Hydrosilylation vs. dehydrogenative silylation of styrene catalysed by iron(0) carbonyl complexes with multivinylsilicon ligands - Mechanistic implications

Article abstract of DOI:10.1016/j.jorganchem.2015.04.051

Well-defined iron(0) complexes with multivinylsilicon ligands of the formula [Fe(CO)₃L] and {[Fe(CO)₃]₂L′}, where L = dienes, trienes or vinylfunctional silicones and L′ = tetraenes, are new very active and effective catal

Full text of DOI:10.1016/j.jorganchem.2015.04.051

Journal of Organometallic Chemistry 791 (2015) 58e65
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Hydrosilylation vs. dehydrogenative silylation of styrene catalysed
by iron(0) carbonyl complexes with multivinylsilicon ligands e
Mechanistic implications
Bogdan Marciniec a, b, , Agnieszka Kownacka , Ireneusz Kownacki , Marcin Hoffmann ,
Richard Taylor
a
*
a
a
a
c
Department of Organometallic Chemistry, Faculty of Chemistry, Adam Mickiewicz University in Pozna nꢀ , Umultowska 89b, 61-614 Poznan, Poland
Center for Advanced Technologies, Adam Mickiewicz University in Pozna nꢀ , St. Umultowska 89c, 61-614 Pozna nꢀ , Poland
Core Products Science and Technology, Dow Corning Ltd, Barry, CF63 2YL, United Kingdom
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 October 2014
Received in revised form
Well-defined iron(0) complexes with multivinylsilicon ligands of the formula [Fe(CO) L] and
3
0
0
{
[Fe(CO)
3
]
2
L }, where L ¼ dienes, trienes or vinylfunctional silicones and L ¼ tetraenes, are new very
active and effective catalysts for hydrosilylation and/or dehydrogenative silylation of styrene with
trisubstituted silanes and hydrosiloxanes. All catalytic data and selected kinetic measurements as well as
the information on stoichiometric reactions of the new precursor catalysts with model substrates (fol-
lowed by NMR and GCeMS techniques), supported by the relative energies of transition states and in-
termediates obtained by DFT calculations, permitted proposing a rational mechanism for the catalysis in
alkene hydrosilylation, proceeding according to the Chalk-Harrod and/or modified Chalk-Harrod
mechanisms.
20 April 2015
Accepted 28 April 2015
Available online 22 May 2015
Keywords:
Hydrosilanes
Hydrosilylation
Dehydrogenative silylation
Olefins
© 2015 Elsevier B.V. All rights reserved.
Iron(0) complexes
Introduction
[2], and tertiary silanes [3] as well as dienes [4]. On the other hand,
2
[FeCp(CO) Me] as reported by Nakazawa, appears to be an efficient
The hydrosilylation reaction of carbonecarbon multiple bonds is
commonly used as an efficient tool for the synthesis of a wide range
of organosilanes and polysiloxanes as well as for the addition
curing of silicone fluids. However, literature and patented data
indicate that in the industrial processes, general application of
catalytic systems is limited to compounds of noble metals, partic-
ularly platinum, e.g. Speier, Karstedt catalysts [1]. The possibility of
replacing expensive platinum catalysts with catalytically active
derivatives of cheaper transition metals such as Ru or Fe, is a
promising alternative and a challenge for many research teams.
Recently, Chirik, Ritter groups published results on the application
of bis(imino)pyridine and terpyridine and iminopyridine iron
complexes in the hydrosilylation of alkenes by primary, secondary
catalyst for the dehydrogenative silylation at one vinyl group and
hydrogenation at the second one in 1,3-divinyldisiloxane [5]. The
iron triad (Fe, Ru and also Os) complexes including carbonyls,
competitively catalyse hydrosilylation and dehydrogenative silyla-
tion [1]. The key step of this mechanism is the insertion of an olefin
into MeH or MeSi bonds to give products of hydrosilylation and/or
dehydrogenative silylation [1b] and references therein]. The
desired step of the two alternative reactions is actually the
competitive
b-H transfer from the two ligands (s-alkyl and s-
silylalkyl) in the complex formed during the reaction (see
Scheme 1). A theoretical study has recently explained that the
complexes of Ni, Pd, Pt are preferred in the hydrosilylation of ole-
fins, while the complexes of Fe (also to some extent Co) have been
shown to be extremely favourable catalysts for the dehydrogen-
ative silylation of olefins, particularly with electron withdrawing
substituents [1,6,7]. It is well-recognized that iron complexes such
as [Fe(CO) ] at high temperature and/or upon UV irradiation under
5
mild conditions catalyse both reactions, i.e. dehydrogenative sily-
lation and hydrosilylation [8]. The mechanism proposed by
*
Corresponding author. Department of Organometallic Chemistry, Faculty of
Chemistry, Adam Mickiewicz University in Pozna nꢀ , Umultowska 89b, 61-614
Poznan, Poland. Tel.: þ48 618291990; fax: þ48 618291987.
E-mail address: bogdan.marciniec@amu.edu.pl (B. Marciniec).
http://dx.doi.org/10.1016/j.jorganchem.2015.04.051
0
022-328X/© 2015 Elsevier B.V. All rights reserved.

B. Marciniec et al. / Journal of Organometallic Chemistry 791 (2015) 58e65
59
Scheme 1. Mechanism of hydrosilylation vs. dehydrogenative silylation reaction.
Randolph and Wrighton assumed important role of irradiation in
Catalytic examinations
3
the formation of catalytically active [Fe(CO) ] species via photo-
dissociation of the FeeCO bond [9].
The tests on the catalytic activity of iron(0) complexes were
conducted in glass ampoules (or Schlenk tubes) filled with a
mixture of styrene (3.6 mmol, 0.42 mL), trisubstituted silane
(1.8 mmol or 3.6 mmol), toluene 1 mL and decane (5% of initial
liquid substrates volume) as an internal standard, under argon at-
mosphere. Then to the initial mixture iron(0) catalyst
(1.8$10 mmol) was added. The ampoules (or Schlenk tubes) were
heated to a desired temperature in oil bath. The catalytic reactions
at low temperature were carried out using the kinetic reactor
cooled with chiller cooling system.
We have recently developed a general strategy for the synthesis
of well-defined iron(0) carbonyl complexes ([Fe(CO) (L)]) or
{Fe(CO) L] stabilized by multivinylsilicon ligands (where
L ¼ diene, triene, tetraene, polyvinyl silicon derivatives) (Scheme
). Their structures were determined by spectroscopic and X-ray
3
[
3 2
}
2
ꢀ
2
methods [10a,b].
The aim of this paper is to present a full picture of the catalysis of
hydrosilylation vs. dehydrogenative silylation of styrene by these
new Fe(0) complexes (1e9) (Scheme 2). [Fe(CO)
Fe(CO) ] (11) were used for comparison.
3
(cod)] (10) and
[
5
The distribution of substrates and products, conversion of sub-
strates and the yield of products in all tests were determined by GC
and GCeMS analysis.
Experimental
Kinetic examinations
All syntheses and manipulations were carried out under argon
using standard Schlenk and vacuum techniques. The H NMR
spectra were recorded using a Bruker Ultra Shield NMR spec-
8
trometer (600 MHz) in toluene-d and referred to the residual
Kinetic measurements were carried out in a reactor (4 mL)
equipped with a reflux condenser, inlet for inert gasses and a
magnetic stirrer. Appropriate amounts of the catalyst and reagents
were placed in the reactor. GC and GCMS analyses were carried out
during the reaction. The conversion of substrates and the yield of
products were calculated using the internal standard method (see
Supplementary Information).
1
1
13
protonated solvent peaks ( H
d
H
¼
2.09 ppm (eMe),
C
d
C
¼ 20.05 ppm (eMe) for toluene-d
8
). The volatile compounds
were determined by GCeMS (Varian Saturn 2100T equipped with
capillary column Varian VF-1 Factor Four, 0.26 mm, 30 m). GC an-
alyses were carried out on a Varian 3400 CX series gas chromato-
graph equipped with a capillary column Varian VF-5 Factor Four,
Selected kinetic curves of the dehydrogenative silylation of
styrene at the molar ratio [HSiMe
hydrosilylation of styrene at the molar ratio [HSiMe
2
Ph]: [PhCH ¼ CH
2
] ¼ 1: 2 and
2
Ph]:
3
0 m and TCD. Yield was calculated on the basis of HSiR
3
conver-
[PhCH ¼ CH ] ¼ 1: 1, in the presence of iron(0) complex 1 were used
2
HD H
sion, using the internal standard calculation method according to
to determine kobs and kobs, respectively. The rate constants were
calculated by fitting the experimental concentration 2.303log
[1ꢀcHSi≡] vs. time to an exponential function using the nonlinear
least-squares procedures. The activation energy for both reaction
was calculated from the Arrhenius plot.
the formula Y(%) ¼ S$CHSi≡ (S ¼ selectivity of the product obtained,
C
HSi≡ ¼ conversion of trisubstituted silane (%)). The chemicals were
obtained from the following sources: HSiEt
tamethyltrisiloxane (HSiMe(OSiMe ) from ABCR, decane, styrene,
toluene-d from Sigma Aldrich Co., pentane, toluene, benzene from
3 2
, HSiMe Ph and hep-
3 2
)
8
POCH Gliwice (Poland). All solvents and liquid reagents were dried
and distilled under argon prior to use. The iron(0) complexes used
for catalytic examinations were synthesized according to published
method [10].
Theoretical calculations
Quantum-chemical calculations were performed with the
Gaussian 03 program [11]. Initial geometries of substrates and
Scheme 2. Formulae of iron(0) complexes with multivinyl silicon ligands used for the catalytic examinations.

60
B. Marciniec et al. / Journal of Organometallic Chemistry 791 (2015) 58e65
products were built on the basis of the averaged values of bonds
lengths and valence angles with the help of Marvin package [12]
and were subjected to full geometry optimization. The calcula-
tions were carried out at the most popular for such systems and
probably that is why most often criticized [13] B3LYP/LANL2DZ
level [14,15] being a compromise between the quality of obtained
results and the computation time needed. Vibrational analyses
were undertaken to verify if the optimized stationary point was
indeed a minimum or a saddle point (transition state) [16]. In the
case of transition state (one imaginary frequency) the IRC [17]
calculations were carried out to identify the energy minima on
both sides of the transition state. Interaction energies between the
Fe catalyst and the alkene were calculated with the usual coun-
terpoise correction to account at least in some part for basis set
superposition error [18].
centre, as reported in our previous paper [10b].
The catalytic results (Fig. 1) illustrate high effectiveness of most
of these new iron complexes in the dehydrogenative silylation of
ꢁ
styrene at 80 C leading to the exclusive formation of unsaturated
organosilicon product.
In the presence of some catalysts (1, 4, 5, 9) the total conversion
of substrates was observed already after a few hours. In all exper-
iments performed, complexes 1, 4, 5 and 9 were good catalysts with
1 being the most effective and selective for dehydrogenative sily-
ꢁ
lation at temperature 80 C, while [Fe(CO)
3
(cod)] (10) and [Fe(CO)
5
]
(11) have proven to be moderately active in the studied systems.
ꢁ
However, on decreasing temperature to the range ꢀ15 to þ20
C
under equimolar substrate concentrations, complex 1 turned out to
be active for the hydrosilylation reaction leading to selective for-
mation of saturated hydrosilylation product (A) (Scheme 3 and
Table 1) with good yield, whereas at double excess of olefin, both
reactions occurred.
Stoichiometric reaction of complex 1 with HSiMe
2
Ph and styrene
Comparing the catalytic results presented in Fig. 1 with the data
of DFT computational study, in particular the calculated interaction
energies (see Table 1) of the systems composed of the vinylsilicon
ꢀ
4
In the NMR Young's tube, 45 mg (1.38$10 mol) of iron complex
(
1) and 0.6 mL of toluene-d
8
were placed under an argon atmo-
ꢁ
sphere. The solution obtained was cooled to ꢀ60 C, and then the
3
ligand and the Fe(CO) core, confirmed that complex 1 with
1
tube was placed in a spectrometer. The first H NMR spectrum was
(H O molecule bonded to iron is the most active
2
C¼CHSiMe
2 2
)
ꢁ
ꢀ4
recorded at ꢀ40 C. In the next step, 18.80 mg (1.38$10 mol)
catalyst. In this compound, 1,3-divinyltetramethyldisiloxane
molecule interacts with metallic centre weaker than the other
ꢁ
HSiMe
2
Ph was added to the solution of the complex at ꢀ60 C. The
ꢁ
NMR spectra were recorded starting from ꢀ40 C on increasing
divinylsubstituted silicon derivatives such as (H
(H C]CH) SiPh . Lower compliance of catalytic results with theo-
retical calculations was found for the complexes stabilized by
trivinylsilanesilanes such as (H C]CH) SiMe, (H C]CH) SiPh (for
2 2 2
C]CH) SiMe ,
ꢁ
ꢁ
temperature by about 10 C in each 30 min to reach ꢀ10 C. Thirty
2
2
2
1
minutes after reaching the desired temperature, the next H NMR
spectrum was recorded, and then at this temperature 15 mg
1.38$10 mol) of styrene was introduced to the NMR tube. The
subsequent spectra were recorded after each 30 min on raising
temperature at a rate of 10 C per 30 min, up to room temperature.
2
3
2
3
ꢀ
4
(
more information see Electronic Supplementary Data), which
proved to show higher catalytic activity than expected on the basis
of the higher energy value of the trivinyl silicon ligand interactions
with the metal centre.
ꢁ
Stoichiometric reaction of complex 1 with styrene
The kinetic measurements performed in the temperature range
of 50e70 C for dehydrogenative silylation and of 0e10 C for
ꢁ
ꢁ
In the NMR Young's tube, 32.6 mg (9.2$10ꢀ5 mol) of iron com-
hydrosilylation, followed by GC, enabled the calculation of the
HD
H
plex (1) and 0.6 mL of toluene-d
mosphere, then the first H and C NMR spectra were recorded at
8
were placed under an argon at-
pseudo first-order rate constants kobs and kobs, respectively. The
1
13
basis of both processes was the consumption of the substrate
ꢀ
4
room temperature. At the next step 38.3 mg (3.68$10 mol) of
styrene was introduced to the NMR tube and the reaction was
carried out for 24 h at room temperature. After this time, H and
2
(HSiMe Ph) (see Table 2).
The effect of temperature on the kobs and k oH bs permitted the
HD
1
13
C
estimation of activation energy Ea, for dehydrogenative silylation
NMR spectra were recorded again.
Results and discussion
Catalytic study
(15.8 [kcal/mol], [HSiMe
hydrosilylation (15.9 [kcal/mol], [HSiMe
(Fig. 3).
2
Ph]: [PhCH ¼ CH
2
] ¼ 1: 2) (Fig. 2) and
2
Ph]: [PhCH ¼ CH ] ¼ 1: 1)
2
The reactivity of iron complex 1 was also studied in the reaction
1
with HSiMe
2
Ph in which the H NMR measurements were started
ꢁ
at ꢀ40 C. The first signs of HSiMe
2
Ph oxidative addition to the
ꢁ
All the iron(0)carbonyl complexes (1e11) were tested in the
reaction of styrene with trisubstituted silanes Et SiH, Me PhSiH
and Me(OSiMe SiH. Here it should be mentioned that in the case
initial iron precursor were observed at ꢀ20 C, as at this temper-
3
2
ature the resonance at ꢀ11.34 ppm appeared. An increase in tem-
ꢁ
3
)
2
perature up to ꢀ10 C caused a significant increase in the height of
of binuclear complexes, particularly those stabilized by tetravinyl
substituted cyclosilicon derivatives such as 1,3,5,7-tetravinyl-
this signal, but addition of styrene to the reaction mixture
ꢁ
at ꢀ10 C, led to total disappearance of the resonance line
1,3,5,7-tetramethyl- cyclotetrasiloxane or cyclotetrasilazane more
at ꢀ11.34 ppm with the simultaneous appearance of new lines
at ꢀ9.60 ppm (coming from a new hydride-silyl iron(II) species) as
well as ꢀ0.36 and ꢀ0.46 ppm (derived from alkyl complexes
formed after insertion of styrene molecule into the FeeH bond), as
shown in Scheme 5.
geometrically matched isomeric molecules, present in the initial
silicon reagent, should be involved in the formation of more ther-
modynamically stable iron(0) carbonyl complexes. Therefore, when
in the synthesis of complex 7, the compound 1,3,5,7-tetravinyl-
1,3,5,7-tetramethylcyclotetrasiloxane was used as a ligand only one
Considering the DFT calculations of interaction energies of
isomer in which all vinyl groups are in axial position, can be
involved in the formation of binuclear complex. However, when
iron(II) species [FeH(CO)
(24.8 kcal/mol) and H
C¼CHPh (22.8 kcal/mol) (see Electronic
Supplementary Data, Table S3) it may be assumed that in the first
step of the oxidative addition of HSiMe Ph to the initial iron(0)
carbonyl complex, the formation of hydride-silyl iron(II) species
3
(SiMe
2
Ph)] with (H
2
C¼CHSiMe
2 2
) O
2
1,3,5,7-tetravinyl-1,3,5,7-tetramethylcyclotetrasilazane was used
instead of cyclosiloxane, the solid material used in catalytic tests
can contain two isomeric complexes 8, namely those formed with
the employment of two isomeric cyclosilazanes, in which all vinyl
groups are in axial position or three vinyl groups are in axial po-
sition and one nitrogen atom is involved in stabilization of iron
2
2
with
h -bonded 1,3-divinyltetramethyldisiloxane ligand is
preferred, according to Scheme 4 (see Electronic Supplementary
Data, Fig. 1). However, in the next step, the addition of styrene to

B. Marciniec et al. / Journal of Organometallic Chemistry 791 (2015) 58e65
61
ꢁ
-2
Fig. 1. Dehydrogenative silylation of styrene with trisubstituted silanes. Reaction conditions: [HSiR
toluene, under argon; (a) e 1h.
3
] : [PhCH¼CH ] : [Fe] ¼ 1 : 2 : 10 , Temperature ¼ 80 C, Reaction time ¼ 24h,
2
reaction had an opportunity to react with two additional moles of
unsaturated groups coming from 1,3-divinyltetramethyldisiloxane
ligand, besides the vinyl group coming from styrene which was
added to the reaction mixture according to the molar ratio [Fe]:
[
HSi≡]: [H
2
C¼CHPh] ¼ 1: 1: 1. The GCeMS analysis of the reaction
mixture obtained, unambiguously confirmed the presence of 1,3-
divinyltetramethyldisiloxane derivatives, such us hydrosilylation
and hydrogenation products, e.g. H
Me
Ph (Yield ¼ 8%) (MS data (%): 322.3 (M, 5.2); 307.2 (Mꢀ15,
12.5); 293.3 (8.5); CH CH SiMe OSiMe CH CH SiMe Ph
Yield ¼ 24%) (MS data (%): 309.2 (Mꢀ15, 75.0); 295.2 (23.5)).
Additionally, in order to check for other pathways of trans-
2
C¼CHSiMe
2 2 2 2
OSiMe CH CH Si
2
Scheme 3. General scheme of hydrosilylation vs. dehydrogenative silylation of styrene.
3
2
2
2
2
2
2
(
the reaction mixture containing of the latter iron(II) species results
formation of vinylsilicon-stabilized iron(0) precursors, for example
the substitution process of divinylsilicon ligands with the styrene,
the reaction of complex 1 with the latter at ambient conditions was
also performed and followed by NMR technique. Introduction of
styrene to the solution of complex 1 resulted in significant changes
2
in the formation of hydride-silyl iron(II) complex with
h
-bonded
styrene molecule which formally initiate the formation of silicon
styrene derivatives, as shown in Scheme 5.
Analysis of the reaction mixture obtained under these condi-
tions by GCeMS technique confirmed the formation of the product
1
13
in the H and C NMR spectra with respect to those obtained for
the solution of the initial complex (see Electronic Supplementary
2 2 2
of styrene hydrosilylation (PhCH CH SiMe Ph).
1
The quantitative calculations revealed that in the temperature
Data, Fig. 2). In the H NMR spectrum recorded after addition of
ꢁ
range from ꢀ40 to 0 C, only 21% of the initial amount of HSiMe
2
Ph
styrene to the solution of complex 1 the appearance of new lines at
was consumed. It was only when the reaction mixture had reached
a temperature 10 C, that a rapid conversion of the starting silane at
6.16, 5.91, 5.75, 5.25 as well as 0.17 ppm confirmed the presence of
ꢁ
free vinyldimethylsilyl moiety as part of 1,3-
a
this temperature was observed and it reached a value of 92%. Under
these conditions, only 42% of the initial amount of HSi≡ reagent was
involved in the formation of styrene hydrosilylation product. The
divinyltetramethyldisiloxane ligand bonded to Fe metallic centre.
Moreover, new resonances observed in this spectrum and located
at 3.26 and 2.80 could be assigned to hydrogen atoms coming from
new vinyl groups bonded to the iron metallic centre. In the C NMR
spectrum, significant changes were also observed; particularly
interesting was the appearance of two new singlets at 73.15 and
13
2
point is that in the system studied, HSiMe Ph applied for the
Table 1
6
8.63 ppm, that is in the region of the chemical shifts of carbon
Calculated interaction energies (counterpoise corrected) between the vinylsilicon
atoms, typical of vinyl groups coordinated to the metallic centre,
which may indicate the coordination of styrene to the iron atom
and the formation of a new iron(0) complex. The formation of a
new iron(0) carbonyl complex is also supported by the appearance
3
ligand and Fe(CO) moiety.
Complex
Alkenea
Interaction energy
kcal/mol]
[
1
2
3
4
5
(H
(H
(H
(H
(H
2
2
2
2
2
C¼CHSiMe
2
)
2
O
ꢀ62.8
ꢀ68.5
ꢀ67.4
ꢀ68.8
ꢀ68.8
of a new resonance at 211.82 ppm typical of CO ligands. Addition-
C]CH)
C]CH)
C]CH)
C]CH)
2
2
3
3
SiMe
2
2
ally, the signals coming from free vinyl group of
h
-bonded 1,3-
SiPh
SiMe
SiPh
2
divinyltetramethyldisiloxane ligand at 139.63 (eSiCH¼), 131.84
(
¼CH
2
) and 0.51 (eMe) ppm were also detected. Taking into ac-
count the above-described results of spectroscopic studies, it may
be assumed that in the initial step of the catalytic cycle, the
a
Despite much effort we did not obtain the convergence of SCF calculations for
systems containing two Fe centres.

62
B. Marciniec et al. / Journal of Organometallic Chemistry 791 (2015) 58e65
Table 2
divinyltetramethyldisiloxane coordinated the central Fe atom. In
the lowest energy structure, the C]C ligands were in equatorial
orientation. (for more information see Electronic Supplementary
Data, Table S2).
The above Scheme 5 presents the possible routes for the for-
mation of hydrosilylation and dehydrogenative silylation products
Hydrosilylation (A) vs. dehydrogenative silylation (B) of styrene with HSiMe
catalysed by complex 1 at the given temperatures.
2
Ph
ꢁ
[HSiMe
2
Ph]: [Styrene] Temp. [ C] Time [h] Conversion Yield
k
obs
of HSiR
3
[%] A(B) [%]
H
1: 1
ꢀ15
0
5
48
24
24
24
24
68
68
67
59
100
60(8)
68(0)
67(0)
59(0)
12(88)
e
in the HSiMe
sults of the stoichiometric study, i.e. the NMR spectroscopic data for
the reaction of complex 1 with HSiMe Ph and styrene at different
2
Ph/styrene system, catalysed by iron complex 1. Re-
0.1189
0.1463
0.1870
e
1
2
0
0
2
temperatures, were supported by theoretical findings.
Relative energies of the transition states as well as intermediates
were calculated at DFT level and the energy profile for the catalysis
DS
1: 2
ꢀ15
48
24
48
24
24
24
24
68
80
83
100
100
100
100
46(22)
54(26)
56(27)
10(90)
e
e
e
e
0
of the reaction of HSiMe
plex [Fe(CO) {(H
DFT calculations indicate that when the iron complex (1) is used,
2
Ph with styrene in the presence of com-
O}] (1) is documented in Fig. 4. The
1
2
5
6
7
0
0
0
0
0
3
2
C¼CHSiMe
2 2
)
0(100) 0.0944
0(100) 0.2151
0(100) 0.4698
the energy barriers appear to be the lowest for the reaction of
HSiMe Ph with styrene according to the Chalk-Harrod mechanism,
2
through the insertion of olefin into the FeeH bond. However, when
the reaction proceeds through the insertion of olefin into the FeeSi
bond, the process of dehydrogenative silylation is preferred as the
energy barrier for the formation of dehydrogenative silylation
Reaction conditions: [Fe] ¼ 10ꢀ2, toluene, argon.
H e Hydrosilylation, DS e Dehydrogenative Silylation.
3 2
product and complex [Fe(CO) (H) ] as a result of b-transfer of a
hydrogen atom, is lower than that for the reductive elimination
process and formation of a CeH bond in the hydrosilylation
product.
The results concerning the catalytic activity of iron(0) com-
plexes stabilized by vinyl silicon ligands and the reactivity of
[
Fe(CO)
3
{(H
2
C¼CHSiMe
2
)
2
O}] (1) precursor in equimolar reactions
]” is
/PhCH ¼ CH
with selected substrates, confirmed that the core “[Fe(CO)
involved in the transformations occurring in the HSiR
system (see Scheme 7).
3
3
2
According to the spectroscopic analysis, during the equimolar
reactions, 1,3-divinyltetramethyldisiloxane ligand is removed from
the coordination sphere of iron under the influence of substrates in
3 3
the reaction of HSiR , leading to the formation of complex [Fe(CO) ]
Fig. 2. Arrhenius plot of the dehydrogenative silylation of styrene with dimethyl-
phenylsilane catalysed by complex 1.
(a) which is the formal catalytic intermediate (see Scheme 5). In the
first step, an oxidative addition of hydrosilane (complex (b)) fol-
lowed by olefin coordination, or alternatively the coordination of
olefin (complex (b )) (see Scheme 7) followed by the oxidative
addition of SieH lead to formation of the six-coordinated complex
0
transformation of the starting iron precursors to five-coordinated
iron species may also take place, according to Scheme 6.
(c). In the next step, depending on temperature and substituents at
3
{ -
h2
The DFT calculations performed for the complex [Fe(CO)
O}(CH
silicon, the insertion of olefin either into the FeeH bond (complex
(
H
2
C¼CHSiMe
2
)
2
2
¼CHPh)] (see Scheme 6), in which the
0
(
d)) or into the FeeSi bond (complex (d )) is preferred.
central Fe atom coordinated in square-planar geometry or tetrag-
onal pyramid, during optimization showed a tendency to converge
The DFT calculations have shown that interaction energy be-
2
tween iron centre and
h
-bonded 1,3-divinyltetramethyldisiloxane
towards trigonal bipyramid structure in which the
p electrons of
is higher than in the complex with styrene (see Electronic Sup-
plementary Data, Table S3). We have confirmed formation such
type of complex by NMR, but in the first catalytic run this ligand is
removed from the coordination sphere as a product of dehydro-
genative silylation process, as we reported in previous paper [19].
However, even if somewhat of this complex will remain in solution,
in the real catalytic system, excess of styrene will shift the equi-
librium towards the formation of iron species (c) with bonded
styrene molecule.
C]C moieties from styrene and one terminus of 1,3-
The reaction of styrene with HSiMe
2
Ph, at low temperatures
ꢁ
(
ꢀ15 e 10 C), catalysed by complex 1 showed a very high selec-
tivity for silane addition to the vinyl group in the styrene molecule
for equimolar ratio of substrates. Evidence of the preference for the
hydrosilylation reaction was also obtained from the spectroscopic
2
study of sequential reactions of iron complex 1 with HSiMe Ph and
styrene in equimolar amounts at low temperatures.
These experimental data show that under such reaction condi-
tions, the process of styrene insertion into the FeeH bond (c) takes
place, leading to the formation of complex (d), according to the
Chalk-Harrod mechanism.
Fig. 3. Arrhenius plot of the hydrosilylation of styrene with dimethylphenylsilane
catalysed by complex 1.

B. Marciniec et al. / Journal of Organometallic Chemistry 791 (2015) 58e65
63
2
Scheme 4. The reaction of complex 1 with HSiMe Ph.
2
Scheme 5. Possible routes for the formation of the hydrosilylation and dehydrogenative silylation products in the HSiMe Ph/styrene system in the presence of complex 1.
Scheme 6. The reaction of complex 1 with styrene.
2
Fig. 4. Energy profile for the two possible reactions of styrene with HSiMe Ph, i.e. Hydrosilylation (H) and Dehydrogenative Silylation (DS).

64
B. Marciniec et al. / Journal of Organometallic Chemistry 791 (2015) 58e65
Scheme 7. Mechanism of hydrosilylation (H) vs. dehydrogenative silylation (DS) of styrene at given temperatures.
The results of DFT calculations seem to confirm the thesis for a
is compensated by the energy gained at the next stage (e)/(a), in
which hydrogenation of styrene takes place. When the stoichio-
metric amounts of reagents are taken at a low reaction tempera-
ture, the transformation (a)/(b)/(c)/(d)/(a), i.e. towards the
hydrosilylation product is preferred, because of the high-energy
transformation of this type, in spite of the energy barrier for styrene
insertion into the FeeSi bond (18.0 [kcal/mol]) being slightly lower
than that of styrene insertion into the FeeH bond (19.5 [kcal/mol]),
intermediate (d) is created, due to the energy release which ac-
companies this process (3.5 [kcal/mol]). Another important factor
that supports the preference of SieH addition to styrene under
these conditions, according to the Chalk-Harrod hydrosilylation
mechanism, is a relatively low value of the energy barrier calcu-
lated for (d)/(a) (18.0 [kcal/mol]) transformation (14 kcal/mol)
0
0
barriers of stages (d )/(a) and (d )/(e).
Conclusions
0
3 3 2
The new iron(0) carbonyl catalysts [Fe(CO) L] or [{Fe(CO) } L ]
which is the determining step of the hydrosilylation (see
ꢁ
E
a
¼ 15.9 kcal/mol). At higher temperatures (i.e. above 20 C and
stabilized by multivinylsilicon ligands (where L ¼ silicon dienes,
trienes, tetraenes and polyvinylsilicon derivatives), whose synthe-
ses and structures determined by spectroscopic and X-ray methods
we reported previously [10], were tested in selected reactions of
trisubstituted silanes and hydrosiloxanes with styrene and show
good activity compared to those of the other well-known iron
ꢁ
particularly 50e70 C) at which the insertion of styrene into the
FeeSi bond takes place, the energy barrier (calculated by DFT)
accompanying by the elimination of hydrosilylation product
(
d)/(a) would require higher energy (19.5 kcal/mol) then the
formation of intermediate (d)/(e) with evolution of silylsub-
stituted styrene (16.5 kcal/mol). The latter is the rate-determining
3 5
carbonyls- [Fe(CO) (cod)] and [Fe(CO) ]) in the hydrosilylation or
step for selective dehydrogenative silylation (E
a
¼ 15.8 kcal/mol).
dehydrogenative silylation reactions of styrene.
The results show that depending on temperature and molar
ratio of reagents, the process can run either towards selective for-
mation of hydrosilylation (according to the Chalk-Harrod mecha-
nism) or dehydrogenative silylation (according to the modified
Chalk-Harrod mechanism) products.
The results obtained show that depending on temperature and
molar ratio of reagents, the process can run either towards the
selective formation of hydrosilylation or dehydrogenative silylation
product.
Kinetic measurements of selected reactions as well as NMR
It was also found that with increasing excess of styrene in
3
studies of stoichiometric reactions of [Fe(CO) (DVTMDS)] with the
relation to HSiMe
selectivity of the unsaturated product formed through the
sequence (a)/(b)/(c)/(d )/(e)/(a) increased. However, the
maximum value of energy barrier for the transformation (d )/(e)
2
Ph, although the temperature was low, the
substrates, supported by DFT determination of the energy of rela-
tive transition states and intermediates, enabled us to suggest a
general scheme for the catalysis of hydrosilylation and/or of
dehydrogenative silylation by these Fe(0) carbonyl precursors.
0
0

B. Marciniec et al. / Journal of Organometallic Chemistry 791 (2015) 58e65
65
Acknowledgements
[8] (a) A.N. Nesmeyanov, R. Kh Friedlina, E.C. Chukovskaya, R.G. Petrova,
A.B. Belyavsky, Tetrahedron Lett. 17 (1962) 61;
(
(
b) F. Kakiuchi, Y. Tanaka, N. Chantani, S. Murai, J. Organomet. Chem. 456
1993) 45.
The financial support from Dow Corning Corporation is grate-
fully acknowledged. This research was supported in part by PL-Grid
Infrastructure.
[9] C. Randolph, M.S. Wringhton, J. Am. Chem. Soc. 108 (1986) 3366.
10] (a) M.-S. Tzou, R. Taylor, I. Kownacki, B. Nguyen, B. Marciniec, A. Kownacka, A.
Surgenor, US Pat App. WO 2013081794; (b) A. Kownacka, I. Kownacki,
M. Kubicki, B. Marciniec, R. Taylor, J. Organomet. Chem. 750 (2014) 132.
[
Appendix A. Supplementary data
[11] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Rob, J.R. Cheeseman,
J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene,
X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli,
J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador,
J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain,
O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz,
Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng,
A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen,
M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Gaussian, Inc., Wallingford,
CT, 2003.
[12] http://www.chemaxon.com/products/marvin/.
[13] A. Plutecka, M. Hoffmann, U. Rychlewska, Struct. Chem. 18 (2007) 379.
[14] (a) A.D. Becke, J. Chem. Phys. 98 (1993) 5648;
(b) C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[15] (a) T.H. Dunning Jr., P.J. Hay, in: H.F. Schaefer III (Ed.), Modern Theoretical
Chemistry, vol. 3, Plenum, New York, 1976, pp. pp.1e28;
(b) P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 (1985) 270.
[16] W.J. Hehre, L. Radom, P.V. Schleyer, J. Pople, AB Initio Molecular Orbital
Theory, Wiley-Interscience, 1986.
[17] H.P. Hratchian, H.B. Schlegel, in: C.E. Dykstra, G. Frenking, K.S. Kim, G. Scuseria
(Eds.), Theory and Applications of Computational Chemistry: the First 40
Years, Elsevier, Amsterdam, 2005, pp. 195e249.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2015.04.051.
References
[
1] (a) B. Marciniec, J. Guli nꢀ ski, W. Urbaniak, Z.W. Kornetka, in: B. Marciniec (Ed.),
Comprehensive Handbook on Hydrosilylation, Pergamon Press, Oxford, 1992,
p. 754;
(
(
b) B. Marciniec, H. Maciejewski, C. Pietraszuk, P. Pawlu ꢀc , in: B. Marciniec
Ed.), Hydrosilylation. Comprehensive Review on Recent Advances,
A
Springer, Berlin, 2009, p. 410.
[
[
2] C. Suzanne, C. Bart, E. Lobkovsky, P.J. Chirik, J. Am. Chem. Soc. 126 (2004)
13794.
3] (a) P.J. Chirik, Science 335 (2012) 567;
(
(
(
b) P.J. Chirik, et al., Organometallics 31 (2012) 4886;
c) J. G. P. Delis, P. J. Chirik, A. M. Tondreau, US 2011/0009565 A1;
d) J. G. P. Delis, S. A. Nye, K. M. Lewis, K. J. Weller, P. J. Chirik, A. M. Tondreau,
S. K. Russell, US 2011/0009573 A1;
e) J. G. P. Delis, S. A. Nye, K. M. Lewis, K. J. Weller, P. J. Chirik, A. M. Tondreau,
(
Patent Application: WO 2011/006044 A2; (f) H. Nakazawa, K. Kouji, A. Suzuki,
Y. Nakai, Organometallics 31 (2012) 3825;
(
g) K. J. Weller, C. C. H. Atienza, J. Boyer, P. Chirik, J. G. P. Delis, K. Lewis, S. A.
Nye, US2013/0158281 A1.
[4] J.Y. Wu, B.N. Stanzl, T. Ritter, J. Am. Chem. Soc. 132 (2010) 13214.
[5] H. Nakazawa, J. Am. Chem. Soc. 134 (2011) 804.
[6] M. Zhang, A. Zhang, Appl. Organometal. Chem. 24 (2010) 751.
[7] C.A. Tsipis, C.E. Kefalidis, J. Organomet. Chem. 692 (2007) 5245.
[18] S. Simon, M. Duran, J.J. Dannenberg, J. Chem. Phys. 105 (1996) 11024e11031.
[19] B. Marciniec, A. Kownacka, I. Kownacki, R. Taylor, Appl. Catal. A: General 486
(2014) 230.