Mechanism of substitution reaction on sp2-carbon center with lithium organocuprate

Article abstract of DOI:10.1021/ja046616w

Theory and experiments suggest that the substitution reaction of a lithium dialkylcuprate(I) with an alkenyl bromide takes place through a π-complex (cuprio(III)cyclopropane) that directly breaks down to the alkenylated product rather than via a conventional three-centered transition state. This mechanism is consistent with the broader mechanistic picture of the organocuprate reactions and accounts for the retentive stereochemistry and the kinetic isotope effect observed in the experiments. Copyright

Full text of DOI:10.1021/ja046616w

Published on Web 09/10/2004
Mechanism of Substitution Reaction on sp2-Carbon Center with Lithium
Organocuprate
Naohiko Yoshikai and Eiichi Nakamura*
Department of Chemistry, The UniVersity of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan
Received June 9, 2004; E-mail: nakamura@chem.s.u-tokyo.ac.jp
Scheme 2. Two Pathways of the Reaction of Cuprate 1 with Vinyl
Bromide 2 To Give Propenea
A report in 1967 that a lithium diorganocuprate (R2CuLi)
undergoes substitution reaction with an alkenyl bromide with
retention of stereochemistry1 changed the accepted wisdom that a
nucleophilic substitution on an sp2-hybridized carbon is synthetically
impracticable.2 The new paradigm is now firmly established in
chemistry through subsequent development of catalytic variants.3,4
A generally accepted mechanism of the cuprate reaction involves
an oxidative addition/reductive elimination sequence (Scheme 1).5-7
Scheme 1
There underlie, however, a few questionable assumptions in such
a scheme: The first is the neglect of the polymetallic structure of
R2CuLi that is essential for the reactivity of organocuprate.8,9 In
addition, the mechanism silently assumes direct insertion of the
metal atom into the C-X bond to form the vinyl complex II and,
hence, underrates the role of the π-complex I (note that the σ*-
and the π*-orbitals of the alkenyl bromide responsible for the σ-
and the π-complexation are orthogonal to each other). On the basis
of theoretical10 and experimental studies, we propose here a
significant mechanistic modification, which is consistent with and
adds a new dimension to the modern mechanistic pictures of
organocuprate reactions.8
Experimental and theoretical studies in recent years have shown
that the cyclic dimer 1 (in its solvated form, S ) ethereal solvent)
exists as a stable species in ethereal solution and reacts with
electrophiles (Scheme 2).8,9 The cuprate 1 undergoes a two-point
interaction with vinyl bromide 2 to form a π-complex 3, as it does
with cyclohexenone.11 This π-complex, however, is less favorable
here than in the case of more electrophilic cyclohexenone. This
energetics was confirmed by 13C NMR measurement on a mixture
of Me2CuLi‚LiI and 1-bromocyclooctene in diethyl ether, which
did not show any indication of a π-complex at -20 to -40 °C
(Supporting Information) and afforded directly the coupling product
at >-10 °C (note that conjugate addition usually proceeds below
-50 °C).
a A and B rotations refer to rotation along the Cu-C1 axis. Energies (in
kilocalories per mole) are relative to [1 + 2]. Energy changes are shown
together with arrows.
vinylcopper(III) complex 4 undergoes quick reductive elimination
via TS4-5 to give the mixed cuprate 5 and propene 6.12
There is another alternative rotational mode that has thus far
escaped attention (rotation B), wherein the Cu-C2 bond is retained
until the C-Br bond cleaving TS (TS3-5). Here the Br atom
eliminates by itself without interaction with the Cu atom, which
may look rather surprising but is quite good since the leaving Br
anion keeps strong interaction with the Li atom. The tetracoordi-
nation that still exists in TS3-5 becomes disrupted as the Cu-C1
bond becomes fully formed after the TS (Supporting Information),
and the resulting tricoordinated CuIII complex undergoes spontane-
ous reductive elimination of the vinyl and the Me1 groups without
giving a discrete CuIII intermediate.13 The activation energies of
paths A and B are (∆Eq) 21.5 and 18.0 kcal/mol, respectively, and
comparable to that for the SN2 alkylation reaction of methyl bromide
with 1.14 These values decrease by ca. 4 and 1 kcal/mol by inclusion
of solvent coordination (S ) Me2O) to Li atoms and by consid-
eration of solvent polarity (PCM method, ꢀ ) 4.335 for Et2O),
respectively (see Supporting Information). With inclusion of solvent
effects, path B is still favored by 3-4 kcal/mol over path A (data
in Table S1, Supporting Information).
In the π-complex 3 that may be viewed as an ate complex of a
cuprio(III)cyclopropane,8 the two methyl groups and the vinyl
carbon atoms serve as four anionic ligands in this d8 square-planar
complex. Conversion of 3 to the conventional three-centered TS
(TS3-4 then to 4) keeps the Cu-C1 bond while breaking the Cu-
C2 bond. This process can be viewed as rotation of the Me1CuMe2
moiety along Cu-C1 bond as indicated as rotation A. This motion
creates a vacant site on the metal and forms the Cu-Br bond. The
Structure and orbital analysis shows the similarity and the
difference between the two TSs of the C-Br bond cleavage, TS3-4
and TS3-5 (Figure 1). In both TSs, the Me1-Cu-Me2 moiety is
bent (117.0° and 104.0°), and the C1 atom becomes nearly sp3
hybridized (C2 becomes closer to sp2). The Me1-Cu-Me2 bending
pushes up the energy level of the Cu 3dxz orbital,15 and the
deformation of the C1 center lowers the LUMO level of vinyl
bromide through mixing of the C1dC2 π* and the C1-Br σ* orbitals
9
12264
J. AM. CHEM. SOC. 2004, 126, 12264-12265
10.1021/ja046616w CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
which agree very well with the experimental data in columns 4
and 5. For 1-bromocyclooctene, the conventional three-centered TS
was higher in energy and could not be located as a stationary point.
As one can surmise from TSoct in Figure 1, the counterclockwise
rotation of the Me2Cu moiety in path A suffers from steric repulsion
between the ring structure and the Me1 group.
In summary, we have proposed a new mechanism of “oxidative”
addition between a cuprate and an alkenyl halide on the basis of
theory and experiments. It is first necessary to note that the initially
formed π-complex is not a simple CuI/alkenyl halide complex but
behaves as a cuprio(III)cyclopropane,8 where charge transfer is
taking place from the 3d orbital of the bent Me2Cu moiety to the
π*/σ*-mixed orbital of the deformed alkenyl halide. The subsequent
C-Br bond cleavage in this complex may go through the “three-
centered” or the “eliminative” way, of which the latter is preferred.
The overall mechanistic framework and the cooperation of various
components in the curprate cluster strongly suggest kinship between
this and the conjugate addition reaction (and the carbocupration
reaction as well).18 We expect that the present mechanistic
framework applies not only to the stoichiometric cuprate(I) reaction
but also to reactions involving d10/d8 transition metal catalytic
cycles3 and that useful new designs of reactions will result.
Figure 1. Structures of C-Br bond cleavage TSs (TS3-4, TS3-5, and TSoct
)
with schematic representations of orbital interaction based on the analysis
of fragment orbitals (Supporting Information). The red arrow indicates back-
donative interaction. The numbers refer to bond length (Å), bond angles
(italic), and natural charges (bold and underlined).
Table 1. Calculated and Experimental 12C/13C KIE Values for the
Reaction between Me2CuLi and Vinyl Bromide/
1-Bromocyclooctene
Acknowledgment. We thank the Ministry of Education, Culture,
Sports, Science and Technology of Japan for financial support, a
Grant-in-Aid for Specially Promoted Research and for the 21st
Century COE Program for Frontiers in Fundamental Chemistry.
calcd (A)a
calcd (B)a
calcd (B)b
run 1c
run 2c
C1
C2
C3
C4
C5
C6
C7
C8
1.039
1.015
1.026
1.018
1.024
1.016
1.003
1.000
1.000
1.000
1.001
1.007
1.023(3)
1.015(2)
0.999(1)
1.003(2)
1.000
0.997(2)
0.997(2)
1.004(1)
1.020(4)
1.017(3)
1.000(1)
1.001(1)
1.000
0.999(2)
0.998(2)
1.006(2)
Supporting Information Available: Details of computation and
experiments (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Corey, E. J.; Posner, G. H. J. Am. Chem. Soc. 1967, 89, 3911.
(2) Ando, K.; Kitamura, M.; Miura, K.; Narasaka, K. Org. Lett. 2004, 6, 2461.
(3) (a) Tamura, M.; Kochi, J. K. J. Am. Chem. Soc. 1971, 93, 1487. (b) Tamao,
K.; Sumitani, K.; Kumada, M. J. Am. Chem. Soc. 1972, 94, 4374. (c)
Corriu, R. J. P.; Masse, J. P. Chem. Commun. 1972, 144. (d) Yamamura,
M.; Moritani, I.; Murahashi, S. J. Organomet. Chem. 1975, 91, C39. (e)
Commercon, A.; Normant, J. F.; Villieras, J. J. Organomet. Chem. 1977,
128, 1.
a KIEs calculated for vinyl bromide. b KIEs calculated for 1-bromocy-
clooctene. c Experiments 1 and 2 are reactions carried out to 80.4 and 72.6%
completion, respectively. The C5 atom was taken as an “internal standard”.
Standard deviations in the last digit are shown in parentheses.
(Supporting Information). Thus, essentially the same frontier orbital
interactions (HOMO: Cu 3dxz, LUMO: CdC π*/C-Br σ*) give
TS3-4 and TS3-5. The copper atom is in the T-shaped coordination
geometry indigenous to CuIII oxidation state.13 The T-geometry,
however, is oriented in an opposite way as to the vinyl bromide
moiety and so is the vacant orbital of the CuIII center (Figure 1):
In TS3-4, the vacant site can readily interact with the Br atom,
while in TS3-5, it can keep the interaction with the C2 atom as
long as it is possible. As one can immediately notice the similarity
between TS3-5 and â-elimination reaction (Figure 1 inset; or also
R-elimination) of the metallacyclopropane, we may call the TS3-5
(4) Metal-Catalyzed Cross-Coupling Reactions; Diederich, F., Stang, P. J.,
Eds.; Wiley-VCH: New York, 1998.
(5) Bruckner, R. AdVanced Organic Chemistry: Reaction Mechanisms;
Academic Press: New York, 2002; Chapter 13.
(6) A Cu-Br exchange/SN2 alkylation sequence can be an alternative
mechanism (ref 7). This possibility was discarded by the following
experiment: the reaction between Me2CuLi and trans-â-bromostyrene in
the presence of methyl iodide-d3 resulted in no CD3 group incorporation
into the product (see Supporting Information).
(7) Whitesides, G. M.; Fischer, W. F., Jr.; San Fillipo, J., Jr.; Bashe, R. W.;
House, H. O. J. Am. Chem. Soc. 1969, 91, 4871.
(8) Nakamura, E.; Mori, S. Angew. Chem., Int. Ed. 2000, 39, 3750.
(9) For a recent experimental study, see: Xie, X. L.; Auel, C.; Henze, W.;
Gschwind, R. M. J. Am. Chem. Soc. 2003, 125, 1595.
(10) All calculations were carried out by the B3LYP hybrid density functional
method. For details, see Supporting Information.
“eliminative TS” as opposed to the “three-centered” TS3-4
.
Taking place via either path A or B, the rate-limiting step of the
reaction is the C-Br bond cleavage, and therefore, we should be
able to probe the reaction experimentally through kinetic isotope
effect (KIE). The 12C/13C KIE values were calculated for TS3-4
and TS3-5 by taking 1 and 2 as starting materials. As shown in the
first and the second columns of Table 1, the KIE values for C1
(1.039 and 1.026 for paths A and B, respectively) are significantly
different from each other, reflecting the mechanistic difference and
indicating that the KIE serves as a measure to probe the mechanism.
Therefore, we compared experimental measurement (based on
quantitative 13C NMR measurement)16 and theoretical prediction
of KIE for the reaction between Me2CuLi and 1-bromocy-
clooctene.17 The calculated KIEs for path B (see TSoct in Figure 1)
in column 3 show significant KIEs on C1 (1.024) and C2 (1.016),
(11) For example: (a) Nakamura, E.; Mori, S.; Morokuma, K. J. Am. Chem.
Soc. 1997, 119, 4900. (b) Vellekoop, A. S.; Smith, R. A. J. Am. Chem.
Soc. 1994, 116, 2902.
(12) The product 5 may undergo reorganization to give polymeric MeCu and
Me2CuLi‚LiBr.
(13) (a) Dorigo, A. E.; Wanner, J.; Schleyer, P. v. R. Angew. Chem., Int. Ed.
1995, 34, 476. (b) Snyder, J. P. J. Am. Chem. Soc. 1995, 117, 11025. (c)
Nakamura, E.; Yamanaka, M.; Mori, S. J. Am. Chem. Soc. 2000, 122,
1826.
(14) Mori, S.; Nakamura, E.; Morokuma, K. J. Am. Chem. Soc. 2000, 122,
7294.
(15) Mori, S.; Nakamura, E. Tetrahedron Lett. 1999, 40, 5319.
(16) (a) Singleton, D. A.; Thomas, A. A. J. Am. Chem. Soc. 1995, 117, 9357.
(b) Frantz, D. E.; Singleton, D. A.; Snyder, J. P. J. Am. Chem. Soc. 1997,
119, 3383.
(17) 1-Bromocyclooctene was chosen as a substrate because of its low volatility
and the lack of stereochemical issues.
(18) Mori, S.; Nakamura, E.; Morokuma, K. Organometallics 2004, 23, 1081.
JA046616W
9
J. AM. CHEM. SOC. VOL. 126, NO. 39, 2004 12265