Catalytic Olefin Hydrosilations Mediated by Ruthenium η3-H2Si σ Complexes of Primary and Secondary Silanes

Article abstract of DOI:10.1021/acscatal.8b02161

Unambiguous examples of η³-H₂SiH(R) complexes featuring a terminal Si-H bond have been prepared and examined as possible intermediates in olefin hydrosilation. These species were generated by displacement of the secondary silane ligands in [PhBP^Ph₃]RuH[η³-H₂SiMePh] (1b) (PhBP^Ph₃ = PhB(CH₂PPh₂)₃^- by primary silanes RSiH₃ to generate [PhBP^Ph₃]RuH[η³-H₂SiH(R)] (R = Cy (1d), CH₂CH₂Ph (1e), Trip = 2,4,6-ⁱPr₃C₆H₂ (1f)). Complexes 1d and 1e were characterized in solution, whereas 1f was isolated and studied in detail. Complex 1b is not a competent precatalyst for the hydrosilation of 1-hexene with CySiH₃, whereas comparable conditions gave reasonable yields for the selective anti-Markovnikov hydrosilations of Cl₃SiCH₂CH=CH₂ (89%), p-chlorostyrene (73%), and allyl chloride (70%). The ¹H NMR spectrum of 1f collected at -30 °C displays a downfield signal (δ = 8.26 ppm) for the terminal Si-H bond that suggests electronic similarities between 1f and cationic silylene dihydrides [Cp?(ⁱPr₃P)Ru(H)₂=SiH(R)]⁺ that mediate olefin hydrosilations via the direct insertion of the C=C bond into the terminal Si-H bond. However, further mechanistic considerations, including results on the hydrosilation of p-chlorostyrene with the secondary silane Et₂SiH₂ and [PhBP^Ph₃]RuH[η³-H₂SiEt₂] (1a) as the catalyst, indicate that an insertion mechanism involving a Ru-H (rather than a Si-H) group is possible. DFT investigations of the hydrosilation of several olefins with CySiH₃ using 1d as the catalyst reveal a preferred pathway involving olefin insertion into a Ru-H bond followed by migration of the resulting alkyl group to the silicon atom of an η³-H₂SiH(Cy) ligand. The latter process occurs via an unusual transition state in which a Ru-H-Si linkage acts as a pivot point to facilitate a Si-H bond cleavage/Si-C bond formation step that is otherwise similar to those involving the kite-shaped four-center transition states of σ-bond metathesis. Direct insertion into the Si-H bond is the next lowest accessible pathway.

Full text of DOI:10.1021/acscatal.8b02161

Subscriber access provided by UNIV OF WESTERN ONTARIO
Article
Catalytic Olefin Hydrosilations Mediated by Ruthenium
#3-H2Si #-Complexes of Primary and Secondary Silanes
Mark C. Lipke, Marie-Noelle Poradowski, Christophe Raynaud, Odile Eisenstein, and T. Don Tilley
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b02161 • Publication Date (Web): 24 Oct 2018
Downloaded from http://pubs.acs.org on October 24, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Catalytic Olefin Hydrosilations Mediated by
Ruthenium 3-H2Si -Complexes of Primary and
Secondary Silanes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mark C. Lipke,1,ǂ Marie-Noelle Poradowski,2,$ Christophe Raynaud,2
Odile Eisenstein,*,2 T. Don Tilley*,1
1. Department of Chemistry, University of California, Berkeley, Berkeley, California, 94720-
1460, United States.
2. Institut Charles Gerhardt, UMR 5253 CNRS-UM-ENSCM, Université de Montpellier, cc
1501, Place E. Bataillon, 34095 Montpellier, France.
ǂ Present address: Department of Chemistry and Chemical Biology, Rutgers, The State
University of New Jersey, New Brunswick, New Jersey 08903, United States.
$ Present address: Universitꢀ Lyon 1, CNRS UMR 5246, Institut de Chimie et Biochimie
Molꢀculaires et Supramolꢀculaires, 43 Boulevard du 11 Novembre 1918, F-69622 Villeurbanne,
France.
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 37
1
2
3
4
Abstract
Unambiguous examples of 3-H2SiH(R) complexes featuring a terminal Si–H bond have been
5
6
7
8
prepared and examined as possible intermediates in olefin hydrosilation. These species were
9
generated by displacement of the secondary silane ligands in [PhBPPh ]RuH[3-H2SiMePh] (1b,
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
PhBPPh = PhB(CH2PPh2)3) by primary silanes RSiH3 to generate [PhBPPh ]RuH[3-H2SiH(R)]
3
3
(R = Cy, 1d; CH2CH2Ph, 1e; Trip = 2,4,6-iPr3C6H2, 1f). Complexes 1d,e were characterized in
solution whereas 1f was isolated and studied in detail. Complex 1b is not a competent precatalyst
for the hydrosilation of 1-hexene with CySiH3, whereas comparable conditions gave reasonable
yields for the selective, anti-Markovnikov hydrosilations of Cl3SiCH2CH=CH2 (89%), p-
chlorostyrene (73%), and allyl chloride (70%). The 1H NMR spectrum of 1f collected at -30 °C
displays a downfield signal ( 8.26 ppm) for the terminal Si–H bond that suggests electronic
similarities between 1f and cationic silylene dihydrides [Cp*(iPr3P)Ru(H)2=SiH(R)]+ that mediate
olefin hydrosilations via the direct insertion of the C=C bond into the terminal Si—H bond.
However, further mechanistic considerations, including results on the hydrosilation of p-
chlorostyrene with the secondary silane Et2SiH2 and [PhBPPh ]RuH[3-H2SiEt2] (1a) as catalyst,
3
indicate that an insertion mechanism involving a Ru—H (rather than a Si—H) group is possible.
DFT investigations of the hydrosilation of several olefins with CySiH3 using 1d as a catalyst
reveal a preferred pathway involving olefin insertion into a Ru–H bond followed by migration of
the resulting alkyl group to the silicon atom of an 3-H2SiH(Cy) ligand. The latter process occurs
via an unusual transition state in which a Ru—H—Si linkage acts as a pivot point to facilitate an
Si—H bond cleavage / Si—C bond formation step that is otherwise similar to those involving the
kite-shaped, four-centered transition states of -bond metathesis. Direct insertion into the Si—H
bond is the next lowest accessible pathway.
Keywords: hydrosilation, silane, -complex, silylene, catalysis, mechanism, DFT
2
ACS Paragon Plus Environment

Page 3 of 37
ACS Catalysis
1
2
3
4
Introduction
5
6
7
8
The catalytic addition of an Si—H bond across a C=C double bond is an important
reaction for the formation of C—Si bonded compounds starting from olefins and hydridosilanes
(e.g. H—SiR3, R3 = combinations of alkyl, aryl, H, alkoxy, and/or Cl).1 For example, platinum-
catalyzed olefin hydrosilations are used for cross-linking silicone polymers, and these reactions
are among the largest scale applications of homogeneous catalysis.1 In addition, catalytic C—Si
bond forming reactions are important for laboratory scale syntheses owing to the utility of silanes
in cross coupling reactions,2 as precursors to alcohols,3,4 or as directing groups for C—H
functionalizations.4 Considering the numerous applications of hydrosilation reactions, it is
important to develop new catalysts that offer wider substrate scope, better selectivity, and/or
lower cost than the commonly employed platinum and rhodium catalysts. In this regard, it may
be worthwhile to investigate new catalytic cycles that do not rely on oxidative addition and
reductive elimination steps that are the basis for the classic Chalk-Harrod and modified Chalk-
Harrod mechanisms for platinum- and rhodium-catalyzed olefin hydrosilations.5 In particular, the
utilization of alternative fundamental reaction steps may be essential for developing catalysts
based on abundant first-row transition metals or main-group elements,6 and for hydrosilations of
substrates featuring C—X functional groups (X = halide, e.g. allyl chloride) that can compete
with Si—H bonds for activation at a transition metal center.7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Relevant studies in this laboratory have identified catalytic olefin hydrosilations that
feature a different type of fundamental Si—C bond-forming step: insertion of the olefin directly
into the Si—H bond of LnM=SiH(R) silylene complexes (LnM = [Cp*(iPr3P)RuH2]+ and
[(PNP)IrH]+, PNP = [(2-iPr2P-4-Me-C6H3)2N]−; Scheme 1).8 For the ruthenium-catalyzed
examples, the intermediate silylene complexes possess a substantial degree of {Cp*(iPr3P)Ru[3-
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 37
1
2
3
4
Scheme 1: The Ru(II) to Ru(IV) Continuum in Electrophilically Activated Hydrosilations
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H2SiH(R)]}+ character.8b The Ru(II) 3-H2SiH(R) and Ru(IV) dihydride silylene bonding
descriptions are related via a symmetric double Si—H oxidative addition process, and a
continuum appears to exist between these structure types (Scheme 1).9 These observations raise
the question of how much Ru(IV) character is needed to promote these hydrosilations.
Notably, electrophilic 3-H2SiRR' complexes of the type [PhBPPh ]RuH[3-H2SiRR']
3
(RR' = Et2, 1a; MePh, 1b; Ph2, 1c) lie towards the Ru(II) end of the continuum9a and engage in
hydrosilation reactions with ketones and isocyanides via mechanisms that involve binding of the
unsaturated substrate to the electrophilic silicon center.10 Since these hydrosilations appear to
exclusively involve Ru(II) intermediates, it seemed possible that analogous 3-H2SiH(R)
complexes might mediate olefin hydrosilations that do not involve redox changes at the metal
center. However, initial efforts to prepare 3-H2SiH(Ar) complexes supported by the
[PhBPPh ]RuH fragment resulted in formation of unusual diruthenium hydridosilicate complexes
3
such as {[PhBPPh ]Ru}2[-4,4-H6Si] (2, Scheme 2).11 Thus, mechanistic relationships between
3
(H)2Ru=SiHR and Ru[3-H2SiH(R)] complexes have been difficult to establish given the lack of
definitive examples of the latter type. This contribution describes the successful in situ
generation of stable [PhBPPh ]RuH[3-H2SiH(R)] (R = Cy, 1d; (CH2CH2Ph), 1e) complexes by
3
displacement of PhMeSiH2 from 1b with primary alkyl silanes (e.g. CySiH3, (PhCH2CH2)SiH3,
4
ACS Paragon Plus Environment

Page 5 of 37
ACS Catalysis
1
2
3
4
Scheme 2: Displacement of 3-H2SiRR' Ligands by Primary Silanes
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2). Complex 1d was found to be an effective catalyst for the hydrosilation of several
olefin substrates with CySiH3, including the challenging substrate allyl chloride. A related 3-
H2SiH(Trip) complex (Trip = 2,4,6-iPr3C6H2, 1f) was isolated and fully characterized to confirm
the identity of 1d-f as unambiguous examples of primary silane 3-H2SiH(R) complexes.
Results and Discussion
Hydrosilations Mediated by [PhBP3]RuH(3-H2SiH(Cy)) (1d). Since previous efforts
to prepare 3-H2SiH(Ar) complexes were unsuccessful due to competing redistribution of the
primary aryl silane to form 2 and diaryl silanes (i.e. Ar2SiH2),11 it seemed possible that 3-
H2SiH(alkyl) complexes might be more stable, since alkyl silanes are less susceptible to
redistribution at silicon.12 Thus, efforts to prepare 3-H2SiH(R) complexes focused on
displacement of PhMeSiH2 from 1b using an excess of a primary alkyl silane. Note that it has
previously been demonstrated that secondary silanes readily undergo ligand exchanges that allow
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 37
1
2
3
4
5
6
7
8
9
interconversions of 1a-c, and there does not appear to be a strong preference for binding of one
silane over another (e.g., addition of 1 equiv of PhMeSiH2 to 1c results in an approximately
equal ratio of 1b to 1c).13
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Treatment of 1b with CySiH3 (10 equiv in C6D6) resulted in displacement of PhMeSiH2
1
31
(by H NMR spectroscopy) and formation of a new [PhBPPh ]Ru species with a P{1H} NMR
3
shift (46 ppm) that is similar to those of 3-H2SiRR' complexes 1a-c.9a The new complex (1d) is
1
presumed to be the 3-H2SiH(Cy) -complex [PhBPPh ]RuH[3-H2SiH(Cy)], but the H NMR
3
spectrum displayed a very broad Ru—H resonance ( -5.5 to -8.5) that was not useful for clearly
identifying this new species (see below for further discussion).
Notably, complex 1d was stable for at least 20 h in solution, and this provided the
opportunity to investigate this complex (generated in situ from 1b and CySiH3) as a possible
catalyst for olefin hydrosilation reactions using CySiH3 as the silane substrate (eq 1). Initial
experiments revealed that 1d is not an effective catalyst for the hydrosilation of 1-hexene with
CySiH3 after 20 h at 23 °C or at 60 °C (by 1H NMR spectroscopy in C6D6, Table 1, entries 1 and
1
2). Instead, the H NMR spectra of these reaction mixtures displayed several new Ru—H
resonances, and thus it appears that 1d undergoes decomposition in the presence of 1-hexene.
However, complex 1d is an effective precatalyst for hydrosilations of styrene, p-chlorostyrene,
allyl trichlorosilane, and allyl chloride, with products formed in moderate to good yield after
20 h at 23 °C (Table 1, entries 3 - 9). For each substrate, selective anti-Markovnikov addition is
6
ACS Paragon Plus Environment

Page 7 of 37
ACS Catalysis
1
2
3
4
5
Table 1. Hydrosilation of olefins with CySiH3 using 1b as a precatalyst (eq 1).a
Olefin
1b
Time
Yield
(%)b
Substrate
(mol %)
(h)
6
1
2c
3
2.5
2.5
2.5
20
20
20
< 5
< 5
89
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
5
2.5
2.5
20
73
20
48
20
56
56
30
6c
7
2.5
5
20
65
8
2.5
2.5
20
20
47
70
9d
a) 23 °C in C6D6 using 1.1 equiv CySiH3 b) Yield of anti-Markovnikov hydrosilation product determined by 1H
NMR using a C6Me6 internal standard. c) Heated to 60 °C. d) 5 equiv of CySiH3.
clearly indicated by a pseudo-quartet SiH resonance (3JHH ≈ 3.5 Hz) that results from 3-bond J-
coupling of the product Si—H hydrogens to the three nearest hydrogens (one methine and two
methylene hydrogens). With a 2.5 mol % loading of 1b, the hydrosilation was most effective for
allyl trichlorosilane (Table 1, entry 3) and p-chlorostyrene (entry 4), while the yields were only
moderate for the hydrosilation of styrene (entry 5) and allyl chloride (entry 8). Notably, allyl
trichlorosilane is an uncommon hydrosilation substrate, and the presence of two dissimilar silyl
substituents in the product could make Cl3SiCH2CH2SiH2Cy useful as a synthetic building block.
Effective catalysts for the hydrosilation of allyl chloride are uncommon, and the resulting
1-chloro-3-silylpropane products are potentially useful chemical intermediates.7 Thus, efforts
directed at optimizing these hydrosilation reactions were undertaken. Attempts to improve the
yields for the hydrosilations of allyl chloride and styrene employed longer reaction times (48 h,
entry 5) or higher temperatures (60 °C, entry 6), but were not effective. The use of a sterically
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 37
1
2
3
4
less hindered primary silane substrate, PhCH2CH2SiH3, was also examined. This silane appeared
5
6
to react with 1b to give [PhBPPh ]RuH[3-H2SiH(CH2CH2Ph)] (1e, as indicated by a new
3
7
8
9
31P{1H} NMR resonance at  46 ppm), but resulted in much lower yields of hydrosilation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
products (by H NMR spectroscopy).14 The relatively low yields associated with the
hydrosilation of styrene may be due to competing decomposition of the 3-H2SiH(Cy) complex
1d, which is the only major ruthenium species initially observed in these reaction mixtures by 1H
and P{1H} NMR spectroscopy. Continued monitoring of these reactions reveals that complex
31
1d is converted to several new ruthenium hydride complexes while catalytic activity decreases
until no additional turnover is observed after 20 h (Table 1, entry 5). It is worth noting that 1d
decomposes more rapidly during the hydrosilation of styrene (nearly full decomposition after 16
h), while 20 % of 1d remains after 20 h for the hydrosilation of the more effective p-
chlorostyrene substrate. Thus, the effectiveness of catalysis appears to be correlated to the
presence of 1d, which is presumably the resting state of the catalytic cycle. However, doubling
the loading of 1d to 5 mol % provided only a modest increase in the yield for the hydrosilation of
styrene to 65 %, while 35 % of this alkene substrate remained unconsumed.
The allyl chloride substrate, in contrast, was almost entirely consumed under
1
hydrosilation conditions (95 % conversion by H NMR spectroscopy), and the low yield of
product in this case is due to formation of propene (20 – 25 %), propyl chloride (25 – 30 %), and
a small amount of another side product that was not identified. Selectivity for the desired
hydrosilation product was improved by using an excess of CySiH3 (5 equiv relative to the alkene
substrate), which increased the yield of 1-chloro-3-(CyH2Si)-propane to 70 %. This increase in
yield appears to result primarily from suppression of the formation of the hydrogenation product
1
propyl chloride, which was not observed by H NMR spectroscopy. Notably, this is the first
8
ACS Paragon Plus Environment

Page 9 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
example of an effective hydrosilation of allyl chloride with a primary silane, and the yield is
comparable to the best reported yields for the hydrosilation of allyl chloride with any other
silanes.7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Characterization of [PhBP3]RuH(3-H2SiH(Trip)) (1f). In order to study the
mechanism of the olefin hydrosilation reactions, efforts were made to confirm the identity of the
3-H2SiH(Cy) complex 1d, which was observed as the catalyst resting state (by 31P{1H} NMR)
prior to its decomposition to several unidentified [PhBPPh ]Ru species. The 1H NMR spectra for
3
1d, collected at temperatures between -70 °C to 20 °C, did not exhibit sharp Ru—H or Si—H
resonances that might identify this species. Furthermore, efforts to isolate 1d or grow single
crystals for X-ray diffraction failed, and thus attention turned to preparation of an analogue that
might be easier to identify. It seemed that the broadness of the Ru—H resonance for 1d might be
1
due to rapid exchange on the H NMR timescale of the terminal Si—H hydrogen with the Ru—
H—Si and Ru—H positions, and that use of a bulkier silane might slow these exchange
processes. Thus, 1b was treated with excess (Trip)SiH3 (7 equiv in C6D6), which resulted in
displacement of PhMeSiH2 and formation of 1f (by 1H and 31P{1H} NMR spectroscopy, eq 2).
Complex 1f was isolated as a yellow powder after treatment of 1b with (Trip)SiH3 (5
equiv in toluene), evaporation of solvent, and recrystallization from a fluorobenzene-pentane
solvent mixture. Elemental analysis indicated that samples of 1f prepared in this manner were
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 37
1
2
3
4
5
6
7
8
9
1
slightly impure (Anal. Calcd for 1f: C, 70.51; H, 6.71; Found: C, 69.84; H, 6.51), and H NMR
spectroscopy revealed that the hexahydridosilicate complex 2 was present as a minor impurity
(ca. 5 - 10 %). Thus, even the fairly bulky aryl silane (Trip)SiH3 appears to undergo
redistribution mediated by the [PhBPPh ]RuH fragment. The low solubility and high crystallinity
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
of complex 2 makes this species difficult to separate from samples of other [PhBPPh ]Ru
3
complexes,11 but 2 was found to be inactive as a catalyst for alkene hydrosilation, and it does not
react appreciably within 2 – 10 days with unsaturated species (e.g. XylNC, CO)15 that react
rapidly (<1 min) with 1a-c.10a Thus, the presence of 2 as a minor impurity in 1f is unlikely to
interfere with the investigation of the reactivity of 1f with olefins or other unsaturated substrates.
1
As with 1d, the room temperature H NMR spectrum of 1f displays a broad Ru—H
1
resonance ( -6.50 in benzene-d6), but the H NMR spectrum of 1f at -30 °C contains a much
sharper Ru—H resonance ( -6.51 ppm in toluene-d8, Figure 1b), along with a downfield signal
for the terminal Si—H bond that is observed as a pseudo-septet ( 8.26 ppm, JHH ≈ JPH ≈ 5 Hz,
1
Figure 1a). The 29Si-filtered H{31P} NMR and 29Si-1H{31P} HMBC NMR spectra of 1f
(collected at -70 °C, Figure 1c,d) indicated that both the Si—H and Ru—H resonances are
coupled to a downfield 29Si NMR resonance (29Si  123 ppm; 1JSiH = 216 Hz for Si—H; JSiH = 68
Hz for Ru—H/Ru—H—Si), and these NMR data are consistent with the presence of the 3-
H2SiH(Trip) ligand. The identification of 1f as [PhBPPh ]RuH[3-H2SiH(Trip)] was further
3
supported by the FTIR spectrum of 1f (Nujol mull on NaCl plates), which exhibits two strong,
sharp absorptions that correspond to the terminal Si—H ( = 2091 cm-1) and Ru—H bonds ( =
1895 cm-1). The FTIR spectrum also displays a weaker, broad absorption characteristic of the
Ru—H—Si interactions ( = 1628 cm-1).16 Thus, on the FTIR timescale, the terminal Ru—H,
terminal Si—H, and bridging Ru—H—Si hydrides of 1f are distinguished, while on the slower
10
ACS Paragon Plus Environment

Page 11 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. NMR spectra of 1f in toluene-d8. a) Si—H resonance in the 1H NMR spectra recorded
at -30 °C to + 20 °C. b) Ru—H resonance in the 1H NMR spectra recorded at -30 °C to + 20 °C.
c) 29Si-filtered 1H{31P} NMR spectrum recorded at -70 °C. d) 29Si-1H{31P} HMBC NMR
spectrum recorded at -70 °C.
1H NMR timescale the Ru—H and Ru—H—Si positions were observed as a single Ru—H
resonance, even in spectra collected at -70 C. Interestingly, the observation of a single Ru—H
1
resonance in the H NMR spectra implies that there must be an easily accessible exchange
process that interconverts the Ru—H and Ru—H—Si positions, but which does not allow
exchange of the terminal Si—H bond into these hydride positions (as suggested by the two
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 37
1
2
3
4
5
6
7
8
9
pathways of Scheme 3). Notably, this observation provides experimental support for fluxional
processes of the type determined by DFT calculations for the Fe—H and Fe—H—Si positions of
closely related [PhBPiPr3]FeH(3-H2SiRR') complexes.9c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3: Possible Hydride Exchange Pathways in [PhBPPh3]RuH[3-H2SiH(Trip)] (1f)
1
Mechanistic Investigations. The H NMR resonance for the terminal Si—H bond in 1f
( 8.26 ppm at -70 °C in toluene-d8) is shifted considerably downfield from those typically
observed for Si—H bonds ( 4.49 ppm for TripSiH3 in C6D6).17 This observation suggests the
possibility that the 3-H2SiH(R) complexes may be activated toward direct addition of its
terminal Si—H bond to an olefin (Scheme 4), as is observed for [LnM=Si(R)H]+ complexes
(Scheme 1) which also exhibit highly downfield Si—H resonances that are indicative of their
activated, electrophilic silicon centers.8b-d This possibility is further supported by similarities
between the olefin hydrosilations catalyzed by [PhBPPh ]Ru (vide supra) and
3
12
ACS Paragon Plus Environment

Page 13 of 37
ACS Catalysis
1
2
3
4
5
Scheme 4: Potential Mechanism of Hydrosilation Catalyzed by [PhBPPh3]RuH[3-
H2SiH(Cy)] (1d)
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[Cp*(iPr3P)Ru]+ complexes.8 In particular, both systems exclusively catalyze formation of the
anti-Markovnikov hydrosilation product and are more effective with alkyl- vs. aryl- silanes due
to competing catalytic redistribution for the latter silanes.8b-d Additionally, when PhMeSiH2 was
examined as a substrate for the hydrosilation of p-chlorostyrene using 1b as a catalyst (2.5 mol
1
%), the hydrosilation product was not formed after 24 h at 23 °C (monitored by H NMR
spectroscopy), and heating the reactions to 80 °C resulted in decomposition of 1b. The lack of
catalytic activity with PhMeSiH2 as the silane substrate suggested that the presence of a terminal
Si—H bond in 1d might be important for the olefin hydrosilation reactions catalyzed by this
species, just as the terminal Si—H bond in [Cp*(iPr3P)Ru(H)2=SiH(R)]+ is necessary for olefin
hydrosilations involving these silylene species.8
Some differences in reaction rates of 3-H2SiH(R) complexes 1d-e and hydridosilylene
complexes [Cp*(iPr3P)Ru(H)2=SiH(R)]+ were noted (Scheme 5). For the latter catalysts, the key
steps of the catalytic cycle involve: 1) insertion of the olefin into the terminal Si—H bond to
form [Cp*(iPr3P)Ru(H)2=SiRR']+ silylene complexes, 2) the elimination of two Si—H bonds
from ruthenium to release the secondary silane product, and 3) the activation of two Si—H bonds
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 37
1
2
3
4
Scheme 5: Relative Rates of Si—H Addition vs. Silane Exchange During Hydrosilation
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
of the primary silane substrate to regenerate [Cp*(iPr3P)Ru(H)2=SiH(R)]+. The first step (olefin
insertion into the Si—H bond) occurs very rapidly (< 1 min at 23 °C), while the exchange of the
product silane for an equivalent of the RSiH3 reactant requires heating to 80 °C for appreciable
rates (Scheme 5).8b Thus, the product/substrate exchange process is rate limiting for catalysis
involving [Cp*(iPr3P)Ru(H)2=SiH(R)]+. In contrast, silane-silane exchange reactions involving
the 3-H2SiRR' complexes 1a-f occur rapidly (< 5 minutes, Scheme 5), such that complex 1d
was observed as the catalyst resting state during hydrosilations using CySiH3 (vide supra). Thus,
with 1d as a catalyst, the rate limiting step occurs during the hydrosilation steps (i.e. Si—H
activation/C—Si and C—H bond formation) of the catalytic cycle, rather than for the
silane/product exchange steps.
The differences between the reactivity of 1d and [Cp*(iPr3P)(H)2Ru=SiH(R)]+ with
olefins might be due to the greater electrophilicity of the [Cp*(iPr3P)Ru]+ fragment and the
higher degree of Si—H activation in its complexes with secondary silanes (Scheme 5). These
features would render the silicon center more accessible to reaction with the olefin substrate,
while inhibiting elimination of the secondary silane product. The formally zwitterionic complex
1d would have a less electrophilic ruthenium center that would be weaker at binding and
14
ACS Paragon Plus Environment

Page 15 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
activating the silane Si—H bonds. The electrophilic character of the complexes may be more
important than the overall degree of Si—H activation since it is known that neutral
Cp*(iPr3P)(H)Ru=SiH(R) complexes, which have full silylene character, do not exhibit any
hydrosilation reactivity with olefins.8g
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Further investigation of the mechanism of olefin hydrosilation using [PhBPPh ]RuH(3-
3
H2SiRR') complexes revealed that steric properties of the silane substrate greatly influence the
efficacy of these hydrosilation reactions, and this might be a more important factor than the
presence of an uncoordinated Si—H bond for these catalysts. When complex 1f was treated with
p-chlorostyrene (10 equiv in benzene-d6), no reaction was observed after 2 h at 23 °C and after
20 h 1f had decomposed to form several Ru—H species, but the expected hydrosilation product
1
was not observed by H NMR spectroscopy. Furthermore, the relatively small secondary silane
Et2SiH2 was found to be an effective substrate for the hydrosilation of p-chlorostyrene using 1a
as a catalyst (eq 3). With 5 mol % of 1a, the anti-Markovnikov hydrosilation product was
1
formed in 60 % yield after heating to 60 °C in benzene-d6 for 20 h (determined by H NMR
spectroscopy). Thus, the presence of an uncoordinated Si—H bond in the [PhBPPh ]RuH(3-
3
H2SiRR') complexes is not essential for all catalytic olefin hydrosilation reactivity. Additionally,
complex 1b was found to catalyze H/D exchange between PhMeSiD2 and the vinylic C—H
bonds of p-chlorostyrene (eq 4, 6 turnovers after 20 h at 23 °C), which indicates that alkenes can
undergo reversible insertion with the terminal Ru—H bond of the [PhBPPh ]RuH moiety.
3
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 37
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A possible mechanism for olefin hydrosilation with Et2SiH2 involves insertion of the
olefin into a terminal Ru—H bond to form an alkyl complex, which then engages in Si—H
activation and Si—C bond formation (Scheme 6). A similar mechanism is implicated in the
catalytic hydrosilation of benzophenone with EtMe2SiH, as catalyzed by [PhBPPh ]Ru
3
complexes.10b However, for the hydrosilation of carbonyl compounds with EtMe2SiH, it was
observed that benzene could not be used as a solvent since it rapidly traps the reactive 14-
electron intermediate [PhBPPh ]Ru—H to form [PhBPPh ]Ru(5-C6H7). Given the efficiency of
3
3
benzene as a trap for the ruthenium hydride species, it seems that hydrosilations with Et2SiH2, as
catalyzed by 1a must not involve [PhBPPh ]Ru—H as an intermediate (Scheme 6, right). This
3
Scheme 6: Comparison of Potential Si—C Bond Forming Steps for Hydrosilations
Mediated by [PhBPPh3]RuH[3-H2SiEt2] (1c)
16
ACS Paragon Plus Environment

Page 17 of 37
ACS Catalysis
1
2
3
4
5
6
7
reactive intermediate might be avoided if the Si—C bond forming step of the catalytic cycle
involves a -bond metathesis process similar to that depicted for degenerate Si—H exchange in
8
path a of Scheme 3. In such a process, an 3-H2SiRR' ligand undergoes a concerted process
whereby Si—H bond cleavage and Si—C bond formation occur concurrently, while the second
coordinated Si—H bond remains unchanged. As shown in Scheme 6, this would lead to the
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
formation of an 2-H—SiR3 -complex [PhBPPh ]RuH(2-H—SiR3), which would protect the
3
new terminal Ru—H bond from reaction with benzene until a new equivalent of the silane
displaces the product (in re-formation of 1a, Scheme 6).10b Thus, 3-H2SiRR' ligands may be
important in the mechanism for olefin hydrosilation even for [PhBPPh ]RuH(3-H2SiRR')
3
complexes that do not feature a terminal Si—H bond. The mechanism depicted in Scheme 6 may
also be operative with CySiH3 as a substrate, but the decomposition of 1d over the course of the
reaction prevented more detailed investigations of the kinetics of the hydrosilation reactions.
Computational Investigations. Experimental evidence indicates that a number of
mechanistic pathways may be viable for hydrosilation reactions involving [PhBPPh ]RuH(3-
3
H2SiRR') complexes, and these mechanisms include: i) insertion of the alkene into an Ru—H
bond followed by Si—C bond forming -bond metathesis (Scheme 6), ii) insertion of the alkene
into a terminal Si—H bond (Scheme 4), and iii) insertion of the alkene into an Si—H bond that is
coordinated to ruthenium (see below, Figure 2, TSA'-D'). DFT calculations were employed to
compare the energetic accessibility of these three mechanisms for the hydrosilation of allyl
chloride, 1-hexene, styrene, p-chlorostyrene, and allyl trichlorosilane by an untruncated model
complex 1d-DFT. The results are presented for the first substrate while those for the other
olefins are available in the Supporting Information. Calculations were performed with Gaussian
0918 using the PBE0 hybrid functional19 with dispersion corrections,20 and the Def2-SVP basis
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 37
1
2
3
4
5
6
7
8
9
set (BS1)21 for all atoms, with quasi-relativistic effective core potentials for Ru.22 Energies were
refined by single point calculations using the Def2-TZVPP basis set (BS2)21 and by accounting
for solvent (benzene) effects using the SCM continuum model.23 Gibbs energies were calculated
from harmonic approximation of frequencies. See Supporting Information for additional details.
Geometry optimization of 1d-DFT located an 3-H2SiH(Cy) structure in which the two
bridging hydride ligands have average bond distances (Ru—H = 1.79 Å and Si—H = 1.65 Å) that
are longer than those calculated for the terminal Ru—H (1.625 Å) and Si—H (1.486 Å) bonds.
The ruthenium coordination geometry is octahedral with bond angles that deviate by < 5° from
90°. Silicon exhibits a tetrahedral geometry with a small compression of the H—Si—H bond
angle (102.3°) for the two coordinated Si—H bonds. As expected, this optimized structure for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1d-DFT is very similar to those of previously characterized for [PhBPPh ]Ru(H)(3-H2SiRR')
3
complexes.9a No ruthenium trihydride structure with a silylene group could be located as a
secondary minimum.
The calculations indicated that the lowest energy hydrosilation pathway involves
insertion of the alkene into the terminal Ru—H bond of the catalyst (Figure 2) in a mechanistic
step similar to a key step in the classic Chalk-Harrod cycle for olefin hydrosilation.5 The olefin
18
ACS Paragon Plus Environment

Page 19 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Hydrosilation pathways characterized by DFT calculations. Dotted lines between
atoms are used to illustrate weak interactions in intermediates and bonds being formed/broken in
transition states.
substrate accesses the ruthenium center via the displacement of one of the coordinated Si—H
bonds, forming intermediate A in a step that is mildly endergonic. The remaining coordinated
Si—H bond in A is elongated to 1.76 Å from an initial distance of 1.65 Å in 1d-DFT, indicating
that this Ru—H—Si interaction exhibits flexibility to accommodate the conversion between 3-
H2Si and 2-HSi binding modes of the silane. Insertion of the olefin into the Ru—H bond of A is
predicted to be the rate limiting step, with a Gibbs activation energy (GǂA-B = + 15.4 kcal mol-1)
that is readily accessible at room temperature. The resulting alkyl group, as initially formed
(intermediate B), features a -agostic C—H bond that is then displaced from ruthenium by an
Si—H bond to provide a new 3-H2SiH(Cy) complex (intermediate C). This displacement occurs
with essentially no energy cost. Interestingly, the 3-H2SiH(Cy) ligand of C facilitates a novel
type of -bond metathesis that constitutes a low energy (GǂC-D = + 7.9 kcal mol-1) Si—C bond
forming step for this pathway.
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 37
1
2
3
4
5
6
7
8
9
At the transition state for the Si—C bond-forming process (TSC-D, Figure 3), one of the
coordinated Si—H bonds is cleaved (1.634 Å at structure C, 1.881 Å at TSC-D, 2.401 Å at D)
with concomitant formation of the Si—C bond (2.95 Å at C, 2.422 Å at TSC-D, 1.905 Å at D),
while the other bridging Ru—H—Si interaction is only slightly perturbed. This calculated
process is consistent with the Si—C bond forming -bond metathesis process proposed in
Scheme 6 on the basis of experimental evidence. This unusual transition state is reminiscent of
the four-center transitions states that are well known for -bond metathesis processes involving
Si—H/Si—C interconversions.24 However, TSC-D is an interesting example of a transition state
in which an additional bridging Ru—H—Si interaction is maintained throughout the Si—H/Si—
C interchange.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Structure of TSC-D. Carbon atoms are depicted in dark gray, hydrogen in off-white,
ruthenium in turquoise, and silicon in yellow. The Ru, C, Si, and H atoms undergoing bonding
changes are depicted as spheres. For clarity, all atoms and bonds that do not participate in the
reaction are illustrated as stick (bound to Si) or wireframe (all other atoms) representations.
20
ACS Paragon Plus Environment

Page 21 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
In the structure TSC-D (Figure 3), the silicon center is positioned effectively to interact
with the transferring H and C atoms as well as the ruthenium center, but the geometric
considerations do not give information on the nature of the various bonds and interactions at this
transition state. Further insight into this novel -bond metathesis process was provided by an
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NBO analysis,25 which revealed that TSC-D possesses an [2-HSiH(Cy)]+ fragment with a silicon
center that engages in highly delocalized Si---H and Si---C interactions. Like all of the
intermediates and transition states of this reaction pathway, the ruthenium center of TSC-D
possesses three filled 4d orbitals. There is only slight delocalization of one of these lone pairs
onto the cationic [2-HSiH(Cy)]+ moiety (92% on Ru and 4.3 % on Si) of TSC-D, as shown by
the NLMO description. Notably, this is the largest delocalization of a ruthenium 4d lone pair that
was found for any of the structures on this hydrosilation pathway. Thus, this mechanism
facilitates hydrosilation processes using a ruthenium fragment that is reluctant to undergo
oxidative addition processes, which might be important for avoiding C—Cl bond activation
processes for the allyl chloride substrate. It is worth noting that the 3-H2Si coordination mode of
the silane may also contribute to inhibiting side reactions since the two Si—H bonds occupy
coordination sites at ruthenium, as in intermediate C, that might otherwise lead to C—Cl bond
activation.
Calculations carried out with other olefins (1-hexene, styrene, p-chlorostyrene, and allyl
trichlorosilane) give similar results with variation in the energies of some transition states by less
than 2 kcal mol-1 relative to the results obtained with the allyl chloride (See Supporting
Information). This suggests that the difference in the observed yields does not originate in
change of efficiencies in the productive pathway, but in decomposition pathways that depend on
the nature of the substrate. These decomposition pathways have not been analyzed by
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 37
1
2
3
4
5
6
calculations due to limited experimental information on them. Recent studies on these processes
highlight how challenging they can be without experimental guidance.26
7
8
9
Both of the other two mechanisms that were examined computationally involve the direct
insertion of the substrate C=C bond into an Si—H bond. One of these pathways involves
insertion of the olefin into an Si—H bond that is coordinated to ruthenium – a process analogous
to the mechanisms implicated for the hydrosilation of isocyanides and ketones by 1a-c.10
However, for the hydrosilation of allyl chloride, this type of insertion step was predicted to have
a relatively high Gibbs activation energy (GǂA'-D' = + 26.3 kcal mol-1) that would suggest that it
is not on the active pathway. Lastly, the silylene-like mechanism, involving insertion of the
substrate into the terminal Si—H bond, has a Gibbs activation energy (GǂA'-E = + 20.8 kcal mol-
1) that would be accessible at 23 °C, but which is still significantly higher than that determined
for insertion into the terminal Ru—H bond (GǂA-B = + 15.4 kcal mol-1). Thus, these calculations
favor the mechanism in which the olefin is activated at the ruthenium center, while suggesting
that the 3-H2SiH(Cy) ligand would still be reactive enough to promote direct insertion of the
olefin into the terminal Si—H bond if a terminal Ru—H bond was unavailable.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
An NBO analysis was carried out to understand more deeply why insertion of the olefin
into the Ru—H bond is preferable to insertion into the terminal Si-H bond. In 1d-DFT, the
ruthenium-bound terminal hydride carries an NBO charge of -0.09 and the bridged hydrogens
have a charge of -0.05. The highest negative charge on hydrogen (-0.15) is carried by that of the
terminal Si—H bond. The Ru itself carried a negative charge of -0.96 while the silicon is
positively charged (1.19). Since the Si—H bond of the isolated complex 1d-DFT has the
strongest positive charge on Si and the strongest negative charge on the hydrogen, it might be
expected that the Si—H bond would be more reactive towards the olefin than the Ru-H group,
22
ACS Paragon Plus Environment

Page 23 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
which is not the preference indicated by the relative energies of the transition states TSA-B and
TSA'-E. Therefore, the charges on the isolated system do not help to understand the preferences
indicated by the full energy profiles of possible pathways.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To provide more insight into the reactivity of 1d-DFT with olefins, the transition states
for insertion into the Ru—H and terminal Si—H bonds were analyzed with the second order
perturbation and NLMO methods, which quantitatively describe the delocalization of electron
density from localized lone pairs and bonds into vacant orbitals of the species involved in the
reaction. The values calculated by this analysis, which are very similar for all olefins are given in
the Supporting Information (see pages S7-10). The Ru-bonded H is found to be an extremely
powerful electron donor to the *CC orbital of the olefin, while the corresponding -bond is a
modest donor to the available empty orbitals at Ru. For the insertion into the Si—H bond, Si acts
as strong acceptor to the -bond of the olefin while the Si-bonded-H is a moderate donor. Thus,
the insertion into the Ru—H bond is dominated by the electron donating power of the hydride
while the insertion into the Si—H bond is more under the control of the electron accepting ability
of silicon. The Si—H group is, thus, well suited to react with electron rich olefins if the Ru—H
site is not available.
Conclusion
The 3-H2SiH(Ar) complex [PhBPPh ]RuH[3-H2SiH(Trip)] (1f), the first unambiguous
3
example of a transition metal complex possessing a primary silane bound in the 3-H2SiH(R)
coordination mode, has been prepared and characterized. Previously reported, related structures
exhibit a bonding pattern that is intermediate between [M](3-H2SiHR) and (H)2[M]=SiH(R)
descriptions.8b Therefore, the synthesis and characterization of 1f provides new insight into the
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 37
1
2
3
4
continuum between 3-silane and silylene dihydride structures, and allows for direct
5
6
7
8
interrogation of the properties of terminal Si–H bonds in the 3-silane structure type. Notably,
the 1H NMR spectrum of 1f features a downfield resonance for the terminal Si—H bond, which
supports the possibility that this Si—H moiety may be chemically similar to those in M=SiH(R)
hydridosilylene complexes with respect to electron deficiency at the silicon center.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Additional examples of [PhBPPh ]RuH[3-H2SiH(R)] (R = Cy, 1d; (CH2CH2Ph), 1e)
3
complexes were prepared in situ, and 1d was found to be an effective catalyst for the
hydrosilation of olefins, including the challenging substrate allyl chloride. The primary silane
CySiH3 is a significantly more effective substrate than secondary silanes, and this suggests that
the terminal Si—H bond of 3-H2SiH(R) complexes may be important for facilitating
hydrosilation by a pathway similar to that studied for hydridosilylene complexes.
The 3-H2Si coordination mode may also be important for facilitating a novel type of -
bond metathesis step that is responsible for the key Si—C bond forming step of the catalytic
cycle when using Et2SiH2 as a substrate. In particular, the pathway identified by DFT
calculations involves a concerted Si—H bond cleavage / Si—C bond formation event that
proceeds through a transition state in which a Ru—H—Si interaction serves as a pivot-point for
facilitating this exchange. This transition state is similar to the kite-shaped transition state usually
described for -bond metathesis,24 except that the 3-H2SiRR’ ligand maintains an additional
coordinated Si—H bond while the other Ru—H—Si interaction undergoes cleavage at the Si—H
bond. The extra Ru—H—Si interaction draws electron density away from silicon, which may
facilitate the facile -bond metathesis process by stabilizing the five-coordinate geometry around
silicon at the transition state. This type of pathway was shown by DFT calculations to be favored
for the hydrosilation of allyl chloride, 1-hexene, styrene, p-chlorostyrene and allyl trichlorosilane
24
ACS Paragon Plus Environment

Page 25 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
by 1d, but these computational investigations suggest that direct insertion of the olefin substrate
into the terminal Si—H bond of 1d is also energetically accessible. An NBO analysis reveals that
the preference for insertion in the Ru-H bond is dominated by the electron donating power of the
hydride while reaction with the terminal Si-H bond is controlled by the electrophilicity of the Si
center. Thus, the unusual 3-H2Si coordination mode of silanes may facilitate catalytic
hydrosilation reactions by a variety of pathways that are not possible with more common ligand
types such as silyl or 2-H—SiR3 ligands.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Details
General Considerations. All manipulations of air sensitive compounds were conducted under a
nitrogen atmosphere using standard Schlenk techniques or using a nitrogen atmosphere
glovebox. Proteo solvents were dried using a JC Meyer solvent drying system, and C6D6 was
vacuum transferred from NaK. The secondary silanes Et2SiH2 and PhMeSiH2 were purchased
from commercial sources and used as supplied. The primary silanes CySiH3, PhCH2CH2SiH3,
and (Trip)SiH3 were prepared by reduction of the corresponding trichlorosilanes with LiAlH4.27
Complexes 1a-c were prepared as previously reported.9a
NMR spectra were recorded on Bruker spectrometers at room temperature unless
otherwise noted. Spectra were referenced internally by the residual proton signal relative to
1
tetramethylsilane for H NMR, solvent peaks for 13C{1H} NMR, external 85 % H3PO4 for
29
31P{1H} NMR, and tetramethylsilane for Si-1H HMBC experiments. Assignments of certain
1
13C{1H} NMR signals were made on the basis of H-13C HSQC NMR data. The JSiH values for
Ru—H—Si resonances were determined by examining satellite signals near the main Ru—H
1
resonance in H{31P} NMR spectra or by the Ru—H resonances displayed in 29Si-filtered
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 37
1
2
3
4
5
1H{31P} NMR experiments. Infrared spectra (Nujol mulls, KBr plates) were recorded using a
Nicolet 6700 FTIR spectrometer at a resolution of 2 cm-1. Hydrosilation products were identified
6
7
8
9
13
29
by comparison of multinuclear NMR data (1H, C{1H}, and Si-1H HMBC NMR) to those
previously reported for identical or closely related organosilanes, and by GC-MS. Elemental
analyses were performed by the University of California, Berkeley College of Chemistry
Microanalytical Facility.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[PhBPPh3]RuH(3-H2SiH(Trip)) (1f). Yellow crystals of 1b (59 mg, 0.064 mmol) were
added to a solution of (Trip)SiH3 (70 mg, 0.33 mmol) in 2 mL of toluene, and the resulting
mixture was stirred for 10 min to provide a clear yellow solution. The volatile materials were
removed under vacuum and the resulting yellow oil was dissolved in fluorobenzene (2 mL) to
give a solution that was stirred for 10 minutes before removal of the volatile components. This
latter step was repeated once more, but evaporation was stopped with 0.5 mL of solvent
remaining. This solution was layered with pentane and cooled to – 35 °C. After 1 day, the yellow
crystals that formed were isolated by pipetting away solvent, washing with pentane (3 x 2 mL),
and briefly drying under vacuum to provide 54 mg of slightly impure 1f (81 % yield). Anal
1
Calcd for C58H52OBP3Ru (969.853): C, 70.51; H, 6.51. Found: C, 69.84; H, 6.22. H NMR (C-
6D6, 400 MHz):  8.37 (vbr, 1 H, Si—H), 8.20 (d, J = 7.3 Hz, 2 H), 7.70 (t, J = 7.3 Hz, 2 H), 7.57
(br m, 12 H), 7.44 (tt, J = 7.3 Hz, 1.1 Hz, 1 H), 7.15 (2 H, trip), 6.86 – 6.74 (m, 18 H), 3.81
(septet, J = 6.7 Hz, 2 H, iPr methine), 2.76 (septet, J = 6.8 Hz, 1 H, Pr methine), 1.89 (br, 6 H,
i
B-CH2-P), 1.30 (d, J = 6.7 Hz, 12 H, iPr methyl), 1.18 (d, J = 6.7 Hz, 6 H, iPr methyl), -6.50 (br,
3 H, Ru—H). 13C{1H} NMR (C6D6, 150.893 MHz):  153.33, 153.02, 150.89, 142.67 (m),
135.97, 132.78, 132.59 (m), 124.72, 122.15, 121.82, 35.58, 35.31, 24.87, 24.36. 31P {1H} NMR
26
ACS Paragon Plus Environment

Page 27 of 37
ACS Catalysis
1
2
3
4
(C6D6, 161.976 MHz):  46.4. 29Si NMR (C6D6, - 70 °C, 1H-29Si HMBC: 600 MHz (1H), 119.23
5
6
7
8
MHz (29Si)):  121. IR (cm-1): 2090 (Si—H), 1895 (Ru—H), 1628 (br, Ru—H—Si).
Representative procedure for catalytic hydrosilation reactions. p-Chlorostyrene (15 mg,
0.11 mmol) and CySiH3 (14 mg, 0.12 mmol) were dissolved in C6D6 (0.5 mL) containing C6Me6
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
or ferrocene as an internal standard. A H NMR spectrum of this solution was collected prior to
the addition of 1b (2.5 mg, 0.0027 mmol) in C6D6 (0.2 mL) to provide a pale yellow solution.
The reaction solution was examined by 1H and 31P{1H} NMR spectroscopy within 15 minutes of
adding 1b. After 20 h the solution was again examined by H and P{1H} NMR spectroscopy
1
31
and the yield of cyclohexyl(p-chlorophenethyl)SiH2 was determined by integrating the initial and
1
final H NMR spectra using the internal standard resonance as a reference integral. The product
silane was isolated as a colorless oil by diluting the solution with hexanes, filtering through a
plug of silica, and evaporating solvent under reduced pressure. The isolated product was
characterized by H, C{1H}, Si-1H HMBC, and Si-filtered H NMR spectroscopy and by
1
13
29
29
1
GC-MS. Characterization data for the hydrosilation products are provided below.
1
Cy(p-chlorophenethyl)SiH2. H NMR (C6D6, 600 MHz):  7.12 (d, J = 8.1 Hz, 2 H), 6.73
((d, J = 8.1 Hz, 2 H), 3.73 (q, J = 3.4 Hz, JSiH = 185 Hz, 2 H, Si—H), 2.43 (m, 2 H, Ar-CH2-),
1.64 (m, 5 H), 1.23 – 1.08 (m, 5 H), 0.77 (m, 2 H, Si-CH2-), 0.72 (m, 1 H, methane C—H).
13C{1H} NMR (C6D6, 150.903 MHz):  143.20, 132.21, 129.89, 129.07, 31.64, 29.84, 28.34,
27.31, 21.71, 10.26. 29Si NMR (C6D6, 1H-29Si HMBC: 600 MHz (1H), 119.23 MHz (29Si)):  -23.
GC-MS m/z 252 (M)+, 169, 141, 125, 103.
1-(Cl3Si)-3-(CyH2Si)-propane. Note that this silane was isolated by air-free microscale
distillation rather than filtration through silica. 1H NMR (C6D6, 600 MHz):  3.67 (q, J = 3.5 Hz,
JSiH = 184 Hz, 2 H, Si—H), 1.63 (m, 5 H), 1.50 (m, 2 H, -CH2-), 1.23 – 1.06 (m, 5 H), 1.01 (m, 2
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 37
1
2
3
4
13
H, Cl3Si-CH2-), 0.72 (m, 1 H, Si-CR2H), 0.47 (m, 2 H, methane C—H). C{1H} NMR (C6D6,
5
6
1
150.903 MHz):  29.81, 28.28, 27.85, 27.29, 21.59, 19.49, 11.01. 29Si NMR (C6D6, H-29Si
7
8
9
HMBC: 600 MHz (1H), 119.23 MHz (29Si)):  -24 (SiCyH2 group), 12 (SiCl3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Et2(p-chlorophenethyl)SiH. H NMR (C6D6, 600 MHz):  7.13 (d, J = 7.9 Hz, 2 H),
6.76 (d, J = 7.9 Hz, 2 H), 3.87 (sep, J = 3.2 Hz, JSiH = 179 Hz, 1 H Si—H), 2.41 (m, 2 H, Ar-
CH2-), 0.94 (t, J = 7.9 Hz, 6 H, Et2), 0.74 (m, 2 H Si-CH2-), 0.50 (qd, J = 7.9 Hz, 3.2 Hz, 4 H,
Et2). 13C{1H} NMR (C6D6, 150.903 MHz):  143.74, 132.09, 129.86, 129.06, 30.73, 13.17, 8.72,
3.36. 29Si NMR (C6D6, 1H-29Si HMBC: 600 MHz (1H), 119.23 MHz (29Si)):  − GC-MS m/z -
226 (M)+, 197, 169, 141, 125, 114, 103.
1-chloro-3-(CySiH2)-propane. Note that a small amount of an unidentified side product
was not separated from the sample that was isolated from the catalytic reaction mixture. 1H NMR
(C6D6, 600 MHz):  3.67 (q, J = 3.4 Hz, JSiH = 185 Hz, 2 H, Si—H), 3.08 (t, J = 6.8 Hz, 2 H, -
CH2Cl), 1.63 (m, 5 H), 1.54 (m, 2 H, -CH2-), 1.21 – 1.05 (m, 5 H), 0.68 (m, 1 H, cyclohexyl
methane), 0.49 (m, 2 H, Si-CH2-). 13C{1H} NMR (C6D6, 150.903 MHz):  47.55 (CH2Cl), 29.74,
29.62, 28.30, 21.63, 5.90. 29Si NMR (C6D6, 1H-29Si HMBC: 600 MHz (1H), 119.23 MHz (29Si)):
 − GC-MS m/z 107 (M - cyclohexyl)+, 79, 65, 55, 41, 28, 18. GC-HRMS (EI) calcd for [(1-
chloro-3-(CyH2)-propane) – H] [C9H17SiCl]+: 188.0788, and 190.0759, found: 188.0789, and
190.0762.
Author Information
Corresponding Authors: tdtilley@berkeley.edu, odile.eisenstein@umontpellier.fr
The authors declare no competing financial interest
28
ACS Paragon Plus Environment

Page 29 of 37
ACS Catalysis
1
2
3
4
5
6
Supporting Information
7
8
9
NMR spectra and full computational details. The coordinates of all calculated species are given
as .xyz files. This material is available free of charge at http://pubs.acs.org.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgments
Experimental work was funded by the National Science Foundation under Grant No. CHE-
1566538. MNP acknowledges the computer resources and technical support provided by the
CTMM group at Université de Montpellier and CCIR of ICBMS and P2CHP of Université Lyon
1. We thank Dr. Allegra Liberman-Martin for discussions and experimental assistance. Dr.
Lionel Perrin (University de Lyon 1) is gratefully thanked for discussions on the computational
section.
References
1. (a) Troegel, D.; Stohrer, J. Recent Advances and Actual Challenges in Late Transition Metal
Catalyzed Hydrosilylation of Olefins from an Industrial Point of View. Coordination
Chemistry Reviews, 2011, 255, 1440–1459. (b) Marciniec, B. Catalysis of Hydrosilation of
Carbon-Carbon Multiple Bonds: Recent Progress. Silicon Chem., 2002, 1, 155 - 175. (c)
Brook, M. A. Silicon in Organic, Organometallic, and Polymer Chemistry; Wiley:ꢁ New
York, 2000. (d) Ojima, I.; Li, Z.; Zhu, J. The Chemistry of Organic Silicon Compounds,
Wiley:ꢁ Avon, 1998; Chapter 29. (e) Roy, A. K. A Review of Recent Progress in Catalyzed
Homogeneous Hydrosilation (Hydrosilylation). Adv. Organomet. Chem. 2007, 55, 1 – 59.
29
ACS Paragon Plus Environment