Mechanism of Enolate Transfer between Si and Cu

Article abstract of DOI:10.1002/chem.201800099

Exchange of X (F, Cl, OMe) and a substituted enolate chain between SiMe₃ and various Cu^I complexes was examined. Reaction mechanisms pass through a cyclic transition state in which the reaction coordinate is associated with rotation of the SiMe₃ moiety. The dependence of the thermodynamic and kinetic features on the nature of the active and ancillary ligands was examined. Formation of copper enolate is shown to be favored when stabilized enolates are used. Replacement of F by Cl reverses the preference of the reaction. This is associated with the small difference between the Cu?Cl and Si?Cl bond energies, in contrast to other Si?X bonds, which are systematically stronger than their Cu?X analogues.

Full text of DOI:10.1002/chem.201800099

A Journal of
Accepted Article
Title: Mechanism for enolate transfer between Si and Cu
Authors: Samira BOUAOULI, Kim SPIELMANN, Emmanuel
VRANCKEN, Jean-Marc CAMPAGNE, and Hélène Gérard
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201800099
Link to VoR: http://dx.doi.org/10.1002/chem.201800099
Supported by

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
Mechanism for enolate transfer between Si and Cu
S. Bouaouli,[a] K. Spielmann,[b] E. Vrancken,[b] J.-M Campagne,[b] and H. Gérard*[a]
Abstract: Exchange of X (F, Cl and OMe) and a substituted enolate chain between SiMe3 and various Cu(I) complexes has been examined.
Reaction mechanisms go through a cyclic transition state in which the reaction coordinate is associated to the rotation of the SiMe3 moiety.
The dependence of the thermodynamic and kinetic features to the nature of the active and ancillary ligand was examined. Formation of
copper enolate is shown to be favoured when stabilized enolates are used. Replacement of F by Cl reverses the preference of the reaction.
This was associated to the small difference between the Cu-Cl and Si-Cl bond energies, in contrast to other Si-X bonds which are
systematically stronger than their Cu-X analogues.
Introduction
Silylated enolate are used as latent nucleophiles in many
reactions, their activation being induced through Lewis acids
or bases.[1] In the case of basic activation, Si-O bond breaking
leading to transfer of the enolate toward a copper (or more
generally to a metal) complex is proposed.[2–20] This can be
formally described as a chain transfer between Si to Cu,
enolate exchanging with an X group whose nature depends
on the exact reaction at stake. For instance, in the case of the
copper catalysed asymmetric vinylogous Mukaiyama-aldol
(Cu-CAVM) reaction,[21] which has been at the origin of our
study, a dienolate transfer from a trimethylsilyl (TMS) moiety
toward a Cu(I) complex is proposed in order to form a Cu-
Scheme 1. Proposed mechanism for Cu-CAVM evidencing two Cu-Si chain
transfers in the initiation and regeneration steps.
bonded nucleophile (Scheme 1). This can be carried out
either with X = F (initiation step) or with X = alcoholate
(regeneration step), as this
X is the product of the
condensation. Beyond this peculiar reaction which has
attracted our attention,[22],[23] Cu(I)-Si chain transfers are
extensively encountered in organic synthesis. The Cu-F/Si-
OMe exchange has been observed by NMR and is proposed
in the initiation steps of Cu-catalyzed allylation.[24] In the
opposite, the Cu-enolate / Si-Cl exchange has been used to
trap and characterized the stereochemical outcome in
cuprate chemistry.[25]
Transfers of OR chains between Cu and SiMe3 thus exhibit a
versatility which allowed using trimethylsilyl as a protecting
group or a trapping agent. Understanding of this reversibility
nevertheless remains quite poor considering the extensive
use made for these reactions. Rationalization through Si-X
bond energies is often proposed, especially in association to
the significantly larger strength of the Si-F bond (565 kJ/mol)
when compared to the Si-O (452 kJ/mol) or Si-Cl (371
kJ/mol).[26] Such a rationale should only holds for exchange of
free anions. For exchange between Si and Cu, the bond
energies to Cu should also be taken into account,[27],[28],[29]
and additional effects should be encountered.
We thus undertook
a combined computational and
experimental study of the exchange reaction between SiMe3
and a Cu(I) complex (Scheme 2, X = OMe, F and Cl; OR = OMe
and various conjugated enolates). The mechanism for the
ligand transfer is first described in the case of the identity
reaction,[30] that is for exchange of two identical ligands (X =
OR = OMe) between Cu and Si. The effects of the exchanged
ligand on the various features of the mechanism are then
computationally quantified and the trends obtained are
confirmed experimentally. Finally, the role of the ancillary
ligand at copper, and thus of the exact nature of the [Cu]
center, is discussed.
[a]
[b]
Sorbonne Université, CNRS, Laboratoire de Chimie Théorique,
UMR 7616, F-75005, Paris, France
E-mail: helene.gerard@upmc.fr
Institut Charles Gerhardt UMR 5253 CNRS-UM2-UM1-ENSCM 8,
rue de l’Ecole Normale, 34296 Montpellier Cedex 5, France
Scheme 2. Exchange reaction of X and OR between Si(Me)3 and a coper
complex [Cu] (the exact nature of [Cu] is discussed in the text)
Supporting information for this article is given via a link at the end of the
document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
Results
Study of the identity reaction
In order to streamline each effect separately, we consider at
first an “identity” reaction, which is the exchange of two
identical ligands (OMe) and thus a symmetric system. In this
case, the copper center is taken as Cu(PH3)2. A concerted
cyclic pathway without assistance of the reaction medium is
proposed, which can be paralleled to the proposal of the
literature for the transmetallation steps in cross-coupling
mechanisms.[31]
It starts (Figure 1) with the pre-coordination of the silylated
compound to the copper center through the O atom
(formation of AR). The TS for this step was not searched, as
such an associative mechanism is expected to be strongly
driven by entropic effects which are only poorly reproduced
within a static approach. The second step is a cyclization
through the formation of a bond between the methoxy
ligand at Cu and the Si center. The obtained intermediate,
named RI, exhibits a five-coordinated Si center[32] and a four-
membered Cu-O-Si-O ring. Two short and two long metal-O
bonds are obtained, thus keeping the track of the initial
metal the OMe group is bonded to. An alternative reaction
path from separated reactant to RI was found which was
going through precoordination of the OMe of the copper to
the silyl moiety (thus increasing the coordination number at
Si prior to increasing that at copper). But the associated TS
was found more than 25 kcal mol-1 above TS’ so that this path
was discarded. Continuing the path in Figure 1, the long and
short Si-O (or Cu-O) bonds are exchanged in the next step, so
that the two Cu-O (resp. Si-O) are identical and equal to 2.04
Å (resp. 1.91 Å) in the corresponding transition state, named
TS. The most evident feature of the reaction coordinate along
this step is nevertheless not the bond lengthening /
shortening but it is the substituent permutation of the SiMe3
moiety. It corresponds to the exchange of the OMe group
between apical and equatorial positions at the Si center. The
Si center thus goes from a trigonal bipyramid in RI to a square
based pyramid in TS, undergoing a global rotation of the
SiMe3 moiety.[33] A similar feature has been reported for
“front-side SN2” on SiH3Cl.[34] The end of the mechanism is
then the mirror image of the first two steps (as the
exchanged ligands are identical): the longest Si-O bond is
broken and then dissociation occurs.
Figure 1. The localized reaction pathway for the exchange of OMe and OMe
ligands. The energies
E (in black) and the Gibbs free energies G (in
parenthesis and in blue) are calculated with respect to the energies of
separate reactant species. Distances are given in Å. Color code : H : white ;
C : black ; O : red ; Si : blue : P : violet ; Cu : orange).
As a conclusion, the activation barrier has to be evaluated
between the separated reactants and TS. For all the
calculations that follows, we will thus focus on the
localization of TS with respect to the separated species, the
adducts being given for the sake of completeness.
Modulation of OR ligands
We next examine the exchange between two different
ligands (Scheme 2), using Cu(PH3)2 all along this part as a
model for the copper center [Cu]. The X ligand is kept equal
to OMe. For the OR one, a variety of enolates is examined
(Table 1, O-CH=CH2 (entry 2), O-C(OMe)=CH2 (entry 3) and
O-C(OMe)=CH-CH=CH2 (entry 4)). For the latest, we are
modelling the regeneration step of the Cu-CAVM (Scheme 1).
The thermodynamic data of the reactions are summarized in
Table 1. Gibbs free energies have been evaluated for all
extrema located with respect to the sum of the Gibbs free
energies of the separated reactants
separated products).
(ΔGprod refers to
This leads to an overall stabilization of the reaction path with
increasing conjugation, as G values decrease along the
series G(OMe) > G(O-CH=CH2) > G(O-C(OMe)=CH2) >
G(O-C(OMe)=CH-CH=CH2). Nevertheless, this stabilization is
much more important for the separated products (as
Δ
This reaction is athermal as it exchanges identical ligands.
The structures of TS' and RI have close energies with
differences of about 0.7 kcal/mol so that TS' is slightly below
RI when Gibbs free energies are considered. This step is thus
better described as an inflection point on the potential
energy surface rather than a true minimum and the
localization of RI and TS’ does not affect the conclusion on
the mechanism. In addition, considering the entropic cost for
formation of the adduct, this structure is found to be higher
in Gibbs free energy than the separated reactant.
Δ
Δ
Δ
Δ
evidenced by the ΔGprod values, which varies from 0.0 to -5.6,
-9.9 and -14.2 kcal mol-1, entries 2, 3 and 4 respectively, thus
a maximum stabilization by 14.2 kcal mol-1) than for the
transition state (ΔGTS ranging from 23.6 to 22.6, 19.6 and
18.1 kcal mol-1 respectively, thus a maximum stabilization by
5.5 kcal mol-1).
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
Table 1. Thermodynamic data (Gibbs free energies in kcal mol-1 with respect
to separated X-Cu(PH3)2 and RO-SiMe3) as a function of OR and X for the
reaction depicted in Scheme 2.
OR
OMe
O-CH=CH2
X
ΔGAR
12.2
11.7
11.6
9.5
9.1
10.9
9.6
ΔGTS
23.6
22.6
19.6
18.1
18.5
39.7
36.5
ΔGAP
12.2
4.5
-0.4
-2.6
7.2
ΔGprod
0
1
2
3
4
5
6
7
OMe
OMe
OMe
OMe
F
-5.6
-9.9
-14.2
-0.5
23.4
9.2
O-COMe=CH2
O-COMe=CH-CH=CH2
OMe
OMe
Cl
Cl
33.9
20.2
O-COMe=CH-CH=CH2
Scheme 3. Competition reaction between enolates with benzaldehyde in the
presence of 1 equiv. of CuCl, MeONa and BINAP (For further experimental
details see the supporting information).
Consequently, the conjugation in OR chains has a strong
stabilizing role for products, and to a lesser extend for the TS.
The regeneration step of the Mukaiyama reaction is thus
reversible for shorter chains, whereas it becomes highly
favoured thermodynamically and thus irreversible for longer
and more conjugate chains
Evaluation of their relative kinetics will thus be carried out
using a competitive reactivity protocol: TMS-OC(OEt)=CH2
and TMS-OC(OEt)=CH-CH=CH2 are added with benzaldehyde
in 1:1:1 ratio. In these condition, the majority (B + C = 27+ 46
= 73%) of the reaction products (77% conversion) comes
from addition of the most conjugate species.
Whereas experimental results can be found for the reactivity
of a variety of enolates,[35] no direct comparison to our
computational data could be made. We thus decided to
confirm these results experimentally through Si-OR/Cu-OMe
chain transfer followed by in situ trapping by benzaldehyde,
which is considered as fast compared to the chain transfer
step (Scheme 3). As a consequence, this experimental
procedure allows evaluating the kinetic of the chain transfer
Considering the reaction condition (kinetic control on the
chain transfer step), the ratio between
A (product of the
transfer of TMS-OC(OEt)=CH2) and (B + C) should be equal to:
G (A)  G (B /C)
 
A



TS
TS
 exp 
1


B



C

RT


step, and thus probing
ΔGTS. First, a 1:1 TMS-OR :
Cu(OMe)(BINAP) mixture is considered to be formed through
stoichiometric addition of CuCl, sodium methanolate and the
BINAP (bis(diphenylphosphine)binaphtyle) ligand, used to
promote the double coordination of the phosphine to
copper and keep as close as possible to the computational
bisphosphine model. Transfer of the O-C(OEt)=CH2 moiety
takes place with 85% conversion in 16h at room temperature.
This is fully in line with the 19.6 kcal mol-1 value found for
Thus the larger amount of
for GTS(A) compared to
A/(B+C) experimental ratio (27/73) is associated with a
Δ(ΔGTS) value of 0.3 kcal mol-1, in line with the small value
obtained computationally.
B
Δ
+
C
is in line with a larger value
Δ
GTS(B/C). Let us note that the
Modulation of X ligands
We next examine the role of X by replacing the OMe
group by a halogen, namely F or Cl. Highly different behaviors
are obtained for the two halogens, as illustrated by the Gibbs
free energies reported in Table 1. When X = F or OMe, similar
mechanisms and G values are obtained (compare entries 1
and 5, Table 1). The energy differences between the two
systems ranging from 3 kcal/mol (for AR) to 5 kcal/mol (for
AP). A slight stabilization is thus obtained for the whole
reaction path for fluorine compared to methoxy but, to a first
approach, OMe and F can be proposed to behave similarly.
This conclusion, within the perspective of the Cu-CAVM,
allows proposing that initiation and regeneration steps of
catalytic cycle have similar properties.
Δ
GTS in Table 1 (entry 3). In a second experience carried out
with TMS-OC(OEt)=CH-CH=CH2, in line with Cu-CAVM
published results, cyclic ( ) and linear ( ) products are
obtained, both resulting of the trapping of the copper
enolate by benzaldehyde. Both and come from terminal
B
C
B
C
addition of the dienolate to benzaldehyde, leading to Z and E
alcoholates. The former arrangement allows an additional
lactonisation step resulting in
differenciation between and
B
. Considering that the
B
C occurs at or after the
trapping by benzaldehyde, the evaluation of the chain
transfer step can be obtained from the addition of the and
contents. A global 80% conversion into and within the
B
C
B
C
time and temperature of experience is obtained, showing a
similar advancement of the reaction in the first and second
experiments. This reaction time can thus be considered as an
“end of reaction” in both cases.
A totally different picture, reported in Figure 2 in the case of
OR = OMe, is obtained for X = Cl as a significantly different
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
Scheme 4. Reaction of silylated dienolate with benzaldehyde in the presence
of 1 equiv. of CuCl and BINAP (For further experimental details see the
supporting information).
We next wondered what happens when the most stabilizing
conjugate chain is associated to the destabilizing Cl (Table 1,
entry 7). When looking only at the separated species, the
effects of the chain and of Cl are purely additive, as no
species contained both X and OR at the same time. As a
consequence, Cl effects the same destabilization (by nearly
24 kcal mol-1) for O-COMe=CH-CH=CH2 than for OMe. The
reaction is thus endergonic by 9.21 kcal mol-1 and is thus not
possible. This was confirmed experimentally as no conversion
is observed in absence of NaOMe using the same conditions
as already used (Scheme 4).
On the opposite, nearly no effect of the nature of OR is
observed on the TS, which remains more than 35 kcal mol-1
above the reactant, and thus more than 25 kcal mol-1 above
the products. As a consequence, protection of a conjugated
enolate chain by SiMe3 through reaction with TMSCl, despite
favorable, is most probably difficult as a 27.3 kcal mol-1
activation Gibbs free energy is obtained.
Figure 2. Reaction pathway for the exchange of Cl and OMe between Si(Me)3
and Cu(PH3)2. The energies E (in black) and the enthalpies of Gibbs free
energies G (in parenthesis and in blue) are calculated with respect to the
energies of separate reactant species. Distances are given in Å. Color code :
H : white ; C : black ; O : red ; Si : blue : P : violet ; Cu : orange).
reaction path was obtained. In addition, the overall
transformation is highly endergonic for X = Cl (ΔGprod = +23.4
kcal mol-1, Table 1, entry 6), instead of neutral for X = F or
OMe.
After formation of the adduct through a Cu…O bond, the
cyclic structure exhibiting a pentacoordinated Si center
cannot be formed for the OMe and Cl pair. In fact, no
pentacoordinated Si is thus obtained in this mechanism,
neither with Cl in axial (reactant side) nor in equatorial
(product side) position. This observation is consistent with a
disfavored reaction involving hypervalent Si for X = I reported
in the literature.[32] First row X groups (OMe and F) seems to
favor the formation of hypervalent Si intermediates, in
contrast to Cl. Moreover, the nature of the equatorial
substituents has been shown to modulate the stability of the
pentacoordinated Si structures from stable transition
complex to transition state[36], as well as solvation,[34] or the
apical substituents in the case of pentacoordinated P.[37]
Both coordination of the chlorine atom, rotation of the SiMe3
unit and breaking of the Si-O bond are already completed
before reaching the TS. The high endergonicity of the
reaction is associated to a late TS, much higher in energy
than the one obtained before. As a conclusion, the opposite
reaction, that is transfer from Si to Cu resulting in the
formation of a silylated enolate is highly favorable and
exhibit a small activation. This corresponds to the direction of
electrophilic quench or protection of the alcoholate chain.
To conclude this first part dealing with the effect of the
transferred ligands, it appears that the reaction
thermodynamic is driven by both the nature of X and OR,
whereas the kinetic of the reaction is mostly governed by the
nature of X, with a smaller influence of OR (Scheme 2).
Modulation of the [Cu] center
In the previous paragraph, a neutral electron rich copper
complex was used as it was coordinated to two phosphines.
In this part we aim at examining the role played by the
nature, charge and number of ligands in the Cu/Si
transmetallation. The Gibbs free energy values are reported
in Table 2 and energetic and geometrical data in ESI.
First, changing from two to one phosphine gives rise to a
slight destabilization of TS (+2.2 kcal/mol in TS when going
from bis to mono-phosphine compound) but to no significant
effect on the products stability. In contrast, using a “naked”
copper (no ligand) leads to a major evolution of the reaction
profile. The adducts become very stable (-32.5 kcal/mol for
AR and -21.8 kcal/mol for AP) whereas the energy of the TS
with respect to separated reactant remains unchanged. This
can be easily understood as, the passage from two to zero
phosphine leads to an electron poor cooper which will tend
to interact strongly with any potential ligand.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
As a conclusion, effects of the number and nature of the
ligand on the reaction energy or kinetic could be observed
but remain quite small, except for the stability of the adduct.
In absence of procedure allowing comparison of the
reactivity of the various Cu center in similar conditions, it is
not possible to confirm these tendencies experimentally.
Scheme 5. Reaction of silylated dienolate with benzaldehyde in the presence
of 1 equiv. of CuCl and NaOMe (For further experimental details see the
supporting information).
Table 2. Gibbs free energies (in kcal mol-1 with respect to separated F-[Cu]
and MeO-SiMe3) as a function of the nature of [Cu] for the reaction depicted in
Scheme 2 in the case X = F and OR = OMe.
Discussion
We have thus shown that the thermodynamic of the reaction
of Scheme 2 is strongly influenced by the nature of X, OR and,
to a lesser extent, by the copper ancillary ligands. Indeed, as
shown above, the reaction can range from strongly
exothermic (X = OMe, OR = O-C(OMe)=CH-CH=CH2) to
athermic (X= F, OR = OMe) and even strongly endothermic (X
= Cl, OR = OMe). With these results in hand, we decided to
revisit the commonly used interpretative schemes in
silylation chemistry, which resort only on the Si-Y bond
strength, and try to adapt them in case of coordination of Y
to copper complexes.
The thermodynamic of the reaction was found to be
determined by the Gibbs free energy difference between
separated reactants and products, which is very close to the
energy difference as the reaction does not modify the
number of free molecules between reactant and products.
We also remind that the [Cu]-F values reported below are
systematically overestimated (see Computational Details) so
that only trends with the bond energy differences can be
considered. The reaction results in breaking of Cu-X and Si-
OR bonds, and formation of Si-X and Cu-OR bonds.
Quantitative values for various transferred ligands are
gathered in Table 3 for [Cu] = Cu(PH3)2. Si-Y bonds are
stronger than the [Cu]-Y analogues by values close to 20 kcal
mol-1, except for Cl. In this case, bondings to Cu or Si centers
are similar. This enables understanding the specific behaviour
of Cl when considering the reaction to Scheme 2 for OR =
OMe. A Cu-Cl bond is broken and replaced by a Si-Cl bond of
similar energy. In contrast, the formed Cu-O bond is much
weaker than the initial Si-O one, so that the reaction is thus
endothermic. In contrast, for F, the formed Si-F is stronger
than the broken Cu-F by a value similar to the difference
between the formed Cu-O and the broken Si-O: these two
effects thus compensate and the reaction is quasi athermic.
Let us now examine the effect of the OR group: when
increasing the length of the chains and thus their
conjugation, their bond energies with Cu and Si decrease.
[Cu]
Cu(PH3)2
Cu(PH3)
Cu
Cu(OMe2)
Cu(CO)
ΔGAR
9.1
8.1
-21.5
8.0
3.7
ΔGTS
18.5
20.7
18.0
22.8
18.2
ΔGAP
7.1
6.7
-12.8
5.7
4.9
ΔGprod
-0.5
-0.7
-6.8
-1.5
-3.9
1
2
3
4
5
In this configuration, the reaction seems to be impossible as
the adducts become the most stable point on the reaction
path: the coordination to copper in this species is so strong
that it is impossible to exchange OMe and F.
This computational result was confronted to experiment
following a similar procedure to that described above, but in
absence of BINAP (Scheme 5). The product obtained results
from Cu-CAVM reaction on the α position of the copper-
enolate intermediate, and corresponds to
a different
regiochemistry for quenching but a similar Si to Cu chain
transfer. Let us mention that formation of the branched
product is consistent with previous studies[38] that have
shown that the regiochemistry is strongly ligand dependent.
A significantly lower conversion (comparing Scheme 3 and 5)
is also observed. To put this experimental result in
perspective with our computational one, it is necessary to
discuss the limit of our computational model. Considering an
uncoordinated copper when working in THF is an unrealistic
model as, in ethereal solvent, the coordination of at least one
solvent molecule on the copper center (explicit solvation)
should be taken into account. A model using OMe2 as an
ethereal solvent was thus used. It yields a pathway similar to
that obtained for the system with two phosphine ligands,
with a TS about 4 kcal mol-1 higher in energy. The lower
conversion, when compared to the BINAP containing
experiments, is consistent with a slower but still possible
chain transfer. We have not been able to confirm this
analysis experimentally. Indeed, when trying to carry out the
reaction in non-coordinating solvent (toluene), the low
solubility of the reactants led to no reaction both in presence
and absence of BINAP.
This can be understood when implying the
character of [M]: the more conjugate the lone pair at O is,
the less available to -donate toward [M], and the weakest is
-acceptor
Finally, we computationally examined the replacement of PH3
by a -acceptor ligand, namely CO. The same mechanism is

obtained with energies close to those of the system with two
phosphines are obtained. A slight stabilization, of AR, AP and
products (by less than 2 kcal/mol) is obtained.
the [M]-O bond. The smaller decrease of the [M]-O bond
energy in the case of Cu can be associated to a lesser π-
acceptor character of the copper compared to SiMe3.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
set was used for all the other atoms (Si, F, Cl, O, C, P, H). The validity of
this computational level, in absence of dispersion correction, was
confirmed through energetic and geometric comparison to MP2 (See
Supporting Information). As a conclusion, no dispersion correction was
added. Full geometry optimizations were carried out followed by
analytical frequency calculations within the harmonic approximation to
confirm the nature of the stationary points obtained. The transition states
were given a small displacement along the imaginary frequency and full
geometry optimization was then carried out to ensure the connection
between transition states and the intermediates. Thermal corrections to
obtain the Gibbs free energies were calculated at 298 K and 1 atm using
unscaled frequencies computed in the harmonic approximation. Simple
models of copper complexes have been used which can be viewed as
prototype of various classes of copper or cuprate complexes, in order to
search for general features of the mechanism. In particular, two PH3
ligands have been used to model the tol-Binap ancillary ligands used in
the Mukaiyama reaction. Considering the simplicity of these models, no
implicit solvation was added. Explicit solvation has been taken into
account when needed by including OMe2 molecule as a model of
ethereal solvents. The bond energies E([M]-Y), where [M] is a metal
center and Y a transferred group, are computed using the radical
electronic structure of [M]. and Y., taken in the geometry of the optimized
[M]-Y molecule, as the energy difference between the separated
fragment and the complex, in order to get positive values of E. The value
obtained for Si-O is very close to that reported in standard textbooks
(109.3 kcal mol-1 vs 108.1 kcal mol-1) whereas a significantly larger value
is obtained for Si-Cl and Si-F (155.8 and 112.0 kcal mol-1 respectively,
compared to 135.2 or 88.8 kcal mol-1 in ref [26]. This overestimation is
associated to the computational procedure, which resorts on a mono-
determinental approach to the evaluation of the energy of F and Cl. This
approach overestimates the energy of the atoms, and thus the
associated bond energy. However, the difference between F and Cl is
well reproduced (43.8 kcal mol-1 vs 46.4 kcal mol-1), and this
overestimation is independent of the metal center the atoms are bonded
to. As a consequence, our theoretical values can be used only to
compare the difference between bond energies with different metal but
identical Y.
Table 3. Bond energies (E([M]-Y) in kcal mol-1) for [M] = SiMe3 and Cu(PH3)2
as a function of the examined ligand computed according to the procedure
reported in Computational Details. ECu/Si = E(Y-Cu(PH3)2)- E(Y-SiMe3).
Y \ [M]
F
Cl
OMe
OCHCH2
SiMe3
155.8
112.9
109.3
95.3
Cu(PH3)2
128.7
108.9
82.9
72.7
65.0
 ECu/Si
-27.1
-4.0
-26.4
-22.6
-22.7
-21.1
OCOMe=CH2
OCOMe=CH-CH=CH2
87.7
77.4
56.3
Conclusions
Anionic ligands transfer between SiMe3 and Cu(I) complexes
with various substitution has been examined. Reaction
mechanisms are determined, and go through the formation
of a cyclic transition state in most cases. The reaction
coordinate is associated to the rotation of the SiMe3 moiety.
The thermodynamic and kinetic features dependence on the
nature of the transferred ligand was detailed, whereas no
major effect of the nature of the copper complex could be
evidenced. On one side, variously conjugated alcoholates
have been examined. The greater the conjugation of the
chain, the most thermodynamically favored the formation of
the Copper-enolate, since the Si-O bond is much more
sensitive to delocalization than the Cu-O one. No such effect
could be found on the kinetic of the reaction: the activation
energy is decreased with exothermicity in line with the
Hammond postulate. This effect is found to be moderate, as
shown by the relatively moderate selectivity observed in the
case of the enolate / dienolate competition. On the other
side, the exchanged (F, Cl or OMe) were shown to play a
major role as they define both the kinetic and the
thermodynamic of the reaction. F or OMe are found to favor
formation of copper-enolate and to increase the kinetic of
the reaction, whereas Cl favors the formation of the silyl-
enolate. As evidenced from the following description,
rationalization of the obtained values was systematically
carried out, in order to develop simple reasoning model to
the prediction of further reactivity. We now have in hand a
deeper understanding of the initiation and regeneration
steps of Cu-CAVM and we can now complete the reaction
mechanism by examining the C-C bond formation step of this
reaction, especially in order to better understand the regio-
and enantioselectivity. These results will be published in due
time.
Experimental Details. Unless otherwise specified (see paragraph
below), all commercial products and reagents were used as purchased,
without further purification. Reactions were carried out in round-bottom
flasks equipped with a magnetic stirring bar under argon atmosphere.
NMR spectra were recorded with a Bruker Ultra Shield 400 Plus. 1H
chemical shifts are reported in delta (δ) units in parts per million (ppm)
relative to the singlet at 7.26 ppm for d-chloroform (residual CHCl3). 13C
chemical shifts are reported in ppm relative to the central line of the triplet
at 77.0 ppm for d-chloroform. Splitting patterns are designated as s,
singlet; d, doublet; t, triplet; q, quartet; quint, quintet; m, multiplet; and br,
broad and combinations thereof. All coupling constants (J values) are
reported in Hertz (Hz). Data are reported as follows: chemical shift (δ in
ppm), multiplicity, coupling constants (Hz), integration and attribution.
THF was dried by distillation over sodium metal and benzophenone
under argon. Dichloromethane and diethylether were dried by distillation
over CaH2 under argon. CuCl and MeONa were purchased from Aldrich,
kept under argon atmosphere, and dried over P2O5 before use.
Enolate synthesis: N,N-diisopropylamine (DIPA) (6.2 mL, 44.2 mmol,
1.20 equiv) and freshly distilled THF (70 mL) were added to a flame-dried
flask at 0 °C. Then, n-BuLi (27.6 mL, 44.2 mmol, 1.20 equiv, 1.6 M in
Et2O) was added dropwise over a period of 30 min. The pale yellow
solution obtained was stirred for an additional 15 min and then cooled at
−78 °C, freshly distilled ethyl acetate was added (3.4 mL, 36.8 mmol,
1.00 equiv), and the color turned yellow. The resulting mixture was stirred
at the same temperature for another hour. Thus, freshly distilled
chlorotrimethylsilane (TMSCl) (5.5 mL, 58.9 mmol, 1.60 equiv) was
added dropwise over a period of 10 min. The milky solution obtained was
stirred for 30 min at −78 °C and then allowed to reach room temperature.
The suspension was filtered through oven dried anhydrous Na2SO4, and
the solvent was removed under reduced pressure. The crude product
obtained was diluted with pentane (150 mL), filtered off, and the solvent
Experimental Section
Computational Details. The calculations were carried out with the
Gaussian 09 package of programs[39] using the PW91 functional.[40]
Copper was represented with the effective core potential SDDall from the
Stuttgart group and the associated basis set.[41],[42] The 6-31++G** basis
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
was removed under vacuum; this step was performed twice. Purification
of the residue was performed by distillation (60 °C, 50 mbar), yielding a
colorless oil (2.3 g, 14.3 mmol, 39 %) corresponding at the desired
product which contains a trace amount of C–TMS. 1H NMR (400 MHz,
CDCl3): δ 3.73 (q, J = 7.1 Hz, 2H), 3.18 (d, J = 2.7 Hz, 1H), 3.04 (d, J =
2.6 Hz, 1 H), 1.28 (t, J = 7.1 Hz, 3 H), 0.21 (s, 9 H) ppm. 13C NMR (100
MHz, CDCl3): δ 160.9, 63.3, 60.1, 14.2, 0.0 (3C) ppm. These data are in
accordance with that previously described.[43]
stirred at room temperature for an additional 16 h, the mixture was
treated with HCl 1M (20 mL) and MeOH (10 mL), an extraction of the
aqueous phase was performed with Diethylether (50 mL). The organic
layer was washed with an ammonia solution (3x30 mL) and brine (1x30
mL), the solvent was removed under vacuum and the crude was
analyzed by 1H NMR (conversion: 80 %). Characteristic peaks:
Benzaldehyde: 1H NMR (400 MHz, CDCl3): δ 9.92 (s, 1 H) ppm. Product
B: 1H NMR (400 MHz, CDCl3): δ 6.15 (ddd, J = 9.6, 2.4 and 1.2 Hz, 1H),
Dienolate synthesis: N,N-Diisopropylamine (DIPA) (6.2 mL, 44.2 mmol,
1.20 equiv) and freshly distilled THF (70 mL) were added to a flame-dried
flask at 0 °C. Then, n-BuLi (27.6 mL, 44.2 mmol, 1.20 equiv, 1.6 M in
Et2O) was added dropwise over a period of 30 min. The pale yellow
solution obtained was stirred for an additional 15 min and then cooled at
−78 °C. Freshly distilled 1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-
pyrimidinone (DMPU) (6.0 mL, 44.2 mmol, 1.20 equiv) was added
dropwise to the reaction mixture, giving rise to a milky solution. After 30
min, freshly distilled ester was added (4.6 mL, 36.8 mmol, 1.00 equiv),
and the color turned yellow. The resulting mixture was stirred at the same
temperature for another hour. Thus, freshly distilled chlorotrimethylsilane
(TMSCl) (5.5 mL, 58.9 mmol, 1.60 equiv) was added dropwise over a
period of 10 min. The milky solution obtained was stirred for 30 min at
−78 °C and then allowed to reach room temperature. The formation of a
white suspension in an orange solution was observed. The suspension
was filtered through oven dried anhydrous Na2SO4, and the solvent was
removed under reduced pressure. The crude product obtained was
diluted with pentane (150 mL), filtered off, and the solvent was removed
under vacuum; this step was performed twice. The crude product so
obtained was diluted with pentane and washed with a saturated aqueous
solution of NaHCO3 and brine. The organic layer was dried over
anhydrous Na2SO4, and the solvent was removed under reduced
pressure. Purification of the residue was performed by distillation (100 °C,
0.7 mbar), yielding a colorless oil (5.7 g, 30.5 mmol, 83%) corresponding
to a mixture of the two inseparable Z/E isomers with a ratio of 8:2 in favor
of the Z isomer (1H NMR determination). Z isomer: 1H NMR (400 MHz,
CDCl3): δ 6.46−6.58 (m, J = 10.4 and 17.2 Hz, 1H), 4.80 (dd, J = 1.8 and
17.2 Hz, 1H), 4.58 (dd, J = 1.8 and 10.4 Hz, 1H), 4.43 (d, J = 10.4 Hz,
1H), 3.78 (q, J = 7.0 Hz, 2H), 1.28 (t, J = 7.0 Hz, 3H), 0.21 (s, 9H) ppm.
13C NMR (100 MHz, CDCl3): δ 157.6, 132.5, 106.4, 80.8, 63.3, 14.3, 0.3,
ppm. E isomer: 1H NMR (400 MHz, CDCl3) δ 6.46−6.58 (m, J = 10.4 and
17.2 Hz, 1H), 4.80 (dd, J = 1.8 and 17.2 Hz, 1H), 4.58 (dd, J = 1.8 and
10.4 Hz, 1H), 4.51 (d, J = 10.4 Hz, 1H), 3.89 (q, J = 7.0 Hz, 2H), 1.22 (t, J
= 7.0 Hz, 3H), 0.24 (s, 9H) ppm.13C NMR (100 MHz, CDCl3) δ 154.6,
132.1, 107.3, 88.4, 62.8, 14.8, −0.2 ppm. These data are in accordance
with that previously described.[23]
1
5.46 (dd, J = 11.0 and 4.9 Hz, 1H) ppm. Product C: H NMR (400 MHz,
CDCl3): δ 5.81 (dt, J = 15.7 and 1.6 Hz, 1H), 4.72 (dd, J = 7.7 and 5.2 Hz,
1H) ppm. These data are in accordance with that previously described.[38]
Competition between enolate and dienolate with benzaldehyde:
CuCl (74 mg, 0.74 mmol, 1.0 equiv.) and MeONa (40 mg, 0.74 mmol, 1.0
equiv.) were added in a flame dried round bottom flask, dissolved in
freshly distilled THF (60 mL), and stirred at 50 °C for 2 h. The light brown
suspension obtained was cooled at room temperature and cannulated in
a separately flame dried round bottom flask containing BINAP (463 mg,
0.74 mmol, 1.0 equiv.), 2x5 mL of freshly distilled THF were used to
rinsed the flask. After being stirred for 1 h at room temperature, freshly
distilled benzaldehyde (76 µL, 0.74 mmol, 1.0 equiv) was added, followed
by a freshly prepared mix of dienolate and enolate (1:1 (verified by 1H
NMR), 305 µL, 0.74 mmol of each, 1.0 equiv. of each). After being stirred
at room temperature for an additional 16 h, the mixture was treated with
HCl 1M (20 mL) and MeOH (10 mL), an extraction of the aqueous phase
was performed with Diethylether (50 mL). The organic layer was washed
with an ammonia solution (3x30 mL) and brine (1x30 mL), the solvent
was removed under vacuum and the crude was analyzed by 1H NMR
(conversion: 77 %). Characteristic peaks: Benzaldehyde: 1H NMR (400
MHz, CDCl3): δ 9.92 (s, 1 H) ppm. Product A: 1H NMR (400 MHz,
CDCl3): δ 5.02 (dd, J = 8.8 and 4.2 Hz, 1H) ppm. Product B: 1H NMR
(400 MHz, CDCl3): δ 6.15 (ddd, J = 9.6, 2.4 and 1.2 Hz, 1H), 5.46 (dd, J
= 11.0 and 4.9 Hz, 1H) ppm. Product C: 1H NMR (400 MHz, CDCl3): δ
5.81 (dt, J = 15.7 and 1.6 Hz, 1H), 4.72 (dd, J = 7.7 and 5.2 Hz, 1H) ppm.
These data are in accordance with that previously described.[38, 44]
Reaction with CuCl:
CuCl (74 mg, 0.74 mmol, 1.0 equiv.) and BINAP (463 mg, 0.74 mmol, 1.0
equiv.) were added in a flame dried round bottom flask, dissolved in
freshly distilled THF (60 mL). After being stirred for 2 h at room
temperature, freshly distilled benzaldehyde (76 µL, 0.74 mmol, 1.0 equiv)
was added, followed by freshly distilled dienolate (160 µL, 0.74 mmol, 1.0
equiv). After being stirred at room temperature for an additional 16 h, the
mixture was treated with HCl 1M (20 mL) and MeOH (10 mL), an
extraction of the aqueous phase was performed with Diethylether (50
mL). The organic layer was washed with an ammonia solution (3x30 mL)
and brine (1x30 mL), the solvent was removed under vacuum and the
crude was analyzed by 1H NMR (conversion: 0 %).
Reaction without ligand: CuCl (74 mg, 0.74 mmol, 1.0 equiv.) and
MeONa (40 mg, 0.74 mmol, 1.0 equiv.) were added in a flame dried
round bottom flask, dissolved in freshly distilled THF (60 mL). After being
stirred for 2 h at room temperature, freshly distilled benzaldehyde (76 µL,
0.74 mmol, 1.0 equiv) was added, followed by freshly distilled dienolate
(160 µL, 0.74 mmol, 1.0 equiv). After being stirred at room temperature
for an additional 16 h, the mixture was treated with HCl 1M (20 mL) and
MeOH (10 mL), an extraction of the aqueous phase was performed with
Diethylether (50 mL). The organic layer was washed with an ammonia
solution (3x30 mL) and brine (1x30 mL), the solvent was removed under
vacuum and the crude was analyzed by 1H NMR (conversion: 40 %).
Characteristic peaks: Benzaldehyde: 1H NMR (400 MHz, CDCl3): δ 9.92
(s, 1 H) ppm. Products D: Dia 1: 1H NMR (400 MHz, CDCl3): δ 5.90 (ddd,
J = 17.2, 10.3 and 8.5 Hz, 1H) ppm. Dia 2: 1H NMR (400 MHz, CDCl3): δ
5.62 (ddd, J = 17.2, 10.3 and 8.9 Hz, 1H) ppm. These data are in
accordance with that previously described.[38]
Reaction between enolate and benzaldehyde: CuCl (74 mg, 0.74
mmol, 1.0 equiv.) and MeONa (40 mg, 0.74 mmol, 1.0 equiv.) were
added in a flame dried round bottom flask, dissolved in freshly distilled
THF (60 mL), and stirred at 50 °C for 2 h. The light brown suspension
obtained was cooled at room temperature and cannulated in a separately
flame dried round bottom flask containing BINAP (463 mg, 0.74 mmol,
1.0 equiv.), 2x5 mL of freshly distilled THF were used to rinsed the flask.
After being stirred for
1 h at room temperature, freshly distilled
benzaldehyde (76 µL, 0.74 mmol, 1.0 equiv) was added, followed by
freshly distilled enolate (145 µL, 0.74 mmol, 1.0 equiv). After being stirred
at room temperature for an additional 16 h, the mixture was treated with
HCl 1M (20 mL) and MeOH (10 mL), an extraction of the aqueous phase
was performed with diethylether (50 mL). The organic layer was washed
with an ammonia solution (3x30 mL) and brine (1x30 mL), the solvent
was removed under vacuum and the crude was analyzed by 1H NMR
(conversion: 85 %). Characteristic peaks: Benzaldehyde: 1H NMR (400
MHz, CDCl3): δ 9.92 (s, 1 H) ppm. Product A: 1H NMR (400 MHz,
CDCl3): δ 5.02 (dd, J = 8.8 and 4.2 Hz, 1H) ppm. These data are in
accordance with that previously described.[44]
Reaction between dienolate and benzaldehyde: CuCl (74 mg, 0.74
mmol, 1.0 equiv.) and MeONa (40 mg, 0.74 mmol, 1.0 equiv.) were
added in a flame dried round bottom flask, dissolved in freshly distilled
THF (60 mL), and stirred at 50 °C for 2 h. The light brown suspension
obtained was cooled at room temperature and cannulated in a separately
flame dried round bottom flask containing BINAP (463 mg, 0.74 mmol,
1.0 equiv.), 2x5 mL of freshly distilled THF were used to rinsed the flask.
After being stirred for
1 h at room temperature, freshly distilled
benzaldehyde (76 µL, 0.74 mmol, 1.0 equiv) was added, followed by
freshly distilled dienolate (160 µL, 0.74 mmol, 1.0 equiv). After being
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
[21] G. Bluet, J.-M. Campagne, J. Org. Chem. 2001, 66, 4293–4298.
[22] G. Bluet, B. Bazán-Tejeda, J.-M. Campagne, Org. Lett. 2001, 3, 3807–
3810.
[23] X. Moreau, B. Bazán-Tejeda, J.-M. Campagne, J. Am. Chem. Soc.
2005, 127, 7288–7289.
Acknowledgements
[24] S. Yamasaki, K. Fujii, R. Wada, M. Kanai, M. Shibasaki, J. Am. Chem.
Soc. 2002, 124, 6536–6537.
This work was initiated by the Agence Nationale de la
Recherche (project COPENOL, n° ANR-09-BLAN-0089). The
authors wish to thank S. Halbert for numerous fruitful
discussions.
[25] Y. Minko, M. Pasco, L. Lercher, M. Botoshansky, I. Marek, Nature 2012,
490, 522–526.
[26] C. Elschenbroich, A. Salzer, Organometallic. A Concise Introduction,
Weinheim, 1989.
[27] E. Clot, M. Besora, F. Maseras, C. Mégret, O. Eisenstein, B. Oelckers,
R. N. Perutz, Chem. Commun. 2003, 2, 490–491.
[28] J. L. Vincent, S. Luo, B. L. Scott, R. Butcher, C. J. Unkefer, C. J. Burns,
G. J. Kubas, A. Lledós, F. Maseras, J. Tomàs, Organometallics 2003,
22, 5307–5323.
Keywords: copper • silyl ethers • mechanistic studies • DFT •
substituent effects
[1]
[2]
J. Matsuo, M. Murakami, Angew. Chemie Int. Ed. 2013, 52, 9109–9118.
R. Noyori, I. Nishida, J. Sakata, J. Am. Chem. Soc. 1981, 103, 2106–
2108.
[29] E. Clot, C. Mégret, O. Eisenstein, R. N. Perutz, J. Am. Chem. Soc.
2006, 128, 8350–8357.
[30] I. Union, O. F. Pure, A. Chemistry, Pure Appl. Chem. 1979, 51, 1725–
1801.
[3]
[4]
I. Kuwajima, H. Urabe, J. Am. Chem. Soc. 1982, 104, 6831–6833.
I. Kuwajima, E. Nakamura, M. Shimizu, J. Am. Chem. Soc. 1982, 104,
1025–1030.
[31] M. García-Melchor, A. A. C. Braga, A. Lledós, G. Ujaque, F. Maseras,
Acc. Chem. Res. 2013, 46, 2626–2634.
[5]
G. A. Kraus, B. Roth, K. Frazier, M. Shimagaki, J. Am. Chem. Soc.
1982, 104, 1114–1116.
[32] A. Sugiyama, Y. Ohnishi, M. Nakaoka, Y. Nakao, H. Sato, S. Sakaki, Y.
Nakao, T. Hiyama, J. Am. Chem. Soc. 2008, 130, 12975–12985.
[33] E. P. A. Couzijn, J. C. Slootweg, A. W. Ehlers, K. Lammertsma, J. Am.
Chem. Soc. 2010, 132, 18127–18140.
[6]
[7]
I. Kuwajima, E. Nakamura, Acc. Chem. Res. 1985, 18, 181–187.
M. Murakami, M. Inouye, M. Suginome, Y. Ito, Bull. Chem. Soc. Jpn.
1988, 61, 3649–3652.
[34] A. P. Bento, F. M. Bickelhaupt, Chem. - An Asian J. 2008, 3, 1783–
1792.
[8]
[9]
K. Chen, G. F. Koser, J. Org. Chem. 1991, 56, 5764–5767.
C. Carfagna, A. Musco, G. Sallese, R. Santi, T. Fiorani, J. Org. Chem.
1991, 56, 261–263.
[35] M. Mitani, Y. Osakabe, J. Chem. Soc., Chem. Commun. 1994, 1759–
1760.
[10] F. Agnelli, G. a. Sulikowski, Tetrahedron Lett. 1998, 39, 8807–8810.
[11] T. Iwama, V. B. Birman, S. a Kozmin, V. H. Rawal, Org. Lett. 1999, 1,
673–676.
[36] A. P. Bento, F. M. Bickelhaupt, J. Org. Chem. 2007, 72, 2201–2207.
[37] M. A. Van Bochove, M. Swart, F. M. Bickelhaupt, J. Am. Chem. Soc.
2006, 128, 10738–10744.
[12] J. Chae, J. Yun, S. L. Buchwald, Org. Lett. 2004, 6, 4809–4812.
[13] T. Iwama, V. H. Rawal, Org. Lett. 2006, 8, 5725–5728.
[14] W. Su, S. Raders, J. G. Verkade, X. Liao, J. F. Hartwig, Angew.
Chemie Int. Ed. 2006, 45, 5852–5855.
[38] B. Bazán-Tejeda, G. Bluet, G. Broustal, J.-M. Campagne, Chem. - A
Eur. J. 2006, 12, 8358–8366.
[39] M. J. et al Frisch, Gaussian 09, Revis. A.02 2009.
[40] K. Burke, J. P. Perdew, Y. Wang, Electron. Density Funct. Theory -
Recent Prog. New Dir. 1998, 81–111.
[15] D. S. Huang, J. F. Hartwig, Angew. Chemie Int. Ed. 2010, 49, 5757–
5761.
[41] D. Andrae, U. Häußermann, M. Dolg, H. Stoll, H. Preuß, Theor. Chim.
Acta 1990, 77, 123–141.
[16] S. Yadav, R. Dixit, K. Vanka, S. S. Sen, Chem. Eur. J. 2018, 24, 1269-
1273.
[42] W. Küchle, M. Dolg, H. Stoll, H. Preuss, Mol. Phys. An Int. J. Interface
Between Chem. Phys. 1991, 74, 1245–1263.
[17] R. J. Trovitch, Acc.Chem. Res. 2017, 50, 2842-2852.
[18] X.-P. Zeng, Z.-Y. Cao, X. Wang, L. Chen, F. Zhou, F. Zhu, C.-H. Wang,
J. Zhou, J. Am. Chem. Soc. 2016 138 (1), 416-425.
[43] K. Oisaki, Y. Suto, M. Kanai, M. Shibasaki, J. Am. Chem. Soc. 2003,
125, 5644–5645.
[19] T. Komiyama, Y. Minami, T. Hiyama, ACS Catalysis 2017, 7, 631-651.
[20] J.-S. Yu, F.-M. Liao, W.-M. Gao, K. Liao, R.-L Zuo, J. Zhou, J., Angew.
Chem. Int. Ed., 2015, 54, 7381-7385.
[44] M. Das, D. F. O’Shea, J. Org. Chem. 2014, 79, 5595–5607.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800099
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Enol ether transfer from trimethylsilyl
to copper (I) centers has been
investigated computationally and
experimentally evidencing the
sensitivity of the thermodynamic of
the reaction to the exchanged
counterpart and of its kinetic to the
nature of the enol ether chain and to
the copper center.
S. Bouaouli, K. Spielmann, E.
Vrancken, J.-M Campagne, and H.
Gérard*
Reversing enolate chain transfer
between Si and Cu
Keywords
copper; sillyl ethers; Mechanistic studies; DFT; substituent effects
This article is protected by copyright. All rights reserved.