Regioselectivity of Stoichiometric Metathesis of Vinylsilanes with Second-Generation Grubbs Catalyst: A Combined DFT and Experimental Study

Article abstract of DOI:10.1021/acs.organomet.5b00878

The regioselectivity of metathesis reactions of trisubstituted vinylsilanes H₂C=CHSiR₃ (SiR₃ = SiCl₃, SiCl₂Me, SiClMe₂, SiMe₃, Si(OEt)₃) with the second-generation rut

Full text of DOI:10.1021/acs.organomet.5b00878

Article
pubs.acs.org/Organometallics
Regioselectivity of Stoichiometric Metathesis of Vinylsilanes with
Second-Generation Grubbs Catalyst: A Combined DFT and
Experimental Study
†
†
,†
‡
‡
́
̇
Paweł Sliwa, Kamil Kurleto, Jarosław Handzlik,* Szymon Rogalski, Patrycja Zak,
‡
,‡
̇
Bozena Wyrzykiewicz, and Cezary Pietraszuk*
Faculty of Chemical Engineering and Technology, Cracow University of Technology, Warszawska 24, 31-155 Krakow
‡Faculty of Chemistry, Adam Mickiewicz University in Poznan
, Umultowska 89b, 61-614 Poznan, Poland
S Supporting Information
†
́
, Poland
́
́
*
ABSTRACT: The regioselectivity of metathesis reactions of trisubstituted vinylsilanes H CCHSiR (SiR = SiCl , SiCl Me,
2
3
3
3
2
SiClMe , SiMe , Si(OEt) ) with the second-generation ruthenium alkylidene complex has been studied theoretically, by density
2
3
3
functional theory (DFT), and experimentally. The DFT results indicate that cycloreversion is the rate-determining step and the
formation of a thermodynamically stable ruthenium methylidene complex and PhCHCHSiR is generally preferred. However,
3
the regioselectivity of the process can be also governed by the relative stabilities of the ruthenacyclobutane intermediates, which
depend on the electronic and steric properties of the SiR substituent. Higher stability of α,β-disubstituted ruthenacyclobutanes
3
in comparison to α,α-disubstituted ruthenacyclobutanes is predicted, in contrast to the corresponding intermediates formed
during metathesis of common α-olefins. The stabilizing Ru−C interaction in the ring is strengthened by the electron-donor SiR
β
3
substituent at C . The experiments performed have shown selectivity toward styrene formation for SiR = SiClMe , SiMe ,
β
3
2
3
whereas a preference for the formation of ruthenium methylidene and PhCHCHSiR has been observed for SiR = SiCl ,
3
3
3
SiCl Me, Si(OEt) , in accordance with the theoretical predictions.
2
3
INTRODUCTION
silylstyrene (path a, Scheme 1), which is the opposite of path b
being preferred for α-olefins.
The observed difference in regioselectivity must be a
consequence of steric and electronic effects of the silyl groups
on the properties of the double CC bond. However, no
■
2
,3
Among functional olefins, vinylsilanes make up a group of
compounds that differs from the most common α-olefins in
their reactivity toward ruthenium olefin metathesis catalysts
1
(
Figure 1). It has been shown that the equimolar reactions of
vinylsilanes with first- and second-generation Grubbs catalysts
occur with preferred formation of a methylidene complex and
Scheme 1. Reactivity of Grubbs Catalysts toward
Vinylsilanes
Received: October 16, 2015
Figure 1. Well-defined ruthenium-based olefin metathesis catalysts.
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
7
,8,10,11,20,23
explanation of the mechanism of steric and electronic effects on
regioselectivity has been proposed.
for other 14-electron ruthenium alkylidene species,
7
,8
which can be explained by analysis of molecular orbitals and
by a stabilizing interaction between the hydrogen from the
benzylidene ligand of [Ru]CHPh species and the electron-
The progress in application of metathesis in the chemistry of
vinylsilanes and vinylsiloxanes and in the design of new
catalysts with improved catalytic performance toward organo-
silicon compounds requires a full understanding of the factors
that determine the regioselectivity of vinylsilane addition to
ruthenium alkylidene complexes.
17
deficient ruthenium center. However, such interactions,
involving H, Cl, or O atoms, are observed for both types of
the conformers of the 14-electron [Ru]CHSiR intermedi-
3
ates (Figure S1).
Although a large number of theoretical studies on olefin
metathesis catalyzed by Grubbs-type complexes have been
Considering the reactions between the active ruthenium
alkylidene species and vinylsilanes (1a,b, Scheme 2), pathway b
is thermodynamically favored (Table 1). However, taking into
4
−20
reported so far,
computational works concerning meta-
thesis reactions of vinylsilanes are lacking. Herein, we report a
detailed study of the regioselectivity of cycloaddition of selected
vinylsilanes to the second-generation ruthenium alkylidene
complex, performed with density functional theory (DFT)
calculations. We also have tried to understand how the
a
−1
Table 1. Calculated Gibbs Energies (ΔG , kJ mol ) for
s
Reactions of Vinylsilanes H CCHSiR with Complex 2
2
3
(Scheme 2)
reaction
properties of SiR substituents influence the relative stability
3
of the ruthenacyclobutane intermediates in the competitive
reaction pathways, which may affect the regioselectivity of the
process. Computational work has been supplemented by results
of experimental studies on the reactivity of selected vinylsilanes
in the presence of equimolar amounts of a second-generation
ruthenium alkylidene complex.
reactant
1a
1b
2a
2b
H CCHSiCl
11
13
14
14
12
6
2
3
4
4
3
42
35
35
44
33
2
3
H CCHSiCl Me
0
3
2
2
H CCHSiClMe
2
2
H CCHSiMe
−4
−15
2
3
H CCHSi(OEt)
2
3
a
M06/def2-TZVPP//PBE0/def2-SVP calculations for simulated di-
RESULTS AND DISCUSSION
■
chloromethane solution.
Thermodynamics of the Process. We have considered
two competitive routes of the cycloaddition of vinylsilanes
H CCHSiR (SiR3 = SiCl , SiCl Me, SiClMe , SiMe ,
account that the dissociation of the PCy ligand in the second-
generation Grubbs catalysts is an endergonic proc-
3
2
3
3
2
2
3
4
,5,10,11,14,16,20
Si(OEt) ) to the ruthenium alkylidene complex (pathways a
ess,
which is also confirmed in the present work,
3
and b, Scheme 2). Pathway a leads to the formation of a
the 16-electron ruthenium alkylidene complexes are expected
to dominate in the reaction environment. Therefore, the
thermodynamics of the overall reactions (2a,b, Scheme 2)
should be considered instead. In this case, a clear thermody-
namic preference for pathway a is predicted. This is explained
Scheme 2. Reactions of Vinylsilanes H CCHSiR with 14-
2
3
Electron Intermediate and Complex 2
by a much higher PCy dissociation energy for the ruthenium
3
methylidene complex formed according to pathway a, in
comparison to that of the ruthenium complexes with the SiR3
group in the alkylidene ligands, being the potential products of
pathway b (Table 2). Hence, the ruthenium methylidene
a
−1
Table 2. Calculated Gibbs Energies (ΔG , kJ mol ) for
s
PCy Ligand Dissociation
3
complex
ΔGs
(
(
(
(
(
(
(
H IMes)(Cl) Ru(CHPh)(PCy ) (2)
66
76
30
32
34
18
18
2
2
3
H IMes)(Cl) Ru(CH )(PCy ) (4)
2
2
2
3
H IMes)(Cl) Ru(CHSiCl )(PCy )
2
2
3
3
H IMes)(Cl) Ru(CHSiCl Me)(PCy )
2
2
2
3
ruthenium methylidene complex and silylstyrenes PhCH
H IMes)(Cl) Ru(CHSiClMe )(PCy )
2
2
2
3
CHSiR . The products of pathway b are ruthenium (silyl)-
3
H IMes)(Cl) Ru(CHSiMe )(PCy )
2
2
3
3
alkylidene complexes and styrene.
H IMes)(Cl) Ru(CHSi(OEt) )(PCy )
2
2
3
3
It is well established that 14-electron intermediates, initially
formed after the dissociation of the phosphine ligand (first- and
second-generation Grubbs catalysts) and pyridine or bromo-
pyridine ligands (third-generation Grubbs catalysts) are the
active ruthenium species playing the crucial role in the catalytic
a
M06/def2-TZVPP//PBE0/def2-SVP calculations for simulated di-
chloromethane solution.
complex, after phosphine binding, is predicted to be a stable
thermodynamic product, in accordance with the present
experimental results. In contrast, the calculated phosphine
dissociation energies for the [Ru]CHSiR (PCy ) complexes
4
−22
cycle of olefin metathesis.
Consequently, we assume the
same dissociative mechanism for the metathesis of vinylsilanes.
There are possible rotational isomers of the active [Ru]
CHSiR species, where [Ru] = Ru(Cl) (H IMes) (see Figure
3
3
are much lower than that for the methylidene complex.
Consequently, the ruthenium (silyl)alkylidene complexes are
most unstable among all the ruthenium compounds present in
the reaction environment, because of the easy formation of the
14-electron active species undergoing further metathesis
3
2
2
S1 in the Supporting Information). The perpendicular
orientation of the alkylidene moiety to the mesityl group is
more energetically favored, in comparison to the parallel
orientation. An analogous geometrical preference is predicted
B
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
−
1
Figure 2. Gibbs energy profile (ΔG , kJ mol ) for the reaction of H CCHSiCl with the second-generation Grubbs catalyst 2 (pathways a and b)
s
2
3
(
M06/def2-TZVPP//PBE0/def2-SVP calculations for simulated dichloromethane solution).
−
1
Figure 3. Gibbs energy profile (ΔG , kJ mol ) for the reaction of H CCHSiMe with the second-generation Grubbs catalyst 2 (pathways a and b)
s
2
3
(
M06/def2-TZVPP//PBE0/def2-SVP calculations for simulated dichloromethane solution).
transformations. This explains why such compounds are not
observed experimentally, although the presence of other
products indicates their formation as reaction intermediates.
The initial step of both processes, phosphine ligand
dissociation leading to the active catalyst species [Ru]
CHPh, is characterized by the transition state 2_TS and
2
,3
Regioselectivity of the Process. The calculated pathways
of the reactions of the catalyst 2 with the exemplary vinylsilanes
H CCHSiCl and H CCHSiMe in simulated dichloro-
associated minimum 2_M. The predicted Gibbs energy (ΔG )
s
−1
barrier for this step (89 kJ mol ) is close to the experimental
activation Gibbs energy determined for the PCy ligand
2
3
2
3
3
21,22
−
1
methane solution are shown in Figures 2 and 3, respectively.
The corresponding energy diagrams for other vinylsilanes are
presented in Figures S3−S5 in the Supporting Information.
exchange in toluene solution (96 kJ mol ).
The
corresponding theoretical values reported by other authors
·
−1
16,20
are similar (97−99 kJ mol , toluene)
or clearly higher
C
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
−
1
14
(
123 kJ mol , dichloromethane). The 14-electron ruthenium
reaction of H CCHSiMe and H CCHSiClMe with
2
3
2
2
intermediate can coordinate vinylsilanes to form the respective
π complexes (I_C1a, I_C1b, II_C1a, II_C1b), localized on the
potential energy surface (PES), but this is predicted to be
unstable in terms of Gibbs energy. For both vinylsilane
reactants, the cycloaddition step is slightly kinetically favored in
the case of pathway b, whereas a clear thermodynamic
preference, indicated by the relative stabilities of the
ruthenacyclobutane intermediates (I_CBa, I_CBb, II_CBa,
II_CBb), is seen for pathway a (Figures 2 and 3). However,
cycloreversion, involving the highest-energy transition states
[Ru]CHPh, pathway b might be kinetically preferred,
because the activation Gibbs energy for the cycloreversion
step along pathway a is higher than the overall activation barrier
calculated for pathway b (Figure 3 and Figure S4 in the
Supporting Information; Table 3, entries 3 and 4). On the
other hand, if a pre-equilibrium between the ruthenacyclobu-
tane intermediates and the reactants is assumed, pathway a will
always be favored. Our experimental results indicate the kinetic
preference for pathway b when H CCHSiMe and H C
2
3
2
CHSiClMe are the reactants, although this observation can be
2
(
I_TS2a, I_TS2b, II_TS2a, II_TS2b), is the rate-determining
step. The overall activation barriers, related to the reactants (2
influenced by β-SiR3 elimination being competitive with
cycloreversion along pathway a (vide infra).
The relative stabilities of the ruthenacyclobutanes probably
depend on the steric effects involving the substituents in the
+
H CCHSiR ), are lower for pathway a, suggesting that this
2
3
route is kinetically preferred. This is a general tendency,
predicted for all of the vinylsilanes considered (Table 3). Along
ring and on the electron-donor abilities of the SiR group. In
3
the case of the metathesis reactions between normal olefins,
α,α-disubstituted metallacyclobutane intermediates are more
stable than the corresponding α,β-disubstituted conform-
a
Table 3. Calculated Overall Activation Gibbs Energies (kJ
−1
mol ) for the Reactions of Vinylsilanes H CCHSiR with
2
3
19,24
Complex 2 and the Values Related to the Active 14-Electron
ers.
This can be explained by weaker steric interactions
Species (H IMes)(Cl) Ru(CHPh)
between the substituents in α positions. It has been also shown
2
2
that placing the substituents at C favors electron donation
α
complex 2
[Ru]CHPh
19
from the olefin fragment to the ruthenium alkylidene species.
reactant
2a
2b
1a
1b
Interestingly, our calculations for the metathesis of vinylsilanes
indicate just the opposite trends (Figures 2 and 3 and Figures
S3−S5 in the Supporting Information). The α,β-disubstituted
ruthenacyclobutane complexes (pathway a) are always
predicted to be more stable than their α,α-disubstituted
analogues (pathway b). This might be partially explained by
H CCHSiCl
120
111
107
103
110
128
131
119
118
126
54
57
57
62
54
62
65
53
52
60
2
3
H CCHSiCl Me
2
2
H CCHSiClMe
2
2
H CCHSiMe
2
3
H CCHSi(OEt)
2
3
a
M06/def2-TZVPP//PBE0/def2-SVP calculations for simulated di-
steric interactions between the substituent SiR in an α position
3
chloromethane solution.
and the mesityl group of the NHC ligand (Figure 4). However,
if analogous model compounds with the smallest SiH3
substituent are considered, the calculated Gibbs energy for
the α,α-disubstituted metallacycle is also higher, by 23 kJ mol ,
in comparison to the α,β-disubstituted ruthenacyclobutane.
This result shows that the unexpected energetic preference
predicted for the reaction of vinylsilanes with the second-
generation Grubbs catalyst is mainly of electronic nature.
This effect can be explained in terms of the electron-donor
the cycloreversion path, π-complex structures have been found
on the PES (I_C2a, I_C2b, II_C2a, II_C2b), again being
thermodynamically unstable species. As mentioned above
−1
(
Table 1), the final products of pathway a (PhCHCHSiR3
and ruthenium methylidene complex 4) are thermodynamically
more stable than the products of pathway b (styrene and
[
Ru](CHSiR )(PCy ) complex). Moreover, the calculated
3
3
activation Gibbs energy of the PCy dissociation for 4 (103 kJ
ability of the SiR
on the basis of the calculated NPA charges on the SiR
fragments of the vinylsilane molecules, increases in the order
SiCl < SiCl Me < SiClMe < SiMe < Si(OEt) . The same
3
group. The electron-donor ability, estimated
3
−1
mol ) is even higher than that for 2 (Figures 2 and 3),
confirming that the former is the least reactive catalyst among
the ruthenium alkylidene complexes considered.
3
3
2
2
3
3
Conclusions about the regioselectivity of the process may
change if the catalyst dissociation stage is omitted in the
consideration of the reaction pathway. Such an approach can be
justified when phosphine is removed from the solution, for
instance, by the reaction with CuCl; thus, the 14-electron
ruthenium alkylidene intermediate dominates over the
precursor in the reaction mixture. The dissociation stage is
also less significant in the case of an excess of the vinylsilane
reactant relative to the catalyst concentration. The calculated
energy profiles can be then related to the energy level of the 14-
electron intermediate + H CCHSiR , and the overall
order is obtained for the ruthenacyclobutane intermediates
localized on pathway a. For the ruthenacyclobutanes on
pathway b, the order is slightly changed (Table S1 in the
Supporting Information). The predicted electron donation
from the SiR
fragment is always stronger for the
3
ruthenacyclobutane complexes on pathway a, in comparison
to that of the corresponding intermediates on pathway b. The
same tendency is observed for the charges calculated on the
CH CHSiR fragment (Table 4 and Table S1 in the Supporting
2 3
Information). A stronger donation from the original vinylsilane
moiety to the ruthenium alkylidene species is predicted for the
α,β-disubstituted ruthenacyclobutanes (CBa, pathway a) than
for the corresponding α,α-disubstituted intermediates (CBb,
pathway b). This is in contrast to the theoretical results
reported for the metathesis of common alkenes catalyzed by
2
3
activation barrier of the process can depend on the relative
Gibbs energy of the ruthenacyclobutane intermediate. In the
case of pathway a, the ruthenacyclobutane complex can be
formed in an endergonic or exergonic reaction, depending on
the vinylsilane reactant, whereas its formation according to
pathway b is always endergonic (Figures 2 and 3 and Figures
S3−S5 in the Supporting Information). Consequently, the
formation of a stable ruthenacyclobutane structure in pathway a
may result in the kinetic preference for pathway b. For the
19
ruthenium systems but is in agreement with the aforemen-
tioned opposite relative stabilities of the α,β- and α,α-
disubstituted ruthenacyclobutane intermediates involved in
vinylsilane metathesis (Table 4 and Table S1 in the Supporting
19
Information) and metathesis of normal alkenes.
D
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
2
7
Ru−C interaction is shown to be dominant, an electron-
β
donor substituent at C (Figure 4) should additionally stabilize
β
the ring. The calculated Ru−C distances are indeed a bit
β
shorter in the ruthenacyclobutane intermediates formed along
pathway a (2.20−2.21 Å), in comparison to those for the
intermediates formed along pathway b (2.22−2.23 Å),
suggesting stronger Ru−C bonding in the former species.
β
These observations are additionally confirmed by the analysis of
the Ru−C bond indices. To focus on the electron-donor effect
β
and avoid an overly significant influence of the steric effects, the
ruthenacyclobutane with the most bulky SiR group, Si(OEt) ,
3
3
is omitted in Table 4 (complete results are presented in Table
S1 in the Supporting Information). For the remaining α,β-
disubstituted ruthenacyclobutane intermediates (CBa, pathway
a), a correlation between the electron-donor ability of the SiR3
group and the Ru−C bond indices is clearly seen. It is also
β
predicted that the electron donation from the SiR group and
3
the Ru−C bond index correlate positively, in general, with the
β
energetic stability of the α,β-disubstituted ruthenacyclobutane
complex, although this relationship is not as smooth. In
contrast, such correlations between the calculated charges on
the SiR fragments and the Ru−C bond indices or the relative
3
β
energies are not observed at all for the α,α-disubstituted
ruthenacyclobutanes (CBb, pathway b).
To conclude, the regioselectivity of the reactions between
trisubstituted vinylsilanes H CCHSiR and the Grubbs-type
2
3
ruthenium alkylidene catalyst can be governed by the relative
stability of the ruthenacyclobutane intermediates (Figure 4),
which is affected by the electronic and steric properties of the
SiR moieties. A higher stability of α,β-disubstituted ruth-
3
enacyclobutanes (bearing a silyl group at C ) than α,α-
β
disubstituted ruthenacyclobutanes is always predicted, in
contrast to the corresponding metallacycles formed during
19
metathesis of common α-olefins. The stabilizing Ru−Cβ
interaction is strengthened by the electron-donor SiR
3
substituent at C (pathway a), whereas a destabilizing steric
β
interaction takes place between the bulk SiR substituent at Cβ
3
and the phenyl substituent at C (pathway a) or between the
α
bulk SiR substituent at C and the mesityl group of the NHC
3
α
ligand (pathway b).
Investigation of Equimolar Reactions. In order to verify
the results of calculations concerning the kinetic preference for
pathways a or b, a series of tests were performed in which
catalyst 2 was treated with 3 equiv of vinylsilanes examined
Figure 4. Structures of the ruthenacyclobutane intermediates formed
during the reactions of H CCHSiR with 2: pathway a (on the left)
2
3
and pathway b (on the right).
(
H CCHSiCl , H CCHSiCl Me, H CCHSiClMe ,
2
3
2
2
2
2
H CCHSiMe , H CCHSi(OEt) ). A 3-fold excess of
2 3 2 3
It has been proposed that ruthenacyclobutane complexes are
stabilized by four-center−two-electron α,β-(C−C−C) agostic
bonding in the ring, resulting in σ donation from the C−C−C
vinylsilane was necessary to speed up the reactions, which
hardly proceeded when an equimolar ratio of reagents was
applied. The reactions were performed in NMR tubes and
25,26
1
fragment to the ruthenium center.
As a covalent two-center
monitored by H NMR spectroscopy. In general, the
a
a
Table 4. NPA Charges on the SiR Groups in the Ruthenacyclobutane Intermediates (CBa, CBb), NPA Charges on the
3
a
Vinylsilane Fragments CH CHSiR of the Ruthenacyclobutane Intermediates, Wiberg Bond Indices for the Ru−C Bonds in
the Ruthenacyclobutane Intermediates and the Relative Energies (ΔE, kJ mol ) of the Ruthenacyclobutane Intermediates
2
3
β
,
ab
−1
q(SiR3)
q(CH CHSiR )
P(Ru−Cβ)
ΔE
2
3
SiR3
SiCl3
CBa
CBb
CBa
CBb
CBa
CBb
CBa
CBb
0.526
0.537
0.569
0.573
0.410
0.450
0.447
0.470
0.166
0.194
0.225
0.256
0.097
0.127
0.160
0.191
0.177
0.186
0.198
0.208
0.165
0.165
0.166
0.166
−92
−102
−99
−82
−73
−88
−79
SiCl Me
2
SiClMe2
SiMe3
−110
a
b
PBE0/def2-SVP calculations. Energies related to (H IMes)(Cl) Ru(CHPh) + H CCHSiR .
2
2
2
3
E
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
postreaction mixture contains a variety of components that are
products of pathways a and b, products of vinylsilane
homometathesis (when applied) as well as products of
potential β-silylsubstituted ruthenacyclobutane decomposition
predicted (Table 3). For H CCHSiMe and H C
2
3
2
CHSiClMe , the preferable formation of silylstyrene (via
2
pathway a) is no longer observed, which can be explained by
significant stabilization of the ruthenacyclobutane intermediate
on pathway a, resulting in a higher activation barrier for the
cycloreversion step, in comparison to the overall barrier for
pathway b (Figure 3 and Figure S4 in the Supporting
Information; Table 3, entries 3 and 4). As the initial form of
catalyst 2 and the 16-electron methylidene complex 4 are the
most stable compounds in both pathways, the removal of PCy3
from the reaction environment (via reaction with CuCl),
favoring the formation of the 14-electron ruthenium alkylidene
species, should also additionally favor pathway b (Table 3,
Figures 2 and 3). However, in the presence of CuCl, the
reaction was so fast that the initial stage could not be observed.
In this case, the data from Table 5 (entry 5) describe the
composition of the postreaction mixture.
2
,3,28
via β-SiR elimination (Scheme 3).
3
Scheme 3. Decomposition of β-Silyl-Substituted
Ruthenacyclobutanes via β-SiR Elimination
3
The composition of the reaction mixture and relative
concentrations of individual components depend on the SiR3
group and the catalyst used. In this study it was assumed that a
measure of the kinetic preference for pathway a or b is the
concentration ratio of silylstyrene to that of styrene monitored
in the first stage of the reactions (Scheme 1). It was confirmed
that silylstyrene formed does not undergo any chemical change.
Homometathesis of the styrene transformation that could affect
its concentration and in consequence distort the observed
silylstyrene to styrene ratio proceeds negligibly slowly and can
be ignored in the first stage of the reactions. A reaction of
styrene with the silylcarbene complex leading to silylstyrene
could not be experimentally excluded. However, it was found
that most of the silylcarbene complex formed in the systems
studied is consumed in the reaction with vinylsilane that leads
to the formation of bis(silyl)ethenes and/or 1,1- and 1,3-
The experimental studies confirmed also the stability of
complex 4 predicted by the calculations. In the reaction
conditions applied, complex 4 did not react with any of
vinylsilanes used.
CONCLUSION
■
On the basis of the results of DFT calculations, cycloreversion
is predicted to be the rate-determining step for the reactions
between trisubstituted vinylsilanes H CCHSiR and the
2
3
second-generation Grubbs catalyst. The energy of the rate-
determining transition state for the pathway leading to the
ruthenium methylidene complex and silylstyrene is always
lower than that for the alternative pathway giving ruthenium
29
bis(silyl)propenes-1.
For SiR = SiMe , β-silyl-substituted ruthenacyclobutane
3
3
(
silyl)methylidene complex and styrene. However, the
formed in pathway a (Scheme 1) undergoes cycloreversion,
leading to silylstyrene, but there is also a tendency for
regioselectivity of the process may be governed by the relative
stabilities of the ruthenacyclobutane intermediates on the
competitive pathways, which depend on the electron-donor
2
,3,28
decomposition via β-SiR elimination (Scheme 3).
Being
3
competitive to cycloreversion, β-elimination affects the
ability and steric properties of the SiR substituent. An opposite
3
silylstyrene to styrene ratio. Thisis why for SiR = SiMe it
3
3
trend in the relative stabilities of the silyl-substituted
ruthenacyclobutanes, in comparison to the corresponding
intermediates involved in the metathesis of common α-
was assumed that a measure of the kinetic preference for
pathway a or b is the ratio of the sum of concentrations of
silylstyrene and (phenyl,silyl)-substituted propenes, formed via
β-elimination (Scheme 3), to the concentration of styrene,
monitored in the first stage of the reactions. The concentration
ratios of silylstyrene to styrene were determined on the basis of
19
olefins, is predicted. This trend is manifested by the higher
stability of α,β-disubstituted ruthenacyclobutanes (bearing a
silyl group at C ) than α,α-disubstituted ruthenacyclobutanes
β
and is explained mainly by the electronic effects.
1
H NMR spectrum analysis and are presented in Table 5. The
The theoretically predicted possible kinetic preference for
the formation of (silyl)methylidene complex in the reactions of
second-generation Grubbs catalyst with H CCHSiMe or
observed regioselectivity for the reaction of vinylsilanes with
catalyst 2 is in agreement with the DFT calculation results. For
H CCHSiCl , H CCHSiCl Me, and H CCHSi(OEt) ,
2
3
2
3
2
2
2
3
H CCHSiClMe has been confirmed experimentally. The
2
2
the observed kinetic preference for pathway a has always been
selectivity of the reaction toward the formation of styrene and
1
the silylcarbene complex (undetectable by H NMR spectros-
Table 5. Silylstyrene/Styrene Molar Ratio for Reactions of
copy) has been observed. Thus, we have demonstrated the
possibility of controlling the regioselectivity of the reaction (so
that it would occur via pathway b) by selecting a relatively small
Catalysts 2 with H CCHSiR Measured in a First Stage of
2
3
a
the Reaction
and strong electron-donor SiR substituent. This possibility
should facilitate the design of effective catalysts of metathetic
transformations of vinylsilanes and analogous vinylmetalloids.
time
(h)
[silylstyrene]/
[styrene]
3
entry
H CCHSiR
2
3
b
1
2
3
4
5
6
H CCHSiCl
0.17
0.17
0.17
0.17
0.08
0.17
5.6/1 (3.2/1)
2
3
b
H CCHSiCl Me
1.9/1 (2.6/1)
2
2
b
H CCHSiClMe
0.3/1 (0.17/1)
EXPERIMENTAL SECTION
General Methods and Chemicals. Unless mentioned otherwise,
all operations were performed by using standard Schlenk techniques.
Chemicals were obtained from Aldrich. Alkyl-, aryl-, and alkoxy-
2
2
■
c
H CCHSiMe
0.1/1 (0.4/1)
2
3
d
,
cd
H CCHSiMe (+3 equiv CuCl)
0.01/1 (0.9/1)
2
3
b
H CCHSi(OEt)
2.2/1 (2.6/1)
2
3
substituted vinylsilanes were dried in a Schlenk tube over CaH for 24
2
a
b
Conditions: [Ru]: [CC] = 1:3; 303 K, CD Cl . Measured on the
h, and after that they were distilled under argon. Chloro-substituted
vinylsilanes were distilled prior to use under argon. All vinylsilanes
were additionally degassed by repeated freeze−pump−thaw cycles. All
2
2
c
basis of GC (FID) analysis. ([silylstyrene] + [(phenyl,silyl)-
d
propenes])/[styrene]. 295 K.
F
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
solvents were dried prior to use over CaH and stored under argon
the metathesis reactions studied, Gibbs energy profiles
for the reactions of vinylsilanes (H CCHSiCl Me,
2
over 4 Å molecular sieves. CD Cl was additionally passed through a
2
2
1
2
2
column of alumina. H NMR measurements were performed on
H CCHSiClMe , H CCHSi(OEt) ) with 2, chem-
2
2
2
3
Bruker Avance DRX 600 spectrometer, operating at a frequency of
ical shifts of signals used for determination of selectivity
6
00.13 MHz. All spectra were recorded at 298 K.
Procedure for Equimolar Reactions Study. The equimolar
1
and relevant H NMR spectra, and experimental details
reactions of vinylsilanes with complex 2 were performed under argon
on the determination of regioselectivity via GC
(calibration curves and GC chromatograms) (PDF)
Cartesian coordinates of all computed molecules (XYZ)
1
in Wilmad LPV NMR tubes and monitored by H NMR spectroscopy.
−5
In a typical procedure complex 2 (0.01 g, 1.18 × 10 mol) and
anthracene (internal standard; 0.002 g) were dissolved in 0.65 mL of
1
CD Cl . Then the H NMR spectrum of the starting mixture was
2
2
recorded and 3 equiv of the corresponding vinylsilane was added
under argon using a microliter syringe. The reaction mixture was
monitored by H NMR spectroscopy at 23 °C and by GC (FID)
analysis. The styrene/silylstyrene product ratio was determined on the
basis of olefinic proton integration (NMR) or integration of the
corresponding peaks (GC).
AUTHOR INFORMATION
■
*
*
Corresponding Authors
J.H.: tel, (+48) 12 628 21 96; e-mail, jhandz@pk.edu.pl
C.P.: tel,(+48) 61 829 16 99; e-mail, pietrasz@amu.edu.pl.
1
Notes
Computational Details. All structures were fully optimized using
The authors declare no competing financial interest.
30
the hybrid PBE0 functional and the split-valence def2-SVP basis
3
1
set. The 28 innermost electrons of Ru were replaced by the Stuttgart
32
ACKNOWLEDGMENTS
effective core potential. This methodology was previously shown to
■
be accurate in predicting the geometry of the Grubbs catalyst on the
This paper is dedicated to Professor Bogdan Marciniec on the
occasion of his 75th birthday. The authors thank Dr. Stefan
Kurek, Cracow University of Technology, for helpful
discussions. Financial support from the National Science
Centre (Poland), (Project No. N 204 1851 40) is gratefully
acknowledged. J.H., P.S., and K.K. acknowledge the computing
resources from PL-Grid Infrastructure and Academic Computer
Centre CYFRONET AGH (Grant Nos. MNiSW/
I B M _ B C _ H S 2 1 / P K / 0 0 3 / 2 0 1 3 a n d M N i S W /
IBM_BC_HS21/PK/037/2014).
1
1
basis of test calculations using many popular DFT functionals.
Harmonic vibrational frequencies were calculated for each structure to
confirm the potential energy minimum or the transition state and to
obtain Gibbs energy corrections (T = 298.15 K, p = 1 atm). The
3
3,34
transition states were additionally verified by IRC calculations.
Single-point energy calculations were performed for the optimized
35
geometries by applying the hybrid M06 functional combined with
31,32
the triple-ζ valence def2-TZVPP basis set.
The M06 method was
developed as a general-purpose hybrid meta-GGA functional
recommended for main-group-element and transition-metal thermo-
3
5
chemistry, kinetics, and studies of noncovalent interactions. A good
performance of the M06 method in reproducing energies of metathesis
9
,11,18
REFERENCES
reactions involving ruthenium alkylidene complexes
and the
■
1
0,11
dissociation energy for the second-generation Grubbs catalyst
was
(1) (a) Pietraszuk, C.; Pawluc, P.; Marciniec, B. Metathesis of Silicon-
Containing Olefins. In Handbook of Metathesis; Grubbs, R. H., Wenzel,
A. G., O’Leary, D. J., Khosravi, E., Eds.; Wiley-VCH: Weinheim,
Germany, 2015; Chapter 2.9. (b) Marciniec, B.; Pietraszuk, C.
Metathesis of silicon-containing olefins. In Handbook of Metathesis;
Grubbs, R. H., Ed.; Wiley-VCH: Weinheim, Germany, 2003; Chapter
2.13. (c) Marciniec, B.; Pietraszuk, C. Curr. Org. Chem. 2003, 7, 691−
735.
(2) Pietraszuk, C.; Fischer, H.; Rogalski, S.; Marciniec, B. J.
Organomet. Chem. 2005, 690, 5912−5921.
(3) Pietraszuk, C.; Marciniec, B.; Fischer, H. Organometallics 2000,
19, 913−917.
(4) Cavallo, L. J. Am. Chem. Soc. 2002, 124, 8965−8973.
(5) Adlhart, C.; Chen, P. J. Am. Chem. Soc. 2004, 126, 3496−3510.
(6) Vyboishchikov, S. F.; Thiel, W. Chem. - Eur. J. 2005, 11, 3921−
3935.
(7) Tsipis, A. C.; Orpen, A. G.; Harvey, J. N. Dalton Trans. 2005,
2849−2858.
proved. On the other hand, within the methodology applied, the PBE0
functional predicts more accurate bond distances than M06 for the
11
ruthenium alkylidene complex 2. For this reason, the former has
been chosen for the geometry optimization in this work. Performed
tests indicate that the formation energies for ruthenacyclobutanes
−1
I_CBa and I_CBb slightly change, by 3 and 13 kJ mol , respectively,
if M06-optimized geometries are taken for the single-point calculations
instead of the PBE0-optimized geometries.
36
The polarizable continuum model (PCM) was used to estimate
solvent effects (dichloromethane). The PBE0/def2-SVP Gibbs energy
corrections were added to the PCM single-point energies to obtain a
better estimate of Gibbs energies (ΔG ) of the compounds in the
s
solvent. The reaction pathways in this paper are discussed in terms of
Gibbs energies calculated at the M06/def2-TZVPP//PBE0/def2-SVP
level for simulated dichloromethane solution.
Electronic properties of the vinylsilanes studied and the
ruthenacyclobutane intermediates were analyzed by using the natural
37,38
39
population analysis (NPA)
and Wiberg bond indices at the
(8) Straub, B. F. Adv. Synth. Catal. 2007, 349, 204−214.
(9) Pandian, S.; Hillier, I. H.; Vincent, M. A.; Burton, N. A.;
Ashworth, I. W.; Nelson, D. J.; Percy, J. M.; Rinaudo, G. Chem. Phys.
Lett. 2009, 476, 37−40.
PBE0/def2-SVP level. All calculations were carried out with the
40
Gaussian 09 set of programs. For the graphic presentation of the
41
systems studied, GaussView 5.0 software was used.
(
10) Benitez, D.; Tkatchouk, E.; Goddard, W. A., III Organometallics
ASSOCIATED CONTENT
2009, 28, 2643−2645.
■
́
(
(
11) Sliwa, P.; Handzlik, J. Chem. Phys. Lett. 2010, 493, 273−278.
*
S
Supporting Information
12) Fomine, S.; Tlenkopatchev, M. A. Organometallics 2010, 29,
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.5b00878.
1
580−1587.
(13) Liu, P.; Xu, X.; Dong, X.; Keitz, B. K.; Herbert, M. B.; Grubbs,
R. H.; Houk, K. N. J. Am. Chem. Soc. 2012, 134, 1464−1467.
14) Nunez-Zarur, F.; Solans-Monfort, X.; Rodríguez-Santiago, L.;
Sodupe, M. Organometallics 2012, 31, 4203−4215.
(
̃
Results of the natural population analysis for the
vinylsilanes and ruthenacyclobutane intermediates and
the ruthenacyclobutane relative energies, structures,
selected atomic distances, and relative energies of
conformers for the ruthenium alkylidene products of
(
15) Trzaskowski, B.; Grela, K. Organometallics 2013, 32, 3625−
3
630.
(16) Minenkov, Y.; Occhipinti, G.; Jensen, V. R. Organometallics
2013, 32, 2099−2111.
G
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(
17) Urbina-Blanco, C. A.; Poater, A.; Lebl, T.; Manzini, S.; Slawin,
A. M. Z.; Cavallo, L.; Nolan, S. P. J. Am. Chem. Soc. 2013, 135, 7073−
079.
18) Poater, A.; Pump, E.; Vummaleti, S. V. C.; Cavallo, L. J. Chem.
Theory Comput. 2014, 10, 4442−4448.
19) Paredes-Gil, K.; Solans-Monfort, X.; Rodriguez-Santiago, L.;
Sodupe, M.; Jaque, P. Organometallics 2014, 33, 6065−6075.
20) Minenkov, Y.; Occhipinti, G.; Heyndrickx, W.; Jensen, V. R. Eur.
J. Inorg. Chem. 2012, 2012, 1507−1516.
21) Sanford, M. S.; Ulman, M.; Grubbs, R. H. J. Am. Chem. Soc.
001, 123, 749−750.
22) Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem. Soc.
001, 123, 6543−6554.
7
(
(
(
(
2
(
2
(
́
23) Sliwa, P.; Handzlik, J.; Czelusn
́
iak, I. J. Organomet. Chem. 2014,
7
(
(
(
(
67, 6−15.
24) Handzlik, J. J. Mol. Catal. A: Chem. 2004, 218, 91−100.
25) Suresh, C. H.; Baik, M.-H. Dalton Trans. 2005, 18, 2982−2984.
26) Suresh, C. H. J. Organomet. Chem. 2006, 691, 5366−5374.
27) Rowley, C. N.; van der Eide, E. F.; Piers, W. E.; Woo, T. K.
Organometallics 2008, 27, 6043−6045.
(
28) (a) Pietraszuk, C.; Fischer, H. Chem. Commun. 2000, 2463−
2464. (b) Pietraszuk, C.; Fischer, H. Organometallics 2001, 20, 4641−
4646.
(
29) Pietraszuk, C.; Rogalski, S. Unpublished results.
(
(
30) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158−6170.
31) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3
(
297−3305.
32) Andrae, D.; Hau
Theor. Chim. Acta 1990, 77, 123−141.
33) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154−
161.
34) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523−
527.
̈
ssermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
(
2
(
5
(
(
35) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−241.
36) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2
(
8
(
999−3093.
37) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
99−926.
38) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.
NBO Version 3.1.
(39) Wiberg, K. B. Tetrahedron 1968, 24, 1083−1096.
(40) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision C.01; Gaussian, Inc., Wallingford, CT, 2009.
(
41) Dennington, R.; Keith, T.; Millam, J. GaussView, Version 5;
Semichem Inc., Shawnee Mission, KS, 2009.
H
DOI: 10.1021/acs.organomet.5b00878
Organometallics XXXX, XXX, XXX−XXX