Stoichiometric and catalytic Si-N bond formation using the p-block base Al(NMe2)3

Article abstract of DOI:10.1039/c5dt00662g

The aluminium amide Al(NMe₂)₃ acts as a stoichiometric or catalytic reagent in dehydrogenic Si-N bond formation using amines and silanes. Although of limited substrate scope, this represents the first p-block metal catalytic system for N-H/Si-H dehydrocoupling. The observed catalytic rate law for the formation of aminosilane products in a model study of one of the catalytic reactions suggests a mechanism involving the silane component in the deprotonation of the amine (possibly in the form of a hypervalent silicon hydride).

Full text of DOI:10.1039/c5dt00662g

Dalton
Transactions
View Article Online
View Journal
PAPER
Stoichiometric and catalytic Si–N bond formation
using the p-block base Al(NMe2)3†‡
Cite this: DOI: 10.1039/c5dt00662g
Lucy K. Allen, Raúl García-Rodríguez and Dominic S. Wright*
The aluminium amide Al(NMe2)3 acts as a stoichiometric or catalytic reagent in dehydrogenic Si–N bond
formation using amines and silanes. Although of limited substrate scope, this represents the first p-block
metal catalytic system for N–H/Si–H dehydrocoupling. The observed catalytic rate law for the formation of
aminosilane products in a model study of one of the catalytic reactions suggests a mechanism involving the
silane component in the deprotonation of the amine (possibly in the form of a hypervalent silicon hydride).
Received 13th February 2015,
Accepted 12th March 2015
DOI: 10.1039/c5dt00662g
www.rsc.org/dalton
Interest in Si–N bonded compounds has stemmed from a
1. Introduction
broad range of synthetic organic and materials applications.4,5
Catalytic bond-forming reactions using organotransition metal
complexes are a central area of modern research. Of particular
interest in recent decades has been the use of single-site tran-
sition metal catalysts in a growing range of dehydrogenic main
group element–element bond-forming reactions (eqn (1)).
Primary motivations in this area are the potential to produce
novel types of inorganic polymers and materials, as well as
possible applications in reversible hydrogen storage. Like con-
ventional organometallic catalysis, however, this new field of
‘inorganometallic’ catalysis is dominated by precious metals
(such as palladium and rhodium). For this reason, there has
most recently been a drive to assess the use of more Earth-
abundant main group metals in place of transition metals.
Although less active than transition metal catalysts, the main
group counterparts are capable of catalysing a comparable range
of element–element bond-forming reactions, most widely studied
being homonuclear P–H/P–H1 and heteronuclear B–H/N–H2
coupling. However, completely novel N–H/N–H coupling into N–
N bonds has also been observed using main group reagents, for
which there are few analogous transition metal reactions.3
Aminosilane compounds have traditionally been synthesised
via aminolysis, a process that generates HCl as a side product
(eqn (2)). Efforts to provide an alternative route to Si–N bonds
using N–H/Si–H dehydrocoupling reactions (eqn (3)) have
largely involved catalysts such as (Ph3P)RhCl,6 Cu(I)Cl,7
8
(NHC)Yb[N(SiMe3)2]2 (NHC = N-heterocyclic carbene) and
[(Et2N)U][BPh4].9 Apart from the applications of non-metal
based Frustrated Lewis Pairs (FLPs),10 main group catalysed
Si–N coupling has so far focused on s-block metal reagents.
Harder and coworkers first showed that the CaII complex
Ca(η3-Ph2CNPh)(hmpa)3 (hmpa = hexamethylphosphoramide)
can catalyse the dehydrogenic formation of Si–N bonds,11 in
an analogous way to the YbII complex reported previously.12
Other alkaline metal complexes were later shown to be active
as dehydrocoupling pre-catalysts, notably Mg(TOM) [TOM = tris-
(4,4-dimethyl-2-oxazolinyl)phenylborate)] by Sadow and co-
workers13 and M{N(SiMe3)2}2 (M = Mg, Ca, Sr) by Hill et al.14
Sadow proposed the mechanism shown in Scheme 1 for his
Mg-based catalyst, which has been assumed to be widely appli-
cable to many Si–N dehydrocoupling systems.13 The active
catalyst is thought to be a metal hydride ([M]-H), similar to
that proposed in catalytic N–H/Si–H dehydrocoupling using
[(Et2N)U][BPh4].9 Like the proposed mechanism of amine–
borane dehydrogenation,15 the next step is the protonolysis reac-
tion of the amine (R2NH) with the metal hydride catalyst. The
formation of the metal amide ([M]-NR2) enhances the nucleo-
philicity of the N-centre, leading to the formation of a hyper-
valent nucleophilic silicon intermediate via attack of the
E–H þ H–E′ ! E–E′ þ H–H
R2NH þ R′3SiCl ! R2N-SiR′3 þ HCl
R2NH þ R′3SiH ! R2N-SiR′3 þ H2
ð1Þ
ð2Þ
ð3Þ
University of Cambridge, Department of Chemistry, Lensfield Road, CB2 1EW, UK.
E-mail: dsw1000@cam.ac.uk
†Dedicated to Prof. Ken Wade, FRS.
N-atom of the amide onto R′ SiH. Subsequent β-H elimination
3
regenerates the metal hydride catalyst and gives the aminosilane
product. The overall mechanism is supported by the insight
that nucleophilic species activate Si–N bond formation.16
Despite general consensus concerning the overall details of
the mechanism of N–H/Si–H coupling using s-block catalysts,
‡Electronic supplementary information (ESI) available: Full 1-D and 2-D NMR
characterisation of compounds obtained from catalytic and stoichiometric redac-
tions involving A (see Table 1 of the text), full kinetic studies of the reaction of
Et2NH with PhSiH3, steady-state analysis of metal-hydride and hypervalent
silicon hydride mechanisms. See DOI: 10.1039/c5dt00662g
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
our studies indicate that at least two mechanisms can be fol-
lowed, a metal hydride mechanism and a silyl mechanism,
which may explain some of the major contradictions between
the previous established kinetic data.
2. Results and discussion
2.1 Activity and scope of pre-catalyst A
A range of reactions between various silanes and amines using
Al(NMe2)3 (A) as the pre-catalyst (Scheme 2) were monitored by
examining the Si–H protons of the silane precursors and amino-
silane products in the 1H NMR spectra, with the unambiguous
assignment of the products being made with the aid of 2D
NMR experiments (see ESI 1–4‡). Preliminary studies aimed to
assess the optimum catalytic loading, substrate stoichiometry
and temperature were performed to allow sufficiently fast reac-
tions to be monitored. For this purpose the dehydrocoupling
reaction of Et2NH with PhSiH3 in toluene was used as a model
system. It was found that the use of less than 10 mol% of A
and/or less than a 1 : 2 ratio of the silane to amine at room
temperature led to slow reaction and a large induction period
before H2 evolution was observed (ca. 15 min). A similar induc-
tion period has been seen for the dehydrogenation of amine–
boranes using A, and corresponds to the time needed for sig-
nificant concentration of the active catalyst to build up in the
reaction.15 Apart from gas evolution from the reaction, the for-
mation of H2 was confirmed by the presence of a singlet at
δ 4.5 ppm in a sealed Young’s tap NMR tube.17 On the basis of
these preliminary studies a minimum catalyst loading of
10 mol%, excess of amine and temperature of 70 °C were used
for most of the further studies of the substrate scope.
Table 1 shows the optimum conditions used and the pro-
ducts formed in catalytic and stoichiometric reactions investi-
gated in the current work. As shown in entry 1, pre-catalyst A is
most active for the reaction between PhSiH3 and BnNH2 (Bn =
benzyl). This reaction was the only one investigated for which
a rapid reaction occurs at room temperature, being complete
in 24 h using a 10 mol% loading of A and a 3 : 1 ratio of
BnNH2 : PhSiH3. Similarly to previously reported studies of
Group 2 catalysts, the use of a 1 : 1 precursor ratio gave an
intractable mixture of products in this case.14 Using a 3 : 1
ratio of reactants, the only organically-soluble product is the
bis-substituted aminosilane (BnNH)2SiHPh (1) (29Si NMR
δ −25.0 ppm; see also ESI‡). In addition, an unidentified white
insoluble power is also produced, which has been observed
previously in alkali metal catalysed reactions using the same
substrates.14
Scheme 1 Proposed mechanism for N–H/Si–H dehydrocoupling.
there still remains some debate. In particular, recent studies
by Hill have suggested that sigma-bond metathesis may be a
key component of the catalytic cycle.14 In addition, the deter-
mination of the rate laws governing these reactions has so far
provided only modest support for the mechanism shown in
Scheme 1. Hill has shown that the rate of consumption of the
amine using M{N(SiMe3)2}2 (M = Mg, Ca) is first order in
amine and catalyst, but zero order in silane (eqn (4)).14 This
rate law is entirely consistent with a mechanism involving
deprotonation of the amine by an intermediate hydride as a
key step (as illustrated in Scheme 1). However, an apparent
change in the mechanism is suggested by the observation of a
different rate law for the consumption of the amine for M = Sr,
which is dependent on the silane and amine. Harder has
shown that the overall rate of formation of the aminosilane
products using his catalyst is zero order in amine and first
order in catalyst and silane (eqn (5)).11 The observation that
the same rate law governs the rate of depletion of silane in
Sadow’s system led to the important conclusion that the turn-
over-limiting step involves the interaction of the catalyst with
the silane and that the amine itself is not involved in the tran-
sition state.13
ꢀd½amineꢁ
¼ k½catꢁ½amineꢁ
ð4Þ
dt
þd½aminosilaneꢁ
dt
¼ k½catꢁ½silaneꢁ
ð5Þ
In this paper we report the first p-block metal catalysed
N–H/Si–H dehydrocoupling reactions using the pre-catalyst
Al(NMe2)3 (A), and compare its substrate scope and activity to
the previously reported alkaline earth metal systems. Detailed
exploration of the reaction kinetics allows us to suggest a new
mechanism for this type of dehydrocoupling reaction, impli-
cating a hypervalent silicon hydride in the key protonolysis
step. Taken together with previously reported investigations, Scheme 2 General silicon–nitrogen dehydrocoupling scheme.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Table 1 Al(NMe2)3 catalysed silicon–nitrogen dehydrocoupling
Entry
Silane
Amine
Catalyst loading
Products
Conversion
a
c
1
2
3
4
5
6
7
8
9
PhSiH3
BnNH2
Et2NHb
iPrNH2
10 mol%
10 mol%
30 mol%
30 mol%
(BnNH)2SiHPh
94%
93%
85%
33%
PhSiH2NEt2 + PhSiH(NMe2)(NEt2)
PhSiH(iPrNH)2 + PhSiH(NMe2)(iPrNH) + PhSiH2(NMe2) (trace)
tBuNH2
PhNH2
iPr2NH
HMDS
Et2NH
Et2NH
PhSiH(tBuNH)(NMe2) + PhSiH(NMe2)2 + PhSiH2(NMe2)
c
No reaction
No reaction
No reaction
Ph2SiH(NMe2)
No reaction
Ph2SiH2
tBu2SiH2
100 mol%
10 mol%
Stoichiometric
a Using 1 : 3 molar ratio of silane to amine at 25 °C. b Using 1 : 2 molar ratio of silane to amine at 70 °C. c Using 1 : 1.5 molar ratio of silane to
amine at 70 °C.
The catalytic reaction of PhSiH3 and Et2NH proceeds slowly
at room temperature using a 10 mol% loading of A and a 1 : 2
ratio of silane to amine (ca. 7 days). However, at 70 °C it is
complete in 24 h (entry 2). In contrast to the reaction involving
BnNH2, two products PhSiH2(NEt2) (2a) (70%) and PhSiH-
(NMe2)(NEt2) (2b) (23%), are formed. The first (2a) is the
expected product of 1 : 1 N–H/Si–H coupling, while the un-
expected product 2b can be traced to the initial inductive
phase of the reaction in which pre-catalyst A reacts with
PhSiH3 to form a metal hydride intermediate and the by-
product PhSiH2(NMe2) (which then couples with Et2NH,
Scheme 3). Support for this come from NMR spectroscopic
study of a 1 : 1 mixture of PhSiH3 with A in d8-toluene which,
after immediate evolution of H2 gas, gives a mixture of the
Fig. 1 (a) 1H NMR spectra of the reaction between PhSiH3 (1 equiv.) and
Et2NH (2 equiv.) catalysed by A (10 mol%), showing the formation of
mono- and di-substituted products PhSiH2(NMe2) and PhSiH-
(NMe2)2 (ESI 5a‡). This reaction also leads to the replacement
of the resonance for A (δ = 110 ppm, br) in the room-tempera-
ture 27Al NMR spectrum by a new species (δ 130 ppm, br) (ESI
5b‡), which is also present throughout in situ studies of the
catalytic reactions of Et2NH with PhSiH3.
PhSiH2NMe2 (2a) PhSiH(NMe2)(NEt2) (2b), (b) 29Si–1H HSQC short-range
spectrum correlating the diagnostic 1H signal for the Si–H proton at δ =
5.04 ppm with the 29Si resonance at δ = −18 ppm in PhSiH(NMe2)(NEt2).
(2b), (c) proton coupled 29Si NMR spectrum showing doublet of the
product PhSiH(NMe2)(NEt2) (2b), and (d) 29Si–1H HSQC long-range spec-
trum correlating the silicon resonance at δ = −18 ppm with the 1H reso-
nances at δ = 2.56 (s, NMe2) and 2.90 ppm (q, NEt2) in PhSiH(NMe2)-
(NEt2) (2b). All spectra were acquired at ambient temperature and in d8-
toluene.
Fig. 1 summarizes selected NMR spectroscopic data on the
product mixture of 2a and 2b produced in the catalytic reac-
tion (Table 1, entry 2). PhSiH(NMe2)(NEt2) (2b) was identified
through a series of NMR experiments. The short-range (1JSi–H
=
202 Hz) 29Si–1H heteronuclear single quantum coherence
(HSQC) spectrum (Fig. 1b) shows a cross-peak correlating the
Me-groups of Me2N (δ 2.56 ppm) and the 1H-quartet for the
two CH2-groups of Et2N (δ 2.90 ppm) (Fig. 1d).
1
doublet 29Si resonance at δ −18 ppm (Fig. 1c) with a the H-
singlet for the Si–H proton at δ 5.04 ppm. Additional cross
peaks were observed in the long-range 29Si–1H HSQC spectrum
between the 29Si resonance and the 1H-singlet for the
Similar products to 2b, resulting from catalyst-activation
and containing Me2N-groups, are also formed in the reactions
i
t
of PhSiH3 with PrNH2 and BuNH2 (Table 1, entries 3 and 4).
The proportion of these by-products increases with increased
i
steric bulk of the amine substituent so that for PrNH2 the
final product mixture after 24 h at 70 °C using a 30 mol%
loading of A is composed of PhSiH(iPrNH)2 (30%), PhSiH-
(NMe2)(iPrNH) (54%) and a trace amount of PhSiH2(NMe2).
t
For BuNH2, all of the products contain Me2N-substitution. In
experiments investigating the effects of changing the catalyst
loading on product distribution in the reaction of PhSiH3 with
Et2NH it was found that the ratio of the product (2a) to the
by-product (2b) increases markedly with increased catalyst
Scheme 3 Formation of the unexpected product 2b.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
conditions in the coupling of Et2NH and the secondary silane
Ph2SiH2 (entry 9). It is noteworthy that increasing the catalyst
loading to 100 mol% leads entirely to stoichiometric reaction
of A with Ph2SiH2 (entry 8), via the same mechanism as that
discussed above in Scheme 4.
Direct comparison of the activity and substrate scope of
pre-catalyst A with previously reported Group 2 N–H/Si–H
dehydrocoupling systems is difficult because only limited turn-
over frequency data is available and the fact that only a few of
the coupling reactions reported are directly comparable with
the current study. In respect to rate of conversion, pre-catalyst
A is clearly less active than Sadows catalyst Mg(TOM), which
will couple primary amines and silanes at room temperature
using a 5 mol% loading in 24 h. It can be noted, however, that
only the coupling of primary amines and silanes was explored
in this previous case. Although the substrate scope of Hill’s
catalytic system involving M{N(SiMe3)2}2 (M = Mg, Ca, Sr) is
greater than that of A, being able to couple a broader range of
secondary amines and silanes, the time taken to obtain
product conversion is difficult to compare with the previously
reported study.14 Our overall conclusions are that, although
A is competent at coupling less sterically demanding amines
and silanes, its use in catalysis is limited by side reactions.
Nonetheless, this is the first p-block metal based prototype
and there is clearly scope for catalyst development in this area,
particularly in respect to variation of the steric bulk of the
amide ligands present in the pre-catalyst.
Fig. 2 Kinetic plot showing the product distribution of 2a and 2b
from the reaction of PhSiH3 and Et2NH varying the loading of the pre-
catalyst A.
loading (Fig. 2). It is worth noting that no such products
arising from ligand-redistribution of the pre-catalyst were
observed in the previously reported studies of Group 2
systems.11,13,14 The absence of such by-products presumably
reflects the nature and steric demands of the ligands present
in the pre-catalysts. For example, the activation stage of M{N-
(SiMe3)2}2 (M = Mg, Ca, Sr)14 would give the bulky amine
(Me3Si)2NH (see Scheme 4) which is less reactive than Me2NH.
Supporting this view, we have found that the reaction of Et2NH
with PhSiH3 with (iPr2N)3Al in place of Al(NMe2)3 (A) gives 2a
as the exclusive product (with none of the analogous com-
pound to 2b being observed).
Looking at the remaining reactions involving PhSiH3 and
Ph2SiH2 (Table 1, entries 5–9), it is clear that the activity of
pre-catalyst A is severely limited by the nature of the amine
and, in particular, the steric demands of the amine and silane
precursors. The inability of A to couple PhNH2 with PhSiH3
(entry 5) may be due to the acidity of the higher amine,
causing catalyst decomposition or deactivation. Harder and
Sadow also report a lack of or reduced activity using aniline
derivatives and Group 2 catalysts.11,13 This, however, contrasts
with work by Hill who found that acidic aniline compounds
increased the reactivity of the catalyst.12 The strong steric influ-
ence on reactivity is seen in the inability of A to couple more
bulky secondary amines with PhSiH3 at 70 °C using a 10 mol%
loading (entries 6 and 7) and the lack of activity under these
2.2 Kinetic studies of the reaction of Et2NH with PhSiH3
using pre-catalyst A
We next turned to a detailed investigation of the mechanism
of the dehydrocoupling reaction involving A, noting the rate
laws previously determined for N–H/Si–H dehydrocoupling
using Group 2 reagents.11,13,14 For this purpose, we focused on
the coupling reactions of Et2NH with PhSiH3 using A. In pre-
liminary studies we showed that A does not react significantly
with Et2NH at 25–70 °C in toluene, suggesting that this simple
reaction is part of the catalytic cycle or catalyst activation.
The rates of formation of the combined aminosilane pro-
ducts 2a and 2b in toluene at 80 °C were monitored as a func-
tion of catalyst loading (Fig. 3), silane concentration (Fig. 4)
Scheme 4 Outline of a mechanism involving a hypervalent silicon
hydride as the protonolysis source.
Fig. 3 Kinetic plots for the reaction of PhSiH3 and Et2NH (1 : 2 equiv-
alents) at 80 °C in toluene varying concentrations of the pre-catalyst A.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 4 Kinetic plot for the reaction of PhSiH3 and Et2NH using 10 mol%
of A in toluene varying the concentration of PhSiH3.
1
and amine concentration (Fig. 5), using the H NMR spectro-
scopic Si–H resonances. It is clear from inspection of the raw
data alone that the reaction rate is independent of the concen-
tration of amine, within experimental error (see the duplicate
data for the 3 : 1 molar ratio reaction of Et2NH with PhSiH3 in
Fig. 5).
Fig. 6a and 6b show plots of the initial rates verses pre-cata-
lyst loading and silane concentration, derived from the data
presented in Fig. 3 and 4, respectively. To a very good approxi-
mation the initial rate at a given concentration of pre-catalyst
or silane is equal to the rate of formation of the mono-substi-
tuted product 2a, since the rate of formation of di-substituted
2b is negligible at the beginning of the reaction (as it requires
2a as a precursor). It can be seen from this analysis that the
rate law for the formation of the aminosilane products is the
same as that previously obtained by Harder (eqn (5)),
being first order in silane and pre-catalyst but zero order in
amine. Further analysis also reveals that the rate law for the
consumption of silane is first order in silane and catalyst but
zero order in amine (the same rate law as determined by
Sadow13) (ESI 6‡).
Fig. 6 (a) Initial rates of reaction vs. pre-catalyst loading for the 1 : 2
reaction of PhSiH3 with Et2NH, and (b) initial rates vs. silane concen-
tration with 10 mol% loading of pre-catalyst. The straight lines drawn are
the best-fit ones.
assumed to be the rate-determining step (ESI 7b‡), as pro-
posed by Sadow.13,18
þd½aminosilaneꢁ
¼ k½catꢁ½amineꢁ
ð6Þ
dt
The observed rate law therefore does not appear to be fully
consistent with the mechanism shown in Scheme 1. However,
one possibility is that the silane is intimately involved in the
protonolysis step. Such a mechanism is outlined in Scheme 4
in which a hypervalent silicon hydride is involved. Steady-state
analysis (see ESI 8‡) suggests that the participation of such a
species as the protonolysis source would result in eqn (5) as
the rate law (the experimentally observed rate law for the reac-
tion investigated in the current work using pre-catalyst A).
Significantly, this proposed mechanism is also consistent with
the previous conclusion that the turnover-limiting step
involves the interaction of the catalyst ([M]-H) with the
silane.13
If the aluminium hydride was the protonolysis source in
the mechanism (as in Scheme 1) then the rate law would be
eqn (6) (see steady-state analysis ESI 7a‡). The same rate law
would also operate even if the attack of the N-atom of the
metal amide ([M]-NR2) onto the silane (R′ SiH) (Scheme 1) is
3
The rate of reaction of intermediate metal hydrides with
amines is likely to be a key factor influencing which mechan-
ism dominates [i.e., metal-hydride (Scheme 1) vs. hypervalent
silicon hydride (Scheme 4)]. This is potentially the cause of the
change in mechanism suggested previously by the kinetics of
N–H/Si–H coupling moving from M{N(SiMe3)2}2 (M = Mg, Ca)
(eqn (4) being the rate equation) to Sr{N(SiMe3)2}2 (having a
rate equation that is partially dependent on the silane concen-
tration).14 It is well known that the reaction rates of aluminium
hydrides with amines can be slow,19 potentially making
Fig. 5 Kinetic plot showing reaction of PhSiH3 and Et2NH using 10 mol%
of A in toluene varying the concentration of amine.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
H-transfer to silicon more viable than deprotonation of the
amine. It can be noted also that the hypervalent silicon
hydride anion [PhSiH4]− has been implicated as an intermedi-
ate in a previous study of the hydrosilylation of alkenes using
alkaline earth metal catalysts.20 Also highly relevant to the
current study, the switch between a metal hydride and hyper-
valent silicon hydride mechanism in this previous case is
apparently highly dependent on metal–H bond polarity.20
Fig. 7 Comparison of overall rate of formation of aminosilanes 2a and
2b in the reaction of Et2NH with PhSiH3 using 10 mol% A at 70 °C in
toluene, vented (to allow the H2 generated to escape) and unvented
(in a sealed NMR tube).
3. Conclusions
We have shown for the first time that a p-block metal based
pre-catalyst [Al(NMe2)3 (A)] can be employed in N–H/Si–H
dehydrocoupling reactions. Clearly, however, the substrate
scope of A is limited generally to the least sterically demanding
primary amines and silanes. One obvious problem with the
use of A is the production of by-products resulting from Me2N-
group transfer to the silane. Reduction in coupling rate and
increase in the proportion of these by-products occur with
more sterically-encumbered amines and silanes. The obser-
vation that Me2N-group transfer can be suppressed by increas-
ing the steric bulk of the amido group present in the pre-
catalyst [as seen for Al(NiPr2)3] may be important in the devel-
opment of more selective reagents in this area.
Our measurements of the reaction kinetics of the coupling
reaction of Et2NH with PhSiH3 using A are consistent with
Harder’s previous report using Ca(η3-Ph2CNPh)(hmpa)3. From
these measurements we have suggested that the dependence
of the rate equation on the silane concentration may be due to
the intimate involvement of the silane in the deprotonation of
the amine (potentially in the form of a hypervalent silicon
hydride).
results show that the formation of H2 in a sealed NMR tube
has no effect on the rate (Fig. 7). Therefore all further studies
were undertaken in sealed NMR tubes.
The following data refer to the entries given in Table 1 of
the text;
Entry 1 – PhSiH3 (0.07 ml, 0.60 mmol) followed by BnNH2
(0.21 ml, 1.80 mmol) was added to an NMR tube containing Al
(NMe2)3 (10 mg, 0.06 mmol) in d8-toluene (0.7 ml). Gas evol-
ution was observed immediately. After 24 hours an insoluble
white precipitate was formed and (BnNH)2SiHPh (1) was
observed by NMR spectroscopy. 1H NMR (+25 °C, 500 MHz),
δ/ppm = 1.24 (s, 2H NH), 3.96 (m, 4H CH2), 5.22 (s, 1H, SiH),
7.06–7.24 (m, 10H, Bn), 7.26 (m, 3H, Ph), 7.67 (m, 2H, Ph). 29Si
NMR (+25 °C, 99 MHz), δ/ppm = −25.0 (d).
Entry 2 – PhSiH3 (0.07 ml, 0.60 mmol) followed by Et2NH
(0.10 ml, 1.20 mmol) was added to a solution of Al(NMe2)3
(10 mg, 0.06 mmol) in d8-toluene (0.7 ml). The NMR tube was
heated to 70 °C and hydrogen evolution was observed. After
24 hours two products PhSiH2NEt2 (2a) and PhSiH(NMe2)-
(NEt2) (2b) are observed in 70% and 23% yield, respectively
(the remainder was unreacted PhSiH3). 2a: 1H NMR (+25 °C,
500 MHz), δ/ppm = 0.99 (t, 6H NCH2CH3), 2.84 (q, 4H
NCH2CH3), 5.08 (s, 2H, SiH2), 7.21 (m 3H, Ph), 7.59 (m, 2H,
Ph). 29Si NMR (+25 °C, 99 MHz), δ/ppm = −25.0 (t). 2b:
1H NMR (500 MHz), δ/ppm = 0.99 (t, 6H, NCH2CH3), 2.56 (s,
6H, NCH3) 2.90 (q, 4H NCH2CH3), 5.04 (s, 1H, SiH), 7.24
(m, 3H, Ph), 7.62 (m, 2H, Ph). 29Si NMR (25 °C, 99 MHz)
δ/ppm = −18.0 (d).
4. Experimental section
All chemicals (the silanes PhSiH3 and Ph2SiH2, and the
amines BnNH2, Et2NH, PhNH2, tBuNH2, iPrNH2, iPr2NH,
HMDS) were acquired from Aldrich. Amines were dried by dis-
tillation over CaH2 and stored under dry, O-free N2. The pre-
catalyst Al(NMe2)3 (A) was prepared by the literature method
from AlCl3 and LiNMe2.21 D8-toluene was dried using a
sodium mirror. Most NMR-scale reactions were performed in
Young’s tap thin-walled 528PP NMR tubes using a Bruker
Avance B8–500 MHz spectrometer. For safety, however, kinetic
experiments investigating the variation of Et2NH concentration
in its reaction with PhSiH3 were undertaken in medium-walled
Young’s tap NMR tubes (i.e., a significant pressure is develo-
ped at 80 °C, especially using a large excess of the volatile
amine). In order to test that the rate of reactions were not
affected if undertaken in a sealed system, a background experi-
ment was initially performed following the rate of formation of
the aminosilane products (2a and 2b) of the 2 : 1 reaction of
Et2NH with PhSiH3 at 80 °C in toluene, with and without
venting of the H2 formed as the reaction proceeded. The
i
Entry 3 – PhSiH3 (0.07 ml, 0.60 mmol) followed by PrNH2
(0.08 ml, 0.90 mmol) was added to an NMR tube containing
toluene-d8 (0.7 ml). The tube was charged with Al(NMe2)3
(30 mg, 0.18 mmol) and gas evolution was observed immedi-
ately. After 24 hours PhSiH(iPrNH)2 (3a) and PhSiH(iPrNH)-
(NMe2) (3b) were observed in 31% and 54% conversion,
respectively (the remainder is unreacted PhSiH3).
3a: 1H NMR (+25 °C, 500 MHz), δ/ppm = 0.66 (d, J = 7.5 Hz,
2H, NH, br) 0.99 (dd, J = 6.4 Hz, 1.7 Hz, 12H, (CH3)2CH,
3.14–3.24 (m, 2H (CH3)2CH), 5.10 (t, 2.1 Hz, 1H, SiH),
7.18–7.21 (m, Ph) 7.63–7.65 (m, Ph). 29Si NMR (+25 °C,
99 MHz), δ/ppm = −31.0. 3b: 1H NMR (+25 °C, 500 MHz),
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
δ/ppm = 0.72 (d, J = 6 Hz 2H, NH, br), 1.02 (d J = 6.4 Hz 6H
(CH3)2CH), 3.08–3.12 (m, 1H (CH3)2CH), 5.04 (d J = 2.2 Hz, 1H,
SiH), 7.18–7.21 (m, Ph) 7.55–7.56 (m, Ph). 29Si NMR (+25 °C,
99 MHz), δ/ppm = −24.0.
and R. Waterman, J. Organomet. Chem., 2014, 751, 541–545;
(e) R. J. Less, H. R. Simmonds and D. S. Wright, Dalton
Trans., 2014, 43, 5785–5792.
3 R. J. Less, V. Naseri, M. McPartlin and D. S. Wright, Chem.
Commun., 2011, 47, 6129–6131; T. W. Myers and
L. A. Berben, J. Am. Chem. Soc., 2013, 135, 9988–9990.
4 E. Kroke, Y.-L. Li, C. Konetschny, E. Lecomte, C. Fasel and
R. Riedel, Mat. Sci. Eng: Reports, 2000, 26, 97–199.
5 C. D. F. Konigs, M. F. Muller, N. Aiguabella, H. F. T. Klare
and M. Oestreich, Chem. Commun., 2013, 49, 1506–1508.
6 H. Kono, I. Ojima, M. Matsumoto and Y. Nagai, Org. Prep.
Proced. Int., 1973, 5, 135–139.
7 H. Q. Liu and J. F. Harrod, Can. J. Chem., 1992, 70, 107–
110.
8 W. Xie, H. Hu and C. Cui, Angew. Chem., Int. Ed., 2012, 124,
11303–11306.
t
Entry 4 – PhSiH3 (0.07 ml, 0.60 mmol) followed by BuNH2
(0.10 ml, 0.90 mmol) was added to an NMR tube containing
d8-toluene (0.7 ml). The tube was charged with Al(NMe2)3
(30 mg, 0.18 mmol) and gas evolution was observed immedi-
ately. After 24 hours an PhSiH(tBuNH)(NMe2) (4a), PhSiH-
(NMe2)2 (4b) and PhSiH2(NMe2) (4c) were observed in 17%,
11% and 5% conversion, respectively (the remainder is
unreacted PhSiH3).
1
4a: H NMR (+25 °C 500 MHz), δ/ppm = 1.03 (NH), 1.15 (s,
9H, NtBu) 2.55 (s, 6H, NMe2), 5.13 (d, J = 3 Hz, 1H, SiH),
7.16–7.20 (m, Ph), 7.50–7.58 (m, Ph). 29Si NMR (+25 °C,
99 MHz), δ/ppm = −28.0. 4b: 1H NMR (+25 °C 500 MHz),
δ/ppm = 2.49, (s, 12H, NMe2), 4.97, (s, 1H, SiH2), 7.16–7.20 (m,
Ph) 7.50–7.58 (m,Ph). 29Si NMR (+25 °C, 99 MHz) δ/ppm = −17.
9 J. X. Wang, A. K. Dash, J. C. Berthet, M. Ephritikhine and
M. S. Eisen, J. Organomet. Chem., 2000, 610, 49–57.
4c: 1H NMR (+25 °C, 500 MHz), δ/ppm = 2.43 (s, 6H, NMe2) 10 For example, (a) M. Pérez, C. B. Caputo, R. Dobrovetsky
4.99 (s, 2H, SiH2) 7.16–7.20 (m, Ph), 7.50–7.58 (m,Ph). 29Si
NMR (+25 °C, 99 MHz), δ/ppm = −22.0.
and D. W. Stephan, Proc. Natl. Acad. Sci. U. S. A., 2014, 111,
10917–10921; (b) L. Greb, S. Tamke and J. Paradies, Chem.
Commun., 2014, 50, 2318–2320.
Entry 8 – Ph2SiH2 (0.06 ml, 0.30 mmol) followed by Et2NH
(0.10 ml, 1.20 mmol) were added to a solution of Al(NMe2)3 11 F. Buch and S. Harder, Organometallics, 2007, 26, 5132–
(5 mg, 0.03 mmol) in d8-toluene (0.7 ml). Gas evolution was 5135.
observed upon addition and the NMR tube was heated to 12 Y. Makioka, Y. Tangiguchi, Y. Fugiwara, K. Takaki, Z. Hou
70 °C. Full conversion to Ph2SiHNMe2 (8) was observed. 1H
and Y. Wakatsuki, Organometallics, 1996, 15, 5476–5478.
NMR (+25 °C 500 MHz), δ/ppm = 2.54 (s, 6H) 5.44 (s, 1H) 7.21 13 J. F. Dunne, S. R. Neal, J. Engelkemier, A. Ellern and
(m, 6H) 7.58 (m, 4H). 29Si NMR (+25 °C, 99 MHz) δ/ppm =
−12.0.
A. D. Sadow, J. Am. Chem. Soc., 2011, 133, 16782–16785.
14 M. S. Hill, D. J. Liptrot, D. J. MacDougall, M. F. Mahon and
T. P. Robinson, Chem. Sci., 2013, 4, 4212–4222.
15 R. J. Less, R. L. Melen and D. S. Wright, RSC Adv., 2012, 2,
2191–2199.
Acknowledgements
16 R. J. P. Corriu, D. Leclercq, P. H. Mutin, J. M. Planeix and
A. Vioux, J. Organomet. Chem., 1991, 406, C1–C4.
We thank The EU (ERC Advanced Investigator Grant for D.S.
W., studentship for L.K.A.) and The EU (Marie Curie Intra 17 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb,
European Fellowship for R.G.-R).
A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg,
Organometallics, 2010, 29, 2176–2179.
18 Unfortunately, it was not possible to determine the
equation for the rate of loss of amine with respect to
amine, silane and catalyst concentration because the
Et-resonances for Et2NH overlap with those of the amino-
silane products and the in situ resonances for the catalyst.
Notes and references
1 (a) V. Naseri, R. J. Less, R. E. Mulvey, M. McPartlin and
D. S. Wright, Chem. Commun., 2010, 46, 5000–5002;
(b) K. A. Erickson, L. S. H. Dixon, D. S. Wright and 19 A. E. Nako, S. J. Gates, N. Schadel, A. J. P. White and
R. Waterman, Inorg. Chim. Acta, 2014, 422, 141–145.
M. Crimmin, Chem. Commun., 2014, 50, 9536–9538, and
2 (a) A. J. M. Miller and J. E. Bercaw, Chem. Commun., 2010,
references therein.
46, 1709–1711; (b) D. J. Liptrot, M. S. Hill, M. F. Mahon and 20 F. Buch, J. Brettar and S. Harder, Angew. Chem., Int. Ed.,
D. J. MacDougall, Chem. – Eur. J., 2010, 16, 8508–8515;
2006, 45, 2741–2745.
(c) M. M. Hansmann, R. L. Melen and D. S. Wright, Chem. 21 K. M. Waggoner, M. M. Olmstead and P. P. Power, Poly-
Sci., 2011, 2, 1554–1559; (d) K. A. Erickson, D. S. Wright
hedron, 1990, 9, 257–263.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.