Transition metal-catalyzed hydrosilylation of polybutadiene – The effect of substituents at silicon on efficiency of silylfunctionalization process

Article abstract of DOI:10.1016/j.jcat.2019.01.026

Herein we present the results of our studies on the hydrosilylation of polybutadiene with alkyl, aryl and alkoxysilanes in the presence of platinum and rhodium complexes. In order to select the most efficient catalytic system, which, under optimal conditions, would smoothly promote incorporation of the above-mentioned organosilicon modifiers into polybutadienes of various structures via hydrosilylation, the real-time in-situ FT-IR spectroscopy was used. The silyl-functionalized polymeric products were characterized by NMR analysis and gel permeation chromatography (GPC). It was demonstrated that the stereo-electronic properties of substituents directly bonded to the HSi moiety play a crucial role in formation of desired products, as well as affect the time required for total conversion of organosilicon reagents. Partially modified polymers containing pendant alkoxy groups can be applied as additives to rubber compounds to enhance dispersion of inorganic particles in the polymer matrix, as well as to promote formation of organic-inorganic hybrid materials.

Full text of DOI:10.1016/j.jcat.2019.01.026

Journal of Catalysis 371 (2019) 27–34
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Transition metal-catalyzed hydrosilylation of polybutadiene – The effect
of substituents at silicon on efficiency of silylfunctionalization process
Rafał Januszewski a,b, Ireneusz Kownacki a,b, , Hieronim Maciejewski a,c, Bogdan Marciniec b
⇑
a Faculty of Chemistry, Adam Mickiewicz University in Poznan, Umultowska 89b, 61-614 Poznan, Poland
b Center for Advanced Technology, Adam Mickiewicz University in Poznan, Umultowska 89c, 61-614 Poznan, Poland
c Poznan Science and Technology Park, Adam Mickiewicz University Foundation, Rubiez 46, 61-612 Poznan, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein we present the results of our studies on the hydrosilylation of polybutadiene with alkyl, aryl and
alkoxysilanes in the presence of platinum and rhodium complexes. In order to select the most efficient
catalytic system, which, under optimal conditions, would smoothly promote incorporation of the
above-mentioned organosilicon modifiers into polybutadienes of various structures via hydrosilylation,
the real-time in-situ FT-IR spectroscopy was used. The silyl-functionalized polymeric products were char-
acterized by NMR analysis and gel permeation chromatography (GPC). It was demonstrated that the
stereo-electronic properties of substituents directly bonded to the HSi moiety play a crucial role in for-
mation of desired products, as well as affect the time required for total conversion of organosilicon
reagents. Partially modified polymers containing pendant alkoxy groups can be applied as additives to
rubber compounds to enhance dispersion of inorganic particles in the polymer matrix, as well as to pro-
mote formation of organic-inorganic hybrid materials.
Received 22 October 2018
Revised 17 January 2019
Accepted 21 January 2019
Keywords:
Rubber
Polymer functionalization
Homogeneous catalysis
Dehydrogenative silylation
In-situ FT-IR
Ó 2019 Elsevier Inc. All rights reserved.
1. Introduction
with unique reactivity and properties. For example, the incorpora-
tion of alkoxysilyl groups into a PB chain significantly increases the
Chemical modification of polymers has recently emerged as a
topic of increasing importance in macromolecular chemistry and
can often lead to novel materials with desirable functional groups
which are otherwise inaccessible, difficult or too expensive to pre-
pare by conventional polymerization techniques. Unsaturated
polymers, especially diene-based polymers, are excellent polymers
which can be exploited in addition reactions due to the technolog-
ical importance associated with the parent materials and the reac-
tivity of the carbon-carbon double bonds in the polymer chain [1].
Polybutadiene (PB) is a versatile starting material due to its rela-
tively low costs and high content of easily accessible double bonds,
which enable polymer modification using a wide range of chemical
reactions, such as epoxidation [2], silylative coupling [3], hydrocar-
boxylation [4], hydroformylation [5], hydroboration [6], thiol addi-
tion [7], hydrozirconation and subsequent halogenolysis [8] and
hydrosilylation [1,9]. The above-mentioned modifications are
directly related to the search for the best method for the synthesis
of polymers with controlled degrees of functional groups for
custom-designing specialty polymer systems and building blocks
dispersion of inorganic particles in the polymer matrix [10–12],
and the presence of epoxy rings permits curing with epoxy resins
and obtaining materials with enhanced thermal and mechanical
properties [13–14]. From the wide gamut of catalytic transforma-
tions, the hydrosilylation reaction seems to be very convenient
for the functionalization of polybutadiene, due to its high degree
of unsaturation. This process is characterized by high selectivity
and tolerance of a wide range of functional groups [15–17] and
can be promoted by various TM-based complexes (TM = transition
metal), i.e. platinum [18], rhodium [19], cobalt [20], iron [21], irid-
ium [22] and ruthenium [23].
In the view of literature data, catalytic hydrosilylation has been
used for modification of polybutadienes (PBs), mostly with com-
monly known silanes such as trieothoxysilane [1,24–26], chlorosi-
lanes [25,27–29], alkylarylsilanes [1,25], trialkylsilanes [1,9,25] or
more complex silicon-based reagents, namely silsesquioxanes
[30], siloxanes and polysiloxanes [25,31–32]. In order to incorpo-
rate the above-mentioned modifiers via hydrosilylation, rhodium
Wilkinson’s catalyst [RhCl(PPh3)3] [1,9,28–29,32] platinum Karst-
edt’s catalyst [24,26,30], H2PtCl6 acid [27] and platinum nanoparti-
cles [25] were employed as promotors. Nonetheless, the studies of
silylfunctionalization of butadiene-based polymers, described in
cited references, have been carried out with the use of polyene
⇑
Corresponding author at: Faculty of Chemistry, Adam Mickiewicz University in
Poznan, Umultowska 89b, 61-614 Poznan, Poland.
E-mail address: irekk@o365.amu.edu.pl (I. Kownacki).
https://doi.org/10.1016/j.jcat.2019.01.026
0021-9517/Ó 2019 Elsevier Inc. All rights reserved.

28
R. Januszewski et al. / Journal of Catalysis 371 (2019) 27–34
materials with different molecular weights and different contents
of 1,2- and 1,4-units, as well as different ratios of C@C bonds to
H-Si„ units. Additionally, the reactions reported have been con-
ducted in different solvents and in a wide range of temperatures
[1,9,24–32]. These facts do not allow general conclusions to be
drawn about the reactivity of PBs in catalytic hydrosilylation with
silicon-based compounds, besides no detailed comparison of the
activity of various catalysts has been performed until now.
With regard to the above, in this paper we present a functional-
ization of polybutadiene with well-defined microstructure using of
various alkyl-, alkylaryl- and alkoxysilanes in order to determine
the effect of substituents directly bonded to the silicon atom on
the reactivity in hydrosilylation of PB, in the presence of rhodium
and platinum catalysts. This paper presents the first part of our
studies concerning the influence of the structure of organosilicon
compounds on their reactivity and selectivity in the TM-
catalyzed hydrosilylation of PB. Partially modified polymers seem
to be interesting ingredients in rubber compositions, in particular,
tires production, or a convenient reagent for a subsequent hydro-
genation reaction which would allow obtaining an original family
of saturated materials with different backbone architectures.
total conversion of HSiCl2Me, the mixture was filtered and washed
by pentane. Then the solvent was removed by reduced pressure
and the obtained liquid was purified by trap to trap distillation.
The desired product was obtained with 82% yield. 1H NMR (CDCl3,
298 K, d): 4.80 (m, 1H Si-H), 1.31 (s, 18H C(CH3)3), 0.17 (d, 3H Si-
CH3). 13C NMR (CDCl3, 298 K, d): 73.09 (C-O), 31.89 (C(CH3)3),
1.71 (Si-CH3). 29Si NMR (CDCl3, 298 K, d): À32.22. MS (EI, m/z):
176.3 (11.3), 175.2 (83.0), 121.1 (8.5), 119.1 (84.3), 117.1 (8.4),
80.1 (5.2), 79.2 (5.0), 78.2 (5.8), 77.1 (100), 75.1 (5.0), 63.1 (12.2),
61.1 (45.4), 59.1 (7.0), 57.1 (59.0).
2.2.2. Synthesis of dimethyloctylsilane (HSiMe2Oc) (8)
To a mixture prepared from 10 g (89.11 mmol) of 1-octene,
98.03 mmol of HSiClMe2 and 100 mL of anhydrous THF, Karstedt’s
complex was added (10À4 mol Pt corresponding to 1 mol of 1-
octene). The reaction was refluxed and monitored with gas chro-
matography. After total conversion of alkenes, the mixture was
cooled to room temperature and 62.4 mmol of LiAlH4 was added,
and the mixture was stirred for 1 h at r.t. and overnight under
reflux. The solid was separated by filtration through cannula and
washed by THF. Evaporation of the solvent and distillation of the
obtained liquid by trap to trap method gave a colorless liquid.
The yield was 73%. 1H NMR (CDCl3, d, ppm): 3.84 (m, 1H SiH),
2. Experimental section
1.31–1.27 (12H, CH2), 0.89 (t, 3H CH2CH3), 0.58 (m, SiCH2), 0.07
(d, 6H SiCH3) 13C NMR (CDCl3, d, ppm): 33.40, 32.12, 29.53,
29.44, 24.54, 22.86, 14.34, 14.27, À4.67 29Si NMR (CDCl3, d,
ppm): À13.14 MS (EI, m/z): 157.2 (11.0) [M-15]+, 142.1 (5.4),
127.1 (17.6), 99.0 (5.7), 87.1 (35.7), 85.1 (5.2), 73.1 (14.9), 60.2
(7.2), 59.0 (1 0 0), 57.9 (10.3),
2.1. Methods, techniques and chemicals
All manipulations were carried out under an argon atmosphere
using standard Schlenk’s and vacuum techniques. The 1H NMR, 13
C
NMR and HSQC NMR spectra were recorded on a Bruker Ultra-
shield 300 MHz spectrometer using CDCl3 as the solvent. The 29Si
NMR spectra were recorded using a Bruker Ascend 400 spectrom-
eter. The mass spectra were obtained by GCMS analysis (Bruker
MS320 Triple quad, equipped with a VF-5 Factor four capillary col-
umn (30 m) and a quadrupole detector). In-situ FT-IR measure-
ments were performed on a Mettler Toledo ReactIR 15 equipped
with a DS 6.3 mm AgX DiComp Fiber Probe with a diamond sensor,
and a Hg-Cd telluride detector. For all the spectra, 256 scans were
recorded, with a resolution of 1 cmÀ1 in 1, 2, and 5 min intervals.
GPC analysis was performed using a Agilent 1260 Infinity system
equipped with RI detector and a set Phenogel 10u Linear(2)
300 Â 7.8 mm (100–10,000,000 g/mol) column. The measurements
were carried out with THF as a mobile phase in a flow rate of 1 mL/
min; the column oven temperature was 35 °C and detector tem-
perature 35 °C. All molecular weights (Mn and Mw) and polydis-
persity index (PD) values were calculated on the basis of the
calibration curve using polystyrene standards (Shodex) in the
2.3. General procedure for the hydrosilylation of polybutadiene
All reactions were performed under an inert atmosphere (dry
and deoxygenated argon). The mixture prepared from 1 g of
polybutadiene
(PB)
(Mn = 4331 g/mol,
Mw = 6437 g/mol,
PDI = 1.48 containing 90% of 1,2- units) and 9.4 mL of anhydrous
toluene was heated to 90 °C, then a silicon-based modifier was
added in the amount corresponding to 10% of all carbon-carbon
double bonds in the polymer. After stabilizing the temperature, a
catalyst was added. For reactions catalyzed by platinum com-
plexes, the ratio was [C@C]:[SiH]:[Pt] = [1]:[0.1]:[1 Â 10À4], for
reactions catalyzed by rhodium complexes the ratio was [C@C]:
[SiH]:[Rh] = [1]:[0.1]:[5 Â 10À4] and for iron and cobalt complexes
the ratio was [C@C]:[SiH]:[Rh] = [1]:[0.1]:[10À2]. The conversions
of SiH-containing reagents were determined with an FT-IR probe
in real time by monitoring the area of the band assigned to the
Si-H bond. After completion of the reaction, the solvent was
removed under reduced pressure. The obtained polymers were
examined by NMR and GPC techniques.
range from 1.31 Â 103 to 3.64 Â 106 Da. The [{Rh(
l-Cl)(COD)}2]
and [RhCl(PPh3)3] complexes were prepared according to the pub-
lished methods [33,34]. The chemicals were obtained from the fol-
lowing
sources:
polybutadiene
(PB)
(Mn = 4331 g/mol,
Mw = 6437 g/mol. PDI = 1.48 containing 90% of 1,2- units),
HSiMeCl2, HSiMe2Cl, HSiMe2Ph, HSiMePh2, 1-octene, [Pt2(dvt-
mds)3] (Karstedt’s catalyst), C8H12 (1,5-cis,cis-COD), PPh3, HSiMe2-
Cy. [Co2(CO)8], [Fe2(CO)9] from Sigma Aldrich, H2PtCl6, RhCl3 and
HSi(OEt)3 from ABCR, KOtBu from TCI Chemicals, and HSiMe2(nBu),
HSiMe2(tBu), HSiMe2Bn (Bn = benzyl), HSiMe2(C18H37), HSiMe2(-
OEt), HSiMe(OEt)2 from Fluorochem.
3. Results and discussion
Considering the possible widespread application of functional-
ized polymers, in our present study we focused on the develop-
ment of a convenient and efficient catalytic system for the
functionalization of polybutadiene via hydrosilylation reaction,
which would be the most effective and the simplest catalytic path-
way for the modification of polyene materials with silicon-based
derivatives, allowing the synthesis of a silicon-containing organic
polymer of unique reactivity and properties tailored to specific
applications. Therefore, in the present study, a wide range of
silanes, bearing in their structures alkyl, aryl and alkoxy groups,
were selected as interesting polybutadiene modifiers and
employed as reagents in catalytic tests (see Scheme 1). Addition-
ally, in order to make the catalytic system convenient for a large
2.2. Synthesis of silanes
2.2.1. Synthesis of methylbis(tert-butoxy)silane (HSiMe(OtBu)2) (3)
To the ice-cooled dispersion prepared from 20 g (178.2 mmol)
of KOtBu and 300 mL of pentane, 8.4 mL (80.70 mmol) HSiCl2Me
was added dropwise. Then the reaction mixture was kept at room
temperature and monitored by gas chromatography analysis. After

R. Januszewski et al. / Journal of Catalysis 371 (2019) 27–34
29
Scheme 1. Hydrosilylation of polybutadiene with various (the selected) organofunctional silanes.
scale process, the simplest homogeneous catalysts were chosen
and examined, such as Pt-Karstedt (C1), Pt-Speier (C2) and [RhCl
(PPh3)3] (C3). The application and activity of the above-
mentioned TM-complexes in the catalytic functionalization of
unsaturated polymers with „SiH derivatives have been reported
in a few patents and papers [1,9,10,24–32]. Additionally, we
decided to examine a binuclear rhodium complex [{Rh(m-Cl)
(COD)}2] (C4), known as active hydrosilylation catalyst
[17,19,35], which has never before been applied as a hydrosilyla-
tion catalyst of butadiene-based materials. Also, carbonyl com-
plexes of cobalt(0) [Co2(CO)8] (C5) and iron(0) [Fe2(CO)9] (C6)
were tested, because of the low costs of the metal precursors.
Previous reports describing the hydrosilylation of butadiene-
based polymers have revealed that only the vinyl moieties in the
polymer structure are active in this process, even in the excess of
organosilicon modifier [1,9,10,24,26,28–32]. As a starting material,
we selected commercially available polybutadiene, with average
molecular weight of 6437 g/mol and number-average molecular
weight of 4331 g/mol, containing 90% vinyl units and 10% 1,4-cis
and 1,4-trans moieties. We decided to functionalize only 10% of
unsaturated units in the polymer structure due to the long time
required for total silylation of C@C bonds [1,36]. Moreover, it has
been proven that conducting the polybutadiene modification pro-
cess for a very long period of time in the presence of TM-
complexes can led to the gelation of the mixture [1,37,38]. Further-
more, according to our idea, partially modified polymers can be
applied in their original purpose i.e. for the production of tires,
while complete saturation of multiple bonds excludes the use of
the product as a component in the rubber mixtures. Catalytic tests
were performed under an inert atmosphere due to the sensitivity
of the complexes, as well as ethoxysilanes for the moisture, and
monitored using in-situ Fourier-transform infrared spectroscopy,
providing specific information about reaction progress, thanks to
the high intensity of the band deriving from the silicon-hydrogen
bond in the infrared spectrum. The in-situ FT-IR method is com-
monly used to study a wide range of reactions [39–41]. However,
it has not been utilized for monitoring the functionalization of
polybutadiene before and it seems to be the simplest method
allowing the comparison of catalysts. Exemplary graphs illustrat-
ing the disappearance of the SiH band are presented in Fig. 1. In
order to find the most active catalytic system for the modification
of polybutadiene with silicon-based compounds, we selected 2 and
1 for the model reactions, also the simplest reagents representing
two types of silanes.
The catalytic activity of the homogeneous catalysts C1-C4 has
been demonstrated in model reactions between polybutadiene
and 1, as the simplest H-Si„ containing a coupling agent. The
results summarized in Table 1 and presented in Fig. 2 show that
rhodium complexes and platinum Karstedt’s catalyst promoted
complete conversion of the silicon reagent 1. However, C2 enabled
92% conversion of 1 after 15 h and conducting the reaction for
another 9 h did not increase the consumption of organosilicon
reagent. On the other hand, C5 and C6 showed no reactivity even
in presence of a high loading of catalyst precursor. It should also
be noted that C1 proved to be very active at the initial reaction
stage, and after one hour the conversion of 1 achieved 56%. How-
ever, at the same time interval, the precursor C4 was able to trans-
form 68% of the initial amount of 1 to the desired product, whereas
rhodium Wilkinson’s complex ([RhCl(PPh3)3]) and Speier’s catalyst
(H2PtCl6/iPrOH), commonly used hydrosilylation reaction promot-
Fig. 1. 3D graph illustrating the decay of the characteristic band assigned to SiH bond, recorded in real time during hydrosilylation of polybutadienie with HSi(OEt)3 (1)
catalyzed by C4.

30
R. Januszewski et al. / Journal of Catalysis 371 (2019) 27–34
Table 1
Hydrosilylation of polybutadiene with 1.
HSiR3
Catalyst
Si-H conversion [%]
Time
Product
HSi(OEt)3
(1)
C1
C2
C3
C4
C4
C5
99
92
99
99
–
12 h
15 h
7 h 20 min
5 h
24 h
PB1
–
–
–
24 h
85–90%. However, for consumption of the remaining 10–15%, addi-
tional ꢀ5–7 h or more were required [42]. Moreover, we found that
higher concentrations of C1 in the reaction system led to the gela-
tion of the mixture. Crosslinking of (EtO)3Si-functionalized polybu-
tadiene has been also reported by Guo and Rempel in the presence
of Wilkinson’s complex [1]. The post-reaction mixtures were ana-
lyzed by NMR techniques. All spectra were collected at room tem-
perature in a CDCl3 solvent. The spectra showed the complete
disappearance of Si-H signal at 4.28 ppm, confirming the complete
conversion of silane, which was consistent with the FT-IR measure-
ments. The structure of PB1 was confirmed by 1H and HSQC NMR
techniques. Furthermore, on the recorded spectra, the appearance
of new resonance lines was noted at 0.73 ppm in the 1H NMR spec-
trum and 10.17 ppm in the 13C NMR spectrum, both of which were
assigned to the Si-CH2 moiety (all NMR spectra are available in
supporting information). Exemplary 1H NMR spectra of starting
materials and PB1 are presented in Fig. 3. Moreover, for all exper-
iments IR spectra showed the appearance of a new band in the
range of ꢀ840–730 cmÀ1, which was assigned to the Si-alkyl
(SiCH2R) group (Fig. 4). This fact also confirmed incorporation of
1 into polybutadiene chain.
Fig. 2. Kinetic plots for the hydrosilylation of PB with 1.
ers, showed very poor performance when compared with C1 and
C4, giving only 35% conversion. Finally, the total conversion of 1
was achieved after 5 h using C4 and approximately 7 h when C3
was applied, while in case of C1, 12 h for total incorporation of 1
into polymer chain was required.
Fast consumption of reagents at the initial stage of the reaction
in C1-catalyzed hydrosilylation of alkenes with POSS compounds
has been reported by Walczak et al. They described that, after
20 min, the initial amount of substrates underwent conversion in
Hydrosilylation tests of PB with 2 showed that the ethyl sub-
stituents directly bonded to the silicon atom have a significant
impact on the reactivity of the H-Si moiety. Catalytic activity of
Pt-based promoters (C1-C2) was very poor and the reactions per-
formed for 24 h did not provide high conversions of 2. However,
rhodium complexes revealed higher activity, compared with those
Fig. 3. Stacked 1H NMR spectra of (a) polybutadiene, (b) triethoxysilane and (c) PB1.

R. Januszewski et al. / Journal of Catalysis 371 (2019) 27–34
31
of C1-C2 (Table 2). It should be underlined that the catalytic sys-
tem proposed by us was more active than the commonly used
Wilkinson’s catalyst. Moreover, the times required for obtaining
high conversions of silanes were much longer than for 1. This
observation could be explained by the electronic properties of
the alkyl groups. This effect was previously described for the
hydrosilylation of 1-hexene with isosteric alkyl and alkoxy substi-
tuted silanes. The reactivity of the silanes decreases in the follow-
recorded 2D spectrum, a proton-carbon correlation was found at
the intersection of the resonance line assigned to one olefinic pro-
ton located at 5.77 ppm on the 1H NMR axis and the line located at
152.88 ppm on the 13C NMR axis assigned to the Si-CH@ moiety.
This suggests that, in the examined system, the formation of an
unsaturated moiety is a result of the occurrence of the dehydro-
genative silylation process. Tremont et al. have reported that, dur-
ing silylfunctionalization of polybutadiene with HSiCl(iPr)2, an
unsaturated organosilicon material was formed, in contrast to
the reaction with HSiMe2Cl which led to the formation of a
hydrosilylation product only [29]. Therefore, it seems that bulky
substituents directly bonded to the silicon atom play a crucial role
in the formation of this unsaturated moiety. The sterically crowded
OtBu groups resulted in a reduction in activity and selectivity of 3
compared to that of 1.
ing order: HSi(OEt)3 > HSi(OEt)2 Pr > HSi(OEt)iPr2 > HSiiPr3. This
i
effect was observed for the reactions catalyzed by [RhCl(PPh3)3]
as well as [Co2(CO)8]. It seems that the replacement of alkoxy
groups in alkyl units significantly reduces the reactivity of silanes
and it is related with the decrease in the electron-withdrawing
properties of alkyl groups. The influence of the substituents
attached to the silicon atom on the reactivity of the Si-H moiety
has been studied for simple alkenes, alkynes and vinylsilanes
[16]. However, it has not yet been described for polymeric
reagents.
Reactions between PB and 3 were carried out in the same con-
ditions as for the previous modifiers (1 and 2). The platinum-based
catalysts (C1-C2) showed poor activity when compared with those
of the rhodium complexes (Table 3). Unexpectedly, only C4 pro-
moted total conversion of silicon material, but to achieve the com-
plete disappearance of Si-H band, 10 h and 20 min were required.
The appearance of new peaks in the 1H NMR spectrum confirmed
the formation of a silylfunctionalized polymeric product. However,
detailed spectroscopic analysis of the isolated material with the
use of the HSQC NMR technique showed an unexpected outcome
that has never before been reported in literature for butadiene-
based materials modified with alkoxysilanes. Namely, in the
However, HSiMe(OtBu)2 (3) contains only two alkoxy groups,
while HSi(OEt)3 three groups (1). The presence of highly elec-
tronegative atoms bonded to silicon can significantly change the
electronic properties of the silicon reagent, strongly affecting the
process selectivity and consequently favoring the course of dehy-
drogenative silylation reactions. In order to study this effect, a ser-
ies of silicon compounds containing different numbers of alkoxy
substituents were selected, namely HSiMe2(OEt), HSiMe(OEt)2,
HSi(OEt)3. Functionalization of PB with ethoxy-containing modi-
fiers was carried out in the previously optimized conditions in
the presence of C3 and C4 as the most active promoters enabling
the total conversions of silicon materials, which also do not initiate
the crosslinking of polyene materials. The results summarized in
Table 4 show that C3 was a little less active than C4 proposed by
us. Kinetic curves obtained for functionalization of PB with SiH sil-
icon derivatives 1, 4 and 5 in the presence of C4 are presented in
Fig. 5. Moreover, we found that the number of ethoxy groups
bonded to the silicon atom bearing H-Si„ moiety significantly
affected the rates of the hydrosilylation processes. It should be
strongly emphasized that the activity of ethoxy silanes increases
with the number of alkoxy groups. It could be explained by the
inductive effect which has been previously described for the reac-
tions between simple alkenes and chlorosilanes. L.A. Efremova
et al. and M. Capka and co-workers have reported that electron-
withdrawing substituents attached to silicon stabilize the Si-
metal intermediate. This is illustrated by the following reactivity
observed in the hydrosilylation of 1-heptene: HSiCl3 > HSiCl2-
Et > HSiClEt2 > HSiEt3 and 1-octene HSiMeCl2 > HSiCl2Et > HSiClEt2
[16,43–44]. The obtained data showed that alkoxy groups exhibit
the same behavior as the above-described chlorine atom. Further-
more, in contrast to the reaction with 3, NMR analysis of isolated
silylfunctionalized PB products showed that in the examined sys-
tems only the desired hydrosilylation product was formed. This
allowed exclusion of the influence of inductive effect on the forma-
tion of dehydrogenative silylation products and it seems clear that
Fig. 4. Imposed starting and post reaction mixture FT-IR spectra for the hydrosi-
lylation reaction of PB with 1.
Table 2
Hydrosilylation of polybutadiene (PB) with 2.
HSiR3
Catalyst
Si-H conversion [%]
Time
Product
HSiEt3
(2)
C1
C2
C3
C4
C5
C6
9
24 h
24 h
20 h 10 min
15 h 50 min
24 h
PB2
13.5
97
97
–
–
–
–
24 h

32
R. Januszewski et al. / Journal of Catalysis 371 (2019) 27–34
Table 3
Hydrosilylation of polybutadiene with 3.
HSiR3
Catalyst
Si-H conversion [%]
Time
Product
HSiMe(OtBu)2
(3)
C1
C2
C3
C4
8
11 h 50 min
12 h
13 h
PB3
9.5
76
99
10 h 20 min
Table 4
Hydrosilylation of polybutadiene with 4 and 5.
HSiR3
Catalyst
Si-H conversion [%]
Time
Product
HSiMe(OEt)2
(4)
C3
C4
99
99
9 h 40 min
7 h 50 min
PB4
HSiMe2(OEt)
(5)
C3
C4
99
99
13 h 20 min
12 h 20 min
PB5
between the silane structure and reactivity, as well as the tendency
to create dehydrogenative silylation products. We selected a wide
range of silanes containing n-alkyl chains, tert-butyl moiety and
six-membered rings such as cyclohexyl and phenyl. The results
summarized in Table 5 show that the structure of the silane
strongly affects its reactivity. It can be seen that the reactivity of
HSiMe2(n-alkyl) (6, 8 and 9) decreases with increasing number of
carbon atoms in the n-alkyl chain (Table 5, Fig. 6). On the other
hand, the presence of a tert-butyl group resulted in a significant
reduction of silicon compound reactivity and 39 h were required
to achieve 95% conversion of 7 in the presence of C4 precursor.
The differences in the reactivity are closely related to the structure
of silanes and their steric properties. Moreover, NMR analyses of
post-reaction mixtures showed the selective formation of hydrosi-
lylation products in all the above-mentioned catalytic systems.
This means that replacement of the substituent ‘‘R” in HSiMe2R
with a more bulky one does not affect the selectivity of the reac-
tion, but has a significant effect on the reactivity of the silane.
Subsequently, we studied the influence of six-membered rings
on reactivity. Silanes 10 and 12 were selected as reactants contain-
ing different organic rings directly attached to the silicon atom,
exhibiting different electronic properties. Also, HSiMe2Bn (11)
(Bn = benzyl) was used as PB modifier in order to study the influ-
ence of the methylene spacer between the phenyl ring and
Fig. 5. Kinetic plots for hydrosilylation of PB with HSiRn(OEt)3-n (1, 4 and 5).
steric hindrance caused by bulky groups is the only factor deter-
mining selectivity.
Therefore, we decided to study the effect of the substituents
bonded to the silicon atom, in particular to find the correlation
Table 5
Hydrosilylation of polybutadiene with dimethylalkyl- and methylarylsilanes.
HSiR3
Catalyst
Si-H conversion [%]
Time
Product
6
C3
C4
99
99
9 h
4 h 10 min
PB6
7
C3
C4
92
95
48 h
39 h
PB7
8
C3
C4
99
99
5 h 40 min
4 h 40 min
PB8
9
C3
C4
97.5
99
12 h 20 min
6 h 30 min
PB9
10
C3
C4
95
99
15 h
11 h 20 min
PB10
11
12
C3
C4
99
99
6 h 40 min
4 h 40 min
PB11
PB12
C3
C4
99
99
4 h 30 min
2 h 50 min

R. Januszewski et al. / Journal of Catalysis 371 (2019) 27–34
33
4. Conclusions
The above-described results seem to imply that the substituents
bonded to the HSi„ unit play a crucial role and determine the
reactivity of silanes in the TM-catalyzed hydrosilylation of poly-
mers. It should be strongly emphasized that the behavior of
organosilicon reactants during the functionalization of polymers
is quite similar to that of the molecular systems described in the
cited references, and similar trends were observed. Silanes bearing
strongly electronegative substituents (alkoxy) and phenyl groups
are more active than similar trialkyl-substituted silanes. Moreover,
the presence of two sterically crowded groups led to the formation
of a dehydrogenative silylation product. We also found that the
reactivity of silanes decreased with increasing ordinariness of the
carbon atom attached to the silicon. On the other hand, the per-
formed experiments showed that the established catalytic condi-
tions for a binuclear Rh(I) complex resulted in promotion of
mainly total conversions of reagents and enabled the incorporation
of silanes into polymer chains in a relatively short time compared
to the other TM-catalysts, and the catalytic system proposed by us
was much more effective than the widely-known rhodium Wilkin-
son’s complex or Pt-Karstedt’s catalyst. To the best of our knowl-
edge, this is the first paper describing a systematic and detailed
study of the influence of a modifier structure on its reactivity in
the TM-catalyzed modification of polymer. Furthermore, GPC
experiments showed that the presence of the Rh(I) catalyst led to
the formation of only desired products, without side reactions such
as crosslinking or chain scission.
Fig. 6. Kinetic plots for hydrosilylation of PB with dimethyl(n-alkyl)silanes
(HSiMe2(n-alkyl)) (6, 8 and 9).
Acknowledgments
The authors wish to thank the National Science Centre, Poland,
for their financial support UMO-2016/23/N/ST5/00335. This
research was supported by Synthos S.A.
Fig. 7. Kinetic plots for hydrosilylation of PB with a six-membered ring containing
dimethylsilanes.
Appendix A. Supplementary material
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.jcat.2019.01.026.
dimethylsilyl moiety, on the efficiency of hydrosilylation. The
results compiled in Table 5 and Fig. 7. show that, as expected,
the presence of the phenyl group resulted in the activation of the
silane. The inductive effect appears to be rate-limiting in the
hydrosilylation of polybutadiene, as illustrated by the following
reactivity sequence HSiMe2Ph > HSiMe2Bn > HSiMe2Cy. The
obtained data confirmed the significant effect of the phenyl group
in comparison to the cyclohexyl moiety, which has been previously
reported for the hydrosilylation of 1-hexene with H3SiR (reactivity
decreases in the following order Ph > Cy > Bu) [16,45]. The greater
reactivity of 11 than 10 could be explained by the steric hindrance
of the tertiary carbon bonded to the silicon atom. A similar effect
was observed for silanes 6 and 7. This means that the reactivity
decreases with increasing ordinariness of the carbon atom directly
attached to the HSi group.
References
[1] X. Guo, R. Farwaha, G. Rempel, Macromolecules 23 (1990) 5047–5054.
[2] C.-C. Peng, V. Abetz, Macromolecules 38 (2005) 5575–5580.
[3] B. Marciniec, E. Małecka, J. Gulin´ ski, M. Grunwald-Wyspian´ ska, M.
Lewandowski, Can. J. Chem. 79 (2001) 775–779.
[4] A.N. Ajjou, H. Alper, Macromolecules 29 (1996) 1784–1788.
[5] S. Im-Erbsin, P. Prasassarakich, G.L. Rempel, J. Appl. Polym. Sci. 93 (2004) 854–
869.
[6] C.B. Wagner, A. Studer, Eur. J. Org. Chem. (2010) 5782–5786.
[7] L. Lotti, S. Coiai, F. Ciardelli, M. Galimberti, E. Passaglia, Macromol. Chem. Phys.
210 (2009) 1471–1483.
[8] J. Zheng, F. Liu, Y. Lin, Z. Zhang, G. Zhang, L. Wang, Y. Liu, T. Tang,
Macromolecules 45 (2012) 1190–1197.
[9] X. Guo, G.L. Rempel, Macromolecules 25 (1992) 883–886.
[10] C. Doring, S. Thiele, D. Heidenreich US 2016/0369015 A1 (2013) (to Trinseo
Europe GmbH).
The in-situ FT-IR measurements revealed that the reaction time
required for the complete reagents conversion as well as the rate of
hydrosilylation process strongly depend on the silanes steric and
electronic properties, which are directly associated with the type
and character of radicals bonded to the silicon atom. Moreover,
GPC experiments of obtained polymers showed higher Mn and
Mw values of organosilicon-functionalized polymers with refer-
ence to the starting material (PB). The Mn values increased in
the range from 17% (PB7) to 39% (PB9). Furthermore, a correlation
between the molar mass of silanes and the molecular weight of
synthesized polymers was observed. The Mn, Mw and PDI values
of all polymers (PB1-PB12) are provided in ESI.
[11] S.-J. Park, K.S. Cho, J. Colloid Interface Sci. 267 (2003) 86–91.
[12] G. Colucci, A. Di Gianni, R. Bongiovanni, A. Priola, F. Torello, S. Bracco, P.
Sozzani, Macromol. Symp. 296 (2010) 550–556.
[13] R. Auvergne, M. Desroches, S. Clerc, S. Carlotti, S. Caillol, B. Boutevin, React.
Funct. Polym. 72 (2012) 393–401.
[14] R. Pandit, J. Giri, G.H. Michler, R. Lach, W. Grellmann, B. Youssef, J.M. Saiter, R.
Adhikari, Macromol. Symp. 315 (2012) 152–159.
[15] B. Marciniec, H. Maciejewski, C. Pietraszuk, P. Pawluc, Hydrosilylation: A
Comprehensive Review on Recent Advances, Springer, 2009.
[16] B. Marciniec, J. Gulinski, W. Urbaniak, Z.W. Kornetka, Comprehensive
Handbook on Hydrosilylation, Pergamon Press, 1992.
[17] R. Januszewski, I. Kownacki, H. Maciejewski, B. Marciniec, J. Organomet. Chem.
846 (2017) 263–268.
[18] I. Kownacki, B. Marciniec, K. Szubert, M. Kubicki, M. Jankowska, H. Steinberger,
S. Rubinsztajn, Appl. Catal. A 380 (2010) 105–112.

34
R. Januszewski et al. / Journal of Catalysis 371 (2019) 27–34
[19] R. Kukawka, R. Januszewski, I. Kownacki, M. Smiglak, H. Maciejewski, Catal.
Commun. 108 (2018) 59–63.
[20] J. Sun, L. Deng, ACS Catal. 6 (2016) 290–300.
[21] B. Marciniec, A. Kownacka, I. Kownacki, R. Taylor, Appl. Catal. A 486 (2014)
230–238.
[32] K. Baum, J.C. Baum, T. Ho, J. Am. Chem. Soc. 120 (1998) 2993–2996.
[33] R.J. Angelici, Inor. Synth., John Wiley & Sons, Inc., 1990.
[34] S. Komiya, in: Synthesis of Organometallic Compounds: A Practical Guide,
Wiley, New York, 1997, p. 442.
[35] R. Januszewski, I. Kownacki, H. Maciejewski, B. Marciniec, A. Szyman´ ska, Eur. J.
[22] I. Kownacki, B. Marciniec, A. Macina, S. Rubinsztajn, D. Lamb, Appl. Catal. A 317
Inorg. Chem. (2017) 851–856.
(2007) 53–57.
[36] D. Jeyaprakash, S. Samuel, R. Dhamodharan, C.K. Ober, J. Polym. Sci. Part A
Polym. Chem. 38 (2000) 1179–1183.
[37] J.K. Mertzweiller, H.M. Tenney, US 3334076 A (1963) (to ExxonMobil Research
and Engineering Co.).
[23] M. Zaranek, B. Marciniec, P. Pawluc´, Org. Chem. Front. 3 (3) (2016) 1337–1344.
[24] R. Matmour, T.J. Joncheray, Y. Gnanou, R.S. Duran, Langmuir 23 (2007) 649–
658.
[25] B.P.S. Chauhan, B. Balagam, Macromolecules 39 (2006) 2010–2012.
[26] R. Matmour, T.J. Joncheray, Y. Gnanou, R.S. Duran, J. Colloid Interface Sci. 311
(2007) 315–321.
[27] A. Iraqi, S. Seth, C.A. Vincent, D.J. Cole-Hamilton, M.D. Watkinson, I.M. Graham,
D. Jeffrey, J. Mater. Chem. 2 (1992) 1057–1064.
[28] F. Wurm, F.-J. López-Villanueva, H. Frey, Macromol. Chem. Phys. 209 (2008)
675–684.
[29] S.J. Tremont, P.W. Collins, W.E. Perkins, R.L. Fenton, D. Forster, M.P. McGrath, G.
M. Wagner, A.F. Gasiecki, R.G. Bianchi, J. Jacquelyn, J. Casler, C.M. Ponte, J.C.
Stolzenbach, P.H. Jones, J.K. Gard, W.B. Wise, J. Med. Chem. 36 (1993) 3087–
3097.
[38] F.L. Ramp, E.J. Dewitt, L.E. Trapasso, J. Polym. Sci. Part A Polym. Chem. 4 (1966)
2267–2279.
[39] H. Maciejewski, J. Karasiewicz, A. Dutkiewicz, M. Dutkiewicz, K. Dopierała, K.
Prochaska, React. Funct. Polym. 83 (2014) 144–154.
[40] H. Stachowiak, J. Kaz´mierczak, K. Kucin´ ski, G. Hreczycho, Green Chem. 20
(2018) 1738–1742.
[41] J.-Y. Ye, Y.-X. Jiang, T. Sheng, S.-G. Sun, Nano Energy 29 (2016) 414–427.
[42] M. Walczak, A. Franczyk, B. Marciniec, Chem. Asian J. 13 (2018) 181–186.
[43] L.A. Efremova, Z.V. Belyakova, K.K. Popkov, S.A. Golubtsov, Zh. Prikl, Spektr. 19
(1973) 367.
[44] M. Capka, P. Svoboda, V. Bazant, V. Chvalovsky, Collect. Czech. Chem. Commun.
36 (1971) 2785.
[30] B.X. Fu, A. Lee, T.S. Haddad, Macromolecules 37 (2004) 5211–5218.
[31] A.E. Ciolino, O.I. Pieroni, B.M. Vuano, M.A. Villar, E.M. Vallés, J. Polym. Sci. Part
A Polym. Chem. 42 (2004) 2920–2930.
[45] M. Mejstrikova, R. Rericha, M. Kraus, Collect. Czech. Chem. Commun. 39 (1974)
135.