Remarkably slow rotation about a single bond between an sp 3-hybridised carbon atom and an aromatic ring without ortho substituents

Article abstract of DOI:10.1002/chem.200802127

A series of polycycles was prepared by using a three-component Joullie-Ugi reaction. The rate of rotation about the bond between a highly hindered bridgehead and a phenyl ring with no ortho substituents was measured by using, in general, variabletemperatu

Full text of DOI:10.1002/chem.200802127

FULL PAPER
DOI: 10.1002/chem.200802127
Remarkably Slow Rotation about a Single Bond between an sp3-Hybridised
Carbon Atom and an Aromatic Ring without ortho Substituents
Sarah Murrison,[a] David Glowacki,[a] Christian Einzinger,[b] James Titchmarsh,[a]
Stephen Bartlett,[a] Ben McKeever-Abbas,[c] Stuart Warriner,*[a] and Adam Nelson*[a]
Abstract: A series of polycycles was
prepared by using a three-component
Joulliꢀ–Ugi reaction. The rate of rota-
tion about the bond between a highly
hindered bridgehead and a phenyl ring
with no ortho substituents was mea-
sured by using, in general, variable-
temperature HPLC. The rate of rota-
tion was highly dependent on substitu-
tion and rotamer half-lives of up to
21 h at 298 K were observed. Insights
into the effect of substitution on the
rate of rotation were gleaned through
electronic structure calculations on
closely related derivatives. Rotamers
resulting from restricted rotation about
carbon atom and a phenyl ring with no
ortho substituents were isolated for the
first time, and the equilibration of the
separated rotamers was followed by
using analytical HPLC. It was demon-
strated, for the first time, that a highly
hindered environment for the sp3-hy-
bridised atom is sufficient for slow
bond rotation about a single bond be-
tween sp3- and sp2-hybridised carbon
atoms.
a
bond between an sp3-hybridised
Keywords: amides atropisomer-
·
ism · electronic structure · multi-
component reactions · polycycles
Introduction
hydroanthracenes 2, the barrier to bond rotation may be in-
creased through selective destabilisation of the TS.[3]
Barriers to rotation about single bonds between sp2- and
sp3-hybridised carbon atoms are, in general, so small that
isomers arising from restricted rotation cannot be isolated at
room temperature.[1] However, in some highly hindered sys-
tems, such as 9-arylfluorene 1, bond rotation is sufficiently
slow that two rotamers may be isolated.[2] Here, the slow
rate of bond rotation may be ascribed to the hindered envi-
ronment of the sp3 carbon atom and the 2,6-substitution pat-
tern of the phenyl ring. In some cases, such as 9-aryl-9,10-di-
Herein, we describe some compounds that exhibit slow
rotation about a bond between sp2- and sp3-hybridised
carbon atoms. In each compound, the sp2-hybridised carbon
is part of a phenyl ring without any ortho substituents. Re-
markably, however, in some cases the rate of bond rotation
is sufficiently slow that the individual rotamers may, none-
theless, be isolated.
[a] S. Murrison, Dr. D. Glowacki, J. Titchmarsh, Dr. S. Bartlett,
Dr. S. Warriner, Prof. A. Nelson
School of Chemistry, University of Leeds
Leeds, LS2 9 JT (UK)
Fax : (+44)113-343-6565
E-mail: a.s.nelson@leeds.ac.uk
s.l.warriner@leeds.ac.uk
[b] Dr. C. Einzinger
Institute of Applied Synthetic Chemistry, Getreidemarkt 9/163
1060 Vienna (Austria)
Results and Discussion
[c] Dr. B. McKeever-Abbas
Polycyclic compounds 9–12 were each prepared by using
two three-component reactions (Scheme 1). Triazines 3 and
4 were prepared by Suzuki reactions involving the appropri-
ate boronic acids. Condensation of a triazine (3 or 4) with
AstraZeneca, Hurdsfield Industrial Estate
Macclesfield, SK10 2NA (UK)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802127.
Chem. Eur. J. 2009, 15, 2185 – 2189
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2185

Scheme 1. Preparation of polycyclic compounds 9–12.
diallylamine and either cyclopentanone or valeraldehyde
gave polycyclic imines 5–8 in good to excellent yield.[4]
A
Joulliꢀ–Ugi reaction[5] involving an imine (5–8), an isocya-
nide and a carboxylic acid was then used to prepare polycyc-
lic compounds 9–12 (see Table 1). In each case the reaction
Table 1. Preparation of polycyclic compounds 9–12 by using a three com-
ponent Joulliꢀ–Ugi reaction.
Imine
Product
Ar
R1
R2
Yield [%]
5
9a
9b
9c
9d
9e
9 f
9g
10a
10b
11
m-nitrophenyl
m-nitrophenyl
m-nitrophenyl
m-nitrophenyl
m-nitrophenyl
m-nitrophenyl
m-nitrophenyl
p-cyanophenyl
p-cyanophenyl
m-nitrophenyl
p-cyanophenyl
cHex
cHex
cHex
cHex
cHex
cHex
Bu
cHex
cHex
cHex
cHex
Ph
Me
iPr
tBu
p-MeOC6H4-
p-O2NC6H4-
Ph
Ph
iPr
Ph
Ph
92
82
76
64
53
71
58
73
70
72
94
Figure 1. Expansion of regions of the 500 MHz 1H NMR spectra of the
polycycles 9a (left) and 10a (right). For polycycle 9a, signals for two ro-
tamers are present, whereas for 10a, the sides of the p-cyanophenyl ring
are diastereotopic. On the NMR timescale, rotation about the single
bond between the sp3-hybridised bridgehead carbon and the substituted
phenyl ring is slow.
6
7
8
12
was >98:<2 diasteroselective, with the isocyanide attacking
the less hindered face of the polycyclic imine. The relative
configurations of 10a and 10b were determined by X-ray
crystallography (see the Supporting Information), and that
of 9g was established by observing diagnostic NOE interac-
tions.
For polycyclic compounds 9–12, slow rotation about the
single bond between the sp3-hybridised bridgehead carbon
and the substituted phenyl ring was immediately apparent.
Scheme 2. Interconversion of the syn and anti rotamers of polycyclic
compounds 9 and 11. The configurations of the rotamers of 9g were de-
termined through observation of diagnostic NOE interactions.
1
The 500 MHz H NMR spectra of m-nitrophenyl-substituted
polycycles 9a–g and 11 indicated that two rotamers were
present (see Figure 1 and Scheme 2). Similarly, for ana-
logues 10a–b and 12, two pairs of diastereotopic protons
(H2/H6 and H3/H5) were observed on the p-cyanophenyl
ring. These observations both implicate slow rotation about
the bond between the sp3-hybridised bridgehead carbon and
the substituted phenyl ring.
The rate of interconversion of the rotamers of polycyclic
compounds 9a–g (Scheme 2) was studied by using variable-
temperature HPLC[6] (vt-HPLC) at 5 K intervals. In each
case, peaks for each of the rotamers were observed at low
temperature (Figure 2). However, coalescence of the peaks
was observed as the rate of equilibration increased with
temperature. Forward and reverse rate constants were deter-
mined by analysis of the chromatograms, which were ac-
quired in triplicate. Our results are summarised in Tables 2
and 3.
2186
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2185 – 2189

Slow Rotation about a Single Bond
FULL PAPER
278 K. The rate of equilibration of both rotamers was deter-
mined at 288 K in water/acetonitrile by using analytical
HPLC (see Figure 3). The rate constant at 288 K determined
by this method compared extremely well with that deter-
mined by using vt-HPLC (equilibration: krot =8.7ꢀ0.6ꢂ
10ꢁ5 sꢁ1; vt-HPLC: krot =8.4ꢀ0.2ꢂ10ꢁ5 sꢁ1).
Figure 2. Chromatograms for polycycle 9a determined by using variable-
temperature analytical HPLC in acetonitrile/water.
Table 2. Populations of the syn and anti rotamers of polycyclic com-
pounds 9 and 11 (in which Ar=m-nitrophenyl). Unless otherwise indicat-
ed, the populations were determined in water/acetonitrile.[a]
syn:anti[b] K[b,c]
DH8[d]
DS8[d]
DG8[b]
[kJmolꢁ1
]
[Jmolꢁ1 Kꢁ1
]
[kJmolꢁ1
]
9a 65:35
0.53ꢀ0.05
0.50ꢀ0.01
0.61ꢀ0.01
0.53ꢀ0.05
0.37ꢀ0.01
0.53ꢀ0.05
0.47ꢀ0.01
0.79ꢀ0.07[e]
ꢁ1.7ꢀ1.6
ꢁ2.8ꢀ0.4
ꢁ1.8ꢀ0.4
ꢁ3.1ꢀ0.5
ꢁ6.9ꢀ1.2
ꢁ2.5ꢀ0.8
ꢁ4.3ꢀ0.3
ꢁ11ꢀ7
ꢁ15ꢀ1
ꢁ10ꢀ1
ꢁ16ꢀ2
ꢁ29ꢀ4
ꢁ13ꢀ3
ꢁ20ꢀ1
1.6ꢀ0.2
1.70ꢀ0.05
1.24ꢀ0.04
1.6ꢀ0.1
9b 67:33
9c 62:38
9d 66:34
9e 70:30
9 f 66:34
9g 68:32
11 56:44
2.43ꢀ0.03
1.61ꢀ0.06
1.84ꢀ0.05
0.6ꢀ0.2
[f]
[f]
–
–
Figure 3. Equilibration of the syn (top) and anti (bottom) rotamers of
polycycle 9a at 288 K in acetonitrile/water.
[a] See Table 1 for the substitution of each compound. [b] At 298 K.
[c] For the definition of K, see Scheme 2. [d] Unless otherwise indicated,
derived from the populations of the rotamers, determined by HPLC, in
the slow and intermediate exchange regimes. [e] Determined by
500 MHz 1H NMR spectroscopy in [D6]DMSO by integrating the signals
corresponding to each rotamer in the slow exchange regime. [f] Not de-
termined.
The interconversion of the rotamers of 11 was much
faster than those of polycycles 9a–g (Table 3). The rotamers
of 11 interconverted rapidly on the HPLC timescale, and
only one peak was, therefore, observed by analytical HPLC;
thus, variable-temperature NMR experiments were used to
study the interconversion process. The half-life of the syn
rotamer of 11, extrapolated to 298 K, was about 3 s (com-
pared with 0.77 h for its constrained analogue 9a).
Table 3. Kinetic data for rotation about the single bond between the sp3-
hybridised bridgehead carbon and the m-nitrophenyl ring.[a]
[c]
[c,e]
Solvent
Method[b]
krot
DGf°[c,d]
[kJmolꢁ1
t =
2
1
[ꢂ10ꢁ6 sꢁ1
]
G
]
[h]
0.77
2.2[g]
21
2.0
2.3
0.26
1.1
0.6
9a
9a
9b
9c
9d
9e
9 f
9g
11
H2O/MeCN
H2O/MeCN
H2O/MeCN
H2O/MeCN
H2O/MeCN
H2O/MeCN
H2O/MeCN
H2O/MeCN
[D6]DMSO
vt-HPLC
HPLC[f]
250ꢀ6
87ꢀ6[g]
9ꢀ7
94.7ꢀ0.8
93ꢀ1[g]
101ꢀ4
97ꢀ1
vt-HPLC
vt-HPLC
vt-HPLC
vt-HPLC
vt-HPLC
vt-HPLC
vt-NMR
94ꢀ6
84ꢀ5
97.5ꢀ0.8
92.1ꢀ0.5
95ꢀ1
720ꢀ80
180ꢀ10
310ꢀ5
200000[h]
94ꢀ1
77[i]
0.0008
[a] See Table 1 for the substitution of the polycyclic compounds. [b] See
the Supporting Information for the temperature ranges used in the vt-
HPLC studies. [c] At 298 K unless otherwise specified. [d] For the for-
ward direction, that is, conversion of the syn into the anti rotamer.
[e] Half-life of the anti rotamer. [f] The rotamers were separated by prep-
arative HPLC at 278 K; the equilibration of both rotamers at 288 K was
followed by analytical HPLC. [g] At 288 K. [h] Extrapolated to 298 K on
the assumption that DSf° is small. [i] At the coalescence temperature.
We quantified the distortion of the amide with the R2 sub-
stituent by using independent parameters[7] that have been
developed. These parameters describe the pyramidalisation
To validate the results obtained by using vt-HPLC, we
also studied the re-equilibration of the rotamers of 9a. For
this compound, the rotamers were separated by HPLC at
at the nitrogen (cN) and carbon atoms (cC) and the torsion
2
ꢁ
about the C N bond (t).The crystal structures of 10a (R =
Ph) and 10b (R2 =iPr) provide experimental evidence for
Chem. Eur. J. 2009, 15, 2185 – 2189
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2187

A. Nelson, S. Warriner et al.
amide distortion (Table 4), although the observed distortion
was smaller than in the minimised structures (Table 5) of
similarly substituted compounds (compare compounds 13a
and 10a, R2 =Ph, with compounds 13c and 10b, R2 =iPr).
Table 4. Analysis of the amide groups with the R2 substituent in the crys-
tal structures of polycyclic compounds 10a and 10b.[a]
Bond lengths [ꢃ]
Distortion parameters [8]
ꢁ
N C(O)
C=O
cN
cC
t
10a
1.372
1.371
1.363
1.368
1.232
1.239
1.240
1.238
ꢁ19.4
ꢁ10.8
10.3
ꢁ10.6
ꢁ0.8
3.6
6.6
7.6
10b[b]
ꢁ2.7
ꢁ9.6
Figure 4. Minimised structure (a) and TS structure (b) for polycycle 13b.
An animation for the bond-rotation process is provided in the Supporting
Information.
3.3
11.0
[a] See Table 1 for the substitution of the polycyclic compounds. [b] The
asymmetric unit contains two molecules from one enantiomeric series
and one molecule from the other; these molecules populate significantly
different conformations in the crystal.
ꢁ0.5), which implies that its conjugation with the carbonyl
group was greater in the TS than in the ground state. The
calculated barrier heights for polycycles 13a (R2 =Ph), 13e
(R2 =p-MeOC6H5) and 13 f (R2 =p-O2NC6H5) reproduced
this trend, with lower barriers with a more electron-rich R2
aryl group (Table 5). Furthermore, the calculations support-
ed the view that the R2 aryl ring was more conjugated with
the carbonyl group in the TS (the dihedral angles were 15,
12 and 338 for 13a, 13e and 13 f, respectively) than in the
minimised structures (the dihedral angles were 48, 44 and
508 for 13a, 13e and 13 f, respectively).
Table 5. Distortion of the amide group with the R2 substituent and barri-
er heights for bond rotation for polycyclic compounds 13a–f and 14 de-
termined by using the B3LYP hybrid functional and a 6-31G* basis set.[a]
Distortion parameters[b] [8]
Free-energy barrier
[kJmolꢁ1
cN
cC
t
N
]
13a
13b
13c
13d
13e
13 f
14
ꢁ23.9
ꢁ17.5
ꢁ18.0
ꢁ22.5
ꢁ24.7
ꢁ22.8
ꢁ35.9
0.3
1.2
3.4
16.7
14.9
18.1
17.1
18.1
16.1
26.1
93.4
104.2
105.0
105.9
89.5
93.9
83.9
ꢁ0.1
0.1
The steric influence of R2 was also important, with slower
bond rotation with smaller groups (compare 9b, R2 =Me,
with 9c, R2 =iPr, and 9d, R2 =tBu; Table 3). However, all of
our DFT calculations, which also used alternative function-
als and basis sets (see the Supporting Information), predict
13b–d to have very similar free-energy barriers. Analysis of
the origins of this discrepancy is complicated given the very
small differences in the free-energy barrier for these com-
pounds. However, a comparison of the minimised structures
of polycycles 13a–d revealed that the amide distortion is
greater with the larger R2 groups (Ph, iPr and tBu) than
with R2 =Me (compare 13a, 13c and 13d with 13b;
Table 5). Because bond rotation requires the amide to dis-
tort dramatically (see Figure 4 and the Supporting Informa-
tion), the compounds with larger R2 groups may have more
destabilised ground states, which result in a faster rate of
bond rotation.
0.6
1.3
[a] See Table 1 for the substitution of the polycyclic compounds. [b] Cal-
culated for the minimised structure.
The fusion of a cyclopentane ring onto the nucleus of the
polycycles had a remarkable effect on the rate of bond rota-
tion: krot was about 1000 times larger for 11 than for its
more constrained analogue 9a. The electronic structure
theory results similarly show a reduced barrier height for 14
compared with 13a, and permit a structural explanation for
its origin. Inspection of the transition-state (TS) structures
of 13a–f shows that there are three significant steric interac-
tions that contribute to the barrier heights (see Figure 4). As
the phenyl ring rotates, the most significant interactions are
between hydrogen atoms Ha and Hc, and between hydrogen
Hb and carbonyl oxygen O. During the rotation of the
phenyl ring, the carbonyl group rotates out of conjugation,
which leads to increased steric interaction between carbonyl
oxygen O and hydrogen Hd. The removal of the fused cy-
clopentane ring (as in 14) means that the carbonyl oxygen
encounters less steric repulsion, which results in a smaller
barrier.
Conclusions
Previously, rotamers arising from restricted rotation about
bonds between sp2- and sp3-hybridised carbon atoms have
only been isolated when the sp2-hybridised atom is part of
an aryl ring with ortho substituents. The rate of equilibration
between the rotamers of polycycles 9 is remarkable because
the m-nitrophenyl ring does not have any such substituents.
We have therefore shown for the first time that a highly hin-
dered environment for the sp3-hybridised carbon is sufficient
for slow bond rotation. The sp3-hybridised bridgehead
The electronic nature of an aromatic R2 substituent had a
small but significant influence on krot (compare 9a, R2 =Ph,
with 9e, R2 =p-MeOC6H5, and 9 f, R2 =p-O2NC6H5;
Table 3). Bond rotation was significantly faster with an elec-
tron-rich aryl substituent, R2, (Hammett[8] parameter 1ꢂ
2188
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 2185 – 2189

Slow Rotation about a Single Bond
FULL PAPER
of comparisons with calculations using other basis sets and functionals
are described in the Supporting Information.
carbon atoms in polycycles 9 are exceptionally hindered,
being both fully substituted and flanked by three highly sub-
stituted atoms. Indeed, our ability to study the dynamic
properties of highly hindered polycycles 9 at all probably
stemmed from the intramolecular nature of the final step of
the Joulliꢀ–Ugi reaction;[9] we had previously been unable
to prepare similar compounds by reduction of the imines 5
and acylation under forcing reaction conditions.
Acknowledgements
We thank AstraZeneca, EPSRC, the University of Leeds and the Techni-
cal University of Vienna for funding, Dr. J. Ouzman and Mr C. Kilner
for X-ray crystallography and Dr. Oliver Trapp, Ruprecht-Karls-Univer-
sitꢄt Heidelberg, for kindly making DCXplorer available to us.
Experimental Section
˘
[1] M. Oki in Topics in Stereochemistry, Vol. 14 (Eds: N. L. Allinger,
Variable-temperature HPLC: Samples of the polycycles were analysed
by using vt-HPLC with a Phenomenex Hyperclone column [ODS (C18),
250ꢂ4.6 mm) and a Dionex analytical HPLC system equipped with a
diode array detector. 20 mL samples of the polycycles (ca. 2 mm in metha-
nol) were analysed at 5 K intervals in triplicate (see the Supporting Infor-
mation for the ratio of acetonitrile/water used in each case). At each
temperature, K and krot were extracted by analysing chromatograms with
DCXplorer.[10] The results in Tables 2 and 3 were obtained by weighted
fitting of the data to appropriate equations (see the Supporting Informa-
tion).
E. L. Eliel, S. H. Wilen), Wiley, New York, 1983, pp. 1–81.
[2] a) M. Nakamura, M. Oki, Tetrahedron Lett. 1974, 15, 505–508;
b) M. T. Ford, T. B. Thompson, K. A. J. Snoble, J. M. Timko, J. Am.
Chem. Soc. 1975, 97, 95–101.
[3] M. Nakamura, M. Oki, Bull. Chem. Soc., Jpn. 1980, 53, 2977–2980.
˘
˘
[4] a) W. J. Bromley, M. Gibson, S. Lang, S. A. Raw, A. C. Whitwood,
R. J. K. Taylor, Tetrahedron 2007, 63, 6004–6014; b) S. A. Raw,
R. J. K. Taylor, J. Am. Chem. Soc. 2004, 126, 12260–12261.
[5] a) M. M. Bowers, P. Carroll, M. M. Joulliꢀ, J. Chem. Soc. Perkin
Trans. 1 1989, 857–865; b) T. M. Chapman, I. G. Davies, B. Gu,
T. M. Block, D. I. C. Scopes, P. A. Hay, S. M. Courtney, L. A. Mc-
Neill, C. J. Schofield, B. G. Davis, J. Am. Chem. Soc. 2005, 127, 506–
507.
[6] a) J. Veciana, M. I. Crespo, Angew. Chem. 1991, 103, 85–88; Angew.
Chem. Int. Ed. Engl. 1991, 30, 74–76; b) A. C. Spivey, P. Charbon-
neau, T. Fekner, D. H. Hochmuth, A. Maddaford, C. Malardier-Jug-
root, A. J. Redgrave, M. A. Whitehead, J. Org. Chem. 2001, 66,
7394–7401.
[7] J. D. Dunitz, F. K. Winkler, Acta Crystallogr. Sect. B 1975, 31, 251–
263.
[8] L. P. Hammett, J. Am. Chem. Soc. 1937, 59, 96–103.
[9] a) I. Ugi, G. Kaufhold, Just. Liebigs Ann. Chem. 1967, 709, 11–28;
b) S. Marcaccini, T. Torroba, Nat. Prot. Nat. Toxicants Food 2007, 2,
632–639.
Separation and re-equilibration of the rotamers of 9a: The rotamers of
polycycle 9a were separated by semi-preparative HPLC at 278 K [Phe-
nomenex Hyperclone column ODS (C18); 250ꢂ10 mm; gradient elution
74:26!79:21, acetonitrile/water, 58C]. Samples of the rotamers were col-
lected in vials cooled to 278 K. The re-equilibration process was followed
over 21 h by HPLC analysis of aliquots of the rotamers stored at 288 K
[Phenomenex Hyperclone column ODS (C18); 250ꢂ4.6 mm]. The results
in Table 3 were obtained by fitting the data to appropriate equations (see
the Supporting Information).
Computational methods: TS structures were determined by performing
constrained potential-energy surface (PES) scans about the torsional co-
ordinate and then initiating TS optimizations in the region of the energy
maximum. Frequency calculations were performed for all stationary
points and confirmed whether they were minima or TSs. The frequency
calculations were also used to obtain DSf°, which was significant to an ac-
curate calculation of DGf°. It has been shown that gas-phase PESs may
be subject to considerable error when investigating solution-phase sys-
tems if there is substantial zwitterionic character associated with the TS
dynamics.[11] However, as this is not the case for the molecules investigat-
ed in this study, we did not implement solvent continuum models. Details
[10] a) O. Trapp, Anal. Chem. 2008, 80, 189–198; b) O. Trapp, J. Chro-
matogr. B 2008, 875, 42–47.
[11] V. K. Aggarwal, J. N. Harvey, J. Richardson, J. Am. Chem. Soc.
2002, 124, 5747–5756.
Received: October 14, 2008
Published online: January 20, 2009
Chem. Eur. J. 2009, 15, 2185 – 2189
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2189