Atropisomerism in monopyrroles

Article abstract of DOI:10.1016/S0957-4166(02)00481-0

As observed by NMR, iodopyrroles 1a and 1b (ethyl and methyl 3,5-dimethyl-4-[(1′-iodo-2′,2′-dimethyl)propyl]pyrrole-2- carboxylate) and a variety of related derivatives with iodine replaced by methoxy 2, thiomethyl 3, acetic acid esters 4, propionic acid ester 5 or malonic esters 6 exhibit restricted rotation about the C(4)-C(1′) bond due to the bulky tert-butyl group and an ortho effect from the sterically crowded 3,5-dimethylpyrrole. Most of the compounds, which are members of the rare class of atropisomers due to restricted rotation about an sp³-sp² C-C bond, undergo diastereomeric enrichment by preparative TLC and crystallization. From dynamic NMR studies of the enriched diastereomers one can determine kinetic and thermodynamic parameters associated with the atropisomerism, e.g., ΔG? ～24 kcal/mol for 1 and 5 (313 K), ～22 kcal/mol for 3 (273 K), and ～25 kcal/mol for 6 (313 K) in C₂D₂Cl₄ solvent.

Full text of DOI:10.1016/S0957-4166(02)00481-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 13 (2002) 1721–1732
Atropisomerism in monopyrroles
Stefan E. Boiadjiev and David A. Lightner*
Department of Chemistry, University of Nevada, Reno, NV 89557-0020, USA
Received 6 June 2002; accepted 13 August 2002
Abstract—As observed by NMR, iodopyrroles 1a and 1b (ethyl and methyl 3,5-dimethyl-4-[(1%-iodo-2%,2%-dimethyl)propyl]pyrrole-
-carboxylate) and a variety of related derivatives with iodine replaced by methoxy 2, thiomethyl 3, acetic acid esters 4, propionic
2
acid ester 5 or malonic esters 6 exhibit restricted rotation about the C(4)ꢀC(1%) bond due to the bulky tert-butyl group and an
ortho effect from the sterically crowded 3,5-dimethylpyrrole. Most of the compounds, which are members of the rare class of
3
2
atropisomers due to restricted rotation about an sp –sp CꢀC bond, undergo diastereomeric enrichment by preparative TLC and
crystallization. From dynamic NMR studies of the enriched diastereomers one can determine kinetic and thermodynamic
‡
parameters associated with the atropisomerism, e.g., DG ꢀ24 kcal/mol for 1 and 5 (313 K), ꢀ22 kcal/mol for 3 (273 K), and
ꢀ
25 kcal/mol for 6 (313 K) in C D Cl solvent. © 2002 Elsevier Science Ltd. All rights reserved.
2 2 4
1
. Introduction
Although 1a was characterized, only its melting point
and combustion analysis data were published. NMR
data were absent and not mentioned. We repeated the
synthesis and isolation of pure 1a and also prepared
methyl ester 1b, both of which are useful compounds in
the synthesis of highly hindered bilirubins and bili-
Atropisomerism is usually associated with biaryl sys-
tems, where bond rotation about the sp –sp CꢀC bond
is sufficiently restricted so as to lead to separable con-
formers. Typically, this means a free energy barrier of
6.2 kcal/mol at 300 K and a half-life of 1000 s.
Restricted rotation about an sp –sp CꢀC bond has
been observed less often. In systems involving fluore-
2
2
1
,2
1
2
verdins. Interestingly, the H NMR spectrum of 1a was
2
3
3
nes, with sufficiently large substituents judiciously
placed to interfere with bond rotation, atropisomers
1,2,4
have been isolated.
One of the simplest examples has
only one aromatic ring: the di-tert-butyl compound of
5
Fig. 1A, whose atropisomers have been isolated. There
are far fewer examples of atropisomerism of the biaryl
6
type among pyrrole compounds, a research area of
current interest to us. In the following, we report what
2
we believe to be the first examples (Fig. 1B) of sp C–
3
sp C atropisomerism in a monopyrrole.
2
. Results and discussion
2.1. Iodopyrroles 1a and 1b
7
Over 25 years ago, Khan and Plieninger reported on
the synthesis of 1a from a b-free pyrrole 7a by reaction
with pivaldehyde in the presence of HI (Scheme 1).
Figure 1. (A) Isolable atropisomers with one aromatic ring.
(
B) Atropisomeric monopyrroles with restricted rotation
about the C(4)ꢀC(1%) bond. The absolute configuration at
C(1%) is arbitrary.
*
Corresponding author. Fax: 775-784-6804; e-mail: lightner@
scs.unr.edu
0
957-4166/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(02)00481-0

1
722
S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1
Figure 2. Partial H NMR spectra of 1a showing (A)
diastereotopic –OCH – hydrogens and (B) showing signal
2
doubling. (Signal doubling is also observed in 1b.)
Scheme 1.
consistent with diastereotopic hydrogens from the CH2
group of the ethyl ester unit (Fig. 2A), and all of the
1
3
proton (Fig. 2B) and C NMR signals (Fig. 3) were
doubled in both 1a and 1b. Since 1a and 1b possess but
a single stereogenic center, a second element of chirality
(
namely an axis of chirality along the C(4)ꢀC(1%) bond)
must be present in order to account for the apparent
diastereomers seen by NMR, which suggests that 1a
and 1b are mixtures of diastereomeric atropisomers
(
Fig. 4). However, chromatography (TLC) failed to
separate isomers of 1a and 1b, and induced decomposi-
tion. (These iodo compounds are stable as solids but
rather unstable or reactive in solution.)
13
Figure 3. Partial C NMR APT spectra of 1a showing signal
doubling in the high field aliphatic region (top) and in the
pyrrole ring carbon region (bottom). The carbon signals are
also doubled in 1b.
1
Integration of the H NMR signals indicated an 88:12
ratio in 1a and a 7:3 ratio in 1b. Crystallization of 1b
1
changed the ratio from 7:3 to 93:7, as determined by H
obtained
by
crystallization
from
hexane–
NMR integration, but after standing in CDCl solvent
3
dichloromethane with slow evaporation of the
dichloromethane at 0°C. Similarly, a highly enriched
for 72 hours at room temperature, the ratio returned to
7
:3, which we take to represent an equilibrated mixture.
(
93:7) sample of 1b was obtained by crystallization
(
The 88:12 ratio observed for 1a, a ratio that we
from hexane–dichloromethane (by evaporation of the
latter at 0°C). These diastereomerically enriched sam-
ples were used to study the rates of interconversion of
the atropisomers.
attribute to the lower reaction temperature used in its
synthesis, drifted to 7:3 under the same experimental
conditions.) Near diastereomerically pure 1a (99:1) was

S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1723
Figure 4. The diastereomers of methyl 3,5-dimethyl-4-[(2%2%-dimethyl-1%-iodo)propyl]-pyrrole-2-carboxylate 1b, as viewed from
C(4) to C(1%). There are two atropisomeric pairs: (A) and (B), and two enantiomeric pairs: (aR,R)+(aS,S) and (aS,R)+(aR,S).
2
.2. Methyl ether 2 and methyl thioether 3
were observed in the corresponding mother liquors.
Chromatography (TLC or radial) failed to separate the
atropisomers of 4a and 4b, thus precluding a kinetic
study of their interconversion. This necessitated the
introduction of an ester residue larger than acetic acid
ester vicinal to the C(1%) stereogenic carbon center.
Treatment of 1b with methanol gave methyl ether 2b, via
solvolysis of the labile iodide; whereas the corresponding
7
ethyl ester 2a had been reported previously. Both 2a and
2
b exhibited an ꢀ7:3 ratio of two sets of signals in their
1
13
H NMR spectra, and the C NMR signals were
doubled. Again, apparent thermodynamic equilibrium of
atropisomeric diastereomers was achieved at a 7:3 ratio
at room temperature. Although the methyl ethers are
much more stable than their iodo precursors and can be
handled easily in solution or in the solid state, neither
gave evidence for separation on TLC. Nor could
diastereomeric enrichment be achieved by crystallization.
2
.4. 2-Propionate ester 5
Alkylation as above of the anion of methyl propionate
with iodide 1b gave a 95% yield of 5 as a mixture of
diastereomers. GC–MS indicated two components with
identical MS fragmentation patterns. Separation by
radial chromatography on silica gel afforded 54% of a
less polar component 5-lp and 40% of a more polar 5-mp.
The NMR spectra were consistent with configurational
diastereomers, each consisting of two atropisomers.
Unlike the other pyrroles of this study, 5 has a second
stereogenic center (in the 2-propionic ester group). NMR
nicely distinguished diastereomers 5-lp and 5-mp, with
the groups near the stereogenic a-carbon of the aliphatic
Bubbling methanethiol through a solution of 1b in
tetrahydrofuran was strongly exothermic and led to a
1
13
quantitative conversion to thioether 3. Its H and
C
NMR spectra indicated a 73:27 mixture of diastereomers,
and preparative TLC afforded an 85:15 diastereomeric
enrichment. Crystallization at 0°C from pre-cooled meth-
ylene chloride and hexane, while removing the methylene
chloride in a stream of air at 0°C gave crystals further
enriched to d.r. of 98:2, and these were used for dynamic
NMR measurements.
ester side chain showing the largest chemical shift differ-
1
ences. For example, in the H NMR spectrum, the a-CH
3
protons of 5-lp appeared at 0.83 and 0.85 ppm, but in
-mp they appeared at 1.37 and 1.38 ppm; and the
5
aliphatic ester methoxy protons of 5-lp were seen at 3.81
and 3.82 ppm, while those of 5-mp were at 3.288 and
2
.3. Acetic acid esters 4
13
3
.294 ppm. Similarly, in the C NMR spectrum, the
Alkylation of the anions of methyl or t-butyl acetate
generated from reaction of LDA at −78°C to −55°C in
THF in the presence of HMPA with 1a gave 88 and 82%
yields of 4a and 4b, respectively. Both purified products
afforded crystalline fractions whose NMR spectra indi-
cated a 65:35 mixture of atropisomers. The same ratios
a-CH of 5-lp appeared at 38.87 and 39.09 ppm, while that
of 5-mp appeared at 42.30 and 42.81 ppm; and the
aliphatic ester carbonyl carbon of 5-lp was seen at 178.92
and 179.26 ppm while in 5-mp it resonated at 176.35 and
176.72 ppm.
8

1
724
S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
Both 5-lp and 5-mp solidified slowly on standing, but
-lp was more prone to crystallization. Recrystallization
at ambient temperature of the almost 1:1 mixture of
-lp atropisomers (828 mg) afforded 184 mg (22%) of a
5
5
single atropisomer. It had mp 128–129°C. Recrystalliza-
tion of the more polar diastereomer 5-mp was much
more difficult. From 621 mg of the mixture of atropiso-
1
mers, a fraction (477 mg, 77%) was obtained, whose H
NMR still showed two atropisomers present in an
ꢀ
60:40 ratio. Several recrystallizations at room tem-
perature or at 0°C from hexane, hexane–ethyl acetate
or hexane–dichloromethane (evaporating the
dichloromethane at 0°C) afforded only equilibrated
mixtures of atropisomers (ratios 69:31). The same equi-
librium ratios were found in the mother liquors of
5
-mp. The solid fraction of 5-mp had mp 105–106°C.
Thus, the higher melting point of one of the rotamers
of 5-lp suggests that during the crystal packing the
lattice preferentially accommodates only one atropiso-
mer. In the case of lower melting, more polar 5-mp,
once crystallization begins, there is no distinction
between atropisomers, and both are included in the
crystal in a ratio typical of equilibration. Despite
numerous attempts, TLC separation of atropisomers
within each of diastereomers 5-lp and 5-mp could not
be achieved. The single atropisomer obtained for 5-lp
was used for dynamic NMR measurements in a study
of the kinetics of atropisomerization.
1
Figure 5. Partial H NMR spectrum of pure 6a in CDCl3
showing the diastereotopic malonic ester methine signals at
time 0, and after standing for 72 h at 25°C. The emergence of
the new signals corresponds to the formation of an atropiso-
mer by rotation about the C(4)ꢀC(1%) bond (see Figs. 1 and
4
). The equilibrium reached after 72 h corresponds to a 7:3
ratio of atropisomers, the same as that observed in the
synthesis of 6a, before chromatographic separation. Similar
clear changes may be observed for all the NMR signals, but
those for the malonic ester methine, the C(1%)ꢀH, the NꢀH
and the C(5)ꢀCH protons are nicely separated.
3
2
.5. Malonic esters 6
TLC on silica gel at 5°C into a less polar, crystalline
fraction (mp 119–120°C) and an oily, more polar frac-
tion. Remarkably, the less polar, crystalline atropiso-
mer crystallized spontaneously from the ꢀ7:3 (oily)
atropisomer mixture. The oil, after removal of the
crystals, again produced crystals as the thermodynami-
cally less favored atropisomer converted to the more
crystalline atropisomer, which was removed by crystal-
lization—a conversion akin to an asymmetric transfor-
mation of the second kind. The crystalline atropisomer
of 6b thus isolated was 96% diastereomerically pure by
Reaction of 1a with sodio diethyl malonate, prepared
from diethyl malonate by reaction with sodium ethox-
ide in ethanol or sodium hydride in THF afforded a
high yield of 6a. In contrast, 1b gave mainly 2b (and
30% of 6b) when reacted with sodio dimethyl malonate
in refluxing methanol and a 93% yield when the alkyla-
tion was carried out in THF using the sodio malonate
generated using NaH in THF. Solvolysis of 1b seems to
be comparatively more rapid in refluxing methanol
than that of 1a in refluxing ethanol. Consequently,
NaH was used as the base in the syntheses of 6b and 6c.
Product 6a was found to be a 69:31 mixture of
1
H NMR. The crystalline, less polar fraction (mp 119–
1
20°C) was sufficiently diastereomerically pure for
kinetic studies of its atropisomerization.
1
atropisomers, as indicated by H NMR. The atropiso-
mers were separated by preparative TLC, applying a
very low load of <10 mg on each 20×20 cm plate made
with an 0.75 mm thick silica gel layer and developed
twice with hexane–cyclohexane–ethyl acetate (4:4:2 by
vol.) at room temperature. The zone of diastereomer
overlap on the TLC plate was discarded and the less
polar fraction was obtained 97% diastereomerically
pure after crystallization at 0°C. In contrast, the more
polar fraction did not crystallize. The latter, a thick oil
reverted to a 7:3 mixture of atropisomers at room
temperature. The less polar crystalline (mp 79–80°C)
material was stable in the crystal but slowly reverted to
a 7:3 mixture of atropisomers when dissolved in
C D Cl solvent at 25°C (Fig. 5). We used it for a
Similarly, the diisopropyl malonate ester 6c was
obtained in 90% yield as a thick oily mixture of
atropisomers (66:34). Separation of the diastereomers
was achieved by preparative TLC on silica gel and
recrystallization at 0°C from ethyl acetate–hexane to
afford a crystalline less polar isomer (mp 98–99°C) of
sufficient purity (98%) for
a
kinetic study of
atropisomerization.
2
.6. Atropisomerism
The process of interconversion of atropisomeric
diastereomers of 1–6 (as shown for 1b in Fig. 4) may be
expressed as a reversible isomerization equilibrium:
2
2
4
comprehensive
interconversion.
kinetic
study
of
diastereomer
k
1
A X A%
k
2
Malonic ester 6b was obtained as a 65:35 oily mixture
of atropisomers which was separated by preparative
The equilibrium constant K may thus be expressed as:

S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1725
[
[
A%]eq [A] −[A]
k
merism of 6a). The validity of the k and k data was
0
eq
1
1
2
=
= =K
(1)
A]eq
[A]eq
k
affirmed by the observation that the independent
kinetic and thermodynamic methods led to identical
DG° values.
2
where [A]eq and [A%]eq are the equilibrium concentra-
tions of the diastereomers and [A] is the initial concen-
0
tration of one diastereomer present at the start of an
isomerization (t=0). Here, the isomerization occurs by
rotation about the pyrrole C(4)ꢀC(1%) carbonꢀcarbon
bond, corresponding to a configurationally labile axis
of chirality.
Using the approach outlined above, we were able to
calculate a comprehensive set of kinetic and thermody-
namic data for the atropisomerism of 1–6. For exam-
ple, with a 93:7 atropisomerically enriched sample of
6
b, the progressive increase of the minor atropisomer
over time is displayed in Fig. 6. From these and similar
From the rate law for a reversible isomerization:
d[A]
data from the pyrrole NH, C(3)ꢀCH , and C(5)ꢀCH ,
3
3
C(1%)ꢀH and malonate a-methine signals, the integrated
1
=
−k [A]+k [A%]
(2)
1
2
( H NMR) ratios were conveniently followed and mea-
dt
one may derive the integrated rate expression:
A] −[A]
sured in CDCl CDCl solvent at 293, 313, 333 and 353
2
2
K. The NMR integration data for 1b at 313 K and
equilibrium constants are shown in Table 1, from which
[
0
eq
we compute the values of −ln([A]−[A] ) found in Table
ln
=(k +k )t
(3)
eq
1
2
[A]−[A]eq
2.
which is the equation used for treating the NMR
intensity data below.
1
Using the data from Table 2 for the C(1%)ꢀH H NMR
signal of 1b, a plot of −ln([A]−[A] ) versus time (Fig. 7)
shows excellent linearity over the range of measure-
ments. Similarly good data fits were found for plots of
eq
Combining Eqs. (1) and (2) gives:
d[A]
dt
k1
−k [A]+ [A%]=−k [A](1−[A%])/[A]K
−ln([A]−[A]eq) versus time for the NꢀH, C(3)ꢀCH and
3
=
(4)
1
1
C(5)ꢀCH signals. From the slopes of best fit lines for
3
K
all four signals measured (Tables 1 and 2), one may
from which it is clear that while the reaction has an
determine (k +k ) for the atropisomerism, and from the
1
2
initial rate of k [A], it slows as A% accumulates. Conse-
corresponding equilibrium constants (K), the mean rate
1
−5 −1 −4 −1
quently, only the NMR integral intensities from the
constants: k =7.21×10
s
and k =1.51×10
s
were
1
2
initial stages were used. As [A] approaches [A] , the
determined. It is also possible to compute the free
energies of activation for the forward and back reac-
eq
constructed linear plots become unreliable. When [A%]/
‡
‡
[
A]=K, the reaction is at equilibrium, and the observed
tions, DG and DG from k and k using the Eyring
1
2
1
2
9
‡
rate falls to zero. Since the diastereomeric groups are
anisochronous, and their integral ratios do not change
when the sample is at equilibrium, no changes in the
NMR spectra (e.g. coalescence) should be observed
over time at moderately elevated temperatures.
equation: the mean DG =24.28 kcal/mol and the
1
‡
2
mean DG =23.82 kcal/mol at 313 K.
Similar kinetic studies were carried out for 3, 5 and 6
(Table 3). We could not achieve sufficient
diastereomeric separation beyond the 7:3 mixture of 2
needed to carry out such studies with 2. It is clear from
the data for 3 that when the C(1%) group is as small as
For practical purposes [A] and [A%] are determined by
1
integration of relevant well-separated H NMR signals
‡
and followed over time at fixed intervals and at con-
trolled temperature. Typically, several pairs of signals
were followed, and the mean values of rate constants
were used for subsequent energy calculations. The con-
centrations at equilibrium [A]eq and [A%]eq were deter-
mined at time ‘infinity’, i.e. when no further spectral
methoxy, DG is too small to promote facile separation
of atropisomers at ambient temperature to 0°C. While
the methylthio group affords barely sufficient steric
hindrance to atropisomerism about the C(4)ꢀC(1%)
bond, it is clear that iodo substituent provides a sub-
stantially larger barrier to free rotation, presumably
changes were found. Plots of −ln([A]−[A] ) versus time
because the CꢀI bond is longer (ꢀ2.2 A
bond (ꢀ1.8 A), and the CꢀO bond is even shorter
). A methylene group adjacent to C(1%), as in
,
) than the CꢀS
eq
(
(
in s) were found to be linear, with slopes equal to
,
10
k +k ). Since the ratio k :k can be calculated from the
(ꢀ1.4 A
,
1
2
1
2
equilibrium concentrations (integrals), the individual
4, is apparently too small to slow the rotational isomer-
ization sufficiently. The bulkier 2-propionic ester group
of 5 is, perhaps surprisingly, slightly less effective than
values of k and k could be obtained. And from such
1
2
‡
values of k, the free energy of activation (DG ) could be
‡
3
determined from the Eyring equation DG =RT[23.76−
iodo in raising the barrier. From the large ( J=11.3
9
ln(k/T)]. From kinetics, the difference between the free
Hz) vicinal spin–spin NMR coupling constant between
C(1%) and C(a)ꢀH of both atropisomers of 5, we assume
an anti-periplanar orientation of the two hydrogens and
very likely a gearing effect in the restricted rotation
about the C(1%)ꢀC(a) bond. The malonic esters are
significantly more effective in raising the activation
energy, but there is little advantage in increasing the
‡
energies of activation (DG ) for the forward and back-
ward isomerizations is equal to the Gibbs free energy
for the equilibrium, DG°, which may also be deter-
mined from the ratio k /k . From thermodynamic con-
1
2
siderations, DG° may be found directly from the
equilibrium concentrations, determined by NMR and
from DH° and DS° obtained by plotting R ln K ver-
size of the malonic ester (CO R) R-group; cf. 6a, 6b
eq
2
sus 1/T at four different temperatures (for the atropiso-
and 6c.

1
726
S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1
Figure 6. Partial H NMR spectra of 6b showing the changes of (left) the malonate methine hydrogen at 4.06, 3.94 ppm; and
right) the pyrrole ring methyls at C(3), 2.28 ppm, and at C(5), 2.23, 2.21 ppm, over time (h) in C D Cl at 313 K.
(
2
2
4
1
Table 1. H NMR integration ratios for the NꢀH, C(1%)ꢀH, C(3)ꢀCH and C(5)ꢀCH diastereotopic hydrogens of 1b,
3
3
determined over an 8 h equilibration at 313 K in CDCl CDCl solvent
2
2
Measurement
Time (s)
NMR signal intensity measured
NꢀH
C(1%)ꢀH
C(3)ꢀCH3
C(5)ꢀCH3
2.15
la=8.59
8.50
5.22
5.16
2.38
2.30
2.22
1
2
3
4
5
6
7
8
9
660
2460
4260
6060
7860
88.7
82.4
78.1
75.3
73.4
71.9
71.0
70.5
70.0
69.7
69.7
69.6
69.5
69.3
69.2
69.1
69.2
69.2
11.3
17.6
21.9
24.7
26.6
28.1
29.0
29.5
30.0
30.3
30.3
30.4
30.5
30.7
30.8
30.9
30.8
89.7
82.6
77.5
74.1
72.0
70.4
69.3
68.7
68.2
67.9
67.7
67.5
67.4
67.3
67.3
67.3
67.3
67.3
10.3
17.4
22.5
25.9
28.0
29.6
30.7
31.3
31.8
32.1
32.3
32.5
32.6
32.7
32.7
32.7
32.7
9.8
17.0
22.0
25.4
27.8
29.4
30.5
31.2
31.7
32.0
32.3
32.5
32.5
32.6
32.7
32.7
32.7
90.2
83.0
78.0
74.6
72.2
70.6
69.5
68.8
68.3
68.0
67.7
67.5
67.5
67.4
67.3
67.3
67.3
67.3
90.2
82.7
77.6
74.1
71.8
70.2
69.1
68.4
67.9
67.7
67.4
67.2
67.2
67.1
67.1
67.1
67.0
67.1
9.8
17.3
22.4
25.9
28.2
29.8
30.9
31.6
32.1
32.3
32.6
32.8
32.8
32.9
32.9
32.9
33.0
9660
11460
13260
15060
16860
18660
20460
22260
24060
25860
27660
29460
10
11
12
13
14
15
16
17
b,c
[
A]eq
a
b
c
l, ppm.
Where [A]eq=[dominant isomer] at equilibrium.
Average value of Keq=0.477 at 313 K.

S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1727
Table 2. Values of [A]−[A]eq (=D) and ln([A]−[A] ) (=ln) for the dominant diastereomer of 1b during the course of its
eq
equilibration in CDCl CDCl at 313 K.
2
2
Measurement
1H NMR signals
NꢀH
C(1%)ꢀH
C(3)ꢀCH3
C(5)ꢀCH3
D
ln
D
ln
D
ln
D
ln
1
2
3
4
5
6
7
8
9
19.5
13.2
8.9
6.1
4.2
2.7
1.8
1.3
0.8
0.5
2.9704
2.5802
2.1861
1.8083
1.4351
0.9933
0.5878
0.2624
−0.2231
−0.6931
22.4
15.3
10.2
6.8
4.7
3.1
2.0
1.4
0.9
0.6
3.1091
2.7279
2.3224
1.9169
1.5476
1.1314
0.6931
0.3365
−0.1054
−0.5108
22.9
15.7
10.7
7.3
4.9
3.3
2.2
1.5
1.0
0.7
3.1311
23.1
15.6
10.5
7.0
4.7
3.1
2.0
1.3
0.8
0.6
3.1398
2.7473
2.3514
1.9459
1.5476
1.1314
0.6931
0.2624
−0.2231
−0.5108
2.7537
2.3702
1.9879
1.5892
1.1939
0.7885
0.4055
0.0000
−0.3567
1
0
of R ln Keq versus 1/T gave DH°=0.86 kcal/mol and
DS°=1.40 cal/deg/mol (Fig. 9). The DG° values from
DG°=DH°−TDS° match up well with those determined
from ln(k /k ): 0.45 versus 0.45 (293 K), 0.42 versus
1
2
0
0
.42 (313 K), 0.39 versus 0.38 (333 K) and 0.37 versus
.37 (353 K), respectively.
3
. Concluding comments
A new class of atropisomeric pyrroles with restricted
2
3
rotation about an sp –sp carbonꢀcarbon bond has
been prepared and analyzed. Kinetic studies indicate
11
that the ease of rotational isomerism about the
C(4)ꢀC(1%) bond of 1–6 correlates with the size of the R%
group and bond lengths at C(1%) of Scheme 1: CH-
(
CO R) >CHMeCO MeꢀI>SCH >CH CO R>OCH .
2 2 2 3 2 2 3
Atropisomers 1–6 slowly reach a ꢀ7:3 equilibrium at
room temperature. For the isolation of atropisomers
that are stable at room temperature, the size of the R%
group must be enlarged, such as when the malonic ester
tertiary hydrogen of 6a or 6b is replaced by methyl.
That raises the activation barrier to atropisomerism to
Figure 7. Plot of −ln([A]−[A] ) versus time for iodide 1b in
eq
1
C D Cl at 313 K from the H NMR data for the C(1%)ꢀH
2
2
4
signal of Table 2.
12
ꢀ
32 kcal/mol at 382 K in C D Cl .
2 2 4
A more complete study of the atropisomerism of 6a
was conducted: dynamic NMR study at four different
4
. Experimental
temperatures (293, 313, 333 and 353 K) in order to
All NMR spectra were obtained on a Varian Unity
Plus spectrometer operating at the H frequency of 500
‡
compute DH and E . As above, at each temperature,
1
a
k and k were determined experimentally and from
1
2
MHz in CDCl solvent (unless otherwise noted). Chem-
‡
‡
3
these data DG1 and DG were calculated (Table 4).
2
ical shifts are reported in l (ppm) referenced to the
‡
‡
Using ln(k/T)=23.76−DH /RT+DS /R, Eyring plots of
1
13
residual CHCl3 H signal at 7.26 ppm, and CDCl
C
4
3
ln(k/T) versus 1/T gave parallel straight lines for k and
1
1
signal at 77.00 ppm. The H NMR spectra in C D Cl
2
2
‡
k , and from the slope, one finds DH =20.95 and 20.23
1
2
were referenced to the residual H signal at l=5.94
ppm. For the dynamic NMR measurements, the probe
temperature was controlled by a standard unit of the
Unity Plus system. A J-modulated spin-echo experi-
kcal/mol for the forward and back isomerizations,
respectively (Fig. 8). From the intercepts, one finds
‡
DS =−13.59 and −14.53 cal/deg/mol for the forward
and reverse reactions, respectively. Similarly, using
1
3
ment (Attached Proton Test) was used to assign
C
ln k=−E /RT+ln A, Arrhenius plots of ln k versus 1/T
NMR spectra. The underlined NMR signals belong to
the dominant diastereomer throughout. Melting points
were taken on a Mel-Temp capillary apparatus and are
uncorrected. Gas chromatography–mass spectrometry
analyses were carried out on a Hewlett–Packard 5890A
a
gave E =21.58 and 20.86 kcal/mol for the forward and
a
reverse reactions. The A factors, computed from the
10
10
intercepts, are 1.95×10 and 1.21×10 for the forward
and back reactions, respectively (Fig. 8). Finally, a plot

1
728
S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
Table 3. Kinetic and thermodynamic parameters calculated for the atropisomerism of pyrroles at 313 K
Table 4. Kinetic and thermodynamic parameters for 6a
reagents and HPLC grade solvents were dried and
13
from variable temperature dynamic NMR experiments
purified following standard procedures. Ethyl 3,5-
dimethyl-1H-pyrrole-2-carboxylate 14was synthesized
according to a literature procedure.
Temperature (K)
Experimentala 293
313
333
353
4
.1. Methyl 3,5-dimethyl-1H-pyrrole-2-carboxylate, 7b
−
−
6
6
−5
−5
−4
−4
−4
−3
k1
1.505×10
1.741×10
3.448×10
0.5056
25.17
24.75
0.42
1.361×10
2.433×10
0.5596
25.46
25.07
0.39
8.232×10
1.465×10
0.5619
25.77
25.36
0.41
k2
Keq
DG1
DG‡2
3.261×10
0.4615
24.95
24.50
0.45
To an acetate buffer [NaOH (36 g, 0.9 mol) and glacial
acetic acid (375 mL)] was added a solution of dimethyl
malonate (160 mL, 1.40 mol) in acetic acid (80 mL),
and the mixture was cooled in an ice bath. A solution
of NaNO (203 g, 2.94 mol) in H O (320 mL) was
added over 3 h with gentle stirring, then the mixture
was allowed to warm overnight to ambient tempera-
ture. Sodium chloride (250 g) was added, and after
stirring for 45 min, the mixture was extracted with
diethyl ether (4×250 mL). After evaporation of the
ether solvent, the resulting crude dimethyl oximinoma-
‡
‡
DDG
DG°
b
2
2
0.45
0.42
0.38
0.40
a
−1
‡
k in s ; DG1 (free energy of activation for the forward reaction),
‡
DG2 (free energy of activation for the reverse reaction) and DG°
equilibrium free energy) in kcal/mol.
Calculated from ln Keq=−DG°/RT.
(
b
15
lonate solution was used immediately in the following
step.
capillary gas chromatograph (30 m DB-1 column)
equipped with Hewlett–Packard 5970 mass selective
detector. Radial chromatography was carried out on
Merck silica gel PF254 with gypsum preparative layer
grade, using a Chromatotron (Harrison Research, Inc.,
Palo Alto, CA). The same type of silica gel was used for
preparative TLC on 20×20 cm glass plates with layer
thickness of 0.75 mm. Combustion analyses were car-
ried out by Desert Analytics, Tucson, AZ. Commercial
To a mechanically-stirred mixture of pentane-2,4-dione
(
(
103 mL, 1.00 mol), zinc (196.2 g, 3.00 g), acetic acid
950 mL) and anh. sodium acetate (205 g, 2.50 mol),
preheated to 80°C, was added a solution of the above
oxime in acetic acid (30 mL). The rate of addition was
such as to maintain the internal temperature at 80–85°C
(3 h). Then the mixture was heated under reflux for 3 h

S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1729
and poured while hot into 12 L of ice-water. After 1 h
at 0°C, the product was collected by filtration, washed
with H O (3×1 L) and dried under vacuum. Recrystalliza-
2
tion from CH OH–H O afforded pyrrole 7b (100.8 g,
3
2
1
6
6%); mp 98–99°C; H NMR: l 2.24 (3H, s), 2.30 (3H,
4
s), 3.82 (3H, s), 5.79 (1H, d, J=2.7 Hz), 8.75 (1H, brs)
ppm; C NMR: l 12.77, 12.79, 50.81, 111.18, 117.37,
13
+
’
1
1
28.97, 133.10, 162.50 ppm; MS: m/z (%) 153 [M ](100),
38 (13), 122 (82), 121 (85), 93 (51). Anal. calcd for
C H NO : C, 62.73; H, 7.24; N, 9.14. Found: C, 62.72;
8
11
2
H, 7.49; N, 9.04%.
4
.2. General procedure for the syntheses of iodides 1a
and 1b
Acetic anhydride (60 mL) was added slowly to 57%
hydriodic acid (60 mL) with occasional cooling to
maintain an internal temperature of ꢀ40–42°C. Then
5
0% hypophosphorus acid (H PO , 6 mL) was added,
3 2
followed by b-free pyrrole, 7a or 7b (30 mmol). After the
pyrrole had completely dissolved and the temperature had
been lowered to 30–32°C, pivalaldehyde (4.3 mL, 40
mmol) was added over 10 min with vigorous stirring while
maintaining the temperature. After stirring for an addi-
tional 25 min, the mixture was slowly (10 min) diluted
with ice-cold water (450 mL). The product was collected
by suction filtration, washed with cold water (5×50 mL)
and dried overnight under vacuum (P O ) to afford near
2
5
quantitative yields of iodides 1a and 1b.
4
1
.2.1. Ethyl 3,5-dimethyl-4-(2%,2%-dimethyl-1%-iodopropyl)-
H-pyrrole-2-carboxylate, 1a. Compound 1a was isolated
in 94% yield; mp 161–163°C (atropisomeric ratio 88:12)
7
1
(
lit. mp 168–169°C); H NMR: l 1.08, 1.09 (9H, 2×s),
1
2
5
.347, 1.348 (3H, 2×t, J=7.1 Hz), 2.21, 2.27 (3H, 2×s),
Figure 8. Best fit straight lines for the forward (ꢀ) and back
ꢁ) atropisomerization of 6a: (upper) Eyring plot at four
3
.36, 2.43 (3H, 2×s), 4.29, 4.30 (2H, ABX , J=7.1 Hz),
3
(
13
.21, 5.26 (1H, 2×s), 8.74, 8.81 (1H, 2×br.s) ppm;
C
‡
different temperatures, R=0.9998. The slope=−DH /R and the
NMR: l 11.00, 12.31, 14.51, 14.85, 16.04, 28.72, 28.98,
9.24, 39.39, 43.14, 43.88, 59.91, 59.94, 115.95, 118.74,
‡
intercept=23.76+DS /R. (lower) Arrhenius plot at four differ-
3
ent temperatures, R=0.9998 and R=0.9999 for k1 and k2,
121.56, 122.24, 125.71, 128.67, 129.15, 133.38, 161.71,
161.90 ppm.
respectively. The slope=−E /R and the intercept=ln A.
a
Careful crystallization of 500 mg of the mixture from
above from 6 mL of dichloromethane at 0°C and 6 mL
of hexane (by partial evaporation of CH Cl at 0°C using
2
2
a stream of air) led to isolation of highly diastereomer-
ically enriched (d.r.=99:1) diastereomer 1a (214 mg) mp
1
1
57–159°C and H NMR corresponding to the underlined
signals above.
4
.2.2. Methyl 3,5-dimethyl-4-(2%,2%-dimethyl-1%-iodo-
propyl)-1H-pyrrole-2-carboxylate, 1b. Compound 1b was
isolated in 86% yield; mp 134–135°C (atropisomeric ratio
1
7
2
5
:3); H NMR: l 1.075, 1.083 (9H, 2×s), 2.21, 2.27 (3H,
×s), 2.35, 2.43 (3H, 2×s), 3.825, 3.830 (3H, 2×s), 5.20,
13
.26 (1H, 2×s), 8.74, 8.81 (1H, 2×br.s) ppm; C NMR:
l 10.97, 12.39, 14.76, 16.10, 28.73, 28.99, 39.25, 39.40,
4
1
2.95, 43.67, 51.05, 51.07, 115.76, 118.55, 121.65,
22.33, 125.94, 128.91, 129.12, 133.35, 161.92, 162.08
ppm. Careful crystallization from 0°C pre-cooled
dichloromethane and hexane (by evaporation of the
CH Cl at 0°C) led to isolation of a sample highly enriched
Figure 9. Best-fit straight line (R=0.992) for the temperature
dependence of R ln Keq versus 1/T for 6a. The slope=−DH°/R
and the intercept=DS°/R.
2
2
in one atropisomer (93:7), mp 137–138°C. Anal. calcd for

1
730
S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
C H INO : C, 44.71; H, 5.77; N, 4.01. Found: C, 44.96;
H, 6.00; N, 4.05%.
4.5. General procedure for the syntheses of 4a, 4b and
5
13
20
2
4
.3. Methyl 3,5-dimethyl-4-(2%,2%-dimethyl-1%-
A solution of methyl or t-butyl acetate (30 mmol), or
methyl propionate in anhydrous THF (18 mL) was added
methoxypropyl)-1H-pyrrole-2-carboxylate, 2b
slowly via syringe to a N -protected solution of LDA
2
A solution of iodide 1b (1.05 g, 3 mmol) in methanol (50
mL) was heated under reflux for 1 h. After cooling, the
(freshly prepared from dry isopropylamine (4.1 mL, 31.2
mmol) in THF (35 mL) and n-BuLi in hexane (2.5 M,12
mL, 30 mmol) at −78°C. After stirring for 1 h at the same
temperature, a solution of the iodide (3.63 g, 10 mmol
mixture was diluted with CHCl (100 mL), washed with
3
H O (3×50 mL), dried (MgSO ), filtered and the solvent
2
4
8
was evaporated under vacuum. The residue was purified
of 1a or 3.49 g, 10 mmol of 1b) and anhydrous HMPA
by
radial
2
chromatography
on
silica
gel
(3.48 mL, 20 mmol) in THF (25 mL) was added over 15
min. The mixture was stirred for 1 h at −78°C, then the
temperature was raised to −55°C over 30 min. The
(CH Cl :CH OH=100:0.5 v/v) and recrystallization
2
3
from EtOAc/hexane to afford 2b (0.39 g, 50%); mp
1
1
20–121°C (atropisomeric ratio 67:33); H NMR: l 0.91
reaction was quenched with sat. aq. NH Cl (20 mL). The
4
(
(
9H, s), 2.18, 2.24 (3H, 2×s), 2.30, 2.38 (3H, 2×s), 3.14
mixture was diluted with 0.2% aq. HCl (150 mL) and the
3H, s), 3.83 (3H, s), 3.89 (1H, s), 8.68, 8.69 (in concd
product was extracted with CHCl3 (4×50 mL). The
13
soln 9.40, 9.48) (1H, 2×br.s) ppm; C NMR: l 11.50,
combined organic extracts were washed with H O (4×50
2
1
5
1
2.31, 12.44, 13.53, 26.49, 26.74, 37.94, 50.81, 50.83,
6.61, 56.69, 85.59, 86.73, 116.61, 117.43, 118.33, 118.82,
28.02, 129.06, 131.53, 132.41, 162.32, 162.63 ppm. MS:
mL), dried over anhydrous MgSO and after filtration
4
through a short pad of silica gel, the solvent was
evaporated under vacuum. The residue was purified by
radial chromatography on silica gel eluting with hexane:
EtOAc=100:10–100:20. Theresultingpurefractionswere
recrystallized from hexane–EtOAc to afford pure
pyrroles.
+
’
m/z (%) 253 [M ] (4), 222 (3), 221 (5), 196 (90), 164 (100).
Anal. calcd for C H NO : C, 66.37; H, 9.15; N, 5.53.
Found: C, 66.65; H, 9.17; N, 5.61%.
1
4
23
3
4
.4. Methyl 3,5-dimethyl-4-(2%,2%-dimethyl-1%-methylthio-
propyl)-1H-pyrrole-2-carboxylate, 3
4
.5.1. Ethyl 3,5-dimethyl-4-[2%,2%-dimethyl-1%-(methoxy-
Methanethiol (CAUTION, highly toxic, stench) was
bubbled through a solution of 1.05 g (3 mmol) iodide 1b
in anhydrous THF (15 mL) for 10 min. The effluent
condenser end was attached to a 15 cm long trap filled
carbonylmethyl)propyl]-1H-pyrrole-2-carboxylate,
4a.
Obtained in 88% yield from 1a; mp 95–98°C (atropisomer
1
ratio 65:35). H NMR: l 0.91, 0.92 (9H, 2×s), 1.33, 1.34
(3H, 2×t, J=7.1 Hz), 2.21, 2.28 (3H, 2×s), 2.32, 2.39 (3H,
3
2
with saturated aqueous Pd(OAc) followed by a bleach
2×s), 2.69, 2.72 (1.4 H, ABX, J=10.3 Hz, J=14.8 Hz;
3 2 3
2
trap, and the gas flow was discontinued when copious
yellow precipitate was formed in the first trap. The
reactionishighlyexothermicleadingtoTHFreflux, which
was maintained for 30 min. more by external heating.
After cooling, the mixture was diluted with chloroform
J=6.5 Hz, J=14.8 Hz), 2.75, 2.82 (0.6 H, ABX, J=5.6
2 3 2
Hz, J=15.4 Hz; J=10.0 Hz, J=15.4 Hz), 3.02, 3.08
3
3
(1H, 2×ABX, J=5.6, 10.0 Hz; J=6.5, 10.3 Hz), 3.49,
3.52 (3H, 2×s), 4.26, 4.27 (2H, 2×ABX , J=7.1 Hz), 8.62,
3
3
13
8.71 (1H, 2×br.s) ppm. C NMR: l 11.21, 12.37, 13.47,
14.45, 14.46, 14.84, 28.15, 28.37, 34.41, 34.57, 35.97,
36.00, 42.27, 43.63, 51.27, 51.29, 59.47, 59.52, 116.44,
117.23, 120.85, 121.13, 126.81, 129.30, 129.77, 132.13,
161.79, 161.88, 173.66, 173.95 ppm. MS: m/z (%) 309
[M ](22), 252(100), 206(55), 164(41), 147(9). Anal. calcd
for C H NO : C, 65.99; H, 8.80; N, 4.53. Found: C,
(
100 mL), washed with water (3×100 mL) and dried over
anhydrous MgSO . After filtration and evaporation, the
4
residue was purified by radial chromatography (gradient
hexane:ethyl acetate=85:15–75:25) and recrystallization
+
’
(
1
EtOAc–hexane) to afford 0.78 g (97%) of 3; mp 131–
1
33°C (atropisomeric ratio 73:27); H NMR: l 1.02, 1.03
17
27
4
(
9H, 2×s), 1.80, 1.82 (3H, 2×s), 2.19, 2.25 (3H, 2×s), 2.46,
66.15; H,8.82; N, 4.69%.
2
8
1
5
1
.52 (3H, 2×s), 3.66, 3.73 (1H, 2×s), 3.826, 3.829 (3H, 2×s),
13
.70, 8.71 (1H, 2×br.s) ppm; C NMR: l 11.23, 12.46,
3.31, 14.63, 15.45, 15.76, 28.66, 28.86, 37.99, 38.04,
0.88, 50.90, 54.81, 56.33, 116.13, 117.64, 118.89, 119.54,
28.16, 129.26, 131.61, 131.93, 162.10, 162.28 ppm. MS:
4.5.2. Ethyl 3,5-Dimethyl-4-[2%,2%-dimethyl-1%-(t-butoxy-
carbonylmethyl)propyl]-1H-pyrrole-2-carboxylate,
4b.
Obtained in 82% yield; mp 103–106°C (atropisomer ratio
1
m/z (%) 222 (18), 221 (21), 212 (91), 190 (21), 180 (100),
46 (18). Anal. calcd for C H NO S: C, 62.41; H, 8.61;
65:35). H NMR: l 0.91, 0.92 (9H, 2×s), 1.19, 1.22 (9H,
1
2×s), 1.336, 1.339 (3H, 2×t, J=7.1 Hz), 2.20, 2.27 (3H,
14
23
2
3
N, 5.20. Found: C, 62.50; H, 8.90; N, 5.47%.
2×s), 2.33, 2.40(3H, 2×s), 2.58, 2.65(1.4H, ABX, J=11.6
2
3
2
Hz, J=14.0 Hz; J=5.8 Hz, J=14.0 Hz), 2.63, 2.70 (0.6
3
2
3
2
Preparative TLC of 3 (180 mg) on 18 20×20 cm plates
H, ABX, J=6.0 Hz, J=14.2 Hz; J=10.5 Hz, J=14.2
3 3
(
hexane:EtOAc=85:15) led to isolation of
a
Hz), 2.94, 3.00 (1H, 2×ABX, J=6.0, 10.5 Hz; J=5.8,
3
diastereomerically enriched sample (110 mg, d.r.=85:15).
Crystallization of this fraction from pre-cooled methylene
chloride and hexane by removing the former in a stream
of air at 0°C gave 65 mg of further enriched to 98:2 (in
pre-cooled to −10°C CDCl ) atropisomer of 3, mp
1
11.6 Hz), 4.27, 4.28 (2H, 2×ABX , J=7.1 Hz), 8.52, 8.61
3
1
(1H, 2×br.s) ppm. H NMR ((CD ) SO): l 0.84, 0.85 (9H,
3
2
2×s), 1.11, 1.12 (9H, 2×s), 1.25 (3H, br.t, J=7.0 Hz), 2.09,
2.16 (3H, 2×s), 2.23, 2.29 (3H, 2×s), 2.53, 2.59 (1.4 H,
3
2
3
2
ABX, J=10.9 Hz, J=14.2 Hz; J=5.8 Hz, J=14.2
3 2
3
1
29–130°C and H NMR corresponding to the underlined
Hz), 2.57, 2.61 (0.6 H, ABX, J=5.8 Hz, J=14.0 Hz;
3 2
signals above.
J=10.4 Hz, J=14.0 Hz), 2.87, 2.92 (1H, 2×ABX,

S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1731
3
3
J=5.8, 10.4 Hz; J=5.8, 10.9 Hz), 4.14, 4.17 (2H,
4.6. Ethyl 3,5-dimethyl-4-[2%,2%-dimethyl-1%-(diethoxycar-
bonylmethyl)propyl]-1H-pyrrole 2-carboxylate, 6a
3
2
×ABX , J=7.0 Hz), 10.97, 11.04 (2H, 2×br.s) ppm.
3
1
3
C NMR: l 11.41, 12.54, 13.57, 14.49, 14.52, 15.00,
7.65, 27.66, 28.18, 28.42, 35.94, 36.03, 36.04, 36.14,
2.70, 44.35, 59.49, 59.53, 79.62, 79.71, 116.28, 117.15,
21.09, 121.40, 127.55, 129.63, 130.01, 131.90, 161.78,
2
4
1
1
1
2
5
1
1
To abs. ethanol (40 mL) under an N
blanket sodium
2
(
690 mg, 30 mmol) was added slowly. After all of the
Na had reacted, a solution of diethyl malonate (5.12
mL, 32 mmol) in ethanol (3 mL) was added and the
mixture was stirred for 15 min. Then iodide 1a (2.18 g,
13
61.79, 172.60, 172.94 ppm. C NMR ((CD ) SO): l
3
2
1.22, 11.81, 13.31, 14.20, 14.48, 14.51, 27.25, 27.28,
8.00, 28.21, 35.46, 35.49, 35.62, 35.64, 42.38, 43.87,
8.58, 58.65, 78.81, 78.86, 115.44, 116.30, 119.72,
20.16, 126.25, 128.14, 130.59, 132.36, 160.70, 160.71,
6
mmol) and ethanol (8 mL) were added and the
mixture was heated under reflux for 30 min. After
cooling, the product was partitioned between Et O/
2
+
’
H O (2×100 mL/150 mL). The organic layer was
2
71.80, 171.94 ppm. MS: m/z (%) 351 [M ] (26), 294
100), 238 (76), 194 (55), 148 (11), 57 (72). Anal. calcd
for C H NO : C, 68.34; H, 9.46; N, 3.99. Found: C,
washed with 1% HCl (2×50 mL), H O (4×50 mL),
2
(
dried (MgSO ), filtered, and the solvent was removed
4
2
0
33
4
under vacuum. Radial chromatography purification
6
8.36; H, 9.70; N, 4.16%.
(
hexane:ethyl acetate=8:2, v/v) afforded pyrrole 6a as
1
a thick oil (2.18 g, 92%; atropisomer ratio 69:31). H
4
.5.3. Methyl 3,5-dimethyl-4-[2%,2%-dimethyl-1%-(1%%-
NMR: l 0.88, 0.89 (3H, 2×t, J=7.1 Hz), 0.90, 0.91
methoxycarbonylethyl)propyl]-1H-pyrrole-2-carboxylate,
(
9H, 2×s), 1.289, 1.293 (3H, 2×t, J=7.1 Hz), 1.33,
5
. Isolated in 95% yield as a mixture of two
1
2
3
4
8
1
3
6
.34 (3H, 2×t, J=7.1 Hz), 2.26, 2.29 (3H, 2×s), 2.33,
diastereomers. They were separated by radial chro-
matography on silica gel (hexane:ethyl acetate gradient
from 90:10 to 80:20 v/v) to afford less polar
diastereomer 5-lp as a thick oil (1.65 g, 54%; atropiso-
mer ratio 59:41). H NMR: l 0.83, 0.85 (3H, 2×d,
J=7.0 Hz), 0.86 (9H, s), 2.24 (3H, s), 2.29, 2.30 (3H,
.34 (3H, 2×s), 3.49, 3.58 (1H, 2×d, J=11.6 Hz), 3.78,
.83 (2H, 2×ABX ), 3.96, 4.08 (1H, 2×d, J=11.6 Hz),
3
.19, 4.21 (2H, 2×ABX ), 4.25, 4.26 (2H, ABX ), 8.43,
3
3
13
.83 (1H, 2×br.s) ppm. C NMR: l 11.28, 12.64,
3.40, 13.88, 14.50, 14.52, 14.76, 15.69, 28.64, 28.92,
6.68, 36.74, 44.80, 46.53, 54.18, 54.67, 59.59, 59.68,
0.97, 61.07, 61.57, 61.58, 116.77, 117.25, 120.31,
1
2
×s), 2.95, 3.05 (1H, 2×dq, J=11.3, 7.0 Hz), 3.00, 3.07
(
1H, 2×d, J=11.3 Hz), 3.692, 3.694 (3H, 2×s), 3.81,
121.00, 127.04, 129.68, 129.87, 132.46, 161.70, 161.75,
13
3
.82 (3H, 2×s), 8.71, 8.77 (1H, 2×br.s) ppm.
C
168.00, 168.08, 169.62, 169.87 ppm. MS: m/z (%) 395
+
’
NMR: l 11.45, 12.70, 14.06, 15.18, 19.13, 19.22, 28.73,
[M ] (14), 338 (100), 265 (4), 220 (56), 174 (41),
146(9). Anal. calcd for C H NO : C, 63.77; H, 8.41;
2
5
1
1
8.89, 36.79, 36.84, 38.87, 39.09, 47.82, 49.19, 50.82,
0.87, 51.64, 51.67, 116.65, 117.10, 120.95, 121.53,
27.08, 129.06, 130.03, 132.18, 161.92, 162.04, 178.92,
21
33
6
N, 3.54. Found: C, 63.83; H, 8.43; N, 3.59%.
+
’
79.26 ppm. MS: m/z (%) 309 [M ] (8), 252 (100), 220
A single atropisomer with 97% purity was obtained
after preparative TLC on silica gel (cyclohexane: ethyl
acetate=8:2 v/v) and recrystallization at 0°C from
ethyl acetate–hexane (mp 79–80°C), and was used for
dynamic NMR measurements.
(
27), 196 (34), 164 (34). Anal. calcd for C H NO : C,
17 27 4
6
4
5.99; H, 8.80; N, 4.53. Found: C, 66.20; H, 8.95; N,
.65%.
Chromatographic separation of atropisomers of 5-lp
was not observed. Crystallization at room temperature
from hexane gave 22% recovery of solid single
atropisomer corresponding to 5-lp, mp 128–129°C
which was used for kinetic measurements.
4
.7. General procedure for synthesis of 6b and 6c
To an N -protected suspension of sodium hydride (900
mg, 30 mmol, 80% oil) in anhydrous THF (40 mL) at
2
1
0°C was added a solution of malonate ester (32
The more polar fractions afforded diastereomeric 5-mp
mmol) in THF (3 mL) over 10 min. After stirring for
a further 10 min, the corresponding iodide 1a or 1b (6
mmol) and THF (5 mL) were added and the mixture
was heated under reflux for 30 min. After cooling, the
mixture was poured into cold 1% aq. HCl (100 mL)
and diethyl ether (200 mL). The aqueous layer was
(
1.24 g, 40%) as a thick oil (atropisomer ratio
ꢀ
60:40). Atropisomer separation of 5-mp was not
found either by chromatography or by recrystalliza-
tion, which yielded solid fraction with mp 105–106°C
1
(
atropisomer ratio 69:31). H NMR: l 0.97, 0.99 (9H,
2
2
×s), 1.37, 1.38 (3H, 2×d, J=6.8 Hz), 2.17, 2.25 (3H,
×s), 2.33, 2.40 (3H, 2×s), 2.71, 2.79 (1H, 2×d, J=11.3
extracted with Et O (2×50 mL) and the combined
ethereal extracts were washed with H O until neutral
2
2
Hz), 3.14, 3.26 (1H, 2×dq, J=11.3, 6.8 Hz), 3.288,
(
3×100 mL). After drying over anhydrous MgSO4,
3
2
1
4
1
1
.294 (3H, 2×s), 3.79, 3.80 (3H, 2×s), 8.43, 8.55 (1H,
filtration and evaporation, the crude product was
purified by radial chromatography on silica gel eluting
with hexane: ethyl acetate=10:1 to 8:2 to afford
diastereomeric mixtures of pyrroles 6.
13
×br.s) ppm. C NMR: l 10.98, 12.33, 14.00, 15.38,
9.67, 19.73, 30.40, 30.79, 35.20, 35.45, 42.30, 42.81,
7.99, 49.65, 50.72. 50.76, 50.96, 116.16, 116.79,
23.04, 123.50, 126.91, 129.12, 129.45, 131.53, 161.97,
+
’
62.03, 176.35, 176.72 ppm. MS: m/z (%) 309 [M ]
(
6), 252 (100), 220 (30), 196 (36), 164 (36). Anal. calcd
4.7.1. Methyl 3,5-dimethyl-4-[1%-(dimethoxycarbonyl)-
methyl-2%,2%-dimethylpropyl]-1H-pyrrole-2-carboxylate,
6b. Obtained in 93% yield as a thick oil (atropisomer
for C H NO : C, 65.99; H, 8.80; N, 4.53. Found: C,
6
1
7
27
4
6.09; H, 8.72; N, 4.50%.

1
732
S. E. Boiadjie6, D. A. Lightner / Tetrahedron: Asymmetry 13 (2002) 1721–1732
1
ratio 65:35). H NMR: l 0.89, 0.90 (9H, 2×s), 2.26, 2.29
3H, 2×s), 2.32, 2.33 (3H, 2×s), 3.31, 3.34 (3H, 2×s),
Acknowledgements
(
3
3
8
1
5
1
1
.48, 3.58 (1H, 2×d, J=11.8 Hz), 3.75 (3H, s), 3.79,
.81 (3H, 2×s), 4.00, 4.13 (1H, 2×d, J=11.8 Hz), 8.61,
.78 (1H, 2×br.s) ppm. C NMR: l 11.11, 12.42, 14.52,
5.47, 28.54, 28.80, 36.52, 36.57, 45.29, 46.88, 50.76,
0.83, 52.04, 52.07, 52.60, 52.62, 53.66, 54.13, 116.58,
17.05, 119.91, 120.57, 126.82, 129.63, 130.05, 132.91,
61.98, 162.15, 168.34, 168.49, 169.95, 170.19 ppm. MS:
We thank the National Institutes of Health (HD-17779)
for support of this research. Stefan E. Boiadjiev is on
leave from the Institute of Organic Chemistry, Sofia,
Bulgaria.
1
3
References
+
’
m/z (%) 353 [M ] (10), 322 (4), 296 (100), 264 (20), 220
(
6
3
27), 196 (38), 164 (26). Anal. calcd for C H NO : C,
1.17; H, 7.70; N, 3.96. Found: C, 61.36; H, 7.86; N,
.98%.
1. Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry
of Organic Compounds; J. Wiley and Sons: New York, 1994;
Chapter 14.
18
27
6
2
. Oki, M. In Topics in Stereochemistry; Allinger, N. L.; Eliel,
(
A single atropisomer of 96% d.e. (mp 119–120°C) was
obtained after preparative TLC on silica gel at 5°C
E. L.; Wilen, S. H., Eds.; J. Wiley and Sons, 1983; Vol. 14,
pp. 1–81.
(
hexane: ethyl acetate=7:3) and recrystallization at 0°C
3. Kessler, H. Angew. Chem., Int. Ed. Engl. 1970, 9, 219–235.
4. Oki, M. The Chemistry of Rotational Isomers; Springer-Ver-
from ethyl acetate–hexane.
(
lag, 1993.
5. Lomas, J. S.; Dubois, J. E. J. Org. Chem. 1976, 41,
4
.7.2. Methyl 3,5-dimethyl-4-[1%-(diisopropoxycarbonyl)-
methyl-2%,2%-dimethylpropyl]-1H-pyrrole-2-carboxylate,
c. Obtained in 90% yield as a thick oil (atropisomer
3033–3034.
6. Falk, H. The Chemistry of Linear Oligopyrroles and Bile
Pigments; Springer-Verlag: Wien, 1989.
7. Khan, S. A.; Plieninger, H. Chem. Ber. 1975, 108, 2475–
2477.
8. MacPhee, J. A.; Dubois, J-E. J. Chem. Soc., Perkin 1 1977,
694–696.
9. G u¨ nther, H. NMR Spectroscopy; J. Wiley & Sons: New
York, 1994; pp. 335–389.
10. In Handbook of Chemistry and Physics, 82nd ed. (2001–
2002); Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 2001.
11. An alternative mechanism was suggested in review. Inter-
conversion of diastereomers might also be achieved in 1 by
a dissociative mechanism rather than simple CꢀC bond
isomerization. The dissociation mechanism would also
require C(4)ꢀC(1%) bond rotation—or recombination with
inversion at C(1%).
6
1
ratio 66:34). H NMR: l 0.75, 0.83 (3H, 2×d, J=6.3
Hz), 0.89, 0.91 (9H, 2×s), 0.98, 1.00 (3H, 2×d, J=6.3
Hz), 1.23, 1.24 (3H, 2×d, J=6.3 Hz), 1.27, 1.28 (3H,
2
2
2
2
×d, J=6.3 Hz), 2.26, 2.28 (3H, 2×s), 2.33, 2.34 (3H,
×s), 3.48, 3.58 (1H, 2×d, J=11.5 Hz), 3.79, 3.81 (3H,
×s), 3.88, 4.00 (1H, 2×d, J=11.5 Hz), 4.63, 4.67 (1H,
×septet, J=6.3 Hz), 5.03, 5.04 (1H, 2×septet, J=6.3
1
3
Hz), 8.43, 8.61 (1H, 2×br.s) ppm. C NMR: l 11.32,
1
2
5
1
1
2.68, 14.79, 15.74, 20.56, 20.87, 21.07, 21.26, 21.39,
1.41, 21.49, 28.65, 28.93, 36.73, 36.80, 44.03, 45.76,
0.78, 50.85, 54.75, 55.25, 68.21, 68.41, 68.91, 68.93,
16.50, 117.01, 120.68, 121.35, 127.35, 129.95, 130.18,
32.62, 161.94, 162.07, 167.50, 167.51, 169.07, 169.33
+
’
ppm. MS: m/z (%) 409 [M ] (11), 352 (92), 310 (2), 266
43), 224 (78), 222 (43), 206 (100), 174 (58), 57 (24).
Anal. calcd for C H NO : C, 64.52; H, 8.62; N, 3.42.
(
12. Boiadjiev, S. E.; Lightner, D. A. Monatsh. Chem. 2002, in
press.
2
2
35
6
Found: C, 64.69; H, 8.90; N, 3.56%.
13. Perrin, D. D.; Armarego, W. L. F. Purification of Labora-
tory Chemicals, 3rd ed.; Pergamon Press: UK, 1988.
14. Robinson, J. A.; McDonald, E.; Battersby, A. R. J. Chem.
Soc., Perkin Trans. 1 1985, 1699–1709.
15. May, D. A., Jr.; Lash, T. D. J. Org. Chem. 1995, 57,
4820–4828.
A single atropisomer of 98% d.e. (mp 99–100°C) was
obtained after preparative TLC on silica gel (hex-
ane:ethyl acetate=8:2 v/v) and recrystallization at 0°C
from ethyl acetate–hexane.