Intramolecular asymmetric heck reactions: Evidence for dynamic kinetic resolution effects

Article abstract of DOI:10.1021/ol0606132

Enantioselectivity of the cyclization of 1 varies at different stages in the reaction. X-ray crystallography has shown that 1 exists as enantiomerically pure (M) and (P) chiral helical structures defined by the relative orientations of the arene, amide, a

Full text of DOI:10.1021/ol0606132

ORGANIC
LETTERS
2006
Vol. 8, No. 14
2917-2920
Intramolecular Asymmetric Heck
Reactions: Evidence for Dynamic
Kinetic Resolution Effects
Michael C. McDermott,† G. Richard Stephenson,*,† David L. Hughes,† and
Andrew J. Walkington‡
Wolfson Materials and Catalysis Centre, School of Chemical Sciences and Pharmacy,
UniVersity of East Anglia, Norwich, NR4 7TJ, U.K., and GlaxoSmithKline, Medicines
Research Centre, Gunnels Wood Road, SteVenage, Hertfordshire SG1 2NY, U.K.
g.r.stephenson@uea.ac.uk
Received March 13, 2006 (Revised Manuscript Received May 10, 2006)
ABSTRACT
Enantioselectivity of the cyclization of 1 varies at different stages in the reaction. X-ray crystallography has shown that 1 exists as enantiomerically
pure (M) and (P) chiral helical structures defined by the relative orientations of the arene, amide, and alkene. The relative rates of interconversion
of the rotamers of 1 have been established, leading to mechanistic proposals to account for the variation of ee based on kinetic resolution
effects.
The Heck reaction1 has become one of the most industrially
important and widely used palladium-catalyzed carbon-
carbon bond forming reactions. The scope and capabilities
of this process have been extensively reviewed.2 Asymmetric
modification3 has made available a powerful procedure for
the induction of asymmetry in the reactions of aryl iodides
and triflates with alkenes, typically involving discrimination
between pairs of prochiral enantiotopic alkenes or the
enantiofaces of prochiral alkenes and dienes.
An unusual example of the asymmetric Heck reaction was
reported by Overman’s group4 as part of the natural product
synthesis of gelsemine. They found that by using a catalyst
prepared from Pd2(dba)3 and (R)-BINAP the addition of
triethylamine gave one diastereoisomer of the product,
whereas addition of Ag3PO4 gave the opposite stereoisomer.
Overman’s group also demonstrated5 that this switch in
stereoselectivity could be employed as an enantioselective
procedure. In the cyclization of substrate 1 with (R)-BINAP
as the chiral auxiliary, the use of PMP6 as the base gave the
(R)-(+) enantiomer of product 2 in 25% ee and the use of
† University of East Anglia.
‡ GlaxoSmithKline U.K.
(1) Heck, R. F. J. Am. Chem. Soc. 1968, 90, 5518-5526. Dieck, H. A.;
Heck, R. F. J. Am. Chem. Soc. 1974, 96, 1133-1136.
(2) Heck, R. F. Organic Reactions; Dauben, W. G., Ed.; Wiley and Sons
Inc.: New York, 1982; Vol. 27, Chapter 2, pp 345-390. Trost, B. M.;
Verhoeven, T. R. ComprehensiVe Organometallic Chemistry; Abel, E. W.,
Stone, F. G. A., Wilkinson, G., Eds.; Pergamon Press: Oxford, 1982; Vol.
8, Chapter 57, pp 854-883. Heck, R. F. ComprehensiVe Organic Synthesis;
Flemming, I., Semmelhack, M. F., Trost, B. M., Eds.; Pergamon Press:
Oxford, 1991; Vol. 4, Chapter 4.3, pp 833-863. Beletskaya, I. P.;
Cheprakov, A. V. Chem. ReV. 2000, 100, 3009-3066.
(3) de Meijere, A.; Meyer, F. E. Angew. Chem., Int. Ed. Engl. 1994, 33,
2379-2411. Boden, C. D. J.; Kojima, A.; Shibasaki, M. Tetrahedron 1997,
53, 7371-7395. Shibasaki, M.; Vogl, E. M. J. Organomet. Chem. 1999,
576, 1-15. Hayashi, M.; Keenan, M.; Loiseleur, O.; Pfaltz, A.; Schmees,
N. J. Organomet. Chem. 1999, 576, 16-22. Dounay, A. B.; Overman, L.
E. Chem. ReV. 2003, 103, 2945-2963. Ohshima, T.; Shibasaki, M.; Vogl,
E. M. AdV. Synth. Catal. 2004, 346, 1533-1552. Bell, H. P.; Ila, H.; Tietze,
L. F. Chem. ReV. 2004, 104, 3453-3516. Guiry, P. J.; McManus, H. A.
Chem. ReV. 2004, 104, 4151-4202.
(4) Mandin, A.; Overman, L. E. Tetrahedron Lett. 1992, 33, 4859-4862.
(5) Ashimori, A.; Overman, L. E. J. Org. Chem. 1992, 57, 4571-4572.
Ashimori, A.; Matsuura, T.; Overman, L. E.; Poon, D. J. J. Org. Chem.
1993, 58, 6949-6951. Ashimori, A.; Bachand, B.; Calter, M. A.; Govek,
S. P.; Overman, L. E.; Poon, D. J. J. Am. Chem. Soc. 1998, 120, 6488-
6499.
(6) 1,2,2,6,6-Pentamethylpiperidene.
10.1021/ol0606132 CCC: $33.50
© 2006 American Chemical Society
Published on Web 06/15/2006

Ag3PO4 gave the (S)-(-) enantiomer in 59% ee, both in high
yields (88-91%).7
(entry 1), the ee of the product 2 was only 9%, but this rose
to 40% ee after 240 min (entry 3). Similarly, at 80 °C, an ee
of 26% at 60 min (entry 9) improved to 48% ee after 180
min (entry 11). The better high-limit enantioselectivity in
this case is ascribed to the more efficient reaction at this
temperature (the reaction had gone to completion after 180
min). In general, the data in Table 1 showed that the observed
ee of the product is best at high conversions (86-100%)
but never reaches the desired high levels of optical purity
because of the low ee product accumulated at the early stages
of the reaction. At 100 °C (entries 12 and 13), the enantio-
selectivity is lower than that at 60 °C (entry 5) and at 80 °C
(entries 8, 10, and 11), although the reactions at 100 °C were
taken more quickly to completion.
Our interest in these asymmetric Heck reactions arose from
investigations of the use of chiral additives to optimize
enantioselectivity8 because the factors that controlled enan-
tioselectivity during the asymmetric cyclization were finely
balanced, thus changes in reaction conditions and/or the use
of additives could profoundly influence the outcome of the
reaction.
We report here preliminary results from a series of
experiments conducted using an Anachem SK233 worksta-
tion9 equipped with “in-line” chiral HPLC analysis (adapted
from an Agilent 1100 HPLC system), in which reaction
times, temperatures, concentrations, and catalyst/Ag3PO4
loadings were varied (Table 1). The data obtained demon-
Despite the wide range of reaction conditions used in our
study, Table 1 (graph) clearly shows that there is a general
tendency for high ee’s in the product at high conversions
and relatively low ee’s at low conversions. These data are
consistent with the presence of two competing influences.
The limiting ee’s at 100 °C are ascribed to catalyst
decomposition, which is a well-known occurrence in Heck
coupling.10 The reaction mixtures blacken noticeably at this
temperature, and if catalytically active Pd(0) particles11 are
generated by the thermal decomposition of the Pd(dba)[(R)-
BINAP] intermediates, a competing racemic cyclization will
build up over time. The low ee’s observed at lower
temperatures at the 60 min stage suggest the presence of a
more complex reaction system than a simple kinetic control
based on two competing pro-(R) and pro-(S) mechanisms.
In all the examples presented in Table 1, Ag3PO4 was used
as the base, and the expected (S)-(-) product was obtained.12
Despite the possibility of competing rac and pro-(R)
pathways, the overall asymmetric induction in our examples
is consistent with published results5,7 on this cyclization
reaction.
In an attempt to gain insight into the transition states in
the coupling process, the substrate 1 was crystallized from
a 3:1 mixture of hexane and DCM and studied by X-ray
diffraction. The structure (5, Figure 1) takes on a helical form
with the bulky ortho-iodide substituent on the aromatic ring
forcing rotation about the C(2)-N(2) bond by 117°.13 The
amide group itself was found to adopt the expected E14-16
conformation, with considerable pyramidalization of the
nitrogen and a C(2)-C(22) torsion angle of 18°.17 The
adjacent carbon-carbon double bond showed a dihedral
Table 1. Asymmetric Cyclizations of 1 under a Variety of
Conditions
catalyst
temp time loading Ag3PO4 concna conversionb eec
entry (°C) (min) (mol %) (equiv) (g mL-1
)
(%)
(%)
1
2
3
4
5
6
7
8
9
10
11
12
13
60
60
10
10
10
5
5
5
5
5
7.5
5/3
5/3
5/3
5/3
1/3
1/3
2
2
1
1
1
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.08
0.08
0.08
0.2
0.2
14
32
39
23
86
17
54
68
51
9
36
40
33
51
17
32
41
26
43
48
38
41
60 150
60 240
60 240
60 240
60 240
80
80
80
80 100
80 180
100
100
60
60
60
7.5
7.5
5
73
100
95
60
60
5/3
1/3
10
100
(10) van Leeuwen, P. W. N. M. Appl. Catal., A Gen. 2001, 212, 61-81.
Koningsberger, D. C.; Sietsma, J. R. A.; Tromp, M.; van der Eerden, A.
M. J.; van Haaren, R. J.; van Leeuwen, P. W. N. M. J. Chem. Soc., Chem.
Commun. 2003, 128-129.
a Concentration of 1. b Measured by HPLC. c Determined by chiral HPLC
using a chiracel OJ column.
(11) Breinbauer, R.; Reetz, M. T.; Wanninger, K. Tetrahedron Lett. 1996,
37, 4499-4502. Reetz, M. T.; Westermann, K. Angew. Chem., Int. Ed.
2000, 39, 165-168.
strated that the enantioselectivity of the product changes as
the reaction proceeds. For example, at 60 °C after 60 min
(12) It is important that Ag3PO4 purchased from Aldrich is used.
(13) C(1)-C(21) torsion angle: (+)-anticlinal, 117°.
(14) Chupp, J. P.; Olin, J. F. J. Org. Chem. 1967, 32, 2297-2303. Curtis,
E.; Mislow, K.; Raban, M.; Shvo, Y. J. Am. Chem. Soc. 1967, 89, 4910-
4917. Itai, A.; Saito, S.; Tomioka, N.; Toriumi, Y. J. Org. Chem. 1995, 60,
4715-4720.
(7) Ashimori, A.; Matsuura, T.; Overman, L. E.; Poon, D. J. J. Am. Chem.
Soc. 1998, 120, 6477-6487.
(8) Sellarajah, S. Ph.D. Thesis, University of East Anglia, 2001.
Sellarajah, S.; Stephenson, G. R. Unpublished results.
(9) Armitage, M. A.; Smith, G. E.; Veal, K. T. Org. Process Res.
DeV. 1999, 3, 189-195. Harre, M.; Tilstam, U.; Weinmann, H. Org.
Process Res. DeV. 1999, 3, 304-318. Information available from
www.reactarray.com.
(15) Siddal, T. H.; Stewart, W. E. Chem. ReV. 1970, 70, 517-551.
(16) Balog, H.; Cass, Q. B.; Curran, D. P.; Degani, A. L. G.; Freitas, L.
C. G.; Grieb, S. J.; Hale, G. R.; Hernandez, M. Z. Tetrahedron: Asymmetry
1997, 8, 3955-3975.
(17) C(2)-C(22) torsion angle: (-)-synclinal, -18°.
2918
Org. Lett., Vol. 8, No. 14, 2006

Figure 1. X-ray structure of (P)-1 (5), representative arrangement of principal planes (6) in (P)-1, and X-ray structure of (M)-1 (7).
angle of 51° with the amide plane.18 This structure can be
characterized on the basis of three principal planes (6, Figure
1), with the overall (P) helical form arising from the sequence
of (P)-axial chirality19 about the aryl-nitrogen bond [C(2)-
N(2)] and also about the carbonyl-alkene [C(21)-C(22)]
bond.
°C shows a difference in rotational broadening indicating
the existence of a second dynamic process with a much lower
rotational barrier. Compared to these two processes, there is
no evidence from the spectroscopic data to establish the
relative ease of rotation about the aryl-nitrogen bond [C(2)-
N(2)]. Fortunately, evidence in the literature16,22,23 allows this
issue to be resolved, as there are many reports of configu-
rationally stable amide atropisomers and it is known24 that
this may be significant in the control of intramolecular Heck
reactions. The interconversion of atropisomers of 1, although
probably accessible at 80 °C, will be far slower than the
rotations studied in the variable-temperature NMR experi-
ment. Because the atropisomer interconversion is slow and
the E/Z interconversion is known to be the main process
revealed by the NMR study, the third dynamic process must
arise from rotation about the carbonyl-alkene bond.
The crystal proved to contain a single enantiomer of this
(P)-helical structure (5, Figure 1). A second crystal from the
sample yielded a structure of the enantiomeric (M)-helical
form (7, Figure 1).20 The presence of pure (P) and (M)
crystals in solid samples of the starting material for the
cyclization (1) offers a possible explanation of the anomalous
low ee’s observed at low conversion because, with the chiral
helical conformations of the aryl iodide, a form of kinetic
resolution may be in operation.
1
The 400 MHz H NMR spectra of 1 gave support to the
possibility that the helical conformations observed in the solid
state are also significant in solution because the signals from
the alkene hydrogen and methyls of the 2-methylbut-2-enyl
group were broadened suggesting hindered rotation on the
NMR timescale. To explore this further, a variable-temper-
ature NMR study of 1 was performed in DMF-d7 at
temperatures between -20 and +80 °C, in steps of 5 °C. At
the low-temperature limit, two distinct signals for the alkene
hydrogen were observed at 5.94. and 5.64 ppm in a ratio of
ca. 1:9. On the basis of peak assignments15,16 for other amide
conformers, the more upfield and much larger peak at 5.64
ppm was identified as the E-amide. The preference for the
E-amide in solution,16 and in the solid state,21,22 is well
established, further supporting these assignments. The rota-
tional barrier for this dynamic process was estimated as 55
kJ mol-1. Comparison of the two signals observed at -20
To remove the atropisomer issue, a preliminary series of
experiments have been performed employing a symmetriza-
tion approach. Using the modified substrate 3, which
contained two 2-methylbut-2-enyl groups as an imide deriva-
tive of 2-iodoaniline, the asymmetric Heck reaction (Scheme
1) gave product 4 with an ee of only 10% (measured using
Scheme 1. Asymmetric Cyclization of Substrate 3
(18) N(2)-C(23) torsion angle: (-)-synclinal, -51°.
(19) Cahn, R. S.; Ingold, C. K.; Prelog, V. Angew. Chem. 1966, 78, 413-
447. Cahn, R. S.; Ingold, C. K.; Prelog, V. Angew. Chem., Int. Ed. Engl.
1966, 5, 385-415, 511.
(20) The chirality of these structures and the E/Z form of the amide can
be defined by the sequence of torsion angles C(1)-C(21), C(2)-C(22),
and N(2)-C(23): (P)-1 +117, -18, -51 [(+)-anticlinal, (-)-synperiplanar,
(-)-synclinal]; (M)-1 -117, +16, +53 [(-)-anticlinal, (+)-synperiplanar,
(+)-synclinal]. The enantiomeric nature of these two helical structures is
apparent from the reversal of signs in all three angles, comparing (P) and
(M) forms.
chiral shift NMR experiments7,8), suggesting that the atro-
pisomeric chiral form of 1 may be a significant factor for
an efficient asymmetric induction in the Heck cyclization.
(23) Izawa, H.; Kitagawa, O.; Taguchi, T. Tetrahedron Lett. 1997, 38,
4447-4450. Fushimi, Y.; Kitagawa, O.; Momose, S.; Taguchi, T. Tetra-
hedron Lett. 1999, 40, 8827-8831. Bennett, D. J.; Blake, A. J.; Cooke, P.
A.; Godfrey, C. R. A.; Pickering, P. L.; Simpkins, N. S.; Walker, M. D.;
Wilson, C. Tetrahedron 2004, 60, 4491-4511. Hata, T.; Kamikawa, K.;
Koide, H.; Uemura, M.; Yoshihara, K. Tetrahedron 2004, 60, 4527-4541.
Curran, D. P.; Geib, S. J.; Petit, M. Tetrahedron 2004, 60, 7543-7552.
(24) Chen, C. H.-T.; Curran, D. P.; Liu, W. J. Am. Chem. Soc. 1999,
121, 11012-11013.
(21) Pedersen, B. F. Acta Chem. Scand. 1967, 21, 1415-1424. Itai, A.;
Kagechika, H.; Saito, S.; Shudo, K.; Toriumi, Y. J. Am. Chem. Soc. 1992,
114, 10649-10650.
(22) Ates, A.; Curran, D. P. J. Am. Chem. Soc. 2001, 123, 5130-5131.
Org. Lett., Vol. 8, No. 14, 2006
2919

The X-ray structure of 3 showed disorder of the iodoarene
group in the crystal. This group lies in one of two coplanar,
overlapping orientations which are shown separately in
Figure 2. There were notable differences compared to the
Scheme 2. Proposed Reaction Pathways
(S) catalytic intermediate is the stereochemically preferred
helix for the carbon-carbon bond forming step. When the
pro-(R)/pro-(S) interconversion of organopalladium interme-
diates is operative, the switch in enantioselectivity arises from
the Ag3PO4-promoted equilibrium improving access to the
otherwise slowly formed oxidative addition product from (P)-
1.
Figure 2. Two structures found in the unit cell of substrate 3.
In summary, we have shown that chiral helical conforma-
tions of 2-iodoanilide 1 interconvert through internal bond
rotations leading to the proposal of a dynamic kinetic
resolution mechanism to account for the switch of enantio-
selectivity.
structure defined for 1, with both amide moieties in the imide
showing the opposite conformation.25 As expected, the two
imide C-N bonds were far longer (1.405 and 1.413 Å) than
the amide bond length in 1 (1.361 Å), reducing steric effects
in this part of the molecule. The long bond length in the
solid state is also presumably present as a true structural
feature in solution, as indicated by sharp NMR signals that
suggest lowered rotational barriers. The two imide structures
observed in the solid state (Figure 2) should rapidly inter-
convert in solution by the facile rotation about the C(21)-
C(22) and C(25)-C(26) bonds.
Acknowledgment. We thank the EPSRC (CASE award)
and GlaxoSmithKline for financial support, the EPSRC Mass
Spectrometry Centre and the National Crystallography
Service [University of Wales (Swansea) (HRMS) and
University of Southampton (data sets)], Dr. Shane Sellarajah
(UEA) for preliminary experiments and procedures for CSR
experiments, Colin MacDonald (UEA) for VT NMR experi-
ments, and Dr. Adrian G. Bateman, Dr. Bryan J. Davies,
Dr. Kathy S. Harwood of Strategic Technologies, and Dr.
Barbara O’Reilly of Analytical Sciences at GlaxoSmithKline
for assistance with the Anachem SK233 system and chiral
HPLC.
In the PMP protocol described by Overman,7 the dynamic
kinetic resolution of interconverting helical iodides (P)-1 and
(M)-1 must favor the oxidative addition to the (M)-helix to
promote the pro-(R) pathway to product 2. For the Ag3PO4
protocol, however, this kinetic resolution cannot account for
the variation of ee with degree of conversion unless the
opposite stereochemical preference is favored in the carbon-
carbon bond forming stages [K(R)/K(S)] of the cycle. We
propose that the addition of Ag3PO4 opens an equilibrium
between the palladium-bound oxidative addition complexes
of the (P)- and (M)-helices [(P)-/(M)-8; Scheme 2]. The route
to the (S)-(-)-2 product would then be favored if the pro-
Supporting Information Available: Procedures for ee
measurements of 2 and 4; synthesis and characterization data
for 3 and 4; 1H NMR spectra of 1 between -20 and 80 °C;
and X-ray crystallographic details for (P)-1, (M)-1, and rac-
3. This material is available free of charge via the Internet
at http://pubs.acs.org.
(25) As the priority of groups changes, this is still labeled E.
OL0606132
2920
Org. Lett., Vol. 8, No. 14, 2006