Studies on the Activation and Hydrosilylation Catalysis of RhCl3(Bu2S)3

Article abstract of DOI:10.1021/acs.inorgchem.7b00029

RhCl₃(Bu₂S)₃ is an industrial precatalyst utilized in the curing of some solventless silicone-release coatings formulations. The catalyst requires no additional inhibitor, in contrast to typical Pt formulations, and so questions arose about how fast the catalyst could trigger curing if it were used in a more activated form. Studies on the activation of RhCl₃(Bu₂S)₃ revealed multiple intermediates, of which [RhCl(Bu₂S)₂]₂ and [RhHCl(SiMe(OSiMe₃)₂) (Bu₂S)₂]₂ were isolated. [RhHCl(SiMe(OSiMe₃)₂)(Bu₂S)₂]₂ is the most activated form of the precatalyst, showing a brief induction period. Various experiments were performed to analyze the nature of the catalyst, including Hg poisoning, addition of poison ligands, and comparisons versus Rh nanoparticles. The data tend to be more consistent with a molecular catalyst than Rh nanoparticles. Data are also provided that support the active catalyst containing a Rh_xCl_x(Bu₂S)_2x fragment, although the exact identity of the active catalyst is not yet determined.

Full text of DOI:10.1021/acs.inorgchem.7b00029

Article
pubs.acs.org/IC
Studies on the Activation and Hydrosilylation Catalysis of
RhCl3(Bu2S)3
Matthew T. Olsen,* Brian D. Rekken, and Donald V. Eldred
Core R&D, Dow Chemical Company, 2200 W. Salzburg Rd, Auburn, Michigan 48611, United States
S
* Supporting Information
ABSTRACT: RhCl3(Bu2S)3 is an industrial precatalyst utilized in the curing of some solventless silicone-release coatings
formulations. The catalyst requires no additional inhibitor, in contrast to typical Pt formulations, and so questions arose about
how fast the catalyst could trigger curing if it were used in a more activated form. Studies on the activation of RhCl3(Bu2S)3
revealed multiple intermediates, of which [RhCl(Bu2S)2]2 and [RhHCl(SiMe(OSiMe3)2) (Bu2S)2]2 were isolated.
[RhHCl(SiMe(OSiMe3)2)(Bu2S)2]2 is the most activated form of the precatalyst, showing a brief induction period. Various
experiments were performed to analyze the nature of the catalyst, including Hg poisoning, addition of poison ligands, and
comparisons versus Rh nanoparticles. The data tend to be more consistent with a molecular catalyst than Rh nanoparticles. Data
are also provided that support the active catalyst containing a RhxClx(Bu2S)2x fragment, although the exact identity of the active
catalyst is not yet determined.
process and the chemical reactions that govern it. Interestingly,
Armstrong et al. showed that pre-reaction of the Rh
tris(sulfide) precatalyst with HSiMe2O(SiMe2O)16SiMe2H led
to a faster curing catalyst.37 The report herein discusses our
efforts to better understand the behavior of Rh sulfide catalysts
in hydrosilylation reactions.
RhCl3(R2S)3 has also been reported as a precatalyst for
various industrially relevant hydrosilylation reactions,37−40
some of which claim improved selectivity versus Pt hydro-
silylation41−43 or improved tolerance of epoxy functional
groups.44 Various Rh complexes have also been evaluated as
potential curing catalysts for hydrosilylation-cross-linked
silicones.45 Related complexes of different metals such as
PtCl2(SR2)2/SnCl2 have been studied for hydrosilylation
catalysis.46 Aside from these, the majority of Rh complexes
with sulfur-containing ligands use the ligand to attach the metal
to a heterogeneous support and have been reviewed else-
where.47
Beyond hydrosilylation, the catalytic activity of RhCl3(SR2)3
has not been extensively explored. The most notable exception
is the hydrogenation of carboxylic acid-containing olefins (such
as maleic, fumaric, and cinnamic acid) by James et al., for which
RhCl3(SEt2)3 is used as a precatalyst.48−51 The catalytic
reaction occurs under somewhat mild conditions (1 atm H2,
70 °C) and requires dimethylacetamide as a solvent; activity is
not observed with benzene as a solvent.50 They also report that
the reaction requires dissociation of SBu2 and that the catalytic
1. INTRODUCTION
The silicones industry sells hydrosilylation-curable, silicone-
based release coatings that are cured at high temperatures with
short dwell times (several seconds) in an oven.1−4 Although the
majority of such release coatings utilize inhibited Karstedt’s
precatalyst1,5,6 (Pt2[(ViMe2O)Si]3 treated with an inhibitor
such as 1-ethynylcyclohexanol),7,8 a small percentage of
coatings is cured with a rhodium catalyst.3 Fundamental studies
on Karstedt’s precatalyst have been performed, although the
exact nature of the catalyst remains debated.9−13 Pt catalysts
generally have the advantage of being exceptionally fast;
however, they are easily inhibited or deactivated by various
species (amines, sulfides, etc.) and have demonstrated substrate
interaction problems with acrylics.4,14,15 In contrast, formula-
tions using a Rh catalyst such as Rh(R2S)3Cl3 are more resistant
to poisoning and are more chemically compatible with various
adhesives.3 However, curing with Rh catalysts is slower and
requires higher temperatures.16 Many other catalysts have been
developed for hydrosilylation, only some of which have been
used in silicone cross-linking, and these have been the subject
of excellent reviews by others.1,17−21 Recent years have seen
exciting advances in first-row transition-metal catalysts.22−34
The Rh curing system was initially developed by Chandra in
the 1970s, and it consists of the precatalyst RhCl3(Bu2S)3.35,36
Chandra examined many Rh precursors, especially those with
sulfide ligands. Unlike Pt, which requires an inhibitor to give it
sufficient working time (“bath life”) in the presence of
formulation substrates, the rhodium sulfide is reasonably stable
without an inhibitor. The activation of this catalyst appears to
be a slow process, and so we aimed to better understand this
Received: January 11, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
reaction is preceded by reduction of RhIII to RhI. In contrast to
metal thioether complexes, metal thiolate complexes are more
commonly studied, as catalysts have been reviewed elsewhere.47
Table 1. Byproducts Observed during the Reaction of
MDHM and MMVi with RhCl3(Bu2S)3 Catalyst
byproduct identity
byproduct yield (%)
M2D(C2H2)DM (cis and trans)
MDViM
1.3, 1.9 (not differentiated)
2. RESULTS AND DISCUSSION
0.8
2.4
2.8
1.0
0.2
10.4
Although it is commercially used and available, we were unable
to find significant basic characterization of the precatalyst.
These data and information on the synthesis of the precatalyst
are available in the Supporting Information (Figures S1−S3).
Catalytic Activity of RhCl3(Bu2S)3 with Model Sub-
strates. To gain more detailed information about the catalytic
performance in a curable siloxane composition, model
substrates were used to examine the rates of hydrosilylation.
1,1,1,3,5,5,5-Heptamethyltrisiloxane (MDHM) and 1,1,1,3,3-
pentamethyl-3-vinyl-disiloxane (MMVi) were reacted in the
presence of RhCl3(Bu2S)3 with toluene as a solvent and
dodecane as an internal gas chromatography (GC) standard.
MDEtM
MM(CH2)2MM
M2D(CH2)2DM2
M2D(CMeH)DM
total
side products are due to dehydrogenative silylation, which is
not uncommon for group 9 transition-metal catalysts.18,53−57
We also observed a trace amount of the Markovnikov
hydrosilylation product. Less expected was a series of products
that appear to arise from functional group exchange between
the Si−H and Si−Vi sources (Scheme 1). The Si−H/Si−Vi
exchange reaction is uncommon but has been reported.57,58
Such a reaction would produce MDViM and MMH, the former
of which is observed but the latter is not, because it is highly
reactive (the byproduct MM(CH2)2MM is observed). These
products do not simply arise from hydrosilylation of impurities
(MMH or MDViM), which were not detected in the unreacted
reagents. We also observed the product of the hydrosilylation
between MDViM and MDHM (e.g., M2D(CH2)2DM2), as well
as MDEtM, which results from hydrogenation of the vinyl
compound. H2 is generated in solution via dehydrogenative
silylation.
Catalyst loadings of 5−150 ppm are typical for hydro-
silylation curing of siloxane polymers, and the number of
catalytic turnovers this corresponds to depends on the polymer
molecular weight.4,52 We selected a catalyst loading of 0.1 mol
% to model a paper coating application. Although Si−H/Si−Vi
ratios of 2:1 are sometimes used, we instead used a 1:1 ratio for
the sake of simplicity (in application, excess Si−H is eventually
hydrolyzed, but this is much slower than the hydrosilylation
rate). We also performed these reactions under inert
atmosphere, although paper coatings applications are per-
formed in open air; future studies could address this difference.
Reactions generally produce ∼90% yield of M2D(CH2)2DM,
the expected anti-Markovnikov product of hydrosilylation (see
Author Information (Notes) for nomenclature). Reaction
monitoring at various temperatures revealed an induction
period that is significant at lower temperatures (Figure 1).
Extrapolating these data indicates that at room temperature in a
polymeric system the induction period should be quite long
(hours), allowing for considerable working time prior to the
onset of curing.
Activation of RhCl3(Bu2S)3 with MMVi and MDHM. To
better understand the activation of the precatalyst
RhCl3(Bu2S)3, we studied its reactivity with model substrates.
RhCl3(Bu2S)3 was unreactive toward MMVi but gradually
reacted with consecutive treatments of MDHM. Varying
mixtures of Rh sulfide species are formed, with MDClM, free
Bu2S, and H2 as byproducts (Scheme 2, Table 2). Treatment of
RhCl3(Bu2S)3 with 1 equivalent of MDHM results in partial
consumption of the starting material (∼32% remains) and the
formation of [Rh(H)2(Cl)(Bu2S)2]2 (RhHa, ∼48%) and
[RhCl(Bu2S)2]2 (∼20%). When a second equivalent of
MDHM is added, the remainder of the starting material is
consumed, and the Rh balance is distributed among RhHa
(∼21%), [RhCl(Bu2S)2]2 (∼35%), and a new Rh hydrides
species, [Rh(H) (Cl)(OSi(OSiMe3)2Me)(Bu2S)2]2 (RhHb,
∼38%). Finally, treatment of the reaction mixture with a
third equivalent of MDHM cleanly converted it to RhHb (75%)
with a small amount of decomposition due to the thermal
instability of RhHb, which will be described later. More details
on the observations during these sequential additions can be
found in the Supporting Information (Section 4).
Several trace byproducts are produced during this reaction
and were identified with GC-MS (Table 1). The most expected
Of these Rh sulfide species, the extent of characterization is
varying. RhHa is most clearly observed in the hydride region of
the 1H NMR (−16.4 ppm, d, JRh−H = 6.4 Hz), but it could only
be isolated as a mixture with other Rh species. Aside from the
1
Rh−H resonance in the H spectrum, there is considerable
overlap of the Rh-sulfide resonances, and its identity is implied
based on the proposed reaction sequence (Scheme 2). A
dimeric structure is proposed from comparison to the other
intermediates that will be discussed.
In contrast to RhHa, [RhCl(Bu2S)2]2 was fully characterized.
When the reaction mixtures were cooled, [RhCl(Bu2S)2]2
precipitates as a crystalline orange solid. This stable material
was characterized by 1H NMR, combustion analysis, and single-
Figure 1. Reaction monitoring by GC of MDHM + MMVi in the
presence of 0.1 mol % RhCl3(Bu2S)3 at the indicated temperature.
B
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Observed Products Arising from the Si−Vi/Si−H Exchange Reaction
Scheme 2. Schematic Illustrating the Proposed Reactions That Occur during the Activation of RhCl3(Bu2S)3
a
Table 2. Mass Balance of Rh-Containing Species upon
Treatment of RhCl3(Bu2S)3 with MDHM and Heated at 55
°C for 45 min Each Step
Rh Mass balance
equivs MDHM
RhCl3(Bu2S)3
RhHa
[RhCl(Bu2S)2]2
RhHb
1
2
3
32%
1%
48%
21%
0%
20%
35%
0%
0%
38%
75%
0%
a
Mass balance is approximate, as [RhCl(Bu2S)2]2 numbers were
obtained via isolated yield, while the other species were measure via 1H
NMR.
crystal X-ray diffraction (Figure 2). [RhCl(Bu2S)2]2 is dimeric,
and unlike many [RhClL2]2 complexes, it features chloride
ligands that are terminal rather than bridging. Instead, two of
the thioether ligands bridge the metals. Bridging thioether
ligands are not uncommon.59
We were able to independently synthesize this complex via
the reaction of Bu2S and [Rh(C2H4)2Cl]2, and the product was
isolated in moderate yield (47%, Scheme 3).
Figure 2. X-ray crystal structure of [Rh(SBu2)Cl]2 with thermal
ellipsoids set at 30%. Hydrogen atoms are removed for clarity.
appeared as a doublet of doublets (JSi−Rh = 30 Hz, JSi−H = 17.5
Hz). The Rh hydride was shown to couple to the Rh silyl group
via a 1H−29Si heteronuclear multiple bond correlation
(HMBC) experiment (Figure 3).
Liquid injection field desorption ionization mass spectrom-
etry (LIFDI-MS) analysis of RhHb showed significant
fragmentation, which primarily arose from the dimeric species
[Rh(H)(Cl)(OSi(OSiMe3)2Me)(Bu2S)2]2 (Figure S13). The
parent peak is not actually observed, but species arising from
Significant characterization of RhHb could be obtained as
well, although it could not be isolated in high purity due to
limited thermal stability and a tendency to form intractable oils.
A hydride resonance is clearly observed in the 1H NMR
spectrum (−18.5 ppm, d, JRh−H = 27.7 Hz), and the 29Si NMR
spectrum indicated the release of 2 equivalent of MDClM
(1,1,1,3,5,5,5-heptamethyl-3-chloro-trisiloxane) as well as
∼0.75 equivalent of a Rh silyl species. The Rh silyl resonance
C
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 3. Schematic Illustrating Two Different Synthetic
Pathways to [RhCl(Bu2S)2]2
tacky oils that could not be separated from the reaction
byproducts. Attempts at further purification resulted in complex
decomposition, because the generation of RhHb is competitive
with its degradation. For this reason, the highest in situ
measured yield we could achieve was ∼75% (at ∼78 mM [Rh],
55 °C for ∼1 h). Approximately 15% of the mass balance was
found in other unidentified Rh−H species, and ∼10% was in
unreacted MDHM. Solutions of RhHb appeared as pale orange
but darkened as reaction times extended beyond 1 h at 55 °C.
1
Monitoring of the decomposition process by H NMR did
not reveal any major intermediates. Rather the growth of free
Bu2S is observed. The orange (RhHb) reaction mixture
becomes dark black/brown and remains soluble; diatomaceous
earth filtration does not remove the color. Analysis of the
reaction mixture at this stage indicated the formation of a third
equivalent of MDClM (Figure 4). Heating at higher temper-
atures (120 min, 110 °C) forms insoluble Rh black and causes
loss of catalytic activity (Section E).
1
Figure 3. Enlarged spectrum from a H−29Si HMBC experiment (x
and y axis, respectively), showing the coupling between the Rh Si.
Note the differences between the 1H reference spectrum (top of
graph) and the 2D spectrum: the 1H NMR spectrum shows a doublet
for the Rh−H resonance and a very small doublet of doublets due to
Figure 4. Equivalents of RhHb and MDClM formed after heating
RhCl3(Bu2S)3 with 3 equivalent of MDHM at 55 °C (∼78 mM [Rh]),
illustrating the competing rates of the synthesis and degradation
reactions.
1
29Si satellites. In the 2D spectrum, the nuclei that are targeted are H
and 29Si, and so the satellite peaks are dominant and give rise to the
observed signal. In the 29Si−1H HMBC experiment, the projection of
1
the H spectrum appears as two overlapping doublets (e.g., a triplet)
rather than a doublet, because the experiment is only showing the
molecules containing 29Si nuclei. In this scenario, the hydride couples
to 29Si and to 103Rh. 29Si−1H coupling was determined from both from
Efforts To Understand the Active Catalyst. A. Induction
Period Comparisons. The induction period of the catalyst
alerted us to two possibilities: first, the induction period may be
due to the reduction of RhIII to RhI and/or the substitution of
Rh−Cl bonds for Rh−H bonds. Second, that the active catalyst
may be nanoparticulate in structure with the induction period
reflecting the agglomeration of a molecular precatalyst into a
higher-order aggregate. Several tests were performed to
differentiate these possibilities.
First, the catalytic competence of various forms of the
partially activated precatalysts were compared (Table 3, Figure
5). The observation of induction periods indicates that none of
these species are part of the catalytic cycle and rather are
precatalysts. The significant reduction in length of the
induction period upon conversion of RhCl3(Bu2S)3 into
[RhCl(Bu2S)2]2 or RhHb indicates that the majority of the
induction period is due to the generation of these species. If
catalysis is actually performed by metal nanoparticles, their
1
the 29Si NMR spectrum as well as from the 29Si satellites in the H
NMR spectrum.
loss of one or multiple Bu2S are. The monomeric fragment
Rh(H) (Cl) (OSi(OSiMe3)2Me) (Bu2S)2 is also observed and
is the largest peak present. It is conceivable that the monomer
and dimer are in equilibrium. Reductive elimination of 1
equivalent of hydrosilane from the dimer is observed to a small
extent, as are higher nuclearity species (Figure S12). In further
support of a loosely bound or dissociated equivalent of Bu2S,
the solution-state 1H NMR spectrum indicates ∼1−1.5
equivalent of free Bu2S (small broad species in the baseline
complicate the integration). The peaks are broadened
indicating exchange with the Rh-bound sulfides (Figure S9).
The overall reaction is shown in eq 4:
Attempted Isolation and Decomposition of RhHb. Our
attempts to isolate RhHb were unsuccessful and resulted in
D
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
conversion into metal nanoparticles. To examine further this
phenomenon we first considered Hg poisoning studies.
Although Hg poisoning tests are not a definitive means of
differentiating nanoparticles and small-molecule catalysts,60,64
they provide an additional line of evidence for consideration.
Others have used the Hg test as an additional data point.60−62
Poisoning of this catalyst by Hg is slow and can only be
observed at lower reaction temperatures with vigorous stirring.
Using a reaction temperature of 50 °C and stirring at 1500 rpm
with multiple stirbars, we added ∼9000 equivalent of Hg (per
Rh) after the induction period had passed. The rate of product
formation gradually decreased (Figure 6). Conditions with less
contact time between the solution and the Hg (100 °C, 500
rpm) did not slow the reaction rate.
Table 3. Observered Induction Period for Various
Precatalysts for the Hydrosilylation of MDHM and MMVi
precatalyst
induction period
RhCl3(Bu2S)3
[RhCl(Bu2S)2]2
RhHb
∼3 min, 70 °C
∼1 min, 70 °C
∼1 min, 50 °C
Figure 5. Reaction monitoring by GC of the reaction of MDHM and
MMVi with RhCl3(Bu2S)3, [RhCl(Bu2S)2]2, and RhHb as precatalysts
at 70 °C (top) and RhHb at 50 °C (bottom).
synthesis from RhHb must be comparatively very fast. This
would differ from some Rh systems studied by others.60−62
B. Attempts at Generating the Active Catalyst. We
attempted to generate the active catalyst (e.g., no induction
period) but have not yet succeeded. Attempts at generating Rh
hydride rapidly in situ via hydride reagents either generated
unstable species or had longer induction periods than RhHb
(see Supplementary Information Section 8). We also pretreated
RhHb with MMVi, knowing that this would trigger product
formation and presumably replicates catalytic conditions
(Supporting Information Section 9). However, this results in
a reaction mixture having an induction period similar to that of
RhHb. We also tested the thermally decomposed mixtures of
RhHb that were discussed in the preceding section; however,
these species had significantly diminished activity (see Section
E below). One possible explanation for the difficulty in
bypassing the induction period could be that the dimeric
species RhHb must dissociate into monomers before it is
catalytically active.
Figure 6. Hg poisoning experiments on the reaction of MDHM +
MMVi as catalyzed by RhCl3(Bu2S)3 at 100 °C (top) and at 50 °C,
with additional stir bars, and prewarmed Hg (bottom).
These results differ from some nanoparticulate catalysts in
which Hg poisoning immediately halts catalysis.11,60,65,66 There
are also examples of homogeneous Rh catalysts being
immediately poisoned by Hg,60 and therefore the Hg test is
one tool of many that should be used.
One possible explanation for the differing poisoning data at
50 versus 100 °C is that the identity of the catalyst changes.
However, we find it likely that higher temperatures would favor
the formation of Rh(0) nanoparticles rather than lower
temperatures. Finke et al. found that Rh(0) particles dominate
by higher temperatures and pressures.60−62
D. Chemical Poisoning Studies: 1,10-Phenanthroline,
SBu2, and Phosphines. We further probed the kinetic
properties of the catalysts by adding poison ligands. The
objective of these tests was to determine the number of active
sites present. Heterogeneous catalysts often are poisoned by
much less than 1 equivalent of poison, because many metal
atoms are buried beneath the surface, and the close proximity of
the metal atoms to each other can allow 1 equivalent of poison
to deactivate nearby active sites.63,67 In contrast, homogeneous
C. Hg Poisoning Studies and Rh Nanoparticle Testing.
Although induction periods are a hallmark of the formation of
metal nanoparticles,63 above we showed that the majority of the
induction period was instead due to the conversion of Rh−Cl
bonds into Rh−H bonds. However, a brief induction period is
still observed for RhHb, and it is possible this originates from its
E
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
catalysts show a stronger correlation between the number of
equivalents of poison and the number of active catalyst
molecules. For these experiments, we used RhHb, because it is
closest to the active catalyst (shortest induction period) and
performed experiments at 50 °C.
We first tested 1,10-phenanthroline (phen), which is a well-
documented poison for both homogeneous and heterogeneous
Rh catalysts.60−62 The poison was added after ∼20% of the
reaction product had formed (∼2 min). Using the method of
initial rates (examining the first ∼3 min of catalysis after
addition of poison), we plotted the relative rate of the reaction
versus the equivalents of poison. A poisoning effect is observed
once more than 0.1 equivalent of phen is added, with greater
poisoning occurring with more poison and complete poisoning
observed when between 0.65 and 1 equivalent is added (Figure
7). Although less than 1 equivalent of phen gives a poisoning
equivalent of phen, whereas their molecular Cp4Rh4 catalyst
requires ∼4 equivalent of phen.60−62 We attempted to make a
similar comparison by testing the same commercially available
∼2 nm Rh nanoparticles (Strem Chemical Co.; stabilized with
PEG dodecylether hydrosol). However, under our catalytic
conditions no reaction was observed. This further supports that
Rh nanoparticles are not the active catalyst generated from
RhCl3(Bu2S)3, although it is also possible that different size/
shape particles are generated in situ in comparison to the
material purchased from Strem.
We next tested the poisoning effect of SBu2 with the
RhCl3(Bu2S)3, which is the ligand initially bound to the
precatalyst. Surprisingly, the reaction profile was unaffected by
up to 25 equivalent of Bu2S, and ∼500 equivalent of Bu2S were
required before a very mild poisoning was observed (Figure 8).
Figure 8. Reaction monitoring by GC aliquots of the hydrosilylation of
MDHM and MMVi catalyzed by RhCl3(Bu2S)3 in the presence of
various amounts of additional Bu2S.
This indicates that additional Bu2S is very weakly binding.
These results explain why the catalyst remains active despite an
activation process that releases an equivalent of free Bu2S.
Similarly, this also explains why the activity of 1/2[RhCl-
(Bu2S)2]2 is similar to that of RhCl3(Bu2S)3, even though the
latter contains less sulfide. The active catalyst only requires 2
equivalent of Bu2S and is unaffected by further excess of
thioether.
Finally we tested catalyst poisoning with the phosphines
PBu3 and 1,1,1-tris(diphenylphosphinomethyl)ethane (tri-
phos). PBu3 poisoned the catalyst weakly, whereas the chelating
triphos poisoning more resembled phen. The details of these
experiments can be found in the Supporting Information
(Section 7).
E. RhxClx(Bu2S)y Versus Soluble Rh Particles. We sought to
determine information about the primary coordination sphere
of the catalyst. Our above studies supported a Rh/Bu2S ratio of
1:2 in the active catalyst. RhHb seems poised to reductively
eliminate chlorosilane and form a reactive thioether-ligated RhI
hydride. However, it seems likely that a chloride ligand remains
on the metal: reactions of RhCl3(Bu2S)3 with varying amounts
of MDHM and MMVi only release 2 equivalent of MDClM
during the conversion of RhCl3(Bu2S)3 into an active catalyst
(Figures S13 and S14). These 2 equivalent of MDClM originate
from the conversion of RhCl3(Bu2S)3 into RhHb (Figure 4).
Our studies support that the chloride ligand remains bound
to Rh until the end of catalysis and that catalysis is significantly
slowed when the RhHb decomposes. The reaction of RhHb
with 1 equivalent of MMVi at 55 °C produced the hydro-
Figure 7. Reaction monitoring by GC aliquots of the hydrosilylation of
MDHM and MMVi catalyzed by RhHb with various equivalents of
phen added at 2 min (top), and the poisoning curve resulting from an
initial rate analysis of the data (bottom).
effect, the number is not significantly less than 1 and is close to
0.5. Thus, these data are not consistent with Rh(0) nano-
particles for which much less than 1 equivalent of phen would
be needed for complete poisoning.60−62 A conceivable
explanation for ∼0.5 equivalent phen causing poisoning is
that the active catalyst is bimetallic and that 1 equivalent of
phen deactivates 1 equivalent of Rh2 dimer.
Finke et al. also showed that the poisoning curve for the
hydrogenation of benzene with Rh nanoparticles resembles an
exponential decay, whereas the reaction catalyzed by a
homogeneous catalyst has a plot with a sigmoidal shape.61
Our poisoning curve is also sigmoidal. More precisely, our data
indicate that phen is a somewhat weakly binding poison toward
the Rh sulfide catalyst, which is more consistent with a
homogeneous molecular catalyst.
Finke et al. showed that ∼2 nm Rh nanoparticles are
significantly poisoned (for hydrogenation catalysis) by only 0.1
F
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 4. Schematic Indicating Species Proposed to Be Encountered during the Thermal Reaction of RhCl3(Bu2S)3 with
MDHM and Subsequent Degradation of the Active Species
silylation product M2D(CH2)2MM with no additional MDClM.
Additionally, when monitoring the catalytic reaction in situ with
29Si NMR, the amount of MDClM released approaches 2
equivalent (see Figures S13−S16). Release of the final
equivalent of MDClM (third equivalent of Cl) is only observed
toward the end of the catalytic reaction and especially after it
has completed. Therefore, we propose that the catalytic
reaction occurs with one Rh−Cl still in place as an ancillary
ligand and that its elimination (forming MDClM) is slow and
represents catalyst deactivation. The proposal of a Rh−Cl
containing catalyst is similar to Wilkinson’s catalyst.68−70
Separate experiments have shown that, under stoichiometric
conditions, the decomposition of RhHb occurs over several
hours and releases the third equivalent of Cl.
However, there remains the possibility that trace Rh particles
or “Rh(Bu2S)2H” species (reductive elimination of MDClM
from RhHb) could be highly active and be the kinetically
dominant catalyst. To test this hypothesis, we attempted to
generate Rh particles in situ.
Our first attempts to study Rh(Bu2S)2H/nanoparticles
involved the thermal decomposition of RhHb. Knowing that
RhHb rapidly decomposes to heterogeneous Rh black at high
temperature, we heated it at 55 °C for ∼3.5 h (Figure 4 and
Scheme 4), resulting in significant darkening of the solution
(but without the formation of heterogeneous species). The
resulting mixture was found to have significantly diminished
activity versus RhHb (Figure 9). Under these conditions the
∼0.5 equivalent of it was observed. The expected Rh byproduct
[RhCl(Bu2S)2]2 was only observed in low quantities (Figure
S12). This is perhaps not surprising, given that [RhCl(Bu2S)2]2
is only a precatalyst (is not part of the catalytic cycle). Similarly,
an induction period, albeit very short, was observed for RhHb,
which supports that it is a precatalyst. Future work will require
more detailed characterization of the reaction mixtures
produced upon stoichiometric reactions of the precatalysts
with MDHM and MMVi.
3. CONCLUSIONS
Detailed studies have been performed on the precatalyst
RhCl3(Bu2S)3 to better understand its role in generating the
active catalyst, its identity, and its decomposition into inactive
forms (Scheme 4). The key data are summarized below:
(a) The long bath life of RhCl3(Bu2S)3 is primarily due to an
induction period in which it reacts with 2 equivalent of
Si−H to form 1/2[RhHCl(SiR3)(Bu2S)2]2.
(b) Preactivating the catalyst by isolating [RhHCl(SiR3)-
(Bu2S)2]2 (e.g., RhHb) can dramatically reduce the time
frame of the reaction, but the active catalyst is not
changed (Figure 10).
Figure 10. Monitoring of progress of the reaction of MDHM with
MMVi using [RhCl(Bu2S)2]2 (RhCl(Bu2S)2, blue) and RhCl3(Bu2S)3
(orange).
Figure 9. Comparison of catalytic activity of RhHb before (blue) and
after (orange) it has been allowed to decompose thermally for 3.5 h at
55 °C.
(c) Other intermediates observed during activation are RhHa
and [RhCl(Bu2S)2]2, the latter having an intermediate
induction period.
(d) Only 2 equivalent of Bu2S per Rh are necessary for the
active catalyst. Approximately 1 equivalent of Bu2S
dissociates as the active species forms.
products of RhHb decomposition have poor catalytic activity.
Further heating/decomposition into heterogeneous Rh black
resulted in a complete loss of activity.
F. Initial Work on Stoichiometric Reactions. We attempted
to further explore the role of RhHb by studying its
stoichiometric reactions with MDHM and MMVi; however,
the reaction mixtures were complex, and we cannot yet propose
a specific catalytic cycle. The reaction of RhHb with MMVi does
release the hydrosilylation product M2D(CH2)2MM, but only
(e) During catalyst activation, two Cl ligands are removed
(forming MDClM), and one remains bound to Rh.
Further heating eventually removes the final Cl, and the
catalyst deactivates.
(f) RhHb slowly eliminates Cl-SiR3, but this reaction is slow
compared to product release.
G
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
0.092 g (0.21 mmol, 35%). H NMR: δ 3.30 (m, 1H), 2.90 (m, 2H),
2.54 (bs, 2H), 2.20 (m, 1H), 1.87 (m, 2H), 1.79 (m, 1H), 1.70 (m,
1H), 1.34 (sextet, JH−H = 7.4 Hz), 1.23 (t, JH−H = 7.4 Hz), 0.95 (t,
JH−H = 7.4 Hz). Anal. Cald. For C32H72Cl2Rh2S4: C, 44.59; H, 8.42; N,
0. Found: C, 44.61; H, 8.19; N, 0%. When a similar experiment was
performed using only 1 equivalent of HTMS, only ∼0.015 g of the
orange precipitate was collected.
(g) Given (d) and (e), the active catalyst should contain the
RhxClx(Bu2S)2x fragment.
(h) RhHb is not catalytically active as evidenced by its brief
induction period.
(i) Various attempts to isolate the active catalyst were
unsuccessful. One possibility is that the dimeric nature of
RhHb requires it to dissociate into monomers before
becoming active.
(j) The Rh sulfide catalysts are tolerant to poisoning. Hg
and Bu2S have limited effects, and polydentate ligands are
needed to stoichiometrically deactivate the catalyst.
(k) The combined data (poison ligand, Hg, Rh nanoparticle
tests) are more consistent with a homogeneous small
molecule catalyst than with Rh nanoparticles. However,
the exact identity of the active catalyst has not yet been
determined.
(l) Various attempts at generating the catalyst from
alternative synthetic routes or allowing RhHb to
decompose in the presence of substrates did not improve
the induction period. Isolating the active catalyst is
nontrivial. The precatalyst easily decomposes into soluble
and insoluble forms which do not further activate
(Scheme 4).
(m) In comparison to Pt (Karstedt’s catalyst), the Rh sulfide
catalyst is a significantly slower catalyst in addition to the
long induction period (Figure 10).
[RhCl(Bu2S)2]2 Synthesis from [RhCl(C2H4)2]2. A solution of
0.512 g (1.32 mmol) of [RhCl(Bu2S)2]2 in 20 mL of tetrahydrofuran
(THF) was treated with a solution of 0.770 g (5.3 mmol) of Bu2S in 5
mL of THF at 5 mL/h via a syringe pump. Approximately 60 min after
the addition was complete, the solvent was removed in vacuo, the dark
black/brown turbid solid was redissolved in 3 mL of toluene, and then
an orange precipitate was formed upon the addition of 30 mL of
hexane. The orange precipitate was collected on a frit (0.485 g) and set
aside, and a second crop was collected; the supernatant was dried in
vacuo until a thick liquid remained, and then 30 mL of hexane was
added. The resultant orange powder was collected on a frit (0.157 g).
The two crops were then combined, redissolved in ∼2 mL of toluene,
and then precipitated with ∼18 mL of hexane. The resultant orange
microcrystalline powder was collected on a frit and dried in vacuo.
Yield: 0.535 g (1.24 mmol, 47%). Product isolated in this way was
slightly impure: Anal. Cald. For C32H72Cl2Rh2S4: C, 43.59; H, 8.42; N,
0. Found: C, 43.64; H, 7.96; N, 0%. Recrystallization from toluene
afforded analytically pure product. Anal. Cald. For C32H72Cl2Rh2S4: C,
44.59; H, 8.42; N, 0. Found: C, 44.43; H, 8.20; N, 0%. Crystals of
sufficient quality for X-ray diffraction were produced by cooling in
toluene as above, or by layered diffusion of hexane into toluene.
RhHCl(SiMe(OSiMe3)2)(Bu2S)2, for Example, RhHb. This
complex was generated in situ as a 0.85 wt % Rh solution in
toluene-d8 as follows: 0.384 g of RhCl3(Bu2S)3 (0.59 mmol) was
dissolved in 6 g toluene-d8 and then was treated with 0.396 g of
1,1,1,3,5,5,5-heptamethyltrisiloxane (1.77 mmol). The resultant red
solution was heated at 55 °C for 1 h, at which point the solution was
pale orange. Integration with an internal standard at this stage
indicated ∼75% yield of the desired product. This reagent solution was
not further purified and was stored at −35 °C between uses. 1H NMR
(toluene-d8): δ 4.92 (m, unreacted MDHM), 3.06 (m, 4H, Rh(Bu2S)),
2.77 (m, 4H, Rh(Bu2S)), 2.32 (t, 6H, free/exchanging Bu2S), 1.65 (m,
8H, Rh(Bu2S)), 1.45 (m, 6H, free/exchanging Bu2S), 1.31 (m, 16H,
free/exchanging Bu2S and Rh(Bu2S)), 0.86 (t, 12H, J = 7.2 Hz,
Rh(Bu2S)), 0.82 (t, 12H, free Bu2S), 0.33 (bs, Rh−SiMe(OSiMe3)2),
0.29 (bs, MeSiCl(OSiMe3)2), 0.15 (bs, RhSiMe(OSiMe3)2 + MeSiCl-
(OSiMe3)2), −18.49 (d, Rh−H, JRh−H = 27.9 Hz).
During peer review, one reviewer expressed skepticism with
regard to conclusion (k). We therefore are sharing these data
with the intention that others can interpret differently.
Additional experiments such as in situ transmission electron
microscopy (TEM) or X-ray absorption fine structure (XAFS)
would provide useful supporting data but were not available
during the course of this work.
4. EXPERIMENTAL SECTION
Unless otherwise indicated, all experiments were performed in an air-
free, nitrogen-filled mBraun glovebox whose sensors indicated an
atmosphere of less than 0.1 ppm of O2 and less than 0.1 ppm of H2O.
RhCl3(Bu2S)3. To a solution of rhodium chloride hydrate (1.000 g,
38−40% Rh, Sigma-Aldrich, assumed molecular formula of
RhCl3(H2O)3) in 10 mL of MeOH was added 1.669 g of Bu2S. The
red solution was heated at 60 °C for 1 h and then dried in vacuo at 85
°C for 2 h. A thick red oil remained. Yield: 2.456 g (99.7%). 1H NMR
(toluene-d8): δ 3.5 ppm (bs, 4H, Rh-SCH2ax), 2.3 (bs, 8H, Rh-
SCH2eq), 1.72 (m, 12H, Rh-SCH2CH2), 1.97 (m, 12H, Rh-
SCH2CH2CH2), 0.95 (m, 18H, Rh-SCH2CH2CH2CH3). 13C NMR
(toluene-d8): δ 37.3 (s, 2C, Rh−SCH2eq), 35.4 (s, 2C, Rh−SCH2ax),
31.0 (s, 2C, Rh-SCH2CH2eq), 30.5 (s, 2C, Rh-SCH2CH2ax), 22.1 (s,
2C, Rh-SCH2CH2CH2eq), 22.0 (s, 2C, Rh-SCH2CH2CH2 ax), 13.8 (s,
2C, Rh-SCH2CH2CH2CH3eq), 13.7 (s, 2C, Rh-SCH2CH2CH2CH3ax).
[RhCl(Bu2S)2]2. To a 20 mL scintillation vial was added 0.4 g (0.62
mmol) of RhCl3(Bu2S)3 followed by 2.00 g of toluene-d8. The mixture
was briefly shaken, and then 0.275 g (1.24 mmol) of 1,1,1,3,5,5,5-
heptamethyltrisiloxane (HTMS) was added. The mixture was briefly
shaken; a stirbar was added, and the mixture was heated at 55 °C for 1
h. At this point the sample was removed from the heating plate, and
NMR Monitoring Reactions. To a 20 mL scintillation vial was
added 0.063 g (0.097 mmol) of RhCl3(Bu2S)3 followed by 1.00 g of
toluene-d8. The mixture was briefly shaken, and then 0.065 g (0.29
mmol) of 1,1,1,3,5,5,5-heptamethyltrisiloxane (HTMS) was added.
The mixture was briefly shaken and then immediately transferred to a
J. Young valve NMR tube. The tube was sealed and stored in a
glovebox freezer at −30 °C until the experiment began. At that point,
it was removed from the glovebox and then inserted into an NMR
1
spectrometer that was preheated to the appropriate temperature. H,
29Si, and 13C NMR spectra were collected as appropriate. Each nucleus
was recorded in a separate experiment to maximize the frequency of
time points. Alternatively, the reaction mixture could be heated outside
of the NMR spectrometer and then inserted into the spectrometer at
various time points for analysis.
Model Catalytic Reactions for Kinetics. A solution containing
9.00 g of toluene, 0.909 g of 1,1,1,3,5,5,5-heptamethyltrisiloxane, 0.712
g of vinylpentamethyldisiloxane, and 0.348 g of dodecane was added to
a 40 mL scintillation vial with a disposable Teflon stirbar and
preheated to the desired reaction temperature while it was stirred at
1500 rpm in a pie-block heating block; a heating block was selected
such that the sides of the block extended ∼1 cm above the top of
reaction solution. An aliquot of the desired precatalyst was injected
into the reaction mixture. Typically, we studied reactions at 0.1 mol %
catalyst loadings (305 μL of a 1 wt % solution of RhCl3(Bu2S)3 in
toluene). Time point aliquots (150 μL) were collected as desired. The
aliquots were immediately diluted with 1.25 mL of xylene and then
1
the H NMR spectrum was checked to ensure full consumption of
HMTS. The sample was then carefully concentrated in vacuo at room
temperature until ∼50% of the solvent had been removed, and an
orange crystalline material became visible. The product was stored at
−30 °C for 3 d and then collected on a frit via vacuum filtration. The
product was then redissolved in ∼3 mL of toluene and precipitated by
the dropwise addition of hexane (∼10 mL). The product was collected
on a frit, and the precipitation was repeated. The orange micro-
crystalline powder was collected on a frit and dried in vacuo. Yield:
H
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
were analyzed by GC. In cases where the catalyst speed was high
enough that some observable amount of reaction occurred in the GC
vial, xylene containing ∼0.0004 M phenanthroline (∼10 equivalent of
phen per equivalent of Rh) was used to stop the reaction; this was
typically only necessary for Pt, and for RhHb it was a precaution.
NMR Monitoring of RhHb with Model Substrates. A solution
of RhHb was generated as indicated above, and to this was added
various amount of 1:1 mixtures of 1,1,1,3,5,5,5-heptamethyltrisiloxane
and vinylpentamethyldisiloxane. After heating if appropriate, the
sample was transferred to a 10 mm Teflon NMR tube with a Teflon
stopper (for 29Si NMR) or a conventional J. Young valve 5 mm NMR
tube for 1H NMR. For the 29Si NMR experiments, ∼0.1 g of
Cr(AcAc)3 was added to the sample to aid in magnetic relaxation. We
have not found this to impact catalysis or the chemistry of the
complexes. When appropriate, internal standards were added;
(4) Jones, L. A.; Schmidt, R. G. In Technology of Pressure-Sensitive
Adhesives and Products; Benedek, I., Feldstein, M. M., Eds.; CRC Press:
Boca Raton, FL, 2009.
(5) Karstedt, B. D. Platinum-vinyl siloxanes. US3715334.
(6) Willing, D. N. Catalysts for the Rection of SiH with Organic
Compounds Containing Aliphatic Unsaturation. US3419593.
(7) Kookootsedes, G. J.; Pluddemann, E. P. Acetylenic Inhibited
Platinum Catalyzed Organopolysiloxane Composition. US3445420.
(8) Marrot, S.; Maadadi, Y. Hydrosilylation Reaction Inhibitors, and
Use Thereof for Preparing Stable Curable Silicone Compositions.
US2014/0329099A1.
(9) Stein, J.; Lewis, L. N.; Gao, Y.; Scott, R. A. In Situ Determination
of the Active Catalyst in Hydrosilylation Reactions Using Highly
Reactive Pt(0) Catalyst Precursors. J. Am. Chem. Soc. 1999, 121, 3693.
(10) Meister, T. K.; Riener, K.; Gigler, P.; Stohrer, J.; Herrmann, W.
A.; Kuhn, F. E. Platinum Catalysis Revisited Unraveling Principles of
Catalytic Olefin Hydrosilylation. ACS Catal. 2016, 6, 1274.
(11) Lewis, L. N.; Lewis, N. Platinum-Catalyzed Hydrosilylation-
Colloid Formation as the Essential Step. J. Am. Chem. Soc. 1986, 108,
7228.
(12) Lewis, L. N. On the Mechanism of Metal Colloid Catalyzed
Hydrosilylation: Proposed Explanations for Electronic Effects and
Oxygen Cocatalysis. J. Am. Chem. Soc. 1990, 112, 5998.
(13) Galeandro-Diamant, T.; Zanota, M.; Sayah, R.; Veyre, L.;
Nikitine, C.; de Bellefon, C.; Marrot, S.; Meille, V.; Thieuleux, C.
Platinum nanoparticles in suspension are as efficient as Karstedt’s
complex for alkene hydrosilylation. Chem. Commun. 2015, 51, 16194.
(14) Speier, J. L. Homogeneous Catalysis of Hydrosilation by
Transition Metals. Adv. Organomet. Chem. 1979, 17, 407.
(15) Webb, J. L.; Ritschard, H. V.; Lambert, J. M. Method and
Intermediates for Preparation of Bis(aminoalkyl)polydiorganosiloxane.
US5026890.
1
(Me3SiO)4Si (M4Q) for 29Si NMR, and trimethoxybenzene for H
NMR.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.7b00029.
X-ray crystallographic information (CIF)
Synthesis and NMR spectra, reaction monitoring data,
poisoning experiments, attempted synthesis and catalysis,
attempted activation, sample and crystal data (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: matthew.olsen@dowcorning.com.
■
(16) http://www.dowcorning.com/content/publishedlit/30-1163-
01.pdf. Date Accessed: Jan 29, 2016.
(17) Marciniec, B.; Maciejewski, H.; Pietraszuk, C.; Pawluc, P.
Hydrosilylation: A Comprehensive Review on Recent Advances; Springer,
2009; Vol. 1.
ORCID
Matthew T. Olsen: 0000-0002-7543-1070
Author Contributions
All authors have given approval to the final version of the
manuscript.
(18) Marciniec, B. Comprehensive Handbook on Hydrosilylation;
Pergamon Press: Oxford, U.K., 1992.
Funding
(19) Roy, A. K. A Review of Recent Progress in Catalyzed
Homogeneous Hydrosilation. Adv. Organomet. Chem. 2007, 55, 1.
(20) Nakajima, Y.; Shimada, S. Hydrosilylation reaction of olefins:
recent advances and perspectives. RSC Adv. 2015, 5, 20603.
(21) Sun, J.; Deng, L. Cobalt Complex-Catalyzed Hydrosilylation of
Alkenes and Alkynes. ACS Catal. 2016, 6, 290.
(22) Tondreau, A. M.; Atienza, C. C. H.; Weller, K. H.; Nye, S. A.;
Lewis, K. M.; Delis, J. G. P.; Chirik, P. J. Iron Catalysts for Selective
Anti-Markovnikov Alkene Hydrosilylation Using Tertiary Silanes.
Science 2012, 335, 567.
(23) Hojilla Atienza, C. C.; Tondreau, A. M.; Weller, K. H.; Lewis, K.
M.; Cruse, R. W.; Nye, S. A.; Boyer, J.; Delis, J. G. P.; Chirik, P. J.
High-Selectivity Bis(imino)pyridine Iron Catalysts for the Hydro-
silylation of 1,2,4-Trivinylcyclohexane. ACS Catal. 2012, 2, 2169.
(24) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. Preparation and
Molecular and Electronic Structures of Iron(0) Dinitrogen and Silane
Complexes and Their Application to Catalytic Hydrogenation and
Hydrosilation. J. Am. Chem. Soc. 2004, 126, 13794.
Funding for this work was provided by Dow Corning
Corporation.
Notes
The authors declare no competing financial interest.
The GE shorthand for silicones is as follows: M = Me3SiO1/2, D
= Me2SiO2/2, T = MeSiO3/2, Q = SiO4/2. The subscript
numerator indicates the number of oxygen linkages shared
between the Si atom and another Si atom. The denominator is
meant to keep track of the appropriate empirical formula. Thus,
the hypothetical siloxane ((MD)3T)4Q is used to name
((Me3SiOMe2SiO)3SiO)4Si.
ACKNOWLEDGMENTS
■
M.T.O. would like to thank Dr. L. Durfee for extensive
discussions on this work as well as Prof. R. Finke for discussions
on the catalyst poisoning experiments.
(25) Chen, C.; Hecht, M. B.; Kavara, A.; Brennessel, W. W.;
Mercado, B. Q.; Weix, D. J.; Holland, P. L. Rapid, Regioconvergent,
Solvent-Free Alkene Hydrosilylation with a Cobalt Catalyst. J. Am.
Chem. Soc. 2015, 137, 13244.
(26) Peng, D.; Zhang, Y.; Du, X.; Zhang, L.; Leng, X.; Walter, M. D.;
Huang, Z. Phosphinite-Iminopyridine Iron Catalysts for Chemo-
selective Alkene Hydrosilylation. J. Am. Chem. Soc. 2013, 135, 19154.
(27) Ibrahim, A. D.; Entsminger, S. W.; Zhu, L.; Fout, A. R. A Highly
Chemoselective Cobalt Catalyst for the Hydrosilylation of Alkenes
using Tertiary Silanes and Hydrosiloxanes. ACS Catal. 2016, 6, 3589.
REFERENCES
■
(1) Lewis, L. N.; Stein, J.; Gao, Y.; Colborn, R. E.; Hutchins, G.
Platinum Catalysts Used in the Silicones Industry. Platinum Metals Rev.
1997, 41, 66.
(2) Troegel, D.; Stohrer, J. Recent advances and actual challenges in
late transition metal catalyzed hydrosilylation of olefins form an
industrial point of view. Coord. Chem. Rev. 2011, 255, 1440.
(3) Eckberg, R. P. In Coatings Technology Handbook; Tracton, A. A.,
Ed.; CRC Press: Boca Raton, FL, 2005.
I
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(28) Buslov, I.; Keller, S. C.; Hu, X. Alkoxy Hydrosilanes As
Surrogates of Gaseous Silanes for Hydrosilylation of Alkenes. Org. Lett.
2016, 18, 1928.
(29) Greenhalgh, M. D.; Frank, D. J.; Thomas, S. O. Iron-Catalysed
Chemo-, Regio-, and Stereoselective Hydrosilylation of Alkenes and
Alkynes using a Bench-Stable Iron (II) Pre-Catalyst (pages 584−590).
Adv. Synth. Catal. 2014, 356, 584.
(30) Wu, J. Y.; Stanzl, B. N.; Ritter, T. A Strategy for the Synthesis of
Well-Defined Iron Catalysts and Application to Regioselective Diene
Hydrosilylation. J. Am. Chem. Soc. 2010, 132, 13214.
(31) Noda, D.; Tahara, A.; Sunada, Y.; Nagashima, H. Non-Precious-
Metal Catalytic Systems Involving Iron or Cobalt Carboxylates and
Alkyl Isocyanides for Hydrosilylation of Alkenes with Hydrosiloxanes.
J. Am. Chem. Soc. 2016, 138, 2480.
(52) Arkles, B.; Larson, G. Silicon Compounds: Silanes and Silicones,
2nd ed.; Gelest, Inc.: Morrisville, PA, 2008.
(53) Onopchenko, A.; Sabourin, S. T.; Beach, D. L. Rhodium(I)-
Catalyzed Hydrosilylation of Styrene. J. Org. Chem. 1983, 48, 5101.
(54) Onopchenko, A.; Sabourin, S. T.; Beach, D. L. Vinyl- and
Allylsilanes fromt he Rhodium(I)-Catalyzed Hydrosilylation of 1-
Alkenes with Trialkylsilanes. J. Org. Chem. 1984, 49, 3389.
(55) Atienza, C. C. H.; Diao, T.; Weller, K. H.; Nye, S. A.; Lewis, K.
M.; Delis, J. G. P.; Boyer, J.; Roy, A.; Chirik, P. J. Bis(imino)pyridine
Cobalt-Catalyzed Dehydrogenative Silylation of Alkenes: Scope,
Mechanism, and Origins of Selective Allylsilane Formation. J. Am.
Chem. Soc. 2014, 136, 12108.
(56) Marciniec, B. Silicon Chem. 2002, 1, 155.
(57) Comprehensive Handbook of Hydrosilylation; Marciniec, B., Ed.;
Pergamon Press: Oxford, England, 1992.
(58) Pitt, C. G.; Skillern, K. R. Vinyl-hydrogen ligand exchange on
silicon. J. Organomet. Chem. 1967, 7, 525.
(59) Murray, S. G.; Hartley, F. R. Coordination Chemistry of
Thioethers, Selenoethers, and Telluroethers in Transition-Metal
Complexes. Chem. Rev. 1981, 81, 365.
(60) Bayram, E.; Linehan, J. C.; Fulton, J. L.; Roberts, J. A. S.;
Szymczak, N. K.; Smurthwaite, T. D.; Ozkar, S.; Balasubramanian, M.;
Finke, R. G. Is It Homogeneous or Heterogeneous Catalysis Derived
from [RhCp*Cl2]2? In Operando XAFS, Kinetic, and Crucial Kinetic
Poisoning Evidence for Subnanometer Rh4 Cluster-Based Benzene
Hydrogenation Catalysis. J. Am. Chem. Soc. 2011, 133, 18889.
(61) Bayram, E.; Finke, R. G. Quantitative 1,10-Phenanthroline
Catalyst-Poisoning Kinetic Studies of Rh(0) Nanoparticle and Rh4
Cluster Benzene Hydrogenation Catalysts: Estimates of the Poison
Kassociation Binding Constants, of the Equivalents of Poison Bound and
of the Number of Catalytically Active Sites for Each Catalyst. ACS
Catal. 2012, 2, 1967.
(32) Sunada, Y.; Noda, D.; Soejima, H.; Tsutsumi, H.; Nagashima, H.
Combinatorial Approach to the Catalytic Hydrosilylation of Styrene
Derivatives: Catalyst Systems Composed of Organoiron(0) or (II)
Precursors and Isocyanides. Organometallics 2015, 34, 2896.
(33) Kownacka, A.; Kownacki, I.; Marciniec, B.; Nguyen, B. T.;
Surgenor, A. E.; Taylor, R. G.; Tzou, M. Hydrosilylation reaction
catalysts and curable compositions and methods for their preparation
and use. US20140296468A1.
(34) Marciniec, B.; Kownacka, A.; Kownacki, I.; Hoffmann, M.;
Taylor, R. G. Hydrosilylation vs. dehydrogenative silylation of styrene
catalyzed by iron(0) carbonyl complexes with multivinylsilicon ligands
- Mechanistic implications. J. Organomet. Chem. 2015, 791, 58.
(35) Chandra, G.; Rowland, C. M. Coating Process. US3960810.
(36) Chandra, G. Rhodium complexes containing sulfur and use as
catalysts. GB1448826.
(37) Armstrong, S. Rhodium catalyst and siloxane coating
composition containing the same. EP0510847A1.
(38) Kropp, P. L.; Bennington, L. D.; Cross, R. P.; Issari, B. Heat
(62) Bayram, E.; Linehan, J. C.; Fulton, J. L.; Szymczak, N. K.; Finke,
R. G. Determination of the Dominant Catalyst Derived from the
Classic [RhCp*Cl2]2 Precatalyst System: Is it Single-Metal Rh1Cp*-
Based, Subnanometer Rh4 Cluster-Based, or Rh(0)n Nanoparticle-
Based Cyclohexene Hydrogenation Catalysis at Room Temperature
and Mild Pressures? ACS Catal. 2015, 5, 3876.
(63) Widegren, J. A.; Finke, R. G. A review of the problem of
distinguishing true homogeneous catalysis from soluble or other metal-
particle heterogeneous catalysis under reducing conditions. J. Mol.
Catal. A: Chem. 2003, 198, 317.
curable silicone compositions. WO2002072704 A2.
(39) Griswold, R. M.; Eckberg, R. P.; Bisaillon, M. J.; Frye, R. L.
Solvent resistant polyurethane. US2007/0093595A1.
(40) Peters, Y. A. Silicone release coatings containing rhodium.
US4961963.
(41) Kleyer, D. L.; Nguyen, B. T.; Hauenstein, D. E.; Davern, S. P.;
Schulz, W. J. Method for controlling hydrosilylation in a reaction
mixture. US5359111.
(42) Farris, D. D.; Rinard, C. J.; Leatherman, M. D. Selective
hydrosilylation method. US7259220B1.
(43) Rinard, C. J.; Farris, D. D.; Leatherman, M. D. Selective
hydrosilylation conditions. US2008/0033136A1.
(44) Eckberg, R. P.; Agars, R. F. Syntheses of epoxysilicones. US
(64) Hagen, C. M.; Widegren, J. A.; Maitlis, P. M.; Finke, R. G. Is It
Homogeneous or Heterogeneous Catalysis? Compelling Evidence for
Both Types of Catalysts Derived from [Rh(η5-C5Me5)Cl2]2 as a
Function of Temperature and Hydrogen Pressure. J. Am. Chem. Soc.
2005, 127, 4423.
(65) Widegren, J. A.; Bennett, M. A.; Finke, R. G. Is It Homogeneous
or Heterogeneous Catalysis? Identification of Bulk Ruthenium Metal
as the True Catalyst in Benzene Hydrogenations Starting with the
Monometallic Precursor, Ru(II)(η6-C6Me6) (OAc)2, Plus Kinetic
Characterization of the Heterogeneous Nucleation, Then Autocatalytic
Surface-Growth Mechanism of Metal Film Formation. J. Am. Chem.
Soc. 2003, 125, 10301.
(66) Weddle, K. S.; Aiken, J. D. I.; Finke, R. G. Rh(0) Nanoclusters
in Benzene Hydrogenation Catalysis: Kinetic and Mechanistic
Evidence that a Putative [(C8H17)3NCH3]+[RhCl4]− Ion-Pair Catalyst
Is Actually a Distribution of Cl− and [(C8H17)3NCH3]+ Stabilized
Rh(0) Nanoclusters. J. Am. Chem. Soc. 1998, 120, 5653.
(67) Hornstein, B. J.; Aiken, J. D. I.; Finke, R. G. Nanoclusters in
Catalysis: A Comparison of CS2 Catalyst Poisoning of Polyoxoanion-
and Tetrabutylammonium-Stabilized 40 6 Å Rh(0) Nanoclusters to
5% Rh/Al2O3, Including an Analysis of the Literature Related to the
CS2 to Metal Stoichiometry Issue. Inorg. Chem. 2002, 41, 1625.
(68) Halpern, J.; Wong, C. S. Hydrogenation of tris-
(triphenylphosphine)chlororhodium(I). J. Chem. Soc., Chem. Commun.
1973, 629.
5258480.
(45) Heidingsfeldova, M.; Capka, M. Rhodium Complexes as
Catalysts for Hydrosilylation Crosslinking of Silicone Rubber. J.
Appl. Polym. Sci. 1985, 30, 1837.
(46) Trofimov, A. E.; Spevak, V. N.; Lobadyuk, V. I.; Skvortsov, N.
K.; Reikhsfel'd, V. O. Hydrosilylation in the presence of platinum(II)
complexes with sulfoxides and sulfides. Zh. Obshch. Khim. 1989, 59,
2048.
́
(47) Bayοn, J. C.; Claver, C.; Masdeu-Bultο, A. M. Homogenous
catalysis with transition metal complexes containing sulfur ligands.
Coord. Chem. Rev. 1999, 193−195, 73.
(48) James, B. R.; Ng, F. T. T. Activation of Hydrogen by Rhodium
Complexes containing Sulphide Ligands. Part II. The Trichlorotris-
(diethy1 sulphide)rhodium(iii)-catalysed Hydrogenation of Maleic
Acid. J. Chem. Soc., Dalton Trans. 1972, 355.
(49) James, B. R.; Ng, F. T. T.; Rempel, G. L. Homogeneous
hydrogenation of olefinic substrates using rhodium complexes
containing sulfur ligands. Inorg. Nucl. Chem. Lett. 1968, 4, 197.
(50) James, B. R.; Ng, F. T. T. Activation of hydrogen by rhodium
complexes containing sulfide ligands. III. Substrate effects and ligand
effects. J. Chem. Soc., Dalton Trans. 1972, 1321.
(51) James, B. R.; Ng, F. T. T. Homogeneous Hydrogenation of
Unsaturated Carboxylic Acids Using Chlororhodate(I) Complexes and
Rhodium(I) Diethylsulfide Complexes. Can. J. Chem. 1975, 53, 797.
(69) Halpern, J.; Okamoto, T.; Zakhariev, A. Mechanism of the
chlorotris(triphenylphosphine) rhodium(I)-catalyzed hydrogenation
J
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
of alkenes. The reaction of chlorodihydridotris(triphenylphosphine)-
rhodium(III) with cyclohexene. J. Mol. Catal. 1977, 2, 65.
(70) Halpern, J. Mechanistic aspects of homogeneous catalytic
hydrogenation and related processes. Inorg. Chim. Acta 1981, 50, 11.
K
DOI: 10.1021/acs.inorgchem.7b00029
Inorg. Chem. XXXX, XXX, XXX−XXX