NMR spectroscopic and computational study of conformational isomerism in substituted 2-aryl-3 H -1-benzazepines: Toward isolable atropisomeric benzazepine enantiomers

Article abstract of DOI:10.1021/jo4013089

Certain 2-aryl-3H-1-benzazepines are conformationally mobile on the NMR time scale. Variable-temperature NMR experiments bolstered by calculations indicate that alkylation of the azepine ring will slow the interconversion of conformational enantiomers mar

Full text of DOI:10.1021/jo4013089

Article
pubs.acs.org/joc
NMR Spectroscopic and Computational Study of Conformational
Isomerism in Substituted 2‑Aryl‑3H‑1-benzazepines: Toward Isolable
Atropisomeric Benzazepine Enantiomers
Keith Ramig,*,† Edyta M. Greer,*,† David J. Szalda,† Sasan Karimi,‡ Allen Ko,† Laura Boulos,†
Jiansan Gu,† Nathan Dvorkin,† Hema Bhramdat,§ and Gopal Subramaniam*,§
†Department of Natural Sciences, Baruch College of the City University of New York, 17 Lexington Avenue, New York, New York
10010, United States
‡Department of Chemistry, Queensborough Community College of the City University of New York, 222-05 56th Avenue, Bayside,
New York 11364, United States
§Department of Chemistry and Biochemistry, Queens College of the City University of New York, 65-30 Kissena Boulevard, Flushing,
New York 11367, United States
S
* Supporting Information
ABSTRACT: Certain 2-aryl-3H-1-benzazepines are conformationally mobile on the NMR time scale. Variable-temperature
NMR experiments bolstered by calculations indicate that alkylation of the azepine ring will slow the interconversion of
conformational enantiomers markedly. DFT studies show that, while the substitution patterns of the aryl groups at C2 and C4 do
not exert large effects on the rate of enantiomerization, alkylation at C5 slows it appreciably. Alkylation at C3 slows
enantiomerization even more, possibly to the extent that isolation of atropisomers might be attempted.
Scheme 1. Enantiomerization of Benzazepine 1 by
Conformational Exchange
INTRODUCTION
■
The 3H-azepine molecule is nonplanar, by virtue of its lone sp3-
hybridized carbon atom. Since the nitrogen atom breaks
symmetry, the molecule exhibits the planar form of chirality,1
even though it lacks a chiral center. Enantiomerization by
conformational exchange (a ring-flip) in this system is facile
even at low temperature, so the two enantiomers are
inseparable from each other. This has been determined from
NMR studies of the ring-flip of various derivatives, where the
energy barriers to enantiomerization range from 10 to 16 kcal/
mol.2−5 There have been two brief reports on the
enantiomerization of their benzo-fused derivatives, 3H-1-
benzazepines. Brooke and Matthews6 have reported the
synthesis and fluxional properties of the highly fluorinated
3H-1-benzazepines 1a and 1b (Scheme 1; stereochemical
descriptors are assigned using the formalism of Eliel and
Wilen7). They noted that the two protons at C3 in 1a appear as
a two-proton singlet in the NMR spectrum at room
temperature. This is an indication that enantiomerization is
rapid in this system. Upon lowering the temperature to −20
°C, the singlet became two very broad one-proton signals. For
1b, however, the protons at C3 each show a broad doublet at
room temperature, and the doublets persist up to 37 °C, the
highest temperature used in that study. Although the energy
barriers to enantiomerization were not calculated, it is obvious
that 1b has the higher of the two. The high energy barrier in 1b
may be attributed to steric interactions between the fluorine
atom at C12 and the proton at C5 developing in the transition
state of the ring-flip.
The other study8 on the enantiomerization of 3H-1-
benzazepines comes from our laboratories. That study dealt
primarily with the synthesis and mechanism of formation of
benzazepines 2a−2c and other benzazepines. We found that
the NMR signal-coalescence temperature of the C3 protons in
Received: June 17, 2013
Published: July 12, 2013
© 2013 American Chemical Society
8028
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
2,4-diphenyl-3H-1-benzazepine (2a) is much lower than that in
its 5-alkyl derivatives 2b and 2c ((Sp) isomers are shown
arbitrarily). We inferred that the activation-energy barrier to
enantiomerization in the alkyl derivatives 2b and 2c must be
higher than that in the parent 2a. The explanation for this could
be due to steric interactions involving groups near the ring
juncture. In the derivatives alkylated at C5, the alkyl group and
the hydrogen atom at C11 must pass closely by each other
during enantiomerization, which would presumably raise the
energy barrier.
Bulky groups near the ring juncture of the benzodiazepine
system raise the energy barrier to enantiomerization, sometimes
imparting sufficient stability for the individual atropisomers to
be isolated and studied. A comparison of benzodiazepines 3a,
3b, and 3c shows that, as the steric bulk of the N1 substituent
increases, so does the activation-energy barrier (Table 1). This
is because the groups on opposite sides of the ring juncture
become close to each other in the planar transition state of the
interconversion, causing steric repulsion. In 3d, where N1 bears
a t-butyl group, the ring-flip is slow enough to allow resolution
of the enantiomers.21 The energy barrier in this case was
estimated to be at least 24 kcal/mol, which appears to be the
lower limit for handling the individual isomers at room
temperature without appreciable racemization. When two
methyl groups are at opposite sides of the ring juncture, as in
3f, the steric repulsion is even greater.17
Below, we report the application of NMR signal-coalescence
techniques and DFT calculations to the study of enantiome-
rization by conformational exchange in a series of 3H-1-
benzazepines. The 1-benzazepine moiety with various sub-
stitution patterns and saturation levels is an important
substructure found in many pharmacologically active mole-
cules.22−25 In some cases, these molecules and closely related
ones that are also chiral exhibit atropisomerism.26−29 Two
stereoisomeric conformers are considered atropisomers if
interconversion by rotation around single bonds is slow enough
to allow their separation and characterization.30 Determining
the atropisomeric properties of a molecule is important, as it is
becoming clear that enantiomerism based on changes in both
planar and axial chirality has an important bearing on the
biological activity of chiral molecules in general,31 not just in
the benzodiazepines mentioned above. Our studies are aimed at
designing a benzazepine that is conformationally rigid enough
to allow isolation of the stereoisomers. This would unveil
previously inaccessible planar-chiral benzazepines, which may
lead to new single-enantiomer benzazepine drug candidates.
Although 1-benzazepines have been little studied, the
structurally related 1,4-benzodiazepines, which are important
pharmaceutical compounds, have received much attention. A
comparison of benzazepine 2a and a typical benzodiazepine,
desmethyldiazepam (3a), highlights their similarity. Both
exhibit planar chirality by virtue of seven-membered rings
that are puckered into boatlike conformations. This close
similarity makes it not surprising that the protons on C3 in
both molecules appear as two-proton singlets in their proton
NMR spectra at room temperature.9,10 If ring-flipping were
slow or not occurring, then each proton would be a doublet, as
they would experience geminal coupling; diazepam (3b)
exhibits this feature in its NMR spectrum at room temper-
ature.10 These phenomena were first noted for benzodiazepines
by Linscheid and Lehn, who used NMR signal-coalescence
measurements to determine energy barriers to enantiomeriza-
tion.11 A series of NMR experiments and calculations studying
this conformational isomerism then appeared.12−17 In general,
the conformational exchange of benzodiazepines is too fast for
the individual enantiomers to be isolated, although, in some
cases, with especially rigid structures, they can be separated by
chiral HPLC.18 Studies of conformational enantiomerism
continue to be important, because it has been noted that
enzymes in the body can distinguish between the conforma-
tional enantiomers of benzodiazepines.10,19,20
RESULTS AND DISCUSSION
■
Seven benzazepines were chosen for the NMR study (Table 2).
Benzazepines 2a,9 2b,8 2c,8 and 2d32 are known compounds,
and were obtained by literature procedures. The (5-methyl-2-
furyl)-substituted benzazepine 2e was prepared by condensa-
tion of 2-fluoroaniline with 2-acetyl-5-methylfuran (Scheme
2).32 The 4-(o-toluyl)-substituted benzazepine 2f was synthe-
Table 1. Energy Barriers to Enantiomerization of Benzodiazepines from the Literature
a
b
c
cpd
R1
R3
R8
R10
Tc (°C)
ΔG⧧
(kcal/mol)
d
ΔG⧧
(kcal/mol)
calc
exp
e
3a
3b
3c
3d
3e
3f
H
H
H
Cl
Cl
Cl
Cl
Cl
Me
−20
118
12.3
10.7−11.3
16.8−17.6
20.2−21.6
f
f
f
Me
i-Pr
t-Bu
H
H
H
18
f
H
H
>185
>100
>21.3
g
H
H
>24
e
Me
H
H
12.5−12.8
26.3
h
h
Me
Me
>180
25.1
a
b
c
NMR signal-coalescence temperature of the protons of the C3 methylene group. Experimental Gibbs free energy of activation. Calculated Gibbs
d
e
f
g
h
free energy of activation. Reference 11. Reference 13. Reference 15. Reference 21. Reference 17.
8029
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
Table 2. NMR Signal-Coalescence Temperatures and Corresponding Enantiomerization Activation Parameters of 3H-1-
Benzazepines 2a−2f
a
b
c
d
e
cpd
Tc (K)
solvent
ΔG⧧ (kcal/mol)
ΔH⧧ (kcal/mol)
ΔS⧧ (cal/K·mol)
Ea (kcal/mol)
2a
2b
2c
2d
2e
2f
233
323
323
248
223
acetone-d6
benzene-d6
benzene-d6
acetone-d6
acetone-d6
acetone-d6
DMSO-d6
9.9 0.2
14.2 0.2
14.3 0.2
10.6 0.2
9.6 0.2
9.6 0.2
12.7 0.2
12.8 0.3
9.5 0.4
8.3 0.2
−1.8 0.7
−4.8 0.6
−5.0 0.7
−4.0 1.5
−4.9 0.7
9.9 0.2
13.3 0.2
12.8 0.3
9.9 0.4
8.8 0.2
228
f
9.8 0.2
f
8.7 0.2
f
−4.3 1.4
8.9 0.3
f
f
2g
a
NMR signal-coalescence temperature of the protons of the C3 methylene group. In all compounds, the location of the signal due to the methylene
b
protons at C3 was confirmed by NOE to the adjacent substituents at the 2- and 4-positions. Gibbs free energy of activation. Values calculated from
c
the rate constants obtained from Tc measurements, and those obtained from Eyring plots, are identical within the error limits. Activation enthalpy.
d
e
f
Activation entropy. Arrhenius activation energy. There was no change in the NMR spectrum up to 373 K, the temperature limit of the probe used.
by the work of Streef and van der Plas, who alkylated 3H-
azepines at C3 in a similar fashion.35 NMR analysis of the crude
reaction mixture containing benzazepine 2g showed only one of
the two possible diastereomers. The pseudoaxial disposition of
the methyl group is indicated by the X-ray crystal structure
(Figure 1).
Scheme 2. Synthesis of (5-Methyl-2-furyl)-Substituted
Benzazepine 2e
sized in three steps from the trimethylsilyl enol ether of 4′-
methylacetophenone (4), 2′-methylacetophenone, and 2-
fluoroaniline (Scheme 3).8,33,34 The 3-methyl-substituted
Scheme 3. Synthesis of 4-(o-Toluyl)-Substituted
Benzazepine 2f
Figure 1. ORTEP diagram of benzazepine 2g ((Rp) (3R) enantiomer
shown).
NMR Spectroscopic Study of the Enantiomerization
Process. Table 2 shows the energy barriers to enantiomeriza-
tion along with error estimates, from experimental NMR
measurements of the protons at C3 as a function of
temperature. For example, at room temperature, the broadened
singlet due to the protons at C3 in phenyl-substituted 2a
appears at 3.38 ppm.9 At lower temperatures, these protons
appear as two doublets. Coalescence occurs at −40 °C. The
Gibbs free energy of activation (ΔG⧧) at the coalescence
temperature was calculated from the coalescence temperature
(Tc),36 and the error limits were estimated using an uncertainty
of 3 K in the measurement of coalescence temperatures. The
Arrhenius activation energy (Ea) was calculated in a similar
fashion (details are in the Experimental Section and the
benzazepine 2g was prepared by deprotonation of benzazepine
2a with LDA, followed by alkylation of the presumed anion
with methyl iodide (Scheme 4). This procedure was suggested
Scheme 4. Synthesis of 3-Methyl-Substituted Benzazepine 2g
8030
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
Table 3. Computed Dihedral Angles and Enantiomerization Activation Parameters of 3H-1-Benzazepines 2a−2h
a
b
c
d
e
cpd
Θ
(deg)
ΔG⧧ (kcal/mol)
ΔH⧧ (kcal/mol)
ΔS⧧ (cal/K·mol)
Ea (kcal/mol)
2a
2b
2c
2d
2e
2f
31.5
9.8
14.0
13.5
10.0
7.5
8.8
12.5
12.8
9.0
−3.1
−5.0
−2.5
−3.4
−3.9
−3.0
−4.1
9.2
13.0
13.1
9.3
f
37.1 (40.7)
38.1
31.6
31.1
6.3
6.7
32.6
9.1
8.2
8.5
f
2g
2h
26.9 (28.8)
49.3
16.4
26.0
15.2
24.6
15.6
−4.7
e
25.0
a
b
c
d
f
C7−C6−C5−C4 dihedral angle. Gibbs free energy of activation. Activation enthalpy. Activation entropy. Arrhenius activation energy. Value in
parentheses was obtained from X-ray crystal data; for 2b, see ref 8.
Supporting Information). As the coalescence temperatures for
benzazepines 2a−2f vary significantly, it was necessary to
employ different deuterated solvents to access these temper-
atures in solution. The compounds with low Tc were studied in
acetone-d6, whereas those that exhibited higher coalescence
temperatures were studied in benzene-d6. Solvent viscosity can
affect the experimental activation-energy values.37 Though
benzene is more viscous than acetone at room temperature, at
the coalescence temperature of 50 °C, the viscosity of benzene
is 0.442 cP, whereas that of acetone at −42.5 °C is 0.695 cP.38
Because these values are similar to each other, the deviations in
Ea coming from viscosity changes due to temperature are
expected to be small.
only changes associated with the ring-flip; no evidence for the
presence of diastereomers, or any other change in the signals
for the other protons, was seen.
The relatively high energy barriers in 5-methyl-substituted 2b
and butylene-substituted 2c are most likely due to the steric
repulsion engendered by passage of the C5 alkyl group and the
C11 hydrogen atom by each other during the ring-flip. A direct
analogy can be drawn between these cases and the
benzodiazepines 3a−3d (see Table 1). This situation is also
similar to the case of fluorinated benzazepine 1b mentioned
earlier, where increased steric bulk at C11 results in a much
higher barrier to ring-flip.
The behaviors of 3-methyl-substituted benzazepine 2g and 3-
methyl-substituted benzodiazepine 3e make an interesting
comparison. The presence of the methyl group at C3 in
benzodiazepine 3e is calculated to add less than 2 kcal/mol to
the ring-flip energy barrier (vs its 3-desmethyl derivative 3a),
while the preferred orientation of the methyl group is
equatorial.13,42,43 On the other hand, we believe that the
flanking phenyl groups of 3-methyl-substituted 2g cause its C3
methyl group to be axial, and inhibit the ring-flip greatly. The
methyl group may suffer massive steric repulsions in the
transition state of the ring-flip, and in the unobserved
pseudoequatorial-methyl isomer.
Computational Study of the Enantiomerization
Process. Gas-phase B3LYP/6-31G(d) calculations of activa-
tion parameters have been conducted for comparison to
experimental values, and for unraveling the intricacies in the
ring-flip processes, especially of 3-methyl-substituted benzaze-
pine 2g, (5-methyl-2-furyl)-substituted 2e, and 4-(o-toluyl)-
substituted 2f. These three compounds contain complicating
structural features that make explanation of trends in their ring-
flip activation energies a challenge. Also, a 5-t-butyl-substituted
benzazepine (2h) is proposed, with a barrier to inversion that
may be high enough to allow separation and study of individual
enantiomers. The data will show that increased boat character
(puckering of the ring) in the azepine moiety corresponds to an
increase in the inversion barrier in some cases. If this
correspondence is a general phenomenon, then benzazepine
enantiomers would be obtainable based on a sufficiently high
degree of ring-puckering. This work relies heavily on the
thorough studies of Carlier et al.,14−17 who characterized the
ring-flip of benzodiazepines both experimentally and computa-
The ΔG⧧ values were also determined from the Eyring
plots,39 which gave the activation enthalpies (ΔH⧧) and
activation entropies (ΔS⧧). We found that this method gave
ΔG⧧ values that matched those from the coalescence
temperature measurements described above. Table 2 shows
that ΔS⧧ values are very small, in the range of −1.8 to −5.0 cal/
K·mol. The dominating activation-enthalpy terms clearly point
to a free-energy barrier arising primarily from potential energy
constraints for achieving the transition state.
Comparison of phenyl-substituted 2a, (2-naphthyl)-substi-
tuted 2d, (5-methyl-2-furyl)-substituted 2e, and 4-(o-toluyl)-
substituted 2f shows that the nature of the aryl groups at C2
and C4 has only a small effect on the energy barrier to ring-flip.
Benzazepine 2f contains an axis of chirality (the C4-phenyl-C
bond), which is a consequence of the presence of the ortho
methyl group on the phenyl group at C4. (Also, two axes of
chirality exist in both (2-naphthyl)-substituted 2d and (5-
methyl-2-furyl)-substituted 2e; the consequences of this will be
discussed later.) This ortho methyl group might be expected to
hinder rotation of the phenyl group during the ring-flip, or even
cause the molecule to exist as a mixture of diastereomers if the
rotation around the axis of chirality is slow enough. An effect of
this sort has been noted in benzodiazepines substituted with
the 2-fluorophenyl group.40,41 The full enantiomerization
process for 4-(o-toluyl)-substituted 2f consists of changes in
both types of chirality, planar and axial. However, the
experimental energy barrier in this case is lower than that of
phenyl-substituted 2a, so the barrier to rotation around the C4-
phenyl-C bond must be lower than the barrier to the ring-flip.
The NMR spectra of 2f at all the temperatures studied shows
8031
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
Figure 2. B3LYP/6-31G(d) optimized minima for benzazepines 2a−2h.
Figure 3. B3LYP/6-31G(d) calculated transition structures of TS-2a to TS-2h.
tionally, using the same basis set used here. Paizs and Simonyi13
have also shown that this basis set reproduces experimental
values best in these types of systems. Both of these research
groups noted that solvent effects are negligible, so we believe
our use of gas-phase calculations is justified.
Table 3 shows the computational activation data for all the
benzazepines from Table 2 (plus t-butyl-substituted benzaze-
pine 2h, which was not synthesized) based on their global
minimum-energy conformers. The structures of the lowest-
energy conformers of benzazepines 2a−2h are displayed in
Figure 2. (See the Supporting Information for full structural
parameters.) The transition structures were determined by
gradually constraining dihedral angles in the azepine rings of
calculations were also used to verify the transition-state
structures.
Transition-state structures are shown in Figure 3. Interest-
ingly, not all of these are planar. Transition-state structures TS-
2a, TS-2c, TS-2d, TS-2f, TS-2g, and TS-2h deviate in differing
degrees from planarity. Whereas structures TS-2a, TS-2c, TS-
2d, TS-2f, and TS-2h are situated closer to the (Sp) reactants,
structure TS-2g is closer to the (Rp) product. (See the
Supporting Information for full structural parameters.)
Similarly, Lam and Carlier15 and Paizs and Simonyi13 have
found nonplanar transition-state structures in the enantiome-
rization process of benzodiazepines. This deviation from
planarity indicates that the interconversion of (Rp) and (Sp)
enantiomers occurs via two equienergetic pathways, which pass
through one of two enantiomeric transition states.
benzazepines 2a−2h, followed by unconstrained optimization.
The transition-state geometries were verified by one imaginary
frequency corresponding to the displacement vectors distorting
the planar or nearly planar geometry of the transition states.
Wherever possible, intrinsic reaction coordinate (IRC)
The computed values of Table 3 compare well with the
experimental values of Table 2, except in the case of (5-methyl-
2-furyl)-substituted 2e, which has a 24% difference between the
experimental and theoretical values of Ea. The theoretical
8032
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
treatment of this benzazepine is complicated by the presence of
multiple similar ground-state conformers, engendered by the
two axes of chirality introduced by the 5-methyl-2-furyl groups.
The conformer of lowest energy was chosen as the basis for the
calculation. The same is true for (2-naphthyl)-substituted 2d
and o-toluyl-substituted 2f. (Details are available in the
Supporting Information.)
appears to have a free-energy barrier to ring-flip greater than 24
kcal/mol.21 The high barrier of 3d has allowed its individual
enantiomers to be synthesized and studied without inter-
conversion.21 We expect that this property observed in the
benzodiazepine series will hold for the benzazepines as well;
that is, the installation of a bulky substituent near the ring
juncture (C5) should be a generally effective way to enhance
the racemization barrier. The synthesis of 5-t-butyl-substituted
benzazepine 2h, and other C5-substituted benzazepines, in
their enantiomerically pure forms will be the subject of future
work.
The 3-methyl derivative 2g is the most vexing computation-
ally and the most interesting experimentally. It is the
benzazepine whose (Rp) and (Sp) isomers are most likely to
be separable without isomerization. Those isomers are
separated by TS-2g, the transition structure for the ring-flip,
shown in Figure 3. Saddle point TS-2g is a transition structure
as it bears one imaginary frequency. The visualization of this
frequency shows the displacement vectors in TS-2g engaged in
out-of-plane motion of the methyl group at C3, to achieve the
minimum geometry of benzazepine 2g shown in Figure 2.
These motions are accompanied by rotation of the phenyl
groups at C2 and C4 away from the emerging methyl group.
On the basis of this data, benzazepine 2g is too sterically
congested for easy passage of the methyl group past the phenyl
groups. The seven-membered ring in TS-2g is significantly
distorted from planarity, and its geometry is reminiscent of the
ring-flip product.
Also included in Table 3 is the dihedral angle of C7−C6−
C5−C4, designated as Θ. A large Θ value indicates a highly
puckered azepine ring. The data indicate that, in certain cases,
there is a strong dependence of Ea on changes in Θ. Thus, it
appears that a highly puckered azepine ring results in a large Ea.
The Θ values for phenyl-substituted 2a, (2-naphthyl)-
substituted 2d, and (5-methyl-2-furyl)-substituted 2e, however,
are identical within the error limits of the computational
method, and Θ for o-toluyl-substituted 2f differs by about a
degree. Although the degrees of puckering for these four are
identical or nearly so, the Ea values vary markedly, in a way that
is difficult to rationalize. The molecules differ from each other
only in the steric bulk of the aryl groups at C2 and C4. The (5-
methyl-2-furyl)-substituted 2e has the smallest aryl groups;
thus, if steric interactions with the azepine ring are important, it
may be expected to have the lowest Ea. Next is phenyl-
substituted 2a; the larger phenyl groups may cause a higher Ea.
The (2-naphthyl)-substituted 2d has the largest aryl groups, but
much of the steric bulk is situated far away from the azepine
ring. Thus, it behaves like phenyl-substituted 2a. The o-toluyl-
substituted 2f bucks this trend entirely. The methyl group of
the o-toluyl group might be expected to interact strongly with
the azepine ring during the ring-flip, yet its Ea is the second
lowest.
o-Toluyl-substituted 2f, as mentioned earlier, has an axis of
chirality. Thus, the full enantiomerization process includes a
ring-flip and a rotation about the axis of chirality. The rotation
can occur in two ways, with the methyl group passing by either
the C5 hydrogen atom or the C3 hydrogen atoms. We have
modeled this process both for the ground-state geometry
(Figure 2) and for the transition state TS-2f (Figure 3). We
found that rotation past the C5 hydrogen atom is more
energetically favorable than the alternative rotation, by 4 kcal/
mol (see the Supporting Information for details). Since the
methyl group passes by the C5 hydrogen atom with a barrier of
6 kcal/mol, and this barrier is smaller than that of the ring-flip
(8.5 kcal/mol), no evidence of the rotation is seen in the
variable-temperature NMR spectrum.
The benzazepines with alkyl groups at C5, 5-methyl-
substituted 2b, butylene-substituted 2c, and 5-t-butyl-substi-
tuted 2h, present a clearer relationship between steric bulk and
Ea. Here, due to steric interactions of the C5 alkyl group with
the C11 hydrogen atom, the azepine rings are highly puckered,
and the puckering varies regularly with the size of the alkyl
groups. The dihedral angles in the global-minimum geometries
of 5-methyl-substituted 2b and butylene-substituted 2c are
37.1° and 38.1°, respectively; the predicted Ea values for them
are 13.0 and 13.1 kcal/mol, respectively. The 5-t-butyl
derivative 2h is highly puckered, with a dihedral angle of
49.3°. The ring-flip is predicted to proceed via TS-2h, where
the azepine ring is significantly distorted from planarity, making
its geometry closer to the (Sp) reactant. The barrier for
racemization of 5-t-butyl derivative 2h observed computation-
ally is 25.0 kcal/mol, which is the highest barrier of the series of
benzazepines 2a−2h. This value is in line with that for the t-
butyl-substituted benzodiazepine 3d from Table 1, which
Calculations were performed to judge the relative stabilities
of (Sp)-(3S)-2g and its ring-flipped diastereomer (Rp)-(3S)-2g.
A ΔG° of 4.2 kcal/mol was found, in favor of (Sp)-(3S)-2g.
Since its ring-flipped diastereomer is of much higher energy,
one would expect an equilibrium ratio of 1400:1 at 20 °C
favoring (Sp)-(3S)-2g; thus, it should be of high diastereomeric
purity, existing almost exclusively as a pair of noninterconver-
tible enantiomers.
CONCLUSION
■
In our efforts to make conformationally stable stereoisomers of
3H-1-benzazepine derivatives for pharmacological applications,
we investigated their fluxional behavior by varying substituents
at C2, C3, C4, and C5, and analyzing the temperature
1
dependences of their H NMR spectra. The rate of exchange
between the conformational enantiomers did not depend
strongly on the size of the substituents at C2 and C4, and was
rapid at room temperature. Activation energies obtained from
variable-temperature NMR data varied from 8.8 to 9.9 kcal/
mol. A simple methyl substitution at C5 slowed the exchange
and increased the coalescence temperature by about 100 °C,
giving an activation energy of 13.3 kcal/mol. Theoretical
calculations of activation energy agreed with these experimental
values. It is suggested that two forms of substitution can slow
ring-flipping appreciably, thus opening the possibility of
isolating pure stereoisomers: placing either a t-butyl group at
C5 or a methyl group at C3 in the pseudoaxial orientation. In
the former case, the calculated energy barrier to the ring-flip is
25 kcal/mol, high enough for possibly separating and studying
the two enantiomers. In the latter case, namely, a pseudoaxial
1
methyl group at C3, the H NMR spectrum did not show the
presence of the conformational diastereomer. Calculations
revealed that the free-energy difference between the two
diastereomers is 4.2 kcal/mol, in favor of the isomer with the
8033
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
evaporation gave 712 mg of an orange viscous oil. Purification by radial
chromatography (silica gel, 2% EtOAc/hexane) gave 236 mg (37%
yield) of benzazepine 2f as a viscous yellow oil. 1H NMR (CDCl3, 400
MHz) δ 7.74 (d, J = 8.2 Hz, 2H), 7.66 (dd, J = 8.1, 1.2 Hz, 1H), 7.48
(dd, J = 8.1, 1.2 Hz, 1H), 7.45 (td, J = 8.1, 1.4 Hz, 1H), 7.30 (m, 2H),
7.26 (td, J = 7.9 and 1.4 Hz, 1H), 7.15−7.23 (m, 4H), 6.79 (s, 1H),
3.31 (s, 2H), 2.39 (s, 3H), 2.32 (s, 3H); 13C NMR (CDCl3, 100 MHz)
δ 157.2 (q), 146.8 (q), 141.5 (q), 140.4 (q), 136.4 (q), 135.8 (q),
135.6 (q), 130.5, 130.2, 129.4 (q), 129.3, 129.2, 128.6, 128.2, 127.9,
127.8, 126.9, 125.9, 123.7, 36.8 (CH2), 21.4 (CH3), 20.5 (CH3);
HRMS (CI-TOF) calcd. for C24H21N m/z 323.1669, found 323.1669.
2,4-Diphenyl-3-methyl-3H-1-benzazepine (2g). A solution of
benzazepine 2a (1.18 g, 3.99 mmol) and 40 mL of anhydrous THF
was cooled to −68 °C, and LDA (4.0 mL, 2.0 M in THF/heptane/
ethylbenzene, 8.0 mmol) was added dropwise, giving a dark-green
solution. After 1 h, CH3I (0.249 mL, 4.00 mmol) was added and the
cold bath was removed. After 1.25 h at r.t., the yellow solution was
partitioned between 30 mL of H2O and 30 mL of Et2O. Drying of the
organic layer over CaCl2 and rotary evaporation gave 1.31 g of crude
product. Purification by radial chromatography (silica gel, 1% EtOAc/
hexane) gave 309 mg (25% yield) of benzazepine 2g as a yellow
crystalline solid, mp 109−111 °C. Slow recrystallization from 20%
EtOAc/hexane gave large hexagonal prisms suitable for X-ray
pseudoaxial methyl group. This high value results in the
presence of a negligible amount of the other isomer.
EXPERIMENTAL SECTION
■
Calculation of Activation Parameters. The rate of exchange, k,
between the pseudoaxial and pseudoequatorial protons was measured
as a function of absolute temperature, T, and ln k was plotted against
1/T and fitted to a straight line (see the Supporting Information for
plots). The Arrhenius activation energy, Ea, was calculated from the
slope of the fitted straight line. Similarly, ln k/T was plotted against 1/
T and fitted to a straight line to obtain the value of ΔH⧧ from the
slope and ΔS⧧ from the y intercept using the Eyring equation.39 To
estimate the error range in the experimental values for Ea, ΔH⧧, and
ΔS⧧, these values were calculated using the fitted equation for all
temperatures to calculate the lower and upper bounds. The ΔG⧧
values obtained from the Eyring plots and those obtained from
coalescence temperature measurements36 agreed with each other,
within the error limits.
Computational Methodology. Calculations were performed
with Gaussian 09.44 Calculations were conducted using DFT with
the exchange-correlation of B3LYP along with the 6-31G(d) basis set.
B3LYP/6-31G(d) performed well based on the structural features
reproduced for 2g obtained via X-ray crystallography as well as X-ray
data of other benzazepines. The optimizations were carried out in the
gas phase, because previous calculations by Lam and Carlier15 showed
that the gas-phase barriers are well-reproduced in DMSO.
Collection, Reduction, and Refinement of X-ray Data. A
crystal (0.50 × 0.50 × 0.40 mm) of C23H19N (2g) was cut from a large
cluster of crystals and mounted on a glass fiber. Diffraction data for 2g
indicated an orthorhombic symmetry and systematic absences
consistent with space groups Aba2 and Cmna. Space group Aba2
was used for the solution and refinement of the structure based on E
statistics. Crystal data and information about the data collection are
provided in the Supporting Information. The structure of 2g was
solved45 by direct methods. Each crystal contains both optical isomers.
Since Mo radiation was used for the collection of intensity data, the
absolute structure could not be reliably determined, so Friedel pairs
were averaged for the refinement of the structure. There is one
molecule of 2g in the asymmetric unit. In the least-squares
refinement,45 anisotropic temperature parameters were used for all
the non-hydrogen atoms. Hydrogen atoms were placed at calculated
positions and allowed to “ride” on the atom to which they were
attached. The isotropic thermal parameter for the hydrogen atoms in
the structure were determined from the atom to which they were
attached.
1
crystallographic analysis. H NMR (CDCl3, 400 MHz) δ 7.86 (dd, J
= 7.7, 1.8 Hz, 2H), 7.63 (d, J = 8.1 Hz, 1H), 7.55 (d, J = 8.1 Hz, 2H),
7.47 (dd, J = 8.0, 1.1 Hz, 1H), 7.34−7.44 (m, 7H), 7.23 (td, J = 7.6,
1.3 Hz, 1H), 7.09 (s, 1H), 5.00 (q, J = 7.2, 1H), 0.96 (d, J = 7.2 Hz,
3H); 13C NMR (CDCl3, 100 MHz) 160.6 (q), 146.0 (q), 142.0 (q),
140.3 (q), 139.7 (q), 130.4, 129.9, 128.7, 128.5, 128.1 (q), 128.0,
127.9, 127.8, 127.3, 127.1, 126.7, 124.0, 41.0 (CH), 12.3 (CH3);
HRMS (CI-TOF) calcd. for C23H19N m/z 309.1512, found 309.1508.
3-Hydroxy-1-(4-methylphenyl)-3-(2-methylphenyl)butan-1-
one (5). To a solution of TiCl4 (13.2 mL, 1.0 M in CH2Cl2, 13.2
mmol) under N2 was added an additional 5 mL of anhydrous CH2Cl2.
The resulting solution was cooled to 2 °C, and a solution of 2′-
methylacetophenone (1.75 mL, 13.2 mmol) in 4 mL of anhydrous
CH2Cl2 was added dropwise, giving a yellow suspension. A solution of
the trimethylsilyl enol ether of 4′-methylacetophenone8 (2.72 g, 13.2
mmol) in 3 mL of anhydrous CH2Cl2 was added dropwise, giving a
dark red cloudy mixture. This was warmed to r.t. and stirred for 1 h.
The mixture was poured into 35 mL of rapidly stirred ice water. The
organic layer was removed, and the aqueous layer was extracted with
CH2Cl2 (2 × 10 mL). The three organic layers were combined and
washed with 2 × 10 mL of 5% NaHCO3 and once with 20 mL of
brine. The organic layer was dried over MgSO4 and suction-filtered
through a short pad of silica gel. Rotary evaporation gave 1.86 g of a
viscous yellow liquid. Purification by radial chromatography (silica gel,
5% EtOAc/hexane) gave 670 mg (19% yield) of aldol 5 as a viscous
yellow liquid, Rf in 10% EtOAc/hexane = 0.27. 1H NMR (CDCl3, 400
MHz) δ 7.83 (d, J = 8.2 Hz, 2H), 7.34 (dd, J = 6.4, 2.0 Hz, 1H), 7.25
(d, J = 8.2 Hz, 2H,), 7.09 (m, 3H), 4.91 (s, 1H), 3.98 (d, J = 17.8 Hz,
1H), 3.28 (d, J = 17.8 Hz, 1H), 2.60 (s, 3H), 2.40 (s, 3H), 1.67 (s,
3H); 13C NMR (CDCl3, 100 MHz) δ 201.1, 144.7 (q), 144.3 (q),
135.3 (q), 134.5 (q), 132.9, 129.4, 128.2, 127.0, 125.7, 125.4, 75.0 (q),
47.9 (CH2), 29.3 (CH3), 22.5 (CH3), 21.7 (CH3); HRMS (CI-TOF)
calcd. for C18H20O2+H m/z 269.1536, found 269.1526.
2,4-Di(5-methyl-2-furyl)-3H-1-benzazepine (2e). A solution of
2-fluoroaniline (0.65 mL, 6.7 mmol), 2-acetyl-5-methylfuran (0.83 g,
6.7 mmol), and p-toluenesulfonic acid monohydrate (40 mg) in 30 mL
of o-xylene was heated at reflux under N2 for 16 h in a Dean−Stark
apparatus. After cooling to r.t., the solution was washed with 5%
NaHCO3 solution and dried over MgSO4. Removal of the solvent
under vacuum gave 1.03 g of crude product, which was purified by
radial chromatography (silica gel, 20% EtOAc/hexane). Isolated was
0.45 g (44% yield) of 2e as a yellow crystalline solid, mp 157−158 °C.
1H NMR (CDCl3, 400 MHz) δ 7.58 (dd, J = 8.1, 0.8 Hz, 1H) 7.46
(2E)-1-(4-Methylphenyl)-3-(2-methylphenyl)but-2-en-1-one
(6). A solution of aldol 5 (193 mg, 0.719 mmol), triethylamine (0.430
mL, 3.09 mmol), and 4-(dimethylamino)pyridine (9 mg, 0.07 mmol)
in 2 mL of anhydrous CH2Cl2 under N2 was cooled to 0−5 °C. A
solution of trifluoroacetic anhydride (0.200 mL, 1.44 mmol) in 2 mL
of anhydrous CH2Cl2 was added dropwise. The resulting solution was
warmed to r.t. and stirred for 20 h. The reaction solution was poured
into 10 mL of rapidly stirred saturated Na2CO3 solution, and the
resulting mixture was partitioned between 10 mL of H2O and 10 mL
of Et2O. The organic layer was separated, and the aqueous layer was
extracted with 10 mL of Et2O. The two organic layers were combined
and dried over MgSO4. Rotary evaporation gave 200 mg of a viscous
yellow oil. Purification by radial chromatography (silica gel, 5%
EtOAc/hexane) gave 141 mg (78% yield) of enone 6 as a viscous
(dd, J = 7.9, 1.2 Hz, 1H), 7.34 (dt, J = 1.6, 7.4 Hz, 1H), 7.21 (s,1H),
7.18 (dt, J = 1.2, 8.2 Hz, 1H), 7.01 (d, J = 3.3 Hz, 1H), 6.53 (d, J = 3.3
Hz, 1H), 6.12 (dd, J = 3.3, 0.9 Hz, 1H), 6.05 (dd, J = 3.0, 1.0 Hz, 1H),
3.09 (s, 2H), 2.39 (s, 3H), 2.34 (s, 3H); 13C NMR (CDCl3, 100 MHz)
δ 156.5 (q), 153.0 (q), 151.9 (q), 151.0 (q), 147.2 (q), 146.6 (q),
130.5, 129.5 (q), 128.5, 126.7, 124.0 (q), 123.9, 121.8, 115.7, 108.7,
108.6, 107.9, 31.0 (CH2), 14.2 (CH3), 13.9 (CH3); HRMS (CI-TOF)
calcd. for C20H17NO2 m/z 303.1254, found 303.1266.
2-(4-Methylphenyl)-4-(2-methylphenyl)-3H-1-benzazepine
(2f). A solution of enone 6 (491 mg, 1.96 mmol), 2-fluoroaniline
(0.380 mL, 3.92 mmol), and p-toluenesulfonic acid monohydrate (20
mg) in 25 mL o-xylene was heated at reflux under N2 for 23 h using a
Dean−Stark apparatus. The solution was cooled to r.t., and washed
with 10 mL of 5% NaHCO3. Drying over MgSO4 and rotary
8034
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
1
yellow oil, Rf in 10% EtOAc/hexane = 0.50. H NMR (CDCl3, 400
MHz) δ 7.87 (d, J = 8.1 Hz, 2H), 7.25 (d, J = 8.1 Hz, 2H), 7.23−7.15
(m, 4H), 6.82 (q, J = 1.4 Hz, 1H), 2.48 (d, J = 1.4 Hz, 3H), 2.40 (s,
3H), 2.35 (s, 3H);13C NMR (CDCl3, 100 MHz) δ 191.2, 157.3 (q),
144.5 (q), 143.3 (q), 136.6 (q), 134.0 (q), 130.5, 129.2, 128.4, 127.7,
127.2, 125.8, 124.0, 21.7 (CH3), 21.6 (CH3), 19.9 (CH3); HRMS (CI-
TOF) calcd. for C18H18O+H m/z 251.1430, found 251.1425.
(13) Paizs, B.; Simonyi, M. Chirality 1999, 11, 651−658.
(14) Carlier, P. R.; Zhao, H.; DeGuzman, J.; Lam, P. C.-H. J. Am.
Chem. Soc. 2003, 125, 11482−11483.
(15) Lam, P. C.-H.; Carlier, P. J. Org. Chem. 2005, 70, 1530−1538.
(16) Carlier, P. R.; Zhao, H.; MacQuerrie-Hunter, S. L.; DeGuzman,
J. C.; Hsu, D. C. J. Am. Chem. Soc. 2006, 128, 15215−15220.
(17) Carlier, P. R.; Sun, Y.-S.; Hsu, D. C.; Chen, Q.-H. J. Org. Chem.
2010, 75, 6588−6594.
ASSOCIATED CONTENT
* Supporting Information
(18) Lee, S.; Kamide, T.; Tabata, H.; Takahashi, H.; Shiro, M.;
Natsugari, H. Bioorg. Med. Chem. 2008, 16, 9519−9523.
(19) Simonyi, M.; Maksay, G.; Kovacs, I.; Tegyey, Z.; Parkanyi, L.;
Kalman, A.; Otvos, L. Bioorg. Chem. 1990, 18, 1−12.
(20) Fitos, I.; Visy, J.; Zsila, F.; Mady, G.; Simonyi, M. Bioorg. Med.
Chem. 2007, 15, 4857−4862.
■
S
Variable-temperature data plots for 2a−2g; H and 13C NMR
1
spectra for 2e−2g, 5, and 6; Arrhenius plots and Eyring plots
for 2a−2f; selected crystallographic data and CIF file for 2g;
additional computational details; complete ref 44. This material
is available free of charge via the Internet at http://pubs.acs.org.
(21) Gilman, N. W.; Rosen, P.; Earley, J. V.; Cook, C.; Todaro, L. J. J.
Am. Chem. Soc. 1990, 112, 3969−3978.
(22) Duffey, M. O.; Vos, T. J.; Adams, R.; Alley, J.; Anthony, J.;
Barrett, C.; Bharathan, I.; Bowman, D.; Bump, N. J.; Chau, R.; Cullis,
C.; Driscoll, D. L.; Elder, A.; Forsyth, N.; Frazer, J.; Guo, J.; Guo, L.;
Hyer, M. L.; Janowick, D.; Kulkarni, B.; Lai, S.-J.; Lasky, K.; Li, G.; Li,
J.; Liao, D.; Little, J.; Peng, B.; Qian, M. G.; Reynolds, D. J.; Rezaei,
M.; Scott, M. P.; Sells, T. B.; Shinde, V.; Shi, Q. J.; Sintchak, M. D.;
Soucy, F.; Sprott, K. T.; Stroud, S. G.; Nestor, M.; Visiers, I.;
Weatherhead, G.; Ye, Y.; Damore, N. J. Med. Chem. 2012, 55, 197−
208.
AUTHOR INFORMATION
Corresponding Author
*E-mail: keith.ramig@baruch.cuny.edu (K.R.), edyta.greer@
baruch.cuny.edu (E.M.G.), gopal.subramaniam@qc.cuny.edu
(G.S.).
■
Notes
The authors declare no competing financial interest.
(23) Becker, A.; Kohfeld, S.; Lader, A.; Preu, L.; Pies, T.; Wieking, K.;
Ferandin, Y.; Knockaert, M.; Meijer, L.; Kunick, C. Eur. J. Med. Chem.
2010, 45, 335−342.
ACKNOWLEDGMENTS
■
We thank Dr. E. Fujita of Brookhaven National Laboratory for
the use of the Bruker Kappa Apex II diffractometer for X-ray
data collection. The experimental assistance of Nataliya Pokeza
and Benjamin Kaplan is much appreciated. We thank the
reviewers for their comments and for suggesting the method
that allowed finding the transition state of compound 2g. The
Professional Staff Congress of the City University of New York
is acknowledged for financial support of this work.
(24) Seto, M.; Aikawa, K.; Miyamoto, N.; Aramaki, Y.; Kanzaki, N.;
Takashima, K.; Kuze, Y.; Iizawa, Y.; Baba, M.; Shiraishi, M. J. Med.
Chem. 2006, 49, 2037−2048.
(25) Jackson, P. F.; Davenport, T. W.; Garcia, L.; McKinney, J. A.;
Melville, M. G.; Harris, G. G.; Chapdelaine, M. J.; Damewood, J. R.;
Pullan, L. M.; Goldstein, J. M. Bioorg. Med. Chem. Lett. 1995, 5, 3097−
3100.
(26) Cole, K. P.; Mitchell, D.; Carr, M. A.; Stout, J. R.; Belvo, M. D.
Tetrahedron: Asymmetry 2009, 20, 1262−1266.
REFERENCES
■
(27) Liu, P.; Lanza, T. J., Jr.; Chioda, M.; Jones, C.; Chobanian, H.
R.; Guo, Y.; Chang, L.; Kelly, T. M.; Kan, Y.; Palyha, O.; Guan, X.-M.;
Marsh, D. J.; Metzger, J. M.; Ramsay, K.; Wang, S.-P.; Strack, A. M.;
Miller, R. J.; Pang, J.; Lyons, K.; Dragovic, J.; Ning, J. G.; Schafer, W.
A.; Welch, C. J.; Gong, X.; Gao, Y.-D.; Hornak, V.; Ball, R. G.; Tsou,
N.; Reitman, M. L.; Wyvratt, M. J.; Nargund, R. P.; Lin, L. S. ACS Med.
Chem. Lett. 2011, 2, 933−937.
(1) Eliel, E. E.; Wilen, S. H. Stereochemistry of Carbon Compounds;
John Wiley and Sons: New York, 1994; pp 1119−1122 and 1166−
1175.
(2) Mannschreck, A.; Rissmann, G.; Vogtle, F.; Wild, D. Chem. Ber.
̈
1967, 100, 335−346.
(3) van Bergen, T. J.; Kellogg, R. M. J. Org. Chem. 1971, 36, 978−
983.
(4) Katritzky, A. R.; Aurrecoechea, J. M.; Quian, K.; Koziol, A. E.;
Palenik, G. J. Heterocycles 1987, 25, 387−391.
(5) Kowalski, D.; Erker, G.; Kotila, S. Liebigs Ann. 1996, 887−890.
(6) Brooke, G. M.; Matthews, R. S. J. Fluorine Chem. 1988, 40, 109−
117.
(7) Referring to ref 1, the (Sp) descriptor of the isomer on the left in
Scheme 1 was assigned in the following way: C2 was chosen as the
pilot atom, the atom of highest priority out of the plane formed by the
benzo group plus C5 and N. From a viewpoint looking down on the
molecule (i.e., so that C2 is between the viewer and the planar
portion), the next three atoms from the pilot atom in order of priority
are marked. Since these atoms, namely, N, C7, and C6, describe a
counterclockwise turn, the (Sp) descriptor is assigned. The assignment
is reversed when the molecule undergoes a ring-flip.
(8) Ramig, K.; Greer, E. M.; Szalda, D. J.; Razi, R.; Mahir, F.; Pokeza,
N.; Wong, W.; Kaplan, B.; Lam, J.; Mannan, A.; Missak, C.; Mai, D.;
Subramaniam, G.; Berkowitz, W. F.; Prasad, P.; Karimi, S.; Lo, N. H.;
Kudzma, L. V. Eur. J. Org. Chem. 2010, 2362−2371.
(9) Kudzma, L. V. Synthesis 2003, 1661−1666.
(28) Tabata, H.; Wada, N.; Takada, Y.; Oshitari, T.; Takahashi, H.;
Natsugari, H. J. Org. Chem. 2011, 76, 5123−5131.
(29) Welch, C. J.; Gong, X.; Schafer, W.; Chobanian, H.; Lin, L.;
Biba, M.; Liu, P.; Guo, Y.; Beard, A. Chirality 2009, 21, E105−E109.
(30) Oki, M. Top. Stereochem. 1983, 14, 1−81.
(31) Clayden, J. Ang. Chem., Int. Ed. 2009, 48, 6398−6401.
(32) Ramig, K.; Alli, S.; Cheng, M.; Leung, R.; Razi, R.; Washington,
M.; Kudzma, L. V. Synlett 2007, 2868−2870.
(33) Mukaiyama, T.; Narasaka, K. Org. Synth. 1987, 65, 6−11.
(34) Narasaka, K. Org. Synth. 1987, 65, 12−16.
(35) Streef, J. W.; van der Plas, H. C. Tetrahedron Lett. 1979, 2287−
2290.
(36) ΔG⧧ = 4.58 Tc(10.32 + log Tc/kc) cal mol−1. Here, Tc is the
c
coalescence temperature and kc is the exchange rate constant at the
coalescence temperature. kc = πΔν/√2, where Δν is the separation of
the two coalescing proton signals in Hz. See: Sandstrom, J. Dynamic
̈
NMR Spectroscopy; Academic Press: New York, 1982.
(37) Anna, J. M.; Ross, M. R.; Kubarych, K. J. J. Phys. Chem. A 2009,
113, 6544−6547.
(38) Weast, R., Ed. CRC Handbook of Chemistry and Physics, 62nd
ed.; CRC Press: Boca Raton, FL, 1981−1982; p F−43.
(39) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic
Chemistry; University Science Books: Herndon, VA, 2006; pp 370−
371.
(10) Blount, J. F.; Fryer, R. I.; Gilman, N. W.; Todaro, L. J. Mol.
Pharmacol. 1983, 24, 425−428.
(11) Linscheid, P.; Lehn, J.-M. Bull. Soc. Chim. Fr. 1967, 992−997.
(12) Sunjic, V.; Lisini, A.; Sega, A.; Kovac, T.; Kajfez, F.; Ruscic, B. J.
Heterocycl. Chem. 1979, 16, 757−761.
8035
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036

The Journal of Organic Chemistry
Article
(40) Finner, E.; Zeugner, H.; Milkowski, W. Arch. Pharm. (Weinheim,
Ger.) 1984, 317, 369−371.
(41) Finner, E.; Zeugner, H.; Milkowski, W. Arch. Pharm. (Weinheim,
Ger.) 1984, 317, 1050−1053.
(42) Alebic-Kolbah, T.; Kajfez, F.; Rendic, S.; Sunjic, V. Biochem.
Pharmacol. 1979, 28, 2457−2464.
(43) Konowal, A.; Snatzke, G.; Alebic-Kolbah, T.; Kajfez, F.; Rendic,
S.; Sunjic, V. Biochem. Pharmacol. 1979, 28, 3109−3113.
(44) Frisch, M.; et al. Gaussian 09, Revision C.01. (See the
Supporting Information for full citation.)
(45) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
8036
dx.doi.org/10.1021/jo4013089 | J. Org. Chem. 2013, 78, 8028−8036