Formal SiH4 chemistry using stable and easy-to-handle surrogates

Article abstract of DOI:10.1038/nchem.2329

Monosilane (SiH₄) is far less well behaved than its carbon analogue methane (CH₄). It is a colourless gas that is industrially relevant as a source of elemental silicon, but its pyrophoric and explosive nature makes its handling and use challenging. Consequently, synthetic applications of SiH₄ in academic laboratories are extremely rare and methodologies based on SiH₄ are underdeveloped. Safe and controlled alternatives to the substituent redistribution approaches of hydrosilanes are desirable and cyclohexa-2,5-dien-1-ylsilanes where the cyclohexa-1,4-diene units serve as placeholders for the hydrogen atoms have been identified as potent surrogates of SiH₄. We disclose here that the commercially available Lewis acid tris(pentafluorophenyl)borane, B(C₆F₅)₃, is able to promote the release of the Si-H bond catalytically while subsequently enabling the hydrosilylation of C-C multiple bonds in the same pot. The net reactions are transition-metal-free transfer hydrosilylations with SiH₄ as a building block for the preparation of various hydrosilanes.

Full text of DOI:10.1038/nchem.2329

ARTICLES
PUBLISHED ONLINE: 24 AUGUST 2015 | DOI: 10.1038/NCHEM.2329
Formal SiH4 chemistry using stable and
easy-to-handle surrogates
Antoine Simonneau and Martin Oestreich
*
Monosilane (SiH4) is far less well behaved than its carbon analogue methane (CH4). It is a colourless gas that is
industrially relevant as a source of elemental silicon, but its pyrophoric and explosive nature makes its handling and use
challenging. Consequently, synthetic applications of SiH4 in academic laboratories are extremely rare and methodologies
based on SiH4 are underdeveloped. Safe and controlled alternatives to the substituent redistribution approaches of
hydrosilanes are desirable and cyclohexa-2,5-dien-1-ylsilanes where the cyclohexa-1,4-diene units serve as placeholders for
the hydrogen atoms have been identified as potent surrogates of SiH4. We disclose here that the commercially available
Lewis acid tris(pentafluorophenyl)borane, B(C6F5)3, is able to promote the release of the Si–H bond catalytically while
subsequently enabling the hydrosilylation of C–C multiple bonds in the same pot. The net reactions are transition-metal-
free transfer hydrosilylations with SiH4 as a building block for the preparation of various hydrosilanes.
ur laboratory recently established ionic transfer hydrosilylation bearing hydride and/or cyclohexa-2,5-dien-1-yl substituents as
as a new strategy in silicon chemistry1–4. Cyclohexa-2,5-dien- potent SiH4 surrogates. Their manifold use in transition-metal-
O
1-yl-substituted silanes I were found to serve as hydrosilane free catalytic as well as transition-metal-catalysed hydrosilylation
precursors when treated with fluoroarylboron Lewis acids, for is demonstrated (Fig. 1c).
example, B(C6F5)3 (1a, Fig. 1a)5,6. This reaction involves two inter-
dependent catalytic cycles. First, borane-mediated hydride abstrac- Results and discussion
tion in the bisallylic position7,8 of I releases the silicon-stabilized Synthesis of SiH4 surrogates. We began by targeting the potential
Wheland complex or benzene-stabilized silicon cation III SiH4 surrogates tetra(cyclohexa-2,5-dien-1-yl)silane (2a) and
(I→II→III) that eventually collapses to liberate hydrosilane IV tri(cyclohexa-2,5-dien-1-yl)silane (2b) (Fig. 2a). We envisioned
and one molecule of benzene (III→IV). The borane Lewis acid their synthesis by reacting SiCl4 or HSiCl3 with a cyclohexa-2,5-
then activates the thus-formed hydrosilane IV9 to perform C=X dien-1-yllithium reagent34. Standard reaction conditions in THF
bond hydrosilylation10–12 through ion pair VI (IV→V→VI)13. We (∼1 M) allowed us to obtain 2a in decent yield, but the same set-
have already shown that reagents I are generally applicable to the up was not suitable for the preparation of 2b in reasonable yield
hydrosilylation of σ- and π-basic substrates1,3 and Si–O dehydro- and purity. We were nonetheless able to establish an improved
genative couplings2. One obvious advantage of this method is the procedure for its synthesis by switching to a nonpolar solvent
possibility to generate in situ the highly flammable Me3SiH gas (n-hexane) and lower concentration (0.4 M). Single crystals of
from an easy-to-handle liquid.
2b suitable for X-ray diffraction analysis were obtained (see
To assess the limits of this strategy, we became interested in its Supplementary Fig. 17 for its molecular structure). We then
implementation for the release of monosilane (SiH4) gas. This smal- turned our attention to the preparation of di(cyclohexa-2,5-dien-
lest member of the hydrosilane family raises particular safety issues 1-yl)silane (2c) (Fig. 2a). We initially applied the procedures
because of its highly pyrophoric, explosive and toxic nature14. The elaborated for the syntheses of 2a and 2b using dihalosilanes as
semiconductor industry has faced several incidents, sometimes electrophiles. However, these approaches led to inseparable
fatal, linked to this chemical15–17, which is intensively used for the mixtures of several silicon-containing compounds, including 2c.
growth of silicon thin films and nanowires via various deposition This forced us to explore a two-step sequence starting with
methods18–23. Indeed, other than in the semiconductor research the reaction of cyclohexa-2,5-dien-1-yllithium with (EtO)2SiCl2
field, the danger of handling SiH4 often results in its banishment to afford 3 quantitatively; reductive Si–O cleavage in 3 with
from laboratories. The reactivity of SiH4 in alkene hydrosilyla- diisobutylaluminium hydride (DIBAL-H) furnished 2c in good
tion24–26, another industrially relevant process, has been poorly yield. It is worth noting that it is possible to handle compounds
investigated. In the 1990s, Itoh and co-workers from the Mitsui 2a–c under air, but rearomatization of one of the cyclohexa-
Toatsu Chemicals Company developed methods for the tran- 1,4-diene moieties occurs if exposed to ambient atmosphere for
sition-metal-catalysed hydrosilylation of alkenes and conjugated prolonged periods of time, yielding the corresponding phenyl-
27–31
dienes with SiH4
and, to the best of our knowledge, these few substituted surrogates.
To verify whether silanes 2a–c could actually be ‘decomposed’ to
articles and patents are the only experimental work related to hydro-
silylation reactions with monosilane (Fig. 1b). Spurred by the intui- SiH4, their behaviour in the presence of catalytic amounts of
tion that safe and practical alternatives to the direct use of SiH4 or its B(C6F5)3 (1a) was monitored by 1H NMR spectroscopy. Fully sub-
generation by disproportionation of hydrosilanes32,33 would be stituted 2a released a hardly discernible amount of SiH4 and decom-
highly desirable and relevant to several areas of silicon chemistry, posed to an unidentified material instead (Supplementary Fig. 2).
we now report our efforts towards achieving this goal. In this However, we found that 2b reacted smoothly with catalyst 1a to suc-
Article, we describe the preparation of novel silanes exclusively cessively give 2c, cyclohexa-2,5-dien-1-ylsilane (2d) and eventually
Institut für Chemie, Technische Universität Berlin, Strasse des 17. Juni 115, 10623 Berlin, Germany. e-mail: martin.oestreich@tu-berlin.de
*
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
1
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2329
ARTICLES
a
One-pot transfer hydrosilylation
SiR3
Previous work: Itoh and co-workers (refs 27-31)
b
c
0.1 mol%
(Ph3P)4Pt
SiH3
SiH2
37%
5%
R
X
H
cat. B(C6F5)3
cat. B(C6F5)3
–C6H6
+
SiH4
H
SiR3
10 bar
80 °C, 3 h
SiR3
(0.33
R
X
2
equiv.)
H
I
X = CH2, O, NR'
Present work
• Safe
• Gaseous
H
Hydrosilane-release
cycle
Hydrosilylation
cycle
SiH4–n
• Easy-to-handle
• Straight forward
syntheses
• Highly flammable
• Explosive
SiR3
SiH4
SiR3
X
n
R
H
• Toxic
2 (n = 2 and 3)
I
R
X
SiR3
H
H
• Applicable in borane-catalysed transfer hydrosilylation
or platinum(0)-catalysed hydrosilylation reactions
and their combination
(C6F5)3B
II
(C6F5)3B
B(C6F5)3
VI
cat. B(C6F5)3
SiH4–n
1a
SiR3
SiR3
R
n
SiR3
cat.
H
+ 2
R
H3–n
Si
B(C6F5)3
(C6F5)3B
X
R
H
H
H
SiR3
IV
cat. [Pt]
R
V
(C6F5)3B
(C6F5)3B
+
C6H6
III
n
Figure 1 | Transfer hydrosilylation: mechanism and implementation for the release of SiH4. a, The concept of transfer hydrosilylation and its mechanism
with R = alkyl, aryl, alkoxy and H. b, Direct hydrosilylation with SiH4. c, Cyclohexa-2,5-dien-1-ylsilanes as safe SiH4 surrogates in metal-free and transition-
metal-catalysed hydrosilylations.
a
b
H
Li•TMEDA
cat.
cat.
cat.
1a
H
(iii) (EtO)2SiCl2
1a
1a
Si
H
Si
H
Si
Si
Si
H
H
OEt
OEt
H
H
H
H
3 (quant.)
(iv) DIBAL–H
2b
2c
2d
(i) SiCl4
(ii) HSiCl3
c
0.5
0.4
0.3
0.2
0.1
0.0
2b
2c
Si
Si
H
Si
H
2d
H
SiH4
2a (52%)
2b (83%)
2c (53%)
after recrystallization
after recrystallization
(X-ray)
cat.
cat.
cat.
B(C6F5)3 (1a)
1a
1a
Decomposition
(Supplementary Fig. 2)
SiH4
(Fig. 2b)
SiH4
(Supplementary
Figs 5 and 6)
0
1
2
3
4
5
6
7
8
Time (h)
Figure 2 | Syntheses and evaluation of cyclohexa-2,5-dien-1-ylsilanes 2a–c as potential SiH4 surrogates. a, Reactions of lithiated cyclohexa-1,4-diene with
various silicon electrophiles allow the preparation of cyclohexa-2,5-dien-1-ylsilanes 2a–c. (i) SiCl4 (0.24 equiv.), THF, −78 °C to room temperature, 3 days.
(ii) HSiCl3 (0.33 equiv.), n-hexane, −78 °C to room temperature, overnight. (iii) (EtO)2SiCl2 (0.5 equiv.), THF, −78 °C to room temperature, overnight.
(iv) DIBAL-H (3.8 equiv.), n-hexane, −78 °C to room temperature, 1 h. b, B(C6F5)3-catalysed stepwise release of SiH4. c, Plot of concentrations of hydrosilanes
2b–d and SiH4 against time in the reaction of 2b with catalytic amounts of 1a. TMEDA, N,N,N′,N′-tetramethylethylenediamine.
SiH4 (Fig. 2b,c as well as Supplementary Figs 3 and 4). A plot of the was observed (Supplementary Figs 5 and 6). The lower rate of con-
concentration of the various species formed during this experiment sumption of 2d compared to 2c or 2b is likely to originate from a
against time is presented in Fig. 2c. Silane 2b was almost immedi- more stable η1-Si–H/B(C6F5)3 complex of 2d as a result of negligible
ately converted into 2c and was no longer detectable after a few steric repulsion. Conversely, steric bulk around the Si‒H bond in 2b
minutes. 2c reached its maximum concentration at the early stage hampers that adduct formation (see its molecular structure,
of the reaction and was consumed within 2 h. After 1 h, 2d was Supplementary Fig. 17), thereby steering the boron Lewis acid
the major species in the reaction mixture and its concentration towards a bisallylic C‒H bond and hence accelerating the transform-
slowly diminished over hours. SiH4 was formed over the course of ation of 2b into 2c. To clarify whether SiH4 inhibits the decompo-
hours and its concentration reached a plateau after 4 h at a concen- sition of compounds 2b–d, we monitored the degradation of 2b in
tration as low as 0.1 mmol ml‒1, indicating the solubility limit of this an open system over a period of 7 h. The SiH4 concentration remained
gas in CD2Cl2 (monosilane is insoluble in most organic solvents). constant at ∼0.05 mmol ml‒1 (vs. ∼0.1 mmol ml‒1 in a closed system),
When the same experiment was repeated with 2c, a similar trend and no significant acceleration was observed (Supplementary Fig. 3).
2
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2329
ARTICLES
observed with tris(pentafluorophenyl)borane (1a). Estimation of
their Lewis acidities relative to archetypical B(C6F5)3 (1a, 100%)
are given according to the established Gutmann–Beckett
method39,40 (percentage values3 in parentheses below catalysts 1
in Table 1; for the determination of the value for 1d, see
Supplementary Table 1). As expected, partially fluorinated 1b
showed reduced reactivity (entry 7). No reaction occurred when
highly sterically encumbered, perfluorinated borane 1d was tested
in combination with 2b (entry 8). Conversely, the reaction
mixture turned into a gel when using less hindered surrogate 2c
(entry 9). The steric situation between 2c and 1d allows for
hydride abstraction (see Fig. 1a, I→II→III), but the resulting boro-
hydride is too congested to rapidly react with the benzene-stabilized
silicon cation intermediate (III→IV). As a result, the silicon cation/
Wheland intermediate might potentially serve as an initiator of cat-
ionic polymerization. Borane 1c, exhibiting intermediate steric
properties, was indeed able to bias the product distribution to a
limited extent. However, 1c was far less reactive than 1a. With
0.35 equiv. of 2b or 2c, di-adduct 6a is now formed together with
tri-adduct 7a (6:94 with 2b or 18:82 with 2c, entries 10 and 11 vs.
0:100 with catalyst 1a, entries 1 and 2). Better conversion and
improved selectivity for 6a were noticed when 2c was used.
Increasing its amount to 0.55 equiv. allowed us to obtain a 52:48
mixture of 6a and 7a, again improving the chemoselectivity
obtained with 1a (entry 12). Changing the SiH4 surrogate to 2b
resulted in a diminished conversion of 4a over the same reaction
time (entry 13).
The results in Table 1 show no significant chemoselectivity
differences between 2b and 2c when full conversion of styrene
(4a) was reached, implying degradation of 2b and 2c to a
common reagent prior to the hydrosilylation event. Time-depen-
dent 1H NMR measurements of the degradation of 2b and 2c
(Fig. 2c and Supplementary Figs 3–6) clearly support this assump-
tion as the half-lives of 2b and 2c were in the range of minutes and
less than 1 h, respectively. Hence, 2d or SiH4 are presumably the
actual hydrosilylating agents. To verify this, we monitored the reac-
tion of 4a and 0.35 equiv. of 2b by means of time-dependent, multi-
nuclear 1H and 29Si NMR experiments (see Table 1, entry 1, Fig. 3
and Supplementary Figs 7–11). In line with what we have previously
observed, SiH4 surrogate 2b is completely degraded by B(C6F5)3
(1a) within a few minutes into its di- and trihydro-substituted con-
geners 2c and 2d, respectively. 2c is also rapidly consumed (∼1 h),
while 2d remains detectable for 4 h. Notably, SiH4 was also detected,
and its concentration reaches a plateau after 1 h (Fig. 3b). Half of the
styrene (4a) is consumed after 1 h, and the intermediate mono- and
di-adducts 5a and 6a are formed during this time. These, as well as
unreacted 4a, are then slowly converted to tri-adduct 7a (Fig. 3c).
We did not detect any silanes with ‘mixed’ substitution patterns,
that is, homobenzyl and cyclohexa-2,5-dien-1-yl groups. It is there-
fore reasonable to assume that homobenzyl-substituted hydro-
silanes 5a and 6a are the actual hydrosilylating agents leading to
6a and 7a, respectively. However, we cannot fully exclude the par-
ticipation of cyclohexa-2,5-dien-1-yl-substituted 2d in the initial
hydrosilylation event, leading to a short-lived intermediate that
eventually releases 5a. Comparison of the observed initial rates of
formation of 5a, 6a and 7a reveals that hydrosilylation of 4a by 5a
is faster than with SiH4 or 6a. While in the latter case this kinetic
difference might merely be the result of different steric situations
(monoalkylation in 5a versus dialkylation in 6a), we believe that
conversion of 4a into 5a is impeded by the poor availability of
monosilane at the early stage of the reaction, resulting in substantial
accumulation of 6a. In another time-dependent NMR experiment
run with 2b in excess (1.35 equiv., Supplementary Figs 12–16),
this rate difference became even more apparent. The solubility
limit of SiH4 was also well demonstrated by a significant decline
in its concentration during the third hour.
Table 1 | Transfer hydrosilylation of styrene (4a).
SiH4–n
5 mol% 1
SiH4
surrogate
+
CH2Cl2
RT
n
4a
2b or 2c
5a (n = 1)
7a (n = 3)
6a (n = 2)
8a (n = 4)
(equiv.)
F
F
F
F
F
F
F
F
F
F
F
F
F
B
X
(F5C6)2B
F
F
F
F
F
F
B
F
3
F
F
3
1a with X = F (100%)
1b with X = H (97%)
1c (96%)
1d (113%)
Gutmann–Beckett Lewis acidities as percentage relative to 1a (100%) in parentheses
Entry SiH4
surrogate
Catalyst Time
(h)
Conversion
of 4a (%)*
Product
distribution
(equiv.)
5a:6a:7a:8a†
1
2
3
4
5
6
2b (0.35)
2c (0.35)
2b (0.55)
2c (0.55)
2b (1.10)
2c (1.10)
2b (0.35)
2b (0.35)
2c (0.35)
2b (0.35)
2c (0.35)
2c (0.55)
2b (0.55)
1a
1a
1a
1a
1a
1a
1b
1d
1d
1c
1c
1c
1c
20
20
20
20
20
20
40
20
20
120
120
80
80
100
100
100
100
100
100
72
0
100
72
0:0:100:0
0:0:100:0
0:23:77:0
18:25:57:0
39:36:25:0
49:31:20:0
0:0:100:0
0:0:0:0
7‡
8§
9||
10
11
12
13
0:0:0:0
0:6:94:0
0:18:82:0
0:52:48:0
0:35:65:0
90
94
82
Conditions: 4a (0.2 mmol),
2
(0.07–0.22 mmol), 1 (0.01 mmol), CH2Cl2 (1.0 M), room
temperature, 20–120 h. Unless otherwise noted, conversion of 2 was always 100%; *measured by
gas chromatography with mesitylene as an internal standard; †determined by integration of
chromatograms, uncorrected; ‡no further conversion of 4a was observed after this period of time;
§no conversion of 2b; ||a gel was obtained.
Optimization of the transfer hydrosilylation. Validated silanes
2b,c were next evaluated as SiH4 precursors in the transfer
hydrosilylation of styrene (4a) with catalyst 1a (Table 1) and we
found that both participated in this process. These reactions went to
full conversion of the starting material 4a with perfect selectivity in
favour of the formation of monohydrosilane 7a (entries 1 and 2).
Notably, and despite the poor solubility of SiH4 in CH2Cl2 (Fig. 2b),
only a slight excess of the required stoichiometry of 2 (0.35 instead
of 0.33 equiv.) and no dilute reaction medium were required to
secure complete consumption of the substrate. Chemoselectivity is
an issue frequently encountered with substitution processes at the
silicon atom decorated with identical leaving groups. The kinetics
are similar for single, double or even triple displacements, and it is
usually steric effects and not the stoichiometry of the reactants that
make the reaction stop at a certain stage35. Accordingly, an increase
in the amount of 2b (0.55 equiv.) also led to dihydrosilane 6a, but
the formation of monohydrosilane 7a still prevails (entry 3).
Essentially the same result was obtained with 0.55 equiv. of 2c,
yielding trihydrosilane 5a along with 6a and 7a (entry 4). Even the
use of over-stoichiometric amounts of the SiH4 surrogate (1.1 equiv.)
did not override the lack of chemoselectivity (entries 5 and 6).
Apart from benchmark catalyst 1a, we also investigated the
performance of other electron-deficient arylboranes in this reaction.
We chose partially fluorinated tris(2,3,5,6-tetrafluorophenyl)
borane (1b)36 to probe the effect of decreased Lewis acidity and
bis(pentafluorophenyl)(nonafluorobiphenyl)borane (1c)37 and tris-
(nonafluorobiphenyl)borane (1d)38 were selected because their aug-
mented steric hindrance could perhaps bias the chemoselectivities
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
3
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2329
ARTICLES
a
H
cat.
cat.
cat.
1a
H
1a
1a
Si
H
Si
H
Si
H
Si
H
H
H
H
H
cat.
1a
cat.
1a
cat.
1a
SiH2
SiH
SiH3
2b
2c
2d
2
3
4a
5a
6a
7a
c
b
0.4
0.3
0.2
0.1
0.0
0.4
4a
5a
6a
7a
2b
2c
2d
0.3
0.2
0.1
SiH4
0.0
0.0
0
2
4
6
8
10
12
0.5
1.0
1.5
Time (h)
Time (h)
Figure 3 | Time-dependent NMR study of the transfer hydrosilylation of 4a with 2b catalysed by 1a. a, Stepwise B(C6F5)3-catalysed release of SiH4
followed by multiple alkene hydrosilylation. b, Plot of concentrations of hydrosilanes 2b–d and SiH4 in the presence of styrene (4a) against time at the early
stage of the reaction. c, Plot of concentrations of starting material 4a and hydrosilylation products 5a, 6a and 7a against time.
Scope and limitations. Encouraged by the results compiled in 4i reacted smoothly to the tri-adducts 7h and 7i, respectively (entries
Table 1, we subjected differently substituted alkenes to these 9 and 10). Internal alkenes such as indene (4j) and less reactive
protocols. As solid 2b is readily accessible in large quantities, we 1,2-dihydronaphthalene (4k) afforded di- rather than tri-adducts,
used it for the substrate scope study (Table 2). Electronically and 6j and 6k were obtained in good to moderate yields (entries
modified styrenes 4a‒4c showed the expected reactivity trends 11 and 12). 1,2-Dialkyl-substituted alkene 4l exhibited poor
(entries 1‒3). An electron-donating methyl group led to a higher reactivity and converted into tri-adduct 7l in low yield after a long
yield and an electron-withdrawing fluorine group to a lower yield reaction time (entry 13). Unlike previous studies1,12, this type of
than the parent styrene. This illustrates well the need for sufficient substrate as well as terminal olefins (for example, 4g) proved to be
π basicity of C=C double bond in the nucleophilic attack at the poorly reactive. This difference might stem from the attenuated
silicon atom in V and the stabilization of the positive charge ability of the SiH3 relative to SiR3 groups to effect stabilization of
developed in VI (see Fig. 1a with X=CR2). To demonstrate the β carbocation developed after transfer of the silicon atom onto
practicality, we performed the preparation of 7a from 2b in good the C=C bond (V→VI, Fig. 1a with X=CR2)41. Finally, we
yield outside the glovebox, using Schlenk techniques with wet and examined
a small panel of trisubstituted double bonds.
non-degassed CH2Cl2 and catalyst 1a briefly exposed to air 1-Methylcyclohex-1-ene (4m) provided di-adduct 6m in good
(footnote † in Table 2). The reaction of the corresponding yield with complete cis selectivity at the cyclohexyl backbone, but
naphthalene 4d gave moderate yield, presumably due to its ease more sterically demanding 1-phenylcyclohex-1-ene (4n) and also
of polymerization (entry 4). The transfer hydrosilylations of 2-methylindene (not shown) did not react at all (entries 14 and 15).
α-substituted styrenes held a few surprises. The reaction of As discussed for transfer hydrosilylation of prochiral 4e (entry 5),
α-methylstyrene (4e) was plagued by substantial decomposition, there is generally no diastereoinduction in multiple additions, and
yielding 26% of the di-adduct 6e rather than the typical tri-adduct statistical mixtures of stereoisomers are obtained with 4i, 4k and 4m
1
7e (entry 5). According to H and 29Si NMR data, 6e is formed as (entries 10, 12, and 14).
a mixture of meso and C2-symmetric compounds, attesting to the
We also attempted the transfer hydrosilylation of alkynes3,42
.
absence of any diastereoinduction in the hydrosilylation of 4e by Terminal alkynes, for example, phenylacetylene and hex-1-yne,
the chiral trihydrosilane intermediate, that is, mono-adduct 5e. In inhibited the reaction completely (not shown), but hex-3-yne (4o)
turn, 1,1-diphenylethylene (4f) participated in the transfer fully converted into trivinylmonohydrosilane 7o (entry 16) with
hydrosilylation with excellent 84% yield of the tri-adduct 7f double bonds of Z geometry. The sensitivity of the method to
(entry 6). Moreover, it was possible to almost completely reverse steric hindrance was further exemplified by the lack of reactivity
the chemoselectivity with this substrate by using 0.55 equiv. of 2b, for diphenylacetylene (not shown).
furnishing the di-adduct 6f in 79% yield (entry 7). This example
corroborates that an adjustment of the stoichiometry will allow Cyclohexa-2,5-dien-1-yl substituents as masked Si–H bonds in
for the control of the chemoselectivity in selected cases. The transition-metal-catalysed hydrosilylation. Although di- and
absence of positive-charge-stabilizing alkene substituents was monohydrosilanes 6 and 7 could be selectively prepared using the
detrimental, and α-olefins such as 4g were not fully converted above methods, trihydrosilanes 5 remained occasionally observed,
into hydrosilylation products. With no branching in the carbon but not accessible chemoselectively. Moreover, α-olefins were too
chain, the tetra-adduct 8g formed chemoselectively in moderate unreactive and were processed to the tetra-adduct 8 after the
yield (entry 8). Conversely, 1,1-dialkyl-substituted alkenes 4h and initial hydrosilylation with SiH4 had occurred. We asked
4
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2329
ARTICLES
Table2 | Transfer hydrosilylation of C–C multiple bonds with SiH4 surrogate 2b.
R1
R1
5 mol% 1a
SiH4–n
+
R2
R2
Si
n
CH2Cl2
RT
R3
R3
H
5 (n = 1)
7 (n = 3)
6 (n = 2)
8 (n = 4)
4
2b (equiv.)
Entry
Substrate
(equiv.)
Time (h) Product
(Yield, %)*
Entry
Substrate
(equiv.)
Time (h) Product
(Yield, %)*
Me
Me
SiH
SiH
nHex
nHex
3
3
X
X
10
4i (0.35)
20
7i (99)‡‡
1
2
3
4a, X = H (0.35)
4b, X = Me (0.35)
4c, X = F (0.35)
20
20
40
7a (91)†
7b (99)
7c (85)‡
SiH2
m
2
m
SiH
11
12
4j, m = 0 (0.55)
4k, m = 1 (0.55)
20
40
6j (83)
6k (55)‡,¶
3
4
4d (0.35)
40
7d (63)§
SiH
R1
R1
3
SiH4–n
13
4l (0.35)
120
7l (42) ‡,††,§§
Ph
Ph
n
R1
R1
5
6
7
4e, R1 = Me (0.35)
4f, R1 = Ph (0.35)
4f, R1 = Ph (0.55)
20
70
20
6e, n = 2 (26)||,¶
7f, n = 3, (84)‡,#
6f, n = 2 (79)**
SiH2
Si
2
nHex
nHex
4
14
15
4m, R1 = Me (0.55)
4n, R1 = Ph (0.55)
80
70
6m (76)¶
8
4g (0.25)
100
8g (50)‡,††
No reaction
Et
Et
SiH
H
3
SiH
3
Et
Et
7o (81)
9
4h (0.35)
20
7h (99)
16
4o (0.35)
20
Conditions: 4 (0.2 mmol), 2 (0.07–0.11 mmol), 1 (0.01 mmol), CH2Cl2 (1.0 M), room temperature, 20–120 h; *isolated yields; †75% yield was obtained at a 0.3 mmol scale in wet, non-degassed CH2Cl2 with 1a
handled in air; ‡full conversion of 4 was not reached; §trace amounts of 6d were also detected by 1H NMR analysis of the crude reaction mixture but could not be isolated; ||decomposition. 5e was also detected in
the crude reaction mixture but could not be isolated; ¶obtained as a 1:1 mixture of meso and C2-symmetric compounds; #2c was used instead of 2b; **trace amounts of 7f were also detected by 1H NMR analysis of
the crude reaction mixture but could not be isolated; ††reaction run on a 0.4 mmol scale; ‡‡obtained as a 3:1 mixture of C1- and C3-symmetric compounds; §§isolated along with trace amounts of 6l.
ourselves whether it would be possible to close these gaps by applied to 10 resulted in clean release of the Si–H bonds,
combining the borane catalysis with transition-metal-catalysed affording 5g in quantitative NMR yield.
hydrosilylation (Fig. 1c). The strategy would be to hydrosilylate,
for example, oct-1-ene (4g) with surrogate 2b or 2c with Conclusion
transition-metal catalysis. If such a reaction were compatible with This work shows that cyclohexa-2,5-dien-1-yl-substituted hydro-
the cyclohexa-2,5-dien-1-yl groups, the Si–H bonds could be silanes 2b and 2c serve as safe and easy-to-handle precursors for
‘deprotected’ with catalytic amounts of B(C6F5)3 (1a) in a the extremely dangerous monosilane (SiH4) gas. The method uses
subsequent step to release the trihydrosilane 5g relevant to their ability to release SiH4 in situ upon treatment with catalytic
material sciences43–47. As a starting point to illustrate this idea, we amounts of B(C6F5)3 (1a) and makes use of the hydrosilylation of
treated surrogate 2b with Karstedt’s catalyst, tris(1,3-divinyl- C–C multiple bonds promoted by the same catalyst. All this
1,1,3,3-tetramethyldisiloxane)diplatinum, in the presence of excess happens in the same pot. This allowed us to perform, for the first
4g and were pleased to obtain hydrosilylation adduct 9 in good time, a comprehensive experimental investigation of alkene hydro-
yield (Fig. 4a). The same procedure applied to 2c containing two silylation with SiH4 in an academic laboratory. The syntheses of
Si–H bonds furnished monohydrosilane 10 in low yield (24%, hydrosilanes that would be difficult to prepare by known methods
not shown). This was substantially improved with dichloro (of the 15 silanes prepared in Table 2, only four had been syn-
(cycloocta-1,5-diene)platinum as catalyst (Fig. 4a). The potential thesized previously) were made possible through practical, metal-
of adducts 9 and 10 to serve as precursors of trihydrosilane 5g free and room-temperature protocols. Moreover, we have also
was verified by catalytic ‘deprotection’ with B(C6F5)3 (1a) in shown here that 2b and 2c are compatible with transition-metal-
CD2Cl2 (Fig. 4b). Merely 30% NMR yield of 5g accompanied by catalysed processes such as platinum(0)-catalysed alkene hydrosily-
significant decomposition to unidentified compounds were lation leaving the cyclohexa-2,5-dien-1-yl substituents untouched.
recorded for the reaction of 9. Conversely, the same conditions Subsequent ‘deprotection’ of the Si–H bonds catalysed by B
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
5
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2329
ARTICLES
13. Sakata, K. & Fujimoto, H. Quantum chemical study of the reaction of
3-(trimethylsilyl)cyclohexa-1,4-dienes with B(C6F5)3. Organometallics 34,
236–241 (2014).
14. DHHS. NIOSH Pocket Guide to Chemical Hazards 279 (Department of Health
and Human Services, US Government Printing Office, 2007).
15. Chen, J. R. Characteristics of fire and explosion in semiconductor fabrication
processes. Process Saf. Prog. 21, 19–25 (2002).
a
3.0–3.2 mol%
Pt2[(Me2SiCH=CH2)2O]3
+
nHex
Si
Si
nHex
RT for 16 h
H
2b
4g (excess)
9 (69%)
16. Chen, J.-R. et al. Analysis of a silane explosion in a photovoltaic fabrication
plant. Process Saf. Prog. 25, 237–244 (2006).
17. Chang, Y.-Y. et al. Revisiting of a silane explosion in a photovoltaic fabrication
plant. Process Saf. Prog. 26, 155–158 (2007).
5.0 mol%
PtCl2(cod)
+
nHex
Si
18. Simmler, W. in Ullmann’s Encyclopedia of Industrial Chemistry Vol. 32,
615–636 (Wiley, 2012).
19. Arkles, B. in Kirk–Othmer Encyclopedia of Chemical Technology (Wiley, 2000);
http://doi.org/fk2mj7
RT for 18 h
Si
H
nHex
H
H
2c
4g (excess)
10 (58%)
20. Schmidt, V., Wittemann, J. V. & Gosele, U. Growth, thermodynamics, and
electrical properties of silicon nanowires. Chem. Rev. 110, 361–388 (2010).
21. Schmidt, V., Wittemann, J. V., Senz, S. & Goesele, U. Silicon nanowires. A review
on aspects of their growth and their electrical properties. Adv. Mater. 21,
2681–2702 (2009).
22. Roca i Cabarrocas, P. Plasma enhanced chemical vapor deposition of silicon thin
films for large area electronics. Curr. Opin. Solid State Mater. Sci. 6,
439–444 (2002).
23. Matsumura, H. Formation of silicon-based thin films prepared by catalytic
chemical vapor deposition (Cat-CVD) method. Jpn J. Appl. Phys. 37,
3175–3187 (1998).
24. Marciniec, B. (ed.) Hydrosilylation (Springer, 2009).
25. Yamamoto, K. & Hayashi, T. in Transition Metals for Organic Synthesis 2nd edn
(eds Beller, M. & Bolm, C.) 167–191 (Wiley, 2004).
b
H
5 mol% 1a
Si
nHex
H
CD2Cl2
Si
nHex
H
RT for 4 h
Z
9, Z = cyclohexa-2,5-dienyl
10, Z = H
5g (30%) from 9
5g (quant.) from 10
Figure 4 | The cyclohexa-2,5-dien-1-yl substituent as a SiH protecting group
in transition-metal-catalysed reactions. a, Platinum(0)-catalysed
hydrosilylation of an α-olefin with SiH4 surrogates leaving the cyclohexa-2,5-
dien-1-yl groups intact. b, B(C6F5)3-catalysed ‘deprotection’ of the Si–H bonds.
26. Ojima, I., Li, Z. & Zhu, J. in The Chemistry of Organic Silicon Compounds
(eds Rappoport, Z. & Apeloig, Y.) 1687–1792 (Wiley, 1998).
27. Ito, M., Abe, T., Takeuchi, A., Iwata, K. & Kobayashi, M. Preparation of
organosilicon compounds. Japanese Patent JP 02045490, A1 9900215 (1990).
28. Itoh, M., Iwata, K., Takeuchi, R. & Kobayashi, M. Hydrosilation of olefins with
monosilane catalyzed by transition metal complexes. J. Organomet. Chem. 420,
C5–C8 (1991).
29. Mitsuzuka, M., Uchiumi, T., Iwata, K. & Ito, M. Preparation of organosilanes
from monosilane and olefins. Japanese Patent JP 07002875, A1 9950106 (1995).
30. Kobayashi, M. & Itoh, M. Hydrosilylation of olefins with monosilane in the
presence of lithium aluminum hydride. Chem. Lett. 25, 1013–1014 (1996).
31. Itoh, M., Iwata, K. & Kobayashi, M. Silylation reactions of olefins with
monosilane and disilane in the presence of a transition metal complex, metal
hydride, and radical initiator. J. Organomet. Chem. 574, 241–245 (1999).
32. Gilman, H. & Miles, D. Disproportionation reaction of diphenylsilane in the
absence of any added catalyst. J. Org. Chem. 23, 326–328 (1958).
33. Voutchkova, A. M. et al. Selective partial reduction of quinolines: hydrosilylation
vs. transfer hydrogenation. J. Organomet. Chem. 693, 1815–1821 (2008).
34. Brandsma, L. & Zwikker, J. W. in Science of Synthesis Vol. 8a (eds Snieckus, V. &
Majewski, M.) 313–327 (Thieme, 2006).
(C6F5)3 (1a) allowed for the preparation of a trihydrosilane derived
from an α-olefin with perfect chemoselectivity compared to the
hydrosilylation of terminal alkenes with SiH4 (Fig. 1b)28.
These results pave the way for the use of cyclohexa-2,5-dien-1-yl
substituents as protecting groups for Si–H bonds in other
transition-metal-catalysed reactions.
Received 23 March 2015; accepted 17 July 2015;
published online 24 August 2015
References
1. Simonneau, A. & Oestreich, M. 3-Silylated cyclohexa-1,4-dienes as precursors
for gaseous hydrosilanes: the B(C6F5)3-catalyzed transfer hydrosilylation of
alkenes. Angew. Chem. Int. Ed. 52, 11905–11907 (2013).
2. Simonneau, A., Friebel, J. & Oestreich, M. Salt-free preparation of trimethylsilyl
ethers by B(C6F5)3-catalyzed transfer silylation by using a Me3SiH surrogate.
Eur. J. Org. Chem. 2077–2083 (2014).
3. Keess, S., Simonneau, A. & Oestreich, M. Direct and transfer hydrosilylation
reactions catalyzed by fully or partially fluorinated triarylboranes: a systematic
study. Organometallics 34, 790–799 (2015).
4. Oestreich, M. & Simonneau, A. Use of cyclohexa-2,5-dien-1-yl-silanes as
precursors for gaseous hydrosilanes. PCT International Patent Application
WO 2015036309, A1 20150319 (2015).
5. Oestreich, M., Hermeke, J. & Mohr, J. A unified survey of Si–H and H–H bond
activation catalysed by electron-deficient boranes. Chem. Soc. Rev. 44,
2202–2220 (2015).
6. Piers, W. E., Marwitz, A. J. V. & Mercier, L. G. Mechanistic aspects of bond
activation with perfluoroarylboranes. Inorg. Chem. 50, 12252–12262 (2011).
7. Webb, J. D., Laberge, V. S., Geier, S. J., Stephan, D. W. & Crudden, C. M.
Borohydrides from organic hydrides: reactions of Hantzsch’s esters with
B(C6F5)3. Chem. Eur. J. 16, 4895–4902 (2010).
8. Gutsulyak, D. V., van der Est, A. & Nikonov, G. I. Facile catalytic hydrosilylation
of pyridines. Angew. Chem. Int. Ed. 50, 1384–1387 (2011).
9. Houghton, A. Y., Hurmalainen, J., Mansikkamäki, A., Piers, W. E. &
Tuononen, H. M. Direct observation of a borane–silane complex involved
in frustrated Lewis-pair-mediated hydrosilylations. Nature Chem. 6,
983–988 (2014).
35. Brook, M. A. Silicon in Organic, Organometallic, and Polymer Chemistry
409 (Wiley, 2000).
36. Ullrich, M., Lough, A. J. & Stephan, D. W. Reversible, metal-free, heterolytic
activation of H2 at room temperature. J. Am. Chem. Soc. 131, 52–53 (2009).
37. Li, L., Stern, C. L. & Marks, T. J. Bis(pentafluorophenyl)(2-perfluorobiphenylyl)
borane. A new perfluoroarylborane cocatalyst for single-site olefin
polymerization. Organometallics 19, 3332–3337 (2000).
38. Chen, Y.-X., Stern, C. L., Yang, S. & Marks, T. J. Organo-Lewis acids as
cocatalysts in cationic metallocene polymerization catalysis. Unusual
characteristics of sterically encumbered tris(perfluorobiphenyl)borane.
J. Am. Chem. Soc. 118, 12451–12452 (1996).
39. Gutmann, V. Solvent effects on the reactivities of organometallic compounds.
Coord. Chem. Rev. 18, 225–255 (1976).
40. Beckett, M. A., Brassington, D. S., Coles, S. J. & Hursthouse, M. B. Lewis
acidity of tris(pentafluorophenyl)borane: crystal and molecular structure of
B(C6F5)3·OPEt3. Inorg. Chem. Commun. 3, 530–533 (2000).
41. Lambert, J. B. The interaction of silicon with positively charged carbon.
Tetrahedron 46, 2677–2689 (1990).
42. Curless, L. D. & Ingleson, M. J. B(C6F5)3-catalyzed synthesis of benzofused
siloles. Organometallics 33, 7241–7246 (2014).
43. Fadeev, A. Y. & McCarthy, T. J. A new route to covalently attached monolayers:
reaction of hydridosilanes with titanium and other metal surfaces. J. Am. Chem.
Soc. 121, 12184–12185 (1999).
10. Parks, D. J. & Piers, W. E. Tris(pentafluorophenyl)boron-catalyzed hydrosilation
of aromatic aldehydes, ketones, and esters. J. Am. Chem. Soc. 118,
9440–9441 (1996).
44. Owens, T. M., Nicholson, K. T., Banaszak Holl, M. M. & Suezer, S. Formation of
alkylsilane-based monolayers on gold. J. Am. Chem. Soc. 124, 6800–6801 (2002).
45. Pelzer, K., Haevecker, M., Boualleg, M., Candy, J.-P. & Basset, J.-M. Stabilization
of 200-atom platinum nanoparticles by organosilane fragments. Angew. Chem.
Int. Ed. 50, 5170–5173 (2011).
11. Rendler, S. & Oestreich, M. Conclusive evidence for an SN2-Si mechanism in the
B(C6F5)3-catalyzed hydrosilylation of carbonyl compounds: implications for the
related hydrogenation. Angew. Chem. Int. Ed. 47, 5997–6000 (2008).
12. Rubin, M., Schwier, T. & Gevorgyan, V. Highly efficient B(C6F5)3-catalyzed
hydrosilylation of olefins. J. Org. Chem. 67, 1936–1940 (2002).
6
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2015 Macmillan Publishers Limited. All rights reserved

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2329
ARTICLES
46. Baudouin, D. et al. Nickel–silicide colloid prepared under mild conditions
as a versatile Ni precursor for more efficient CO2 reforming of CH4 catalysts.
J. Am. Chem. Soc. 134, 20624–20627 (2012).
Author contributions
A.S. and M.O. conceived and designed the experiments. A.S. performed the experiments
and analysed the data. A.S. and M.O. discussed the results and co-wrote the paper.
47. Tuan, H.-Y., Lee, D. C., Hanrath, T. & Korgel, B. A. Catalytic solid-phase seeding
of silicon nanowires by nickel nanocrystals in organic solvents. Nano Lett. 5,
681–684 (2005).
Additional information
Supplementary information and chemical compound information are available in the
online version of the paper. Reprints and permissions information is available online at
www.nature.com/reprints. Correspondence and requests for materials should be
addressed to M.O.
Acknowledgements
This research was supported by the Alexander von Humboldt Foundation (postdoctoral
fellowship to A.S., 2014‒2015) and the Deutsche Forschungsgemeinschaft (Oe 249/11-1).
M.O. is indebted to the Einstein Foundation (Berlin) for an endowed professorship. The
authors thank E. Irran for the X-ray analysis, S. Kemper for expert advice on NMR
spectroscopy, as well as L. Omann and O. Yahiaoui for experimental support
(all TU Berlin).
Competing financial interests
The authors declare no competing financial interests.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
7
© 2015 Macmillan Publishers Limited. All rights reserved