Scope and selectivity of B(C6F5)3-catalyzed reactions of the disilane (Ph2SiH)2

Article abstract of DOI:10.1016/j.jorganchem.2016.02.035

The diverse catalytic activity of the electrophilic borane B(C₆F₅)₃ in reactions of silanes with organic substrates has been exploited extensively in studies that mostly focus on very particular, useful functional group transformations in organic synthesis. This study examines the potential for harnessing the collective, broad scope of these transformations in the preparation of new oligosilane derivatives. Borane-catalyzed hydrosilation, heterodehydrogenative coupling, and dealkylative coupling reactions of a wide range of substrates with the disilane (Ph₂SiH)₂ were investigated. These allowed new mono- and disubstituted disilanes to be prepared that contain Si-O, Si-S, and Si-C linkages. Challenges and opportunities are described that arise from competing "over-reduction" chemistry, and the sensitivity of the catalysis to both the Lewis basicity and steric bulk of the substrates is examined.

Full text of DOI:10.1016/j.jorganchem.2016.02.035

Journal of Organometallic Chemistry 809 (2016) 86e93
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Scope and selectivity of B(C6F5)3-catalyzed reactions of the disilane
(Ph2SiH)2
Peter T.K. Lee, Lisa Rosenberg*
Department of Chemistry, University of Victoria, P.O. Box 3065, Victoria, BC V8W 3V6, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
The diverse catalytic activity of the electrophilic borane B(C6F5)3 in reactions of silanes with organic
substrates has been exploited extensively in studies that mostly focus on very particular, useful func-
tional group transformations in organic synthesis. This study examines the potential for harnessing the
collective, broad scope of these transformations in the preparation of new oligosilane derivatives.
Borane-catalyzed hydrosilation, heterodehydrogenative coupling, and dealkylative coupling reactions of
a wide range of substrates with the disilane (Ph2SiH)2 were investigated. These allowed new mono- and
disubstituted disilanes to be prepared that contain SieO, SieS, and SieC linkages. Challenges and op-
portunities are described that arise from competing “over-reduction” chemistry, and the sensitivity of
the catalysis to both the Lewis basicity and steric bulk of the substrates is examined.
Received 16 February 2016
Accepted 25 February 2016
Available online 2 March 2016
Keywords:
Disilane
SieH activation
Electrophilic borane
Homogeneous catalysis
Functionalized oligosilanes
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
dealkylative coupling (primarily demethanative coupling). Here we
report a wider survey of the scope of all three of these borane-
The vast majority of short-chain oligosilanes contain simple
SieMe or SieAr groups, and it can be challenging to introduce more
varied or functional sidechains [1]. An appealing route to more
structurally diverse oligosilanes involves modifying SieH bonds in
the all-silicon chains that result from metal-catalyzed dehy-
drocoupling of primary or secondary silanes [2]. However, as we
have previously described [3], the SieSi bonds in these catenates
are susceptible to cleavage under “traditional” SieH activation
conditions, for example in the presence of transition metal cata-
lysts, radical initiators, oxidative halogenating reagents, or strong
base. We showed, for a limited number of O- and S-donor sub-
strates, that the Lewis acid B(C6F5)3 catalyzes SieH activation re-
actions of the sym-dihydridodisilanes (Ph2SiH)2 and (Me2SiH)2
with absolute chemoselectivity, leaving the SieSi bonds intact [3].
The chemistry of silanes mediated by B(C6F5)3 is huge and
growing [4]. Scheme 1 summarizes the wide variety of borane-
catalyzed reactions of monosilanes that has been reported in the
literature [5,6]. Of these, we exploited a) hydrosilation (of benzal-
dehyde, p-nitrobenzaldehyde, and thiobenzophenone) and b) het-
erodehydrocoupling (of catechol and aromatic and alkyl thiols) in
our previous work, but we did not explore the utility of c)
catalyzed reactions of the disilane (Ph2SiH)2, which provides gen-
eral insight into the use of this catalyst in the synthesis of struc-
turally complex oligosilane materials. This work provides models
for the potential post-polymerization modification of polysilanes
containing SieH bonds in the repeat unit [7]. Issues of selectivity for
the degree of substitution at this disilane substrate also point to
new strategies for the construction of unsymmetrically derivatized
short-chain oligomers.
2. Results and discussion
2.1. New mono- and disilanes
We screened new substrates for the derivatization of (Ph2SiH)2
by first examining their reactivity with the monosilane Ph2MeSiH,
in which the SieH bond should experience a similar steric and
electronic environment to that in the disilane. The new monosilane
derivatives we prepared are shown in Table 1. Products 1e8 are all
colorless oils, which were characterized by 1H, 13C{1H}, and 29Si{1H}
NMR. The reactions proceed cleanly, except for the hydrosilation of
propionaldehyde, for which competing “over-reduction” chemistry
is observed: borane catalyzes the reaction of the hydrosilation
product Ph2MeSiOPrn (1) and a second equivalent of Ph2MeSiH to
give (Ph2MeSi)2O, presumably with loss of propane. This reaction is
actually a subset of the dealkylative coupling illustrated in Scheme
* Corresponding author.
E-mail address: lisarose@uvic.ca (L. Rosenberg).
http://dx.doi.org/10.1016/j.jorganchem.2016.02.035
0022-328X/© 2016 Elsevier B.V. All rights reserved.

P.T.K. Lee, L. Rosenberg / Journal of Organometallic Chemistry 809 (2016) 86e93
87
solids as indicated. Competing over-reduction of propionaldehyde
in reactions of (Ph2SiH)2 with this substrate led to complex mix-
tures from which we were unable to isolate discrete disilanes, so we
have not included those reactions here. Surprisingly, the reaction of
2-chloroethyl methyl ether with (Ph2SiH)2 to give compound 18
(Table 3) proceeds cleanly, with no evidence for this over-reduction
chemistry, despite being slower than the corresponding reaction
with Ph2MeSiH (vide supra). As discussed below, a monosubstituted
complex could not be isolated via the hydrosilation of 1-hexene,
disubstituted complexes could not be isolated for the hydrosilation
reactions of acetone or cyclohexanone, and the disilane did not
react at all with N-benzylideneaniline.
2.2. Issues affecting the generality of this route to functionalized
disilanes
Scheme 1. Three general reactions of silanes catalyzed by B(C6F5)3.
Our extension of the simple reactions of Ph2MeSiH shown in
Table 1 to the derivatization of (Ph2SiH)2 (Tables 2 and 3) highlights
two issues that can affect strategies for this route to functionalized
oligosilanes, in addition to the possibility of competing over-
reduction reactions such as we described above [11]. First, the
relative Lewis basicity of the substrate can have a large impact on
the conversions obtained (or conditions required), due to potential
catalyst inhibition. We [3] and others [8d,12b] have discussed the
delicate balance of substrate basicity (its binding affinity for the
borane) and nucleophilicity (its ability to attack the silylium-like
1c, and has been reported for other hydrosilation and dehy-
drocoupling reactions that install linear alkoxy sidechains at silicon
[8]. We do not observe comparable over-reduction of 2-chloroethyl
methyl ether in the synthesis of 6, which we attribute to the much
shorter reaction time required for this less Lewis basic substrate to
react (vide infra).
Unlike the heteroatom-containing substrates, the reaction of 1-
hexene to generate Ph2MeSiCH2(CH2)4CH3 (8) required both an
excess of substrate and a switch from toluene or benzene to the
more polar solvent dichloromethane. We suspect that the reaction
is sluggish due to poorer nucleophilicity of the C]C bond for the
proposed silylium intermediate (vide infra), and that a polar solvent
facilitates transient polarization of the double bond during this
critical SieC bond-forming step [9,10].
Tables 2 and 3 show the mono- and disubstituted disilanes we
were able to prepare from the reactions of (Ph2SiH)2 with sub-
strates from Table 1. We have also included products of thiol
dehydrocoupling reactions that we previously reported [3b,7], for
comparison of the requisite conditions. The products shown in
Table 2 are colorless oils; those in Table 3 include both oils and
intermediate generated through an
h
1-silane-borane adduct, vide
infra) in determining the activity of this borane catalyst in the re-
actions of silanes, and these studies provide new illustrations of
this. Second, for silanes such as (Ph2SiH)2, which have more than
one reactive SieH bond, the selectivity for the degree of substitu-
tion (e.g. mono-vs disubstituted) is very sensitive to the steric bulk
of the substrate. Below, we discuss both these issues in the context
of the generally-accepted mechanism for B(C6F5)3-catalyzed SieH
activation, which is shown in Scheme 2 for a carbonyl hydrosilation
reaction [12].
In step a of Scheme 2, the reversible formation of an h
1-silane-
borane complex (A), gives the silicon strong “silylium” character
and renders it highly susceptible to nucleophilic attack (step b). In
Table 1
New B(C6F5)3-catalyzed reactions of Ph2MeSiH (see Scheme 1 for reaction types).a
Substrate
Ph2MeSiX X¼
[Borane]/[SieH]
Time (h)
16
Isolated yield (%)
b
O(CH2)2CH3 (1)
0.04
OCH(CH3)2 (2)
OC6H11 (3)
0.04
0.05
0.05
12
16
1
73
c
OC6H4-p-But (4)
99
OC6H4-p-Me (5)
O(CH2)2Cl (6)
0.04
0.03
1
2
99
83
N(Ph)CH2Ph (7)d
0.04
0.04
16
16
97
74
CH2(CH2)4CH3 (8)e
a
Reactions carried out at RT in toluene, unless otherwise noted.
Quantitative conversion (with respect to silane) to a 76:24 mixture of 1 and the over-reduction product (Ph2MeSi)2O.
Obtained as an 86:14 mixture of 3 and unreacted Ph2MeSiH.
Reaction carried out in benzene.
Reaction carried out in dichloromethane using a four-fold excess of 1-hexene.
b
c
d
e

88
P.T.K. Lee, L. Rosenberg / Journal of Organometallic Chemistry 809 (2016) 86e93
Table 2
Monosubstituted disilanes prepared by B(C6F5)3-catalyzed reactions of (Ph2SiH)2.
X¼
[Borane]/[SiH]total
Time (h)
Isolated yield (%)
86a
OCH(CH3)2 (9)
0.05
0.07
0.05
0.05
0.07
0.05
0.02
16
5
OC6H11 (10)b
67
OC6H4-p-But (11)
OC6H4-p-Me (12)
O(CH2)2Cl (13)d
S(CH2)2CH3 (14)f
SC6H4-p-Me (15)g
16
16
144
16
1
c
99
e
98
90
a
Product contained impurity (<10%) tentatively identified as disubstituted
product (Ph2SiOCH(CH3)2)2.
b
T ¼ 55 ꢀC, six-fold excess of substrate used.
c
Isolated as a 26:67:7 mixture of unreacted (Ph2SiH)2, monosubstituted disilane
11, and disubstituted disilane 16, as determined by 1H NMR.
d
Reaction carried out in d6-benzene.
Isolated as a 15:75:10 mixture of unreacted (Ph2SiH)2, monosubstituted disilane
e
Scheme 2. Generally-accepted mechanism for hydrosilation of carbonyl groups cata-
lyzed by B(C6F5)3.
13, and disubstituted disilane 18, as determined by 1H NMR.
f
Ref. [7].
Ref. [3b].
g
demethanative coupling of the methyl ether derivative of a primary
alcohol, 2-chloroethyl methyl ether, with Ph2MeSiH (Table 1,
compound 6) and (Ph2SiH)2 (Table 3, compound 18). We attribute
this activity to the lower Lewis basicity of the primary ROMe
fragment, relative to primary ROH. The catalyst does not show such
a pronounced difference in activities for the coupling reactions of
aryl methyl ether MeOC6H4-p-Me and the phenol HOC6H4-p-But,
although both O-donor substrates react more slowly, at higher
catalyst loadings, than the softer S-donor thiocresol (e.g. compare
entries in Table 2 for compounds 11 and 12 versus compound 15)
[17].
this case the nucleophile is the O of a generic carbonyl-containing
substrate and the net reaction is an addition; other unsaturated
substrates such as imines [13] and thioketones [3a] behave simi-
larly. However this mechanism can be considered general also for
other nucleophiles including alcohols [14] and silanols [15],
methyl- and other linear alkyl ethers [8a,b], and silyl ethers (or
alkoxysilanes) R'OSiR3 [8d]. These reactions all involve the loss of
either H2 or alkane, RH, from the intermediate ion pair B, in addi-
tion to the silicon-containing product (step c).
Catalyst inhibition was most pronounced for N-benzylidenea-
niline, which did not react at all with (Ph2SiH)2, even when mix-
tures were refluxed in toluene for up to 3 d [18]. This is consistent
with the strong binding affinity of B(C6F5)3 for this imine substrate
reported by Piers et al.; they found that high temperatures, long
reaction times, and/or relatively unhindered silanes were required
to give hydrosilation products [13a]. Presumably nucleophilic
attack of a bulky aryl imine at whatever small amount of activated
silane forms under these conditions is greatly accelerated for less
hindered silanes (vide infra) [19]. Thus the absence of catalytic ac-
tivity for hydrosilation of N-benzylideneaniline by (Ph2SiH)2 not
only highlights the issue of catalyst inhibition, but also indicates the
2.2.1. Catalyst inhibition
The mechanism shown in Scheme 2 shows that the formation of
traditional Lewis adducts can occur (off-cycle step d) in these
borane-catalyzed reactions, but they generally provide a slower
pathway than
h
1-silane adduct formation (step a) and are not
involved in productive catalysis [12d]. If this equilibrium lies suf-
ficiently towards D, catalyst inhibition will occur, since the con-
centration of free borane available to activate the SieH bond is
significantly reduced. For example, primary alcohol substrates
typically bind sufficiently strongly to the borane to inhibit catalytic
dehydrogenative coupling with silanes [14,3b,16]. In this work,
though, the borane catalyst exhibits relatively high activity for
Table 3
Disubstituted disilanes prepared by B(C6F5)3-catalyzed reactions of (Ph2SiH)2.
X¼
[Borane]/[SiH]total
Time
Product
Isolated yield (%)
OC6H4-p-But (16)a
OC6H4-p-Me (17)a
O(CH2)2Cl (18)
0.10
0.08
0.05
0.06
0.04
0.03
16 h
29 d
16 h
72 h
144 h
72 h
white solid
cloudy oil
white solid
colorless oil
colorless oil
white solid
21
b
65
CH2(CH2)4CH3 (19)c
S(CH2)2CH3 (20)d
SC6H4-p-Me (21)d
87
e
94
a
T ¼ 70 ꢀC.
b
Isolated as a 65:35 mixture of the monosubstituted disilane 12 and the disubstituted disilane 17.
Reaction carried out in a ten-fold excess of 1-hexene diluted with 10% dichloromethane (by volume).
Ref. [3b].
c
d
e
Four equivalents of n-propane thiol gave 1:1 mixture of monosubstituted disilane 14 and disubstituted disilane 20.

P.T.K. Lee, L. Rosenberg / Journal of Organometallic Chemistry 809 (2016) 86e93
89
higher steric hindrance associated with the SieH bonds in
(Ph2SiH)2 relative to our model monosilane Ph2MeSiH, since
hydrosilation of the same imine by the latter does occur at room
temperature in the presence of the borane catalyst (Table 1, com-
pound 7).
2.2.2. Steric bulk at the disilane and the substrate
The selectivity of these borane-catalyzed reactions of (Ph2SiH)2
for mono- or disubstituted products is sensitive to the bulk of the
organic substrate. For example, substrates that give new Si-
substituents that are branched beta to silicon (e.g. ketone, phenol,
methyl phenyl ether, and thiophenol) favour formation of the
monosubstituted disilane (compounds 9e12, 15 in Table 2), and it is
challenging to obtain the disubstituted product in these cases. Of
these substrates, disubstituted compounds derived from acetone
and cyclohexanone could not be prepared even with increased
catalyst loadings and temperature, but those containing the
(effectively branched but arguably less bulky) aryl substituents
(eOC6H4-p-But (16) and eSC6H4-p-Me (21), Table 3) were acces-
sible. These results can be rationalized in terms of the relative rates
of nucleophilic attack of the substrate at the silylium centre (step b
in the catalytic cycle in Scheme 2). Scheme 3 shows the proposed
transition state structure for a carbonyl nucleophile, in which the
incoming O-donor is trans-to the departing hydride at silicon [12b-
d,13b]. The Scheme illustrates how the rate of this step will be
sensitive to steric bulk of both the substituents at silicon and the
nucleophile. For the relatively bulky (Ph2SiH)2 this nucleophilic
attack is likely to be rate-determining even for the first substitution
leading to monosubstituted product. Once a new substituent is
installed at one of the two silicons in (Ph2SiH)2, the steric encum-
brance at the remaining SieH bond is certainly enhanced, and
much more so for the branched, relative to the linear, side-chains
introduced. For the methyl phenyl ether substrate, CH3OC6H4-p-
Me, the steric hindrance caused by the presence of the new phe-
noxy substituent in the monosubstituted “intermediate” 12, is
compounded by presence of the methyl group adjacent to the O-
donor in the incoming second equivalent of substrate (Scheme 4).
Thus, in contrast to the relatively facile preparation of the bis-
phenoxy derivative 16 (Table 3), the reaction of two (or more)
equivalents of CH3OC6H4-p-Me with (Ph2SiH)2 gave only partial
conversion (35% estimated by 1H NMR integration) to the disub-
stituted product (Ph2SiOC6H4-p-Me)2 (17), even under forcing
conditions (Table 3). This is discussed further below, in the context
of strategies for the synthesis of unsymmetrically substituted
disilanes.
Scheme 4. Nucleophilic attack of aryloxy substrates at borane-activated, mono-
substituted disilane.
these conditions we observe almost quantitative conversion to
disubstituted product 15, with just trace amounts (~1% by 1H NMR)
of the monosubstituted product; we were unable to identify con-
ditions to allow isolation of the monosubstituted product. These
results point to a barrier to the second substitution that is actually
lower than that for the first, despite the presence of the additional
alkyl substituent in the monosubstituted product. This might be
attributable to an inductive electronic effect of the new alkyl
sidechain on the ease of borane activation of the second SieH bond.
Curiously, in our previous studies of the dehydrogenative
coupling of the linear substrate n-propylthiol by (Ph2SiH)2 we
found conversion to the disubstituted disilane (Ph2SiSCH2CH2CH3)2
to be very slow [3b]; even after 6 d reaction time at relatively high
catalyst loading (for this substrate), we obtained 1:1 mixtures of
the mono- and disubstituted disilanes. It seems unlikely that the
slower second substitution in this case could be due to steric hin-
drance in the monosubstituted disilane 13, despite the larger size of
S relative to C or O, especially since we had no trouble preparing the
analogous S-disubstituted disilane 19 via dehydrogenative coupling
of HSC6H4-p-Me (thiocresol). This may point to a possible damp-
ening electronic effect of the -SPrn substituent on the borane acti-
vation of the residual SieH bond in 13 [20]. We have not yet
explored more forcing conditions such as increased temperature
and catalyst loading in an attempt to push the second substitution
reaction to completion.
3. Conclusions
Borane-catalyzed reactions of poly(hydrido)oligosilanes provide
a wide scope of strategies for the construction of new, functional
molecules containing SieSi bonds. These studies delineate the is-
sues that may arise in exploring new substrates; the isolation of
pure, discrete products will almost inevitably require some careful
optimization of catalyst loading, temperatures, and reaction times,
especially when the products are oils instead of solids (see Section 4
and the Supporting Material).
The selectivity of these disilane derivatization reactions for
mono- or disubstituted products provides an interesting opportu-
nity to prepare novel, unsymmetrically substituted disilanes. We
demonstrated this while assessing the low conversions we ob-
tained for preparation of the disubstituted disilane (Ph2SiOC6H4-p-
Me)2 (17). To show that the presence of one eOC6H4-p-Me group in
the monosubstituted disilane 12 did not inherently deactivate the
remaining SieH bond in this molecule, we examined its reaction
with HOC6H4-p-But in the presence of borane. The reaction
required relatively high catalyst loading, and elevated tempera-
tures, but we were able to isolate the mixed disilane 22 as a
colorless oil in 84% yield (Scheme 5). We continue to explore the
potential of this strategy in preparing new disilanes and larger,
extended macromolecules with mixed substituents.
These steric arguments suggest that it should be relatively easy
to access the disubstituted disilane (Ph2SiX)2 for substrates that
lead to linear sidechains at Si. We did find that careful control of the
stoichiometry in demethanative coupling reactions of 2-
chloroethyl methyl ether is sufficient to allow isolation of either
the monosubstituted product 12 (Table 2) or the disubstituted
product 18 (Table 3). It is also straightforward to obtain the
disubstituted product of hydrosilation of 1-hexene by (Ph2SiH)2
(Table 3, compound 19). In this case, however, a high concentration
of 1-hexene is required to give reasonable conversions (ten-fold
excess of 1-hexene diluted in just 10% CH2Cl2 by volume). Under
R2
R1
‡
R
R
C6F5
C
O
Si
R
H
B
C6F5
C6F5
Scheme 3. Transition state for nucleophilic attack of carbonyl at borane-activated
silicon.

90
P.T.K. Lee, L. Rosenberg / Journal of Organometallic Chemistry 809 (2016) 86e93
4.2. Synthesis of monosilane derivatives
General procedure for compounds 1-7. Substrate was added with
stirring to a toluene or benzene solution of Ph2MeSiH and B(C6F5)3
in a Schlenk flask. The mixture was left stirring under N2, either is a
sealed flask or open to a Nujol bubbler. PPh3 was added to quench
the catalyst and the volatiles were removed under vacuum. Addi-
tion of hexanes or pentane to the resulting oily residue caused
precipitation of the borane-phosphine adduct, which was removed
by filtration through a Celite filter stick. Additional solvent was
added to ensure elution of the product(s) from the filter stick, and
volatiles were removed under vacuum from the combined wash-
ings to give the final isolated product(s).
Scheme 5. Synthesis of a mixed bis(aryloxy)-substituted disilane.
Ph2MeSiOCH2CH2CH3 (1). Ph2MeSiH (0.10 g, 0.52 mmol), B(C6F5)3
(0.009 g, 0.02 mmol), toluene (1 mL), propionaldehyde (0.10 mL,
1.4 mmol); stirred under N2 (closed flask) for 16 h PPh3 (0.005 g,
0.02 mmol), hexanes (4 ꢂ 1 mL). Dried under vacuum at RT for 16 h.
Clear, colorless oil (0.13 g, in 76% purity. The remaining 24% was the
over-reduction product, (Ph2MeSi)2O). For 1: 1H NMR (300 MHz,
4. Experimental
4.1. General details
C6D6)
d
0.58 (s, 3H), 0.84 (t, 3JHH ¼ 6 Hz, 3H, CH3), 1.46e1.57 (m, 2H,
All reactions and manipulations were performed under nitrogen
using conventional glovebox or Schlenk techniques. Toluene,
pentane, hexanes and dichloromethane were degassed by sparging
with nitrogen and dried by passing through columns of activated
alumina in a solvent purification system. d6-Benzene (Aldrich) was
degassed by one freeze-pump-thaw cycle, and vacuum transferred
from sodium/benzophenone. Tris(pentafluorophenyl)borane (Alfa
Aesar), was doubly sublimed. Triphenylphosphine (Alfa Aesar) was
recrystallized from 95% ethanol and sublimed. Cyclohexanone,
acetone (Caledon), and propionaldehyde (Aldrich) were distilled
from anhydrous magnesium sulphate. 2-Chloroethyl methyl ether
(Aldrich) was distilled from calcium hydride. p-Methylanisole
(Aldrich) was sublimed. p-t-Butylphenol (Aldrich) was recrystal-
lized from hexanes and sublimed. 1-Hexene (Aldrich) was distilled
from sodium/benzophenone. N-Benzylideneaniline, prepared from
the condensation of benzaldehyde and aniline in CH2Cl2, was
recrystallized from CH2Cl2 and then sublimed (70 ꢀC, dynamic
vacuum) [21]. Diphenylmethylsilane (Aldrich) was used as
received. (Ph2SiH)2 was prepared by a literature procedure [22].
Florisil® (Aldrich or Caledon) was dried in an oven (135 ꢀC,
ꢁ16 h) and vacuum oven (~0.05 mm Hg, 45 ꢀC, 6 h), before being
taken into the glovebox for use. “Florisil column” refers to a Pasteur
pipette plugged with oven-dried glass wool and filled with ~6 cm of
Florisil® and ~1 cm of sand. Celite (AGP or Aldrich) was dried in an
oven (135 ꢀC, ꢁ16 h) and vacuum oven (~0.05 mm Hg, 45 ꢀC, for at
least 6 h). “Celite filter stick” refers to a Pasteur pipette, plugged
with oven-dried glass wool and filled with ~1 cm of Celite and
~1 cm of sand. “Bomb flask” refers to thick-walled, cylindrical flask
with a 4 mm Teflon needle valve and a side-arm.
OCH2CH2) 3.57 (t, 3JHH ¼ 6 Hz, 2H, OCH2), 7.19e7.23 (overlapping m,
6H, Hm/p-SiPh2), 7.64e7.69 (m, 4H, Ho-SiPh2). 13C NMR (75 MHz,
C6D6)
d
ꢃ2.9 (SiCH3), 10.4 (CH2CH2CH3), 26.1 (OCH2CH2), 65.1
(OCH2), 128.0 (Cm-SiPh2), 129.9 (Cp-SiPh2), 134.6 (Co-SiPh2). 29Si
NMR (99 MHz, C6D6)
C6D6)
d
ꢃ3.8. For (Ph2MeSi)2O: 1H NMR (300 MHz,
d
0.38 (s, 6H, CH3), 7.19e7.23 (overlapping m, 12H, overlaps
H
m/p-SiPh2 in 1, Hm/p-SiPh2), 7.38 (d, 8H, Ho-SiPh2). 13C NMR
(75 MHz, C6D6) 2.9 (SiCH3, overlaps SiCH3 in 2e3), 128.2 (Cm-
SiPh2), 129.8 (Cp-SiPh2), 134.3 (Co-SiPh2). DEPT30 29Si NMR
(99 MHz, C6D6)
ꢃ9.5 [23]. Anal. (calcd for 76% C16H20OSi and 24%
26H26OSi2): C 72.51 (75.31) H 6.99 (7.36) [24].
Ph2MeSiOCH(CH3)2 (2). Ph2MeSiH (0.13 g, 0.65 mmol), B(C6F5)3
d
d
C
(0.012 g, 0.023 mmol), toluene (1 mL), acetone (0.10 mL, 1.4 mmol);
stirred under N2 (closed flask) for 12 h PPh3 (0.006 g, 0.02 mmol),
pentane (4 ꢂ 1 mL). Dried under vacuum at RT for 16 h. Clear,
colorless oil (0.12 g, 73%). 1H NMR (300 MHz, C6D6)
d 0.59 (s, 3H,
3
CH3) 1.11 (d, JHH ¼ 6 Hz, 6H, CH(CH3)2), 4.00 (sept, 1H, OCH),
7.17e7.21 (overlapping m, 6H, Hm/p-SiPh), 7.65e7.70 (m, 4H, Ho-
SiPh). 13C NMR (75 MHz, C6D6)
d 0.0 (SiCH3), 28.0 (OCH), 58.0
(OCH(CH3)2), 130.2 (Cm-SiPh2), 132.0 (Cp-SiPh2), 136.9 (Co-SiPh2).
DEPT30 29Si NMR (99 MHz, C6D6)
C 74.44 (74.95) H 7.72 (7.86).
d
ꢃ6.1. Anal. (calcd for C16H20OSi):
Ph2MeSiOC6H11 (3). Ph2MeSiH (0.10 g, 0.50 mmol), B(C6F5)3
(0.012 g, 0.023 mmol), toluene (1 mL), cyclohexanone (0.060 mL,
0.58 mmol); stirred under N2 (closed flask) for 16 h PPh3 (0.006 g,
0.02 mmol), hexanes (4 ꢂ 1 mL). Dried under vacuum at 70 ꢀC for
2 h. Clear and colorless oil (0.12 g, in 86% purity. The remaining 14%
was unreacted Ph2MeSiH). 1H NMR (300 MHz, C6D6)
d 0.61 (s, 3H,
NMR spectra were obtained on Bruker AVANCE 300 (1H, 13C),
500 (13C, 29Si), or 360 (29Si) spectrometers. Chemical shifts are re-
ported in ppm at ambient temperature. 1H NMR spectra were
SieCH3), 1.01e1.13 (overlapping m, 3H, OCHCH2CHeq and
OCHCH2CH2CHeq), 1.29 (m, 1H, OCHCH2CH2CHax), 1.41e1.54 (m, 2H,
OCHCHeq), 1.57e1.66 (m, 2H, OCHCH2CHax), 1.71e1.80 (m, 2H,
OCHCHax), 3.77 (quint,1H, OeCH), 7.18e7.22 (overlapping m, 6H, Hm/
referenced to residual protonated d6-benzene (d
7.16 ppm). 13C
3
p-SiPh2), 7.69 (d, JHH ¼ 7 Hz, 4H, Ho-SiPh2). 13C NMR (75 MHz,
NMR spectra (run as DEPT 135 experiments unless otherwise
noted) were referenced to solvent peaks. 29Si NMR spectra were
referenced to external TMS at 0 ppm. Melting points were collected
on a Gallenkamp melting point apparatus and are uncorrected.
Elemental analyses were performed by Canadian Microanalyt-
ical Service Ltd., Delta, BC. Almost all products were obtained as
oils, which contained trace unidentified impurities and/or small
amounts of unreacted reagents or by-products. Microanalytical
data is provided nonetheless for all products to give an indication of
bulk purity; in some cases combined and weighted formulae were
used to calculate expected %C, %H. NMR spectra for all samples are
included in the Supplementary data.
C6D6) ꢃ2.1 (SieCH3), 23.9 (OCHCH2CH2), 25.8 (OCHCH2CH2CH2),
35.9 (OCHCH2), 71.3 (OeCH), 128.0 (Cm-SiPh2), 129.8 (Cp-SiPh2),
134.6 (Co-SiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
(calcd for 86% C19H24OSi and 14% C13H14Si): C 74.94 (77.14) H 7.88
(8.06) [24].
d
ꢃ6.1. Anal.
Ph2MeSiOC6H4-p-But (4). Ph2MeSiH (0.12 g, 0.58 mmol), B(C6F5)3
(0.014 g, 0.027 mmol), toluene (1 mL), p-t-butylphenol (0.088 g,
0.59 mmol); stirred under N2 (open to Nujol bubbler) for 1 h PPh3
(0.007 g, 0.03 mmol), pentane (4 ꢂ 1 mL). Dried under vacuum at
70 ꢀC for 2 h. Clear and colorless oil (0.20 g, 99%, containing a trace
amount (<1%) of HOC6H4-p-But as determined by 1H NMR). 1H NMR
(300 MHz, C6D6)
d 0.67 (s, 3H, SiCH3), 1.16 (s, 9H, C(CH3)3),

P.T.K. Lee, L. Rosenberg / Journal of Organometallic Chemistry 809 (2016) 86e93
91
6.92e6.97 (m, 2H, Ho-C6H4-p-But), 7.05e7.09 (m, 2H, Hm-C6H4-p-
33.7 (SiCH2CH2) 128.2 (Cm-SiPh2), 129.4 (Cp-SiPh2), 134.9 (Co-
SiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
ꢃ7.1. Anal. (calcd for
C19H26Si): C 80.68 (80.78) H 9.53 (9.28).
3
But), 7.17e7.20 (m, 6H, Hm/p-SiPh2), 7.70 (d, 4H, JHH ¼ 4 Hz, Ho-
d
SiPh2); 13C NMR (75 MHz, C6D6)
d
ꢃ2.6 (SiCH3), 31.4 (C(CH3)3) 33.9
(C(CH3)3, identified from direct acquisition expt)), 119.6 (Co-C6H4-p-
But), 126.5 (Cm-C6H4-p-But), 128.7 (Cm-SiPh2), 130.2 (Cp-SiPh2),
4.3. Synthesis of disilane derivatives
134.7 (Co-SiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
(calcd for C23H26OSi): C 81.8 (79.71) H 7.93 (7.56).
d
ꢃ3.7. Anal.
Ph2SiHeSi(OCH(CH3)2)Ph2 (9). Procedure was as described for
monosilanes 1e7. (Ph2SiH)2 (0.21 g, 0.58 mmol), B(C6F5)3 (0.030 g,
0.059 mmol), toluene (2 mL), acetone (0.050 mL, 0.68 mmol);
stirred under N2 (closed flask) for 16 h PPh3 (0.015 g, 0.057 mmol),
pentane (6 ꢂ 1 mL). Dried under vacuum at RT for 16 h. Clear and
colorless oil (0.23 g, 86% in >90% purity). A small amount of
isopropoxy-containing impurity was observed by 1H NMR, which
may be the disubstituted product (Ph2SiHOCH(CH3)2)2. When the
experiment was repeated using two equivalents of acetone, the
same ratio of monosubstituted product to isopropoxy-containing
impurity was obtained. Likewise, the addition of a second equiv
of acetone to isolated monosubstituted disilane product showed no
reaction: the monosubstituted disilane with the same quantity of
Ph2MeSiOC6H4-p-Me (5). Ph2MeSiH (0.13 g, 0.63 mmol), B(C6F5)3
(0.014 g, 0.027 mmol), toluene (1 mL), p-methylanisole (0.083 g,
0.68 mmol); stirred under N2 (open to Nujol bubbler) for 1 h PPh3
(0.007 g, 0.03 mmol), pentane (4 ꢂ 1 mL). Dried under vacuum at
RT for 16 h. Clear and colorless oil (0.19 g, 99%, contains trace
unreacted MeOC6H4-p-Me). 1H NMR (300 MHz, C6D6)
d 0.66
3
(SiCH3), 2.02 (p-CH3), 6.80 (d, 2H, JHH ¼ 9 Hz, Ho-OC6H4-p-ME),
6.88 (d, 2H, 3JHH ¼ 8 Hz, Hm-OC6H4-p-Me), 7.18 (m, 6H, Hm/p-SiPh2),
7.69 (d, 4H, 3JHH ¼ 4 Hz, Ho-SiPh2); 13C NMR (75 MHz, C6D6)
ꢃ2.6
d
(SiCH3), 20.4 (p-CH3), 120.0 (Co-OC6H4-p-Me), 128.1 (Cm-SiPh2),
128.2 (Cm-OC6H4-p-Me), 130.2 (Cp-SiPh2), 134.7 (Co-SiPh2). DEPT30
29Si NMR (99 MHz, C6D6)
(78.90) H 6.49 (6.62).
d
ꢃ3.6. Anal. (calcd for C20H20OSi): C 77.52
isopropoxy-containing impurity was recovered. 1H NMR (300 MHz,
3
Ph2MeSiOCH2CH2Cl (6). Ph2MeSiH (0.16 g, 0.79 mmol), B(C6F5)3
(0.011 g, 0.021 mmol), toluene (1 mL), 2-chloroethyl methyl ether
(0.10 mL,1.1 mmol); stirred under N2 (open to Nujol bubbler) for 2 h
PPh3 (0.005 g, 0.02 mmol), pentane (6 ꢂ 1 mL). Dried under vacuum
C6D6)
d
1.08 (d, JHH ¼ 6.0 Hz, 6H, Si(CH(CH3)2)), 4.18 (sept, 1H,
1
Si(CH(CH3)2)), 5.55 (s, JSiH ¼ 183 Hz, 1H, SieH), 7.08e7.16 (over-
lapping m, 12H, Hm/p-SiPh2), 7.65e7.79 (m, 8H, Ho-SiPh2); 13C NMR
(75 MHz, C6D6)
d 25.7 (CH3), 67.4 (OCH), 127.9 (Cm-HSiPh2), 128.1
at RT for 16 h. Clear and colorless oil (0.18 g, 83%). 1H NMR
(Cm-OSiPh2), 128.2 (Cp-HSiPh2), 130.0 (Cp-OSiPh2), 135.4 (Co-
3
(300 MHz, C6D6)
d
0.52 (s, 3H, SiCH3), 3.17 (t, JHH ¼ 6 Hz, 2H,
HSiPh2), 136.5 (Co-OSiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
OCH2CH2Cl), 3.58 (t, 2H, OCH2CH2Cl), 7.17e7.21 (overlapping m, 6H,
d
ꢃ38.2 (SiH), -8.1 (SiOCH(CH3)2). Anal. (calcd for C27H28OSi2): C
H
m/p-SiPh2), 7.60 (d, 3JHH ¼ 4 Hz, 4H, Ho-SiPh2); 13C NMR (75 MHz,
75.88 (76.36) H 6.93 (6.65).
C6D6)
d
ꢃ3.1 (SiCH3), 45.1 (OCH2CH2Cl), 63.8 (OCH2CH2Cl), 128.1
Ph2SiHeSi(OC6H11)Ph2 (10). Procedure was as described for
monosilanes 1e7. (Ph2SiH)2 (0.10 g, 0.28 mmol), B(C6F5)3 (0.019 g,
0.037 mmol), benzene (1 mL), cyclohexanone (0.20 mL, 1.9 mmol);
used “bomb” flask (vide supra), degassed mixture by one freeze-
pump-thaw cycle, heated sealed, evacuated flask in an oil bath
(55 ꢀC) for 5 h PPh3 (0.010 g, 0.037 mmol), hexanes (5 ꢂ 1 mL).
Dried under vacuum at 50 ꢀC for 1 h. Clear and colorless oil (0.084 g,
(Cm-SiPh2), 130.1 (Cp-SiPh2), 134.6 (Co-SiPh2). 29Si NMR (99 MHz,
C6D6)
(6.19).
d
ꢃ1.9. Anal. (calcd for C15H17OClSi): C 64.98 (65.08) H 6.14
Ph2MeSiN(Ph)CH2Ph (7). Ph2MeSiH (0.11 g, 0.55 mmol), B(C6F5)3
(0.012 g, 0.023 mmol), benzene (1 mL), N-benzylideneaniline
(0.10 g, 0.55 mmol); used “bomb” flask (vide supra), degassed
mixture by one freeze-pump-thaw cycle, heated flask in an oil bath
(60 ꢀC) for 48 h PPh3 (0.006 g, 0.02 mmol), hexanes (4 ꢂ 1 mL).
Dried under vacuum at 60 ꢀC for 2 h. Clear and colorless oil (0.20 g,
97%, trace impurities were observed by 1H NMR). 1H NMR
67%). 1H NMR (300 MHz, C6D6)
d 0.73e0.91 (overlapping m, 3H,
OCHCH2CHeq and OCHCH2CH2CHeq), 1.02 (m, 1H, OCHCH2CH2CHax),
1.19e1.30 (m, 2H, OCHCHeq), 1.30e1.41 (m, 2H, OCHCH2CHax),
1.49e1.61 (m, 2H, OCHCHax), 3.93 (quintet, 1H, OeCH), 5.56 ppm (s,
1JSiH ¼ 184 Hz, 1H, SieH), 7.08e7.12 (overlapping m, 8H, Hm-SiPh
and Hm-OSiPh2), 7.14e7.18 (m, 4H, Hp-SiPh and Hp-OSiPh2), 7.68 (d,
(300 MHz, C6D6) d 0.71 (s, 3H, SiCH3), 4.54 (s, 2H, NCH2), 6.94 (t, 2H,
3JHH ¼ 8 Hz, Hm-CH2Ph), 7.00 (t, 2H, 3JHH ¼ 7 Hz, Hm-NPh, overlaps
3
Ho-CH2Ph), 7.01 (d, 2H, JHH ¼ 8 Hz, Ho-CH2Ph, overlaps Hm-NPh),
3JHH ¼ 6 Hz, 4H, Ho-SiPh), 7.78 (d, 3JHH ¼ 7 Hz, 4H, Ho-OSiPh2). 13
C
3
7.04 (t, 1H, JHH ¼ 8 Hz, Hp-CH2Ph), 7.12e7.17 (overlapping m, 9H,
NMR (75 MHz, C6D6) 23.9 (OCHCH2CH2), 25.7 (OCHCH2CH2CH2),
35.9 (OCHCH2), 73.0 (OeCH), 128.1 (Cm-SiPh), 128.2 (Cm-OSiPh2),
129.4 (Cp-SiPh), 130.0 (Cp-OSiPh2), 135.3 (Co-SiPh), 136.5 (Co-
Hp-NPh, Hm-SiPh2, Hp-SiPh2, Ho-NPh), 7.59 (d, 4H, 3JHH ¼ 8 Hz, Ho-
SiPh2). 13C NMR (75 MHz, C6D6)
d
ꢃ1.1 (SiCH3), 53.1 (NCH2), 120.1
(Cp-CH2Ph), 125.6 (Cp-SiPh2), 126.7 (Cm-NPh), 126.8 (Co-CH2Ph),
128.4 (Cp-NPh), 128.6 (Co-NPh), 129.0 (Cm-CH2Ph), 130.1 (Cm-SiPh2),
OSiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
d
ꢃ33.8 (SiH), -8.2
(SiOC6H11). Anal. (calcd for C30H32OSi2): C 77.53 (77.53) H 6.80
(6.94).
135.0 (Co-SiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
(calcd for C26H25NSi): C 81.30 (82.27) H 7.20 (6.64).
d
ꢃ5.3. Anal.
Ph2SiHeSi(OC6H4-p-But)Ph2 (11). (Ph2SiH)2 (0.10 g, 0.27 mmol),
HOC6H4-p-But (0.041 g, 0.27 mmol) and B(C6F5)3 (0.015 g,
0.029 mmol) were combined in toluene (1 mL) in a Schlenk flask.
The solution bubbled and frothed (elimination of H2). The mixture
was degassed with one freeze-pump-thaw cycle, then stirred under
static vacuum at RT for 16 h. Volatiles were removed by evacuation
to give a clear, faintly brown oil, which was washed through a
Florisil column with hexanes (4 ꢂ 1 mL) to remove B(C6F5)3 (and
coloured impurity). Dried under vacuum at 60 ꢀC for 1 h. Clear,
colorless oil (0.095 g; shown by 1H NMR to contain 67% 11, along
with 26% unreacted (Ph2SiH)2 and 7% disubstituted disilane 16). For
Ph2MeSiCH2CH2Bun (8). PhMe2SiH (0.21 g, 1.1 mmol) and
B(C6F5)3 (0.025 g, 0.049 mmol) were combined with 1-hexene
(1.0 mL, 4.0 mmol) and CH2Cl2 (0.5 mL) in a Schlenk flask. The
mixture was stirred under N2 with the flask closed for 16 h. The
volatiles were removed under vacuum, and the resulting oily res-
idue was dissolved in pentane (1 mL) and filtered through a Florisil
column to remove residual B(C6F5)3 (a faint brown color was
retained on the column). The column was washed with pentane
(3 ꢂ 1 mL), and the volatiles were removed under vacuum from the
combined filtrates to give a clear and colorless oil (0.22 g, 74%, trace
impurities were observed by 1H NMR). 1H NMR (300 MHz, C6D6)
11: 1H NMR (300 MHz, C6D6)
d 1.14 (s, 9H, overlaps C(CH3)3 from
1
d
0.51 (s, 3H, SieCH3), 0.86 (t, 3JHH ¼ 7 Hz, 3H, (CH2)5CH3),1.00e1.46
(Ph2SiOC6H4-p-But)2, C(CH3)3), 5.56 (s, 1H, JSiH ¼ 188 Hz, SieH),
6.98e7.00 (overlapping m, 4H, overlaps Hm-C6H4-p-But in
(Ph2SiOC6H4-p-But)2, Ho/m-C6H4-p-But), 7.05e7.13 (overlapping m,
12H, overlaps Hm/p-SiPh2 in (Ph2SiOC6H4-p-But)2 and (Ph2SiH)2, Hm/
p-HSiPh and Hm/p-OSiPh), 7.57 (d, 4H, overlaps Ho-SiPh2 in
(overlapping m, 10H, SieCH2CH2CH2CH2CH2), 7.18e7.23 (over-
lapping m, 6H, Hm/p-SiPh2), 7.53 (d, 3JHH ¼ 4 Hz, 4H, Ho-SiPh2); 13
C
NMR (75 MHz, C6D6)
d
ꢃ4.2 (SiCH3), 14.3 (Si(CH2)5CH3), 14.6
(SiCH2), 23.0 (Si(CH2)4CH2), 24.2 (Si(CH2)3CH2), 31.8 (Si(CH2)2CH2),

92
P.T.K. Lee, L. Rosenberg / Journal of Organometallic Chemistry 809 (2016) 86e93
3
(Ph2SiH)2, JHH ¼ 4 Hz, Ho-HSiPh2), 7.56e7.83 (overlapping m, 4H,
complete conversion to the monosubstituted product, Ph2SiHe-
Si(OC6H4-p-Me)Ph2 (12). More B(C6F5)3 (0.022 g, 0.043 mmol) and
toluene (1 mL) were added, the flask was degassed with one freeze-
pump-thaw cycle, and the mixture was again heated to 70 ꢀC under
static vacuum for 28 d, after which the flask was cooled to RT and
toluene was removed under vacuum to give a light brown, viscous
oil. The oil was washed through a Florisil column with hexanes
(4 ꢂ 1 mL) to remove B(C6F5)3. Volatiles were removed under
vacuum with heating to 60 ꢀC, to give 0.13 g of a cloudy oil that was
Ho-OSiPh2). 13C NMR (75 MHz, C6D6)
d 31.4 (CH3), 33.9 (C(CH3), from
direct acquisition expt) 119.8 (CoeC6H4-p-But), 126.4 (Cm-C6H4-p-
But), 128.2 (Cm-HSiPh2), 128.3 (Cm-OSiPh2), 129.6 (Cp-HSiPh2), 130.3
(Cp-OSiPh2), 135.3 (Co-HSiPh2), 136.5 (Co-OSiPh2). DEPT30 29Si NMR
(99 MHz, C6D6)
34H34OSi2 (12), 26% C24H22Si2 ((Ph2SiH)2), and 7% C44H46O2Si2
(16)): C 79.36 (79.23) H 6.52 (6.57).
d
ꢃ38.3 (SiH), -6.4 (SiAr). Anal. (calcd for 67%
C
Ph2SiHeSi(OC6H4-p-Me)Ph2 (12). Procedure was as described for
1e7. (Ph2SiH)2 (0.30 g, 0.82 mmol), B(C6F5)3 (0.042 g, 0.082 mmol),
toluene (2 mL), p-methylanisole (0.10 g, 0.82 mmol); stirred under
N2 (open to Nujol bubbler) for 16 h PPh3 (0.021 g, 0.080 mmol),
pentane (6 ꢂ 1 mL). Dried under vacuum at 70 ꢀC for 2 h. Clear,
colorless, viscous oil (0.38 g, 99%) in >97% purity. A small amount
a 65:35 mixture of 12 and 17. For 17: 1H NMR (300 MHz, C6D6)
d 1.97
(s, 6H, p-CH3, overlaps p-CH3 in 12), 6.73 (d, 4H, Hm-OC6H4-p-Me,
overlaps Hm-OC6H4-p-Me in 12), 6.93 (d, 4H, Ho-OC6H4-p-Me,
overlaps Ho-OC6H4-p-Me in 12), 7.06e7.12 (overlapping m, 6H, Hm/
p-SiPh2, overlaps Hm/p-SiPh2 in 12), 7.76e7.82 (overlapping m, 8H,
Ho-SiPh2, overlaps Ho-SiPh2 in 12). Mixture was not submitted for
microanalysis.
(~3%) of unreacted (Ph2SiH)2 was detected by 1H NMR. 1H NMR
1
(300 MHz, C6D6)
d
1.99 (s, 3H, p-CH3), 5.56 (s, 1H, JSiH ¼ 187 Hz,
SieH), 6.74 (dt, 2H, 3JHH ¼ 8 Hz, 4JHH ¼ 1 Hz, Ho-OC6H4-p-Me), 6.92
(Ph2SiOCH2CH2Cl)2 (18). (Ph2SiH)2 (0.10 g, 0.27 mmol) and
B(C6F5)3 (0.015 g, 0.029 mmol) were combined with toluene (1 mL)
3
4
(dt, 2H, JHH ¼ 9 Hz, JHH ¼ 1 Hz, Hm-OC6H4-p-Me), 7.04e7.12 (m,
12H, Hm/p-SiPh2), 7.57e7.79 (m, 8H, Ho-SiPh2); 13C 20.4 (p-CH3),
120.1 (Co-OC6H4-p-Me), 128.2 (Cm-Si(H)Ph2), 128.3 (Cm-OSiPh2),
129.6 (Cm-OC6H4-p-Me), 130.1 (Cp-HSiPh2), 130.3 (Cp-OSiPh2), 135.3
(Co-HSiPh2), 136.5 (Co-OSiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
in
a Schlenk flask. 2-Chloroethyl methyl ether (0.050 mL,
0.55 mmol) was added dropwise with stirring, and the resulting
mixture was stirred under N2 (open to the nujol bubbler) for 16 h,
by which time it was cloudy white. The toluene was removed under
vacuum and cold pentane (2 mL) was added, giving a white pre-
cipitate. The precipitate was collected by filtration, washed with
cold pentane (4 mL), and dried under vacuum, giving a white
d
ꢃ38.4 (SiH), -6.3 (SiAr). Anal. (calcd for 97% C31H28OSi2 (12) and
3% C24H22Si2 ((Ph2SiH)2)): C 76.93 (78.75) H 6.09 (5.97).
Formation of Ph2SiHeSi(OCH2CH2Cl)Ph2 (13) monitored by 1H
NMR. (Ph2SiH)2 (0.050 g, 0.14 mmol), CH3OCH2CH2Cl (0.013 g,
0.14 mmol), B(C6F5)3 (0.010 g, 0.020 mmol), and C6D6 (1 mL) were
combined in an NMR tube equipped with a J-Young valve. The NMR
tube was shaken and vented periodically (to release CH4 pressure)
on the vacuum line under N2. 1H NMR signal for the OCH3 group in
CH3OCH2CH2Cl disappeared after 4 h, at which point the relative
ratios of (Ph2SiH)2, 13, and (Ph2SiOCH2CH2Cl)2 (18) were 15:75:10
powdery solid (0.094 g, 65%). 1H NMR (300 MHz, C6D6)
d 3.17 (t,
3JHH ¼ 6 Hz, 4H, OCH2CH2Cl), 3.77 (t, 4H, OCH2CH2Cl), 7.14 (m, 12H,
H
m/p-SiPh2), 7.75 (m, 8H, Ho-SiPh2); 13C NMR (75 MHz, C6D6)
d 45.1
(OCH2CH2Cl), 64.8 (OCH2CH2Cl), 128.2 (Cm-SiPh2), 130.3 (Cp-SiPh2),
135.4 (Co-SiPh2). 29Si NMR (99 MHz, C6D6)
Anal. (calcd for 93% C28H28O2Cl2Si2 (18) and 7% C2H6OSi (grease)): C
63.28 (63.23) H 5.17 (5.36).
d
ꢃ7.7. M.p. 138e140 ꢀC.
respectively. For 13: 1H NMR (300 MHz, C6D6)
2H, CH2CH2Cl), 3.71 (t, 2H, CH2CH2Cl), 5.55 (s, JSiH ¼ 186 Hz, 1H,
SieH), 7.04e7.19 (overlapping m,12H, Hm/p-SiPh2 and Hm/p-OSiPh2),
7.63e7.71 (overlapping m, 8H, Ho-SiPh2 and Ho-OSiPh2).
d
3.11 (t, 3JHH ¼ 6 Hz,
(Ph2SiCH2CH2Bu)2 (19). (Ph2SiH)2 (0.30 g, 0.82 mmol) and
B(C6F3)3 (0.053 g, 0.10 mmol) were combined with 1-hexene
(1.0 mL, 8.0 mmol) and CH2Cl2 (0.1 mL) in a Schlenk. The mixture
was stirred under N2 in a closed flask for 72 h, after which the
volatiles were removed under vacuum. The resulting cloudy oily
residue was washed through a Florisil column with hexanes
(6 ꢂ 1 mL) to remove B(C6F5)3. The volatiles were removed under
vacuum, giving a clear, colorless, viscous oil (0.38 g, 87% in ~99%
purity). 1H/13C NMR spectra showed minor peaks presumed to be
due to the monosubstituted product Ph2SiHeSi(CH2CH2Bu)Ph2
(<1%). However increased reaction time (up to 3 d) and additional
catalyst (up to 0.20 mmol) did not result in this impurity being
1
(Ph2SiOC6H4-p-But)2 (16). (Ph2SiH)2 (0.10 g, 0.27 mmol), p-t-
butylphenol (0.11 g, 0.71 mmol), and B(C6F5)3 (0.028 g, 0.055 mmol)
were combined with toluene (1 mL) in a “bomb” flask. The mixture
was degassed with one freeze-pump-thaw cycle, then heated un-
der static vacuum to 70 ꢀC for 16 h, after which the flask was cooled
to RTand toluene was removed under vacuum to give a light brown,
viscous oil. A 1H NMR spectrum taken at this point shows 100%
conversion to disubstituted product (see Fig. S15a in the SI). The
mixture was washed through a Florisil column with hexanes
(6 ꢂ 1 mL) to remove B(C6F5)3 and B-OC6H4-p-But-containing
byproducts [25]. Volatiles were removed under vacuum with
heating to 60 ꢀC, to give a white oily solid. Pentane (1 mL) was
added to dissolve the crude product, then the volatiles were
removed under vacuum to give a white, powdery solid (0.038 g,
converted to 19. 1H NMR (300 MHz, C6D6)
d
0.82 (t, 3JHH ¼ 7 Hz, 6H,
CH3), 1.07e1.62 (m, 20H, CH2), 7.12e7.21 (overlapping m, 12H, Hm/p
-
SiPh2), 7.59e7.67 (m, 8H, Ho-SiPh2); 13C NMR (75 MHz, C6D6)
d 14.0
(CH3), 14.1 (SiCH2), 22.8 (Si(CH2)4CH2), 24.9 (Si(CH2)3CH2), 31.6
(Si(CH2)2CH2), 33.7 (SiCH2CH2), 128.1 (Cm-SiPh2), 129.2 (Cp-SiPh2),
136.1 (Co-SiPh2). 29Si NMR (99 MHz, C6D6)
d
ꢃ20.3. Anal. (calcd for
21%). 1H NMR (300 MHz, C6D6)
d
1.13 (s, 18H, -Ce(CH3)), 6.98e7.01
C36H46Si2): C 80.97 (80.83) H 9.04 (8.67).
(overlapping m, 8H, Ho/m-C6H4-p-But), 7.07e7.13 (overlapping m,
12H, Hm/p-Ph), 7.81 (d, 3JHH ¼ 7 Hz, 8H, Ho-Ph); 13C NMR (75 MHz,
Ph2Si(OC6H4-p-Me)-Si(OC6H4-p-But)Ph2 (22). Compound 12 was
generated in situ as described above for the attempted prep of 17,
using (Ph2SiH)2 (0.10 g, 0.27 mmol), p-methylanisole (0.040 g,
0.33 mmol), B(C6F5)3 (0.015 g, 0.029 mmol), and toluene (1 mL). The
toluene was removed under vacuum to give an oily residue, to
which p-t-butylphenol (0.44 g, 0.29 mmol), B(C6F5)3 (0.22 g,
0.043 mmol) and toluene (1 mL) were added. The flask was
degassed with one freeze-pump-thaw cycle, and the mixture was
again heated to 70 ꢀC under static vacuum. After 72 h the flask was
cooled to RT and toluene was removed under vacuum to give a light
brown, viscous oil. The oil was washed through a Florisil column
with hexanes (5 ꢂ 1 mL) to remove B(C6F5)3 and coloured impu-
rities. The hexanes was removed under vacuum, and the resulting
residue was heated to 70 ꢀC for 2 h under dynamic vacuum to
C6D6)
d
31.4 (CH3), 33.9 (eCe(CH3) from 13C NMR), 119.8 (Co-C6H4-
p-But), 126.37 (Cm-C6H4-p-But), 128.2 (Cm-SiPh2), 130.2 (Cp-SiPh2),
135.6 (Co-SiPh2). DEPT30 29Si NMR (99 MHz, C6D6)
d
ꢃ11.1. Anal.
(calcd for 81% C44H46O2Si2 (16) and 19% C2H6OSi (grease)): C 78.47
(78.50) H 6.95 (7.02).
(Ph2SiOC6H4-p-Me)2 (17). (Ph2SiH)2 (0.20 g, 0.55 mmol), p-
methylanisole (0.19 g, 0.20 mL, 1.6 mmol), and B(C6F5)3 (0.024 g,
0.047 mmol) were combined with toluene (2 mL) in a “bomb” flask.
The mixture was degassed using one freeze-pump-thaw cycle, then
stirred and heated under static vacuum to 70 ꢀC for 16 h, then
cooled to RT. Toluene was removed under vacuum to give a light
brown, viscous oil. An aliquot removed for 1H NMR showed

P.T.K. Lee, L. Rosenberg / Journal of Organometallic Chemistry 809 (2016) 86e93
93
Reddy, P. Srihari, Org. Biomol. Chem. 4 (2006) 1650; (c) A. Simonneau, M.
Oestreich, Angew. Chem. Int. Ed. 52 (2013) 11905; (d) S. Keess, A. Simonneau,
M. Oestreich, Organometallics 34 (2015) 790.
remove volatiles including residual CH3OC6H4-p-Me, HOC6H4-p-
But, and C6F5H, giving a viscous cloudy oil. (0.14 g, 84%, >99% purity,
<1% residual Ph2SiHeSi(OC6H4-p-Me)Ph2 (12)). 1H NMR: 1.13 (s, 9H,
C(CH3)3), 1.98 (s, 9H, p-CH3), 6.74 (dt, 3JHH ¼ 8 Hz, 4JHH ¼ 2 Hz, 2H,
Hm-OC6H4-p-Me), 6.94 (dt, 3JHH ¼ 8 Hz, 4JHH ¼ 2 Hz, 2H, Ho-OC6H4-
p-Me), 6.98e7.00 (overlapping m, 4H, Ho-OC6H4-p-But, Hm-OC6H4-
p-But), 7.07e7.12 (overlapping m, 12H, Hm-OC6H4-p-But, Hp-OC6H4-
p-But, Hm-OC6H4-p-Me, Hp-OC6H4-p-Me), 7.77e7.83 (overlapping
m, 8H, Ho-SiPh2OC6H4-p-But, Ho-SiPh2OC6H4-p-Me). 13C NMR 20.4
(C(CH3)3), 31.4 (p-CH3), 119.7 (Cm-OC6H4-p-But), 120.1 (Cm-OC6H4-
p-Me),126.4 (Co-C6H4-p-But), 128.2 (overlapping, Cp-SiPh2OC6H4-p-
But, Cp-SiPh2OC6H4-p-Me), 130.0 (Co-OC6H4-p-Me), 130.2 (over-
lapping, Cm-SiPh2OC6H4-p-But, Cm-SiPh2OC6H4-p-Me), 135.6 (over-
lapping, Co-SiPh2OC6H4-p-But, Co-SiPh2OC6H4-p-Me). DEPT30 29Si
(99 MHz, C6D6) ꢃ10.9 (SiOC6H4-p-Me), -11.1 (SiOC6H4-p-But). Anal.
(calcd for 99% C41H40O2Si2 (22) and 1% C31H28OSi2 (12): C 77.59
(79.30) H 6.40 (6.49)).
[10] Preliminary studies indicate that this reaction also proceeds for excess pro-
pene in trace dichloromethane, but we see only trace product (Ph2Me-
SiCH2CH2CH2Ph) from the reaction of excess allylbenzene with Ph2MeSiH, and
no reaction of this substrate with (Ph2SiH)2, either neat or with trace
dichloromethane.
[11] We have explored further the issues and opportunities presented by borane-
catalyzed over-reduction by silanes of various substrates including some S-
donor substrates (see preliminary report in reference 7). Lee P. T. K., Ph.D.
Thesis, University of Victoria, 2015.
[12] (a) D.J. Parks, W.E. Piers, J. Am. Chem. Soc. 118 (1996) 9440; (b) D.J. Parks, J.M.
Blackwell, W.E. Piers, J. Org. Chem. 65 (2000) 3090; (c) D.T. Hog, M. Oestreich,
Eur. J. Org. Chem. (2009) 5047; (d) K. Sakata, H. Fujimoto, J. Org. Chem. 78
(2013) 12505.
[13] (a) J.M. Blackwell, E.R. Sonmor, T. Scoccitti, W.E. Piers, Org. Lett. 2 (2000) 3921;
(b) M. Mewald, M. Oestreich, Chem.-Eur. J. 18 (2012) 14079. The mechanism
for hydrosilation of imines has been shown to be further complicated by
desilylation of the silyliminium intermediate, producing amine that can then
participate in a competing reduction cycle: (c) J. Hermeke, M. Mewald, M.
Oestreich, J. Am. Chem. Soc. 135 (2013) 17537.
[14] J.M. Blackwell, K.L. Foster, V.H. Beck, W.E. Piers, J. Org. Chem. 64 (1999) 4887.
[15] (a) D.Q. Zhou, Y. Kawakami, Macromolecules 38 (2005) 6902; (b) S. Shinke, T.
Tsuchimoto, Y. Kawakami, Silicon Chem. 3 (2007) 243.
Acknowledgements
[16] Harrison, D. J., M.Sc. Thesis, University of Victoria, 2005.
[17] We found evidence for the formation of the phenol-borane adduct in these
studies: 1H and 19F NMR spectra of the crude reaction mixture from the
synthesis of the disubstituted derivative 16 showed signals attributable to
products of the decomposition of this putative adduct, including C6F5H and
“B(OC6H4-p-But)”-containing species (see Supplementary data).
[18] Mixtures resulting from the attempted hydrosilation of N-benzylideneaniline
showed a downfield 1H NMR signal attributable the iminyl proton (N¼CH) of
a putative imine-borane complex at 8.57 ppm in d6-benzene and 8.59 ppm in
d8-toluene, relative to the analogous signal for the free imine, which appears
at 8.14 ppm and 8.08 ppm respectively (see Supplementary data). However,
Piers reports a high field shift of 7.89 ppm for the iminyl proton of the isolated
borane adduct in d8-toluene. J.M. Blackwell, W.E. Piers, M. Parvez,
R. McDonald, Organometallics 21 (2002) 1400.
L.R. thanks the NSERC of Canada for a Discovery Grant. P.T.K.L.
thanks the University of Victoria for a Graduate Fellowship.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2016.02.035.
References
[19] We tried using a less sterically hindered imine, PriN¼C(H)Et, to see if this
might accelerate reaction of small amounts of borane-activated disilane with
the N-nucleophile. The imine was completely consumed in this reaction, but
the disilane did not react at all. We have not yet identified the N-containing
product resulting from this reaction (see Supplementary data). Previous re-
ports of B(C6F5)3-catalyzed hydrosilation and hydrogenation reactions of
imines inevitably employ relatively bulky imine substrates, with N-aryl, -Pri,
-But, or esulfonato groups. (See, for examples: references 13a, 12c, and
P.A. Chase, T. Jurca, D.W. Stephan, Chem. Commun. (2008) 1701.) Presumably
these substrates help to avoid the catalyst poisoning and possible decompo-
sition reactions we describe here.
[1] See, for examples: C. Marschner Struct. Bond. 155 (2014) 163, and references
therein.
[2] (a) J.Y. Corey, Adv. Organometal. Chem. 51 (2004) 1. For reviews of major
routes to oligosilanes, see: (b) M.S. Hill, Struct. Bond. 136 (2010) 189; (c) A.
Feigl, A. Bockholt, J. Weis, B. Rieger, Adv. Polym. Sci. 235 (2011) 1.
[3] (a) D.J. Harrison, R. McDonald, L. Rosenberg, Organometallics 24 (2005) 1398;
(b) D.J. Harrison, D.R. Edwards, R. McDonald, L. Rosenberg, Dalton Trans.
(2008) 3401.
[4] Reviews include: (a) W.E. Piers, A.J.V. Marwitz, L.G. Mercier, Inorg. Chem. 50
(2011) 12252; (b) M. Oestreich, J. Hermeke, J. Mohr, Chem. Soc. Rev. 44 (2015)
2202.
[20] We are uncertain as to how the presence of the propylthiolato group at the
adjacent silicon in 13 might achieve an inhibitory electronic effect on borane
activation of the remaining Si-H group. It is possible that some hyper-
conjugation occurs from the S lone pair to the nascent silylium center in the
borane-silane adduct A (Scheme 2), reducing its susceptibility to nucleophilic
attack.
[5] Reference 4b provides
a comprehensive overview of hydrosilation and
dehydrocoupling reactions catalyzed by B(C6F5)3. More recent reports include
the hydrosilation of (a) nitriles (N. Gandhamsetty, J. Park, J. Jeong, S.W. Park, S.
Park, S. Chang, Angew. Chem. Int. Ed. 54 (2015) 6832) and (b) fulvenes (S.
Tamke, C.G. Daniliuc, J. Paradies, Org. Biomol. Chem. 12 (2014) 9139).
[6] Reference 4b describes B(C6F5)3-catalyzed alkane elimination from ethers,
silyl ethers, and phosphonic and phosphinic esters. This reaction of silyl ethers
(or alkoxy silanes) has had a particularly profound impact on polysiloxane
chemistry, as described in M.A. Brook, J.B. Grande, F. Ganachaud, Adv. Polym.
Sci. 235 (2011) 161.
[7] We have communicated the successful modification of poly(phenylsilane) via
borane-catalyzed hydrosilation and dehydrocoupling reactions of S-donor
substrates. P.T.K. Lee, M.K. Skjel, L. Rosenberg Organometallics 32 (2013) 1575.
[8] See, for examples: (a) V. Gevorgyan, M. Rubin, S. Benson, J.X. Liu, Y. Yama-
moto, J. Org. Chem. 65 (2000) 6179; (b) V. Gevorgyan, M. Rubin, J.X. Liu, Y.
Yamamoto, J. Org. Chem. 66 (2001) 1672; (c) G.B. Bajracharya, T. Nogami, T.
Jin, K. Matsuda, V. Gevorgyan, Y. Yamamoto, Synthesis (2004) 308; (d) J.
Chojnowski, S. Rubinsztajn, J.A. Cella, W. Fortuniak, M. Cypryk, J. Kurjata, K.
Kazmierski, Organometallics 24 (2005) 6077; (e) L.L. Adduci, T.A. Bender, J.A.
Dabrowski, M.R. Gagne, Nat. Chem. 7 (2015) 576; and references therein.
[9] In previous reports of B(C6F5)3-catalyzed hydrosilation of alkenes the use of
halogenated solvents predominates, although aromatic solvents and saturated
alkanes (hexanes) have also been used. (a) M. Rubin, T. Schwier, V. Gevorgyan,
J. Org. Chem. 67 (2002) 1936; (b) S. Chandrasekhar, G. Chandrashekar, M.S.
[21] The 1H and 13C NMR spectra of N-benzylideneaniline thus prepared closely
match spectroscopic data from the literature. P. Eisenberger, A.M. Bailey,
C.M. Crudden J. Am. Chem. Soc. 134 (2012) 17384.
[22] L. Rosenberg, C.W. Davis, J.Z. Yao, J. Am. Chem. Soc. 123 (2001) 5120.
[23] These 1H and 13C NMR chemical shifts differ slightly from those reported for
this disiloxane in d1-chloroform ((a) L. Brakha, J.Y. Becker, Electrochim. Acta
59 (2012) 135) and in benzene (1H only, (b) J. Homer, A.W. Jarvie, J.A. Holt, H.J.
Hickton, J. Chem. Soc. B (1967) 67). However the 29Si signal at ꢃ9.5 ppm is
consistent with the presence of an Si-O bond in this Ph2MeSi-containing
species (see Tables 2, 3 and (c) M.A. Brook, Silicon in Organic, Organometallic,
and Polymer Chemistry. John Wiley & Sons, Inc., New York, 2000, Ch. 1), and
we do not see a signal in the 1H NMR for a putative silanol SiOH group.
[24] Adjustment of the molar ratio of the spectroscopically-observed products
does not improve significantly the fit of the calculated values with this
microanalytical data, suggesting there are one or more impurities in this oil
that are “NMR-silent”.
[25] See Supplementary data for 1H and 19F NMR spectra obtained for an aliquot of
the crude reaction mixture, indicating the presence of these species.