. Ac2O
(1.5 mL) was added to a solution of 1e or 1h (0.25 mmol) in pyridine
(1 mL) at rt under a N2 atmosphere, and the whole was stirred at rt for
3h (for 1i) or 80 °C overnight (for 1f). The reaction mixture was
evaporated under reduced pressure, and then the residue was purified
by silica gel column chromatography to give acetates 1f or 1i.
2-Acetyl-4-((1R,6R)-6-Isopropyl-3-methylcyclohex-2-enyl)-
21
1,3-phenylene diacetate (1f). 92%, liquid: [α]D + 70.8 (c 1.3,
CHCl3); 1H NMR (300 MHz, CDCl3) δ 7.30 (d, J = 8.6 Hz, 1H), 7.03
(d, J = 8.6 Hz, 1H), 5.14 (d, J = 1.0 Hz, 1H), 3.38 (br m, 1H), 2.45 (s,
3H), 2.29 (s, 3H), 2.24 (s, 3H), 2.03 (m, 2H), 1.79−1.65 (m, 4H),
1.58−1.32 (m, 3H), 0.87 (d, J = 6.7 Hz, 3H), 0.79 (d, J = 6.7 Hz, 3H);
13C NMR (75 MHz, CDCl3) δ 199.4 (C), 169.1 (C), 168.6 (C), 145.6
24
1
g, 54%) as a liquid: [α]D + 83.6 (c 1.8, CHCl3); H NMR (300
MHz, CDCl3) δ 6.56 (s, 1H), 6.54 (s, 1H), 5.94 (s, 1H), 5.49 (s, 1H),
3.52 (br d, J = 9.2 Hz, 1H), 2.27 (s, 3H), 2.24 (s, 3H), 2.21−2.01 (m,
2H), 1.88−1.65 (m, 4H), 1.63−1.48 (m, 2H), 1.45−1.20 (m, 1H),
0.88 (m, 6H); 13C NMR (75 MHz, CDCl3) δ 155.4 (C), 139.7 (C),
137.6 (C), 137.0 (C), 125.3 (C), 124.8 (CH), 123.4 (CH), 116.0
(CH), 42.9 (CH), 39.0 (CH), 30.6 (CH2), 27.4 (CH), 23.6 (CH3),
22.1 (CH2), 21.9 (CH3), 20.8 (2 CH3), 16.7 (CH3); IR (NaCl) 3447,
2951, 2925, 2357, 1620, 1571, 1455, 1305, 1282, 1042, 838 cm−1; MS
(EI) m/z (rel intensity, %) 258 (M+, 41), 188 (36), 173 (100), 135
(28); HRMS (EI-TOF) m/z calcd for C18H26O 258.1984, found
258.1982. Anal. Calcd for C18H26O: C, 83.67; H, 10.14. Found: C,
83.48; H, 10.08.
(C), 145.4 (C), 136.7 (C), 135.3 (CH), 131.4 (CH), 127.6 (C), 123.8
(CH), 120.3 (CH), 45.8 (CH), 39.3 (CH), 31.0 (CH3), 29.8 (CH2),
27.3 (CH), 23.4 (CH3), 21.6 (CH3), 21.3 (CH2), 21.1 (CH3), 20.8
(CH3), 16.9 (CH3); IR (NaCl) 2956, 2925, 1770, 1700, 1470, 1367,
1186, 1037 cm−1; MS (EI) m/z (rel intensity, %) 372 (M+, 0.7), 329
(71), 287 (100), 218 (85), 203 (73), 165 (48); HRMS (EI-TOF) m/z
calcd for C22H28O5 372.1937, found 372.1953. Anal. Calcd for
C22H28O5: C, 70.94; H, 7.58. Found: C, 70.77; H, 7.65.
2-((1R,6R)-6-Isopropyl-3-methylcyclohex-2-enyl)-3,5-dime-
thylphenyl acetate (1i). 80%, liquid: [α]D 24 + 138.3 (c 0.4, CHCl3);
1H NMR (300 MHz, CDCl3) δ 6.84 (br s, 1H), 6.62 (br s, 1H), 5.16
General Procedure for the Synthesis of Pivalates 1d,g,j.
PivCl (1 mL) was added to a solution of 1a, 1e or 1h (0.3 mmol) in
pyridine (1 mL) at rt under a N2 atmosphere, and the mixture was
stirred at 80 °C overnight. The reaction mixture was evaporated under
reduced pressure, and the residue was purified by silica gel column
chromatography to give pivalates 1d, 1g or 1j.
(br s, 1H), 3.50 (br s, 1H), 2.31 (s, 3H), 2.27 (s, 3H), 2.21−1.91 (m,
5H), 1.78 (dt, J = 12.7 Hz, J = 2.3 Hz, 1H), 1.66 (s, 3H), 1.51−1.21
(m, 3H), 0.82 (m, 6H); 13C NMR (75 MHz, CDCl3) δ 169.9 (C),
149.5 (C), 138.6 (C), 136.2 (C), 132.6 (C), 131.9 (C), 128.9 (CH),
125.6 (CH), 122.4 (CH), 42.7 (CH), 39.8 (CH), 30.7 (CH2), 27.7
(CH), 23.4 (CH3), 22.5 (CH2), 21.8 (CH3), 21.0 (CH3), 20.7 (2
CH3), 16.8 (CH3); IR (NaCl) 2951, 2920, 1762, 1445, 1362, 1259,
1204, 1127, 1042, 869, 796 cm−1; MS (EI) m/z (rel intensity, %) 300
(M+, 38), 258 (24), 215 (23), 188 (47), 173 (100), 170 (46), 135
(30); HRMS (EI-TOF) m/z calcd for C20H28O2 300.2089, found
300.2100.
2-((1R,6R)-6-Isopropyl-3-methylcyclohex-2-enyl)-1-me-
thoxy-3,5-dimethylbenzene (1k). A suspension of phenol 1h (110
mg, 0.43 mmol), Me2SO4 (0.1 mL, 110 mg, 0.9 mmol), and anhydrous
K2CO3 (147 mg, 1.08 mmol) in Me2CO (6 mL) was stirred under
reflux for 4 h. The reaction was filtered and evaporated under reduced
2-Acetyl-4-((1R,6R)-6-isopropyl-3-methylcyclohex-2-enyl)-
benzene-1,3,5-triyl tris(2,2-dimethylpropanoate) (1d). 79%,
24
liquid: [α]D
+ 45.96 (c 2.08, CHCl3); 1H NMR (300 MHz,
CDCl3) δ 6.71 (s, 1H), 5.67 (s, 1H), 3.44 (br s, 1H), 2.42 (s, 3H),
2.12−1.63 (m, 5H), 1.58 (s, 3H), 1.50−1.16 (m, 28H), 0.81 (d, J = 6.9
Hz, 3H), 0.72 (d, J = 6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ
198.9 (C), 176.3 (2 C), 175.8 (C), 151.2 (C), 146.9 (C), 145.5 (C),
132.9 (C), 128.3 (C), 126.5 (C), 124.5 (CH), 115.3 (CH), 42.2
(CH), 39.2 (C), 39.1 (C), 39.0 (C), 38.1 (CH), 31.6 (CH3), 31.0
(CH2), 27.9 (CH), 27.1 (6 CH3), 27.0 (3 CH3), 23.2 (CH3), 22.9
(CH2), 21.5 (CH3), 15.9 (CH3); IR (NaCl) 2961, 2930, 1760, 1708,
1602, 1478, 1269, 1093, 1037, 755 cm−1; MS (EI) m/z (rel intensity,
J
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
pressure to give a residue that was purified by column chromatography
(petroleum ether) to yield 1k (90 mg, 77%) as a liquid: 1H NMR (300
MHz, CDCl3) δ 6.56 (s, 1H), 6.54 (s, 1H), 5.20 (br s, 1H), 4.14 (br s,
0.75H), 3.74 (br s, 3H), 3.43 (br s, 0.25H), 2.28 (br s, 6H), 2.15−1.95
(m, 2H), 1.90−1.70 (m, 2H), 1.66 (br s, 3H), 1.50−1.30 (m, 2H),
0.80 (d, J = 6.9 Hz, 3H), 0.79 (d, J = 6.9 Hz, 3H); 13C NMR (75 MHz,
CDCl3) δ 138.3 (C), 135.9 (C), 132.8 (C), 128.6 (C), 126.4 (CH),
125.1 (CH), 109.0 (CH), 55.7 (CH3), 42.8 (CH), 36.3 (CH), 30.9
(CH2), 28.0 (CH), 23.3 (CH3), 22.9 (CH2), 21.8 (CH3), 21.2 (CH3),
20.1 (CH3), 16.0 (CH3); MS (EI) m/z (rel intensity, %) 272 (M+, 32),
202 (56), 187 (100), 171 (26), 149 (14); HRMS (EI-TOF) m/z calcd
for C19H28O 272.2140, found 272.2147.
(M,P)-2-((1R,6R)-6-Isopropyl-3-methylcyclohex-2-enyl)-1-
ethyl-3,5-dimethylbenzene (1l). To the phenol 1h (219 mg, 0.85
mmol) in anhydrous CH2Cl2 (5 mL) was added pyridine (0.14 mL,
0.13 mmol, 1.7 mmol) and the solution was cooled to 0 °C. Tf2O
(0.15 mL, 263 mg, 1.1 mmol) was added dropwise and the mixture
was warmed to rt. The reaction was complete within 5 min as shown
by TLC. The mixture was diluted with CH2Cl2 (5 mL), quenched with
10% aq HCl and washed successively with sat NaHCO3, and brine.
After drying on MgSO4 the solvent was evaporated and the residue
was purified by column chromatography on silica gel (petroleum
ether) to give the triflate (279 mg, 85%) as a liquid.23a
To a solution of the triflate (279 mg, 0.72 mmol) in dry THF (5
mL) were added a 5% Pd(PPh3)4 (40 mg, 0.04 mmol) and Et3Al (0.2
mL, 163 mg, 1.44 mmol). The mixture was refluxed under argon
atmosphere for 2 days. The reaction was followed by TLC. The
reaction solution was evaporated to dryness, CH2Cl2 (10 mL) was
added, and the residue was washed with brine. Chromatography on
silica gel (petroleum ether) yielded 1l (45 mg, 23%) as a liquid:23b dr
45/55; 1H NMR (400 MHz, CDCl3) δ 6.78 (s, 0.45H), 6.74 (s,
0.55H), 6.73 (s, 0.45H), 6.71 (s, 0.55H), 5.21 (s, 0.45H), 5.16 (s,
0.55H), 3.57 (br d, J = 10.2 Hz, 1H), 2.82 (dq, J = 14.3 Hz, J = 7.4 Hz,
0.55H), 2.74 (q, J = 7.5 Hz, 0.9H), 2.49 (dq, J = 14.3 Hz, J = 7.4 Hz,
0.55H), 2.23 (s, 3H), 2.18 (s, 3H), 2.10−1.65 (m, 4H), 1.59 (s, 3H),
1.42−1.22 (m, 2H), 1.07 (m, 3H), 0.8−0.7 (m, 6H); 13C NMR (75
MHz, CDCl3) δ 143.9 (C), 143.7 (C), 138.0 (C), 137.6 (C), 137.3
(C), 135.1 (C), 133.5 (C), 132.5 (C), 131.2 (CH), 129.5 (CH), 128.7
(CH), 127.4 (CH), 126.6 (CH), 126.0 (CH), 44.2 (CH), 42.3 (CH),
41.0 (CH), 40.6 (CH), 30.8 (CH2), 30.7 (CH2), 27.9 (CH2), 27.5
(CH), 27.4 (CH), 25.2 (CH2), 23.4 (CH3), 23.3 (CH3), 22.9 (CH2),
22.8 (CH2), 22.1 (CH3), 22.06 (CH3), 20.8 (CH3), 16.8 (CH3), 16.3
(CH3); MS (EI) m/z (rel intensity, %) 270 (M+, 22), 185 (52), 171
(100), 157 (27); HRMS (EI-TOF) m/z calcd for C20H30 270.2348,
found 270.2348.
85 (42); HRMS (EI-TOF) m/z calcd for C33H48O8 572.3349, found
572.3356.
2-Acetyl-4-((1S,2S,3R,6R)-3-isopropyl-6-methyl-7-
oxabicyclo[4.1.0]heptan-2-yl)-1,3-phenylene diacetate (2f).
21
72%, crystalline solid: mp 105−107 °C; [α]D
+ 34.9 (c 0.75,
CHCl3); 1H NMR (300 MHz, CDCl3) δ 7.29 (d, J = 8.6 Hz, 1H), 7.10
(d, J = 8.6 Hz, 1H), 3.04 (br d, J = 9.6 Hz, 1H), 2.81 (br s, 1H), 2.47
(s, 3H), 2.30 (s, 3H), 2.29 (s, 3H), 2.12 (m, 1H), 1.82−1.58 (m, 2H),
1.46−1.17 (m, 6H), 0.78 (d, J = 6.8 Hz, 3H), 0.72 (d, J = 6.8 Hz, 3H);
13C NMR (75 MHz, CDCl3) δ 199.1 (C), 168.9 (C), 168.5 (C), 146.1
(C), 145.5 (C), 134.5 (C), 131.1 (CH), 128.1 (C), 120.8 (CH), 63.6
(CH), 58.3 (C), 40.4 (CH), 38.8 (CH), 31.1 (CH3), 30.8 (CH2), 26.8
(CH), 23.4 (CH3), 21.3 (CH3), 21.1 (CH3), 20.9 (CH3), 17.5 (CH2),
15.8 (CH3); IR (KBr) 3426, 2956, 2941, 1770, 1698, 1370, 1210,
1184, 1037 cm−1; MS (EI) m/z (rel intensity, %) 388 (M+, 0.5), 346
(21), 286 (74), 243 (100), 216 (73), 203 (36), 165 (85); HRMS (EI-
TOF) m/z calcd for C22H28O6 388.1886, found 388.1888. Anal. Calcd
for C22H28O6: C, 68.02; H, 7.27. Found: C, 67.95; H, 7.35.
2-Acetyl-4-((1S,2S,3R,6R)-3-isopropyl-6-methyl-7-
oxabicyclo[4.1.0]heptan-2-yl)-1,3-phenylene bis(2,2-dimethyl-
24
propanoate) (2g). 68%, amorphous solid: [α]D + 54.8 (c 0.95,
CHCl3); 1H NMR (300 MHz, CDCl3) δ 7.31 (d, J = 8.5 Hz, 1H), 7.06
(d, J = 8.5 Hz, 1H), 2.99 (br d, J = 10.9 Hz, 1H), 2.74 (br s, 1H), 2.44
(s, 3H), 2.11 (m, 1H), 1.82−1.65 (m, 2H), 1.50−1.16 (m, 24H), 0.78
(d, J = 6.8 Hz, 3H), 0.68 (d, J = 6.8 Hz, 3H); 13C NMR (75 MHz,
CDCl3) δ 199.1 (C), 176.5 (C), 176.2 (C), 146.0 (C), 145.6 (C),
134.2 (C), 129.1 (CH), 128.9 (C), 120.5 (CH), 63.7 (CH), 58.1 (C),
44.7 (CH), 39.8 (CH), 39.1 (C), 37.9 (CH), 31.5 (CH3), 30.8 (CH2),
27.9 (CH3), 27.1 (CH3), 27.1 (CH3), 27.0 (CH3), 26.7 (CH3), 22.9
(CH3), 21.3 (CH3), 17.8 (CH2), 16.1 (CH3); IR (NaCl) 2956, 1755,
1706, 1473, 1223, 1096 cm−1; MS (EI) m/z (rel intensity, %) 472
(M+, 4), 285 (36), 243 (38), 216 (44), 165 (20), 57 (100); HRMS
(EI-TOF) m/z calcd for C28H40O6 472.2825, found 472.2820.
(M,P)-2-((1S,2S,3R,6R)-3-Isopropyl-6-methyl-7-oxabicyclo-
[4.1.0]heptan-2-yl)-3,5-dimethylphenyl acetate (2i). 90%, amor-
phous solid: dr 59/41; Rf in 8% EtOAc/hexane = 0.27 and 0.24, HPLC
(Silica, 10% EtOAc/hexane) retention times of the epimers (M) 5.7
1
min and (P) 6.1 min dr 67/33; H NMR (300 MHz, CDCl3) δ 6.88
(s, 1H), 6.71 (s, 0.59H), 6.69 (s, 0.41H), 3.25 (d, J = 11.0 Hz, 0.41H),
3.11 (d, J = 11.2 Hz, 0.59H), 2.89 (s, 0.41H), 2.86 (s, 0.59H), 2.40−
2.25 (m, 9H), 2.21−2.03 (m, 1H), 1.89−1.59 (m, 1H), 1.51−1.13 (m,
6H), 1.00−0.70 (m, 7H); 13C NMR (75 MHz, CDCl3) δ 169.5 (C),
169.4 (C), 149.4 (C), 149.2 (C), 138.7 (C), 137.0 (C), 136.9 (C),
130.6 (C), 130.5 (CH), 130.2 (C), 128.8 (CH), 121.8 (CH), 120.3
(CH), 64.5 (CH), 64.0 (CH), 58.7 (C), 58.4 (C), 43.1 (CH), 42.6
(CH), 39.1 (CH), 36.7 (CH), 30.9 (CH2), 30.2 (CH2), 27.6 (CH),
27.5 (CH), 23.1 (CH3), 22.5 (CH3), 21.6 (CH3), 21.5 (CH3), 21.1
(CH3), 21.0 (CH3), 20.8 (CH3), 20.7 (CH3), 18.1 (CH2), 17.7 (CH2),
16.5 (CH3), 15.9 (CH3); IR (NaCl) 2951, 1765, 1450, 1365, 1204,
1039, 869 cm−1; MS (EI) m/z (rel intensity, %) 316 (M+, 19), 256
(27), 231 (59), 213 (100), 186 (37), 173 (56), 135 (70); HRMS (EI-
TOF) m/z calcd for C20H28O3 316.2038, found 316.2037. Anal. Calcd
for C20H28O3: C, 75.91; H, 8.92. Found: C, 75.94; H, 8.97.
(M,P)-2-((1S,2S,3R,6R)-3-Isopropyl-6-methyl-7-oxabicyclo-
[4.1.0]heptan-2-yl)-3,5-dimethylphenyl pivalate (2j). 70%, amor-
phous solid: dr 30/70; Rf in 10% EtOAc/hexane = 0.33 and 0.27,
HPLC (Silica, 10% EtOAc/hexane) retention times of the epimers
(M) 3.8 min and (P) 4.2 min dr 27/73; 1H NMR (300 MHz, CDCl3)
δ 6.86 (s, 1H), 6.60 (s, 0.7H), 6.49 (s, 0.3H), 3.19 (d, J = 11.4 Hz,
0.7H), 3.11 (d, J = 11.9 Hz, 0.3H), 2.91 (s, 1H), 2.49 (s, 0.9H), 2.47
(s, 2.1H), 2.28 (s, 2.1H), 2.27 (s, 0.9H), 2.10 (m, 1H), 1.88−1.10 (m,
16H), 1.01−0.63 (m, 7H); 13C NMR (75 MHz, CDCl3) δ 177.5 (C),
177.3 (C), 150.6 (C), 150.0 (C), 138.9 (C), 137.0 (C), 136.9 (C),
130.5 (C), 130.3 (CH), 128.8 (CH), 121.3 (CH), 120.1 (CH), 64.9
(CH), 64.1 (CH), 58.6 (C), 58.3 (C), 42.3 (CH), 41.9 (CH), 40.7
(CH), 39.1 (C), 37.0 (CH), 30.7 (CH2), 30.3 (CH2), 27.8 (CH3),
27.7 (CH3), 27.3 (CH3), 26.5 (CH3), 23.2 (CH3), 22.5 (CH3), 21.6
(CH3), 21.2 (CH), 20.7 (CH3), 20.4 (CH3), 19.0 (CH2), 18.2 (CH2),
16.3 (CH3), 16.2 (CH3); IR (KBr) 3442, 2951, 2925, 2367, 2341,
1698, 1633, 1447, 1310, 1295, 1259, 1111, 1039, 827, 677 cm−1; MS
General Procedure for the Synthesis of Epoxides 2d,f,g,i−l.
m-CPBA (1.2 equiv) was added to a solution of 1d,f,g,i−l in CH2Cl2 at
rt under a N2 atmosphere, and the whole was stirred for 0.5−1 h. The
reaction mixture was treated with saturated NaHCO3 and extracted
with CH2Cl2. The organic layer was dried and evaporated. The residue
was purified by silica gel column chromatography to give either
rotameric, atropisomeric or single epoxide(s).
(P,M)-2-Acetyl-4-((1S,2S,3R,6R)-3-isopropyl-6-methyl-7-
oxabicyclo[4.1.0]heptan-2-yl)benzene-1,3,5-triyl tris(2,2-dime-
1
thylpropanoate) (2d). 90%, amorphous solid: dr 68/32; H NMR
(300 MHz, CDCl3) δ 6.84 (s, 0.32H), 6.73 (s, 0.68H), 3.13 (d, J =
12.2 Hz, 0.32H), 2.98 (d, J = 12.1 Hz, 0.68H), 2.95 (s, 0.32H), 2.90 (s,
0.68H), 2.44 (s, 2.04H), 2.43 (s, 0.96H), 2.19 (m, 2H), 1.82−0.82 (m,
34H), 0.76 (d, J = 6.9 Hz, 2.04H), 0.75 (d, J = 6.9 Hz, 0.96H), 0.66 (d,
J = 6.9 Hz, 2.04H), 0.65 (d, J = 6.9 Hz, 0.96H); 13C NMR (75 MHz,
CDCl3) δ 199.1 (C), 198.5 (C), 176.8 (C), 176.6 (C), 176.4 (C),
176.2 (C), 175.8 (C), 151.3 (C), 151.1 (C), 147.1 (C), 146.4 (C),
146.2 (C), 126.5 (C), 126.3 (C), 126.2 (C), 115.6 (CH), 114.9 (CH),
64.3 (CH), 64.1 (CH), 58.2 (C), 41.5 (CH), 41.3 (CH), 39.5 (C),
39.4 (C), 39.3 (C), 39.2 (C), 39.18 (C), 39.0 (CH), 38.8 (CH), 31.8
(CH3), 31.6 (CH3), 30.8 (CH2), 29.8 (CH2), 28.1 (CH), 27.9 (CH),
27.4−27.1 (CH3, t-Bu), 23.1 (CH3), 21.4 (CH3), 21.2 (CH3), 19.24
(CH2), 19.2 (CH2), 15.6 (CH3), 15.5 (CH3); IR (NaCl) 2961, 1760,
1708, 1478, 1266, 1088, 1039, 755 cm−1; MS (EI) m/z (rel intensity,
%) 572 (M+, 0.5), 470 (18), 385 (100), 316 (58), 301 (59), 259 (46),
K
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(EI) m/z (rel intensity, %) 358 (M+, 35), 256 (22), 231 (42), 213
(100), 186 (41), 173 (36), 135 (58); HRMS (EI-TOF) m/z calcd for
C23H34O3 358.2508, found 358.2516.
Dynamic Stereochemistry of Chiral Compounds; Royal Society of
Chemistry: Cambridge, U.K., 2008. (c) Clayden, J.; Fletcher, S. P.;
McDouall, J. J. W.; Rowbottom, S. J. M. J. Am. Chem. Soc. 2009, 131,
5331−5343. (d) Gustafson, J. L.; Lim, D.; Miller, S. J. Science 2010,
328, 1251−1255. (e) Gustafson, J. L.; Lim, D.; Barrett, K. T.; Miller, S.
J. Angew. Chem., Int. Ed. 2011, 50, 5125−5129. (f) Leermann, T.;
Broutin, P.-E.; Leroux, F. R.; Colobert, F. Org. Biomol. Chem. 2012, 10,
4095−4102. (g) Leroux, F. R.; Berthelot, A.; Bonnafoux, L.; Panossian,
A.; Colobert, F. Chem.Eur. J. 2012, 18, 14232−14236. (h) Ros, A.;
Estepa, B.; Ramirez-Lopez, P.; Alvarez, E.; Fernandez, R.; Lassaletta, J.
M. J. Am. Chem. Soc. 2013, 135, 15730−15733.
(M,P)-(1R,4R,5S,6S)-4-Isopropyl-1-methyl-5-(2-methoxy-4,6-
dimethylphenyl)-oxabicyclo[4.1.0]heptane (2k). 74%, liquid: dr
1
78/22; Rf in 5% EtOAc/petroleum ether = 0.27 and 0.23; H NMR
(300 MHz, CDCl3) δ 6.61 (s, 1H), 6.56 (s, 0.78H), 6.54 (s, 0.22H),
3.81 (d, J = 10.9 Hz, 0.22H), 3.78 (s, 2.34H), 3.76 (s, 0.66H), 3.07 (d,
J = 10.6 Hz, 0.78H), 2.89 (s, 0.22H), 2.81 (s, 0.78H), 2.31 (s, 1.32H),
2.30 (s, 4.68H), 2.04 (dt, J = 14.2 Hz, J = 3.2 Hz, 1H), 1.85−1.55 (m,
2H), 1.36 (s, 0.66H), 1.34 (s, 2.34H), 1.35−1.25 (m, 3H), 0.81 (d, J =
6.9 Hz, 2.34H), 0.79 (d, J = 6.7 Hz, 0.66H), 0.74 (d, J = 6.9 Hz, 3H);
13C NMR (75 MHz, CDCl3) δ 157.9 (C), 157.7 (C), 137.9 (C), 136.7
(C), 136.6 (C), 127.8 (C), 127.5 (C), 124.7 (CH), 123.5 (CH), 110.0
(CH), 108.9 (CH), 65.7 (CH), 65.1 (CH), 59.0 (C), 58.7 (C), 55.5
(CH3), 55.1 (CH3), 43.2 (CH), 42.4 (CH), 38.5 (CH), 34.8 (CH),
30.5 (CH2), 27.8 (CH), 23.1 (CH3), 22.7 (CH3), 21.7 (CH3), 21.66
(CH3), 21.3 (CH3), 20.8 (CH3), 18.0 (CH2), 17.8 (CH2), 16.5 (CH3),
15.9 (CH3); MS (EI) m/z (rel intensity, %) 288 (M+, 48), 270 (20),
227 (33), 149 (100), 136 (56), 119 (32); HRMS (EI-TOF) m/z calcd
for C19H28O2 288.2089, found 288.2102.
(M,P)-(1R,4R,5S,6S)-4-Isopropyl-1-methyl-5-(2-ethyl-4,6-di-
methylphenyl)-oxabicyclo[4.1.0]heptane (2l). 82%, liquid: dr 44/
56; 1H NMR (300 MHz, CDCl3) δ 6.91 (s, 0.44H), 6.85 (s, 1H), 6.83
(s, 0.56H), 3.26 (d, J = 11.1 Hz, 1H), 2.90 (s, 1H), 2.83 (dq, J = 14.3
Hz, J = 7.4 Hz, 0.56H), 2.65 (q, J = 7.5 Hz, 0.88H), 2.52 (dq, J = 14.3
Hz, J = 7.4 Hz, 0.56H), 2.35 (s, 1.68H), 2.34 (s, 1.32H), 2.27 (s, 3H),
2.11 (m, 1H), 1.82 (m, 1H), 1.63−1.15 (m, 10H), 0.81 (d, J = 6.8 Hz,
1.32H), 0.80 (d, J = 6.8 Hz, 1.32H), 0.74 (d, J = 6.5 Hz, 1.68H), 0.73
(d, J = 6.9 Hz, 1.68H); 13C NMR (75 MHz, CDCl3) δ 143.9 (C),
142.3 (C), 137.4 (C), 136.8 (C), 136.4 (C), 135.9 (C), 135.75 (C),
135.7 (C), 130.9 (CH), 129.0 (CH), 128.8 (CH), 127.5 (CH), 64.83
(CH), 64.8 (CH), 58.9 (C), 58.91 (C), 44.5 (CH), 42.5 (CH), 39.8
(CH), 39.2 (CH), 30.2 (CH2), 30.16 (CH2), 27.7 (CH2), 27.4 (CH),
27.2 (CH), 25.4 (CH2), 22.7 (CH3), 22.6 (CH3), 21.9 (CH3), 20.8
(CH3), 18.1 (CH2), 18.08 (CH2), 16.9 (CH3), 16.6 (CH3), 16.3
(CH3); MS (EI) m/z (rel intensity, %) 286 (M+, 29), 268 (11), 243
(100), 225 (24), 197 (25), 147 (79), 133 (44); HRMS (EI-TOF) m/z
calcd for C20H30O 286.2297, found 286.2311.
(2) (a) Clayden, J.; Moran, W. J.; Edwards, P. J.; LaPlante, S. R.
Angew. Chem., Int. Ed. 2009, 48, 6398−6401. (b) LaPlante, S. R.;
Edwards, P. J.; Fader, L. D.; Jakalian, A.; Hucke, O. ChemMedChem.
2011, 6, 505−513. (c) LaPlante, S. R.; Fader, L. D.; Fandrick, K. R.;
Fandrick, D. R.; Hucke, O.; Kemper, R.; Miller, S. P. F.; Edwards, P. J.
J. Med. Chem. 2011, 54, 7005−7022.
(3) (a) Cano, C.; Golding, B. T.; Haggerty, K.; Hardcastle, I. R.;
Peacock, M.; Griffin, R. J. Org. Biomol. Chem. 2010, 8, 1922−1928.
(b) Yoshida, K.; Itoyama, R.; Yamahira, M.; Tanaka, J.; Loaec, N.;
̈
Lozach, O.; Durieu, E.; Fukuda, T.; Ishibashi, F.; Meijer, L.; Iwao, M. J.
Med. Chem. 2013, 56, 7289−7301.
(4) (a) Takahashi, H.; Wakamatsu, S.; Tabata, H.; Oshitari, T.;
Harada, A.; Inoue, K.; Natsugari, H. Org. Lett. 2011, 13, 760−763.
(b) Tabata, H.; Nakagomi, J.; Morizono, D.; Oshitari, T.; Takahashi,
H.; Natsugari, H. Angew. Chem., Int. Ed. 2011, 50, 3075−3079.
(c) Moure, A.; Sanclimens, G.; Bujons, J.; Masip, I.; Alvarez-Larena, A.;
Per
́
ez-Paya,
́
E.; Alfonso, I.; Messeguer, A. Chem.Eur. J. 2011, 17,
7927−7939. (d) Paul, B.; Butterfoss, G. L.; Boswell, M. G.; Renfrew, P.
D.; Yeung, F. G.; Shah, N. H.; Wolf, C.; Bonneau, R.; Kirshenbaum, K.
J. Am. Chem. Soc. 2011, 133, 10910−10919. (e) Wakamatsu, S.;
Takahashi, Y.; Tabata, H.; Oshitari, T.; Tani, N.; Azumaya, I.;
Katsumoto, Y.; Tanaka, T.; Hosoi, S.; Natsugari, H.; Takahashi, H.
Chem.Eur. J. 2013, 19, 7056−7063. (f) LaPlante, S. R.; Forgione, P.;
Boucher, C.; Coulombe, R.; Gillard, J.; Hucke, O.; Jakalian, A.; Joly,
M.-A.; Kukolj, G.; Lemke, C.; McCollum, R.; Titolo, S.; Beaulieu, P.
L.; Stammers, T. J. Med. Chem. 2014, 57, 1944−1951.
(5) Tietze, L. F.; Schuster, H. J.; von Hof, J. M.; Hampel, S. M.;
Colunga, J. F.; John, M. Chem.Eur. J. 2010, 16, 12678−12682.
(6) Delaye, P.-O.; Lameiras, P.; Kervarec, N.; Mirand, C.; Berber, H.
J. Org. Chem. 2010, 75, 2501−2509.
ASSOCIATED CONTENT
* Supporting Information
■
S
1H and 13C NMR spectra of compounds 1d−l and 2d,f,g,i−l;
1H NMR spectra at low temperature of 1c,d,i−k; NOESY
spectra of 1a,i−l and 2c,d,i−l; VT-HSQC spectra of 1d; VT 1H
NMR spectra of 1a−d,i−l and 2c,d,f,g,i−l; line-shape simulated
spectra of 1c; Eyring plots data of 1c and 2c,j. HPLC
chromatograms of 2i,j, computational data of 1c and 2c,k. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(7) For review articles on recent advances on CBD, see: Mechoulam,
R.; Peters, M.; Murillo-Rodriguez, E.; Hanus,
2007, 4, 1678−1692.
(8) Appendino, G.; Gibbons, S.; Giana, A.; Pagani, A.; Grassi, G.;
Stavri, M.; Smith, E.; Rahman, M. M. J. Nat. Prod. 2008, 71, 1427−
1430.
̌
L. O. Chem. Biodiversity
(9) (a) Ballerini, E.; Minuti, L.; Piermatti, O.; Pizzo, F. J. Org. Chem.
2009, 74, 4311−4317. (b) Wilkinson, S. M.; Price, J.; Kassiou, M.
Tetrahedron Lett. 2013, 54, 52−54.
(10) (a) Tamir, I.; Mechoulam, R. J. Med. Chem. 1980, 23, 220−223.
̆
AUTHOR INFORMATION
Corresponding Author
*E-mail: hatice.berber@univ-reims.fr.
(b) Kane, V. V.; Martin, A. R.; Jaime, C.; Osawa, E. Tetrahedron 1984,
■
40, 2919−2927.
(11) (a) Mechoulam, R.; Shvo, Y. Tetrahedron 1963, 19, 2073−2078.
(b) Uliss, D. B.; Razdan, R. K.; Dalzell, H. C. J. Am. Chem. Soc. 1974,
96, 7372−7374.
Notes
The authors declare no competing financial interest.
(12) (a) Tanaka, H.; Ichino, K.; Ito, K. Chem. Pharm. Bull. 1984, 32,
3747−3750. (b) Ichino, K.; Tanaka, H.; Ito, K. Tetrahedron 1988, 44,
3251−3260. (c) Ichino, K. Phytochemistry 1989, 28, 955−956.
(13) (a) Orjala, J.; Wright, A. D.; Erdelmeier, C. A. J.; Sticher, O.;
Rali, T. Helv. Chim. Acta 1993, 76, 1481−1488. (b) Benosman, A.;
ACKNOWLEDGMENTS
■
We thank C. Petermann and C. Machado for spectroscopic
recordings (NMR and MS, respectively), S. Lanthony for
microanalysis and HPLC technical assistance, and CNRS,
PlANET for financial support.
́
Oger, J.-M.; Richomme, P.; Bruneton, J.; Sevenet, T.; Ito, K.; Ichino,
K.; Hamid, A.; Hadi, A. J. Nat. Prod. 1997, 60, 921−924.
(14) (a) Bhayana, B. J. Org. Chem. 2013, 78, 6758−6762.
(b) Clayden, J.; Worrall, C. P.; Moran, W. J.; Helliwell, M. Angew.
Chem., Int. Ed. 2008, 47, 3234−3237. (c) Eto, M.; Yamaguchi, K.;
Shinohara, I.; Ito, F.; Yoshitake, Y.; Harano, K. Tetrahedron 2010, 66,
898−903.
REFERENCES
■
(1) (a) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic
Compounds; Wiley-Interscience: New York, 1994. (b) Wolf, C.
L
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(15) P, M nomenclature can also be used in sp2−sp3 axial chirality,
and the conformers can be considered as epimers here; the other
centres of chirality are not indicated in their numeral names for clarity.
(16) (a) Sandstrom, J. Dynamic NMR Spectroscopy; Academic Press:
̈
New York, 1982. (b) Shanan-Atidi, H.; Bar-Eli, K. H. J. Phys. Chem.
1970, 74, 961−963.
(17) Line-shape simulation was carried out using Topspin 2.1 Bruker
program; see the Supporting Information for details.
(18) For full reference of Gaussian09 program; see the Supporting
Information.
(19) (a) Peng, C.; Schlegel, H. B. Isr. J. Chem. 1993, 33, 449−454.
(b) Peng, C.; Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J. Comput.
Chem. 1996, 17, 49−56.
(20) Masson, E. Org. Biomol. Chem. 2013, 11, 2859−2871.
(21) (a) Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102,
7211−7218.
(22) The strongest interactions obtained in the both cases (NBO and
NMR) between OMe and C3-H may refer to intramolecular H-
bonding and/or hyperconjugation between O lone-pair electrons and
the antibonding C3-H orbital (Olp → σ*C3‑H). A deep study
(experimental and theoretical) on the exact nature of these
interactions is in progress and a full discussion will follow in a later
report.
(23) For the preparation of (M,P)-1l, see procedures in
(a) Thompson, A. L. S.; Kabalka, G. W.; Akula, M. R.; Huffman, J.
W. Synthesis 2005, 4, 547−550. (b) Hirota, K.; Isobe, Y.; Maki, Y. J.
Chem. Soc., Perkin Trans. 1 1989, 2513−2514.
M
dx.doi.org/10.1021/jo5006069 | J. Org. Chem. XXXX, XXX, XXX−XXX