Hetero-dehydrocoupling of silanes and amines by heavier alkaline earth catalysis

Article abstract of DOI:10.1039/c3sc51797g

The homoleptic alkaline earth hexamethyldisilazides, [M{N(SiMe ₃)₂}₂]₂ (1: M = Mg; 2: M = Ca; 3: M = Sr), have been demonstrated as active pre-catalysts for the cross-dehydrocoupling of Si-H and N-H bonds under mild (25-60 °C) conditions. The reactions are applicable to the coupling of a wide variety of amine and silane substrates and are proposed to occur via a sequence of discrete Si-H/M-N and N-H/M-H metathesis steps. Whereas reactions of dialkyl group 2 species with 2,6-di-iso-propylaniline and phenylsilane delivered a series of well-defined compounds consistent with this rationale, kinetic analysis of the cross-coupling of diethylamine with diphenylsilane provided evidence for a more complex and subtly variable mechanistic landscape. Although reactions performed with all three pre-catalysts presented a number of common features, in every case the calcium species, 2, was found to provide notably superior catalytic activity, an order of magnitude higher than both 1 and 3 and in excess of many previously described benchmark transition metal- or f-element-mediated processes. Variations in overall reaction order, mode of pre-catalyst activation and the nature of the rate determining process are postulated to arise as a consequence of the marked change in M²⁺ radius and resultant charge density as group 2 is descended. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3sc51797g

Chemical Science
View Article Online
View Journal | View Issue
EDGE ARTICLE
Hetero-dehydrocoupling of silanes and amines by
heavier alkaline earth catalysis†
Cite this: Chem. Sci., 2013, 4, 4212
*
Michael S. Hill, David J. Liptrot, Dugald J. MacDougall, Mary F. Mahon
and Thomas P. Robinson
The homoleptic alkaline earth hexamethyldisilazides, [M{N(SiMe3)2}2]2 (1: M ¼ Mg; 2: M ¼ Ca; 3: M ¼ Sr), have
been demonstrated as active pre-catalysts for the cross-dehydrocoupling of Si–H and N–H bonds under mild
(25–60 ꢀC) conditions. The reactions are applicable to the coupling of a wide variety of amine and silane
substrates and are proposed to occur via a sequence of discrete Si–H/M–N and N–H/M–H metathesis steps.
Whereas reactions of dialkyl group 2 species with 2,6-di-iso-propylaniline and phenylsilane delivered a
series of well-defined compounds consistent with this rationale, kinetic analysis of the cross-coupling of
diethylamine with diphenylsilane provided evidence for a more complex and subtly variable mechanistic
landscape. Although reactions performed with all three pre-catalysts presented a number of common
features, in every case the calcium species, 2, was found to provide notably superior catalytic activity, an
order of magnitude higher than both 1 and 3 and in excess of many previously described benchmark
transition metal- or f-element-mediated processes. Variations in overall reaction order, mode of pre-catalyst
activation and the nature of the rate determining process are postulated to arise as a consequence of the
marked change in M2+ radius and resultant charge density as group 2 is descended.
Received 27th June 2013
Accepted 13th August 2013
DOI: 10.1039/c3sc51797g
www.rsc.org/chemicalscience
As archetypal main group compounds containing E–H
bonds, hydrosilanes display diverse reactivity and have found
Introduction
The mediation of main group element–element (E–E) bond
formation, particularly through the homo- and hetero-dehy-
drocoupling of E–H (or E0–H) bonds, is an enterprise that has
commanded attention for over three decades.1 As early as 1972
Keller demonstrated that more complex amidoboranes could be
prepared by a combination of R2N– for H exchange and H2
elimination from the interaction of either lithium primary
amides and B2H6 or alkali metal hydrides and amine boranes.2
Driven by potential hydrogen storage applications, the dehy-
drocoupling of these latter species and ammonia borane has
achieved recent prominence,3 while recent advances describing
a plethora of Si–Si,4 P–P1c,d,5 and B–B6 bond forming reactions
have signposted that a greatly enhanced generality for such
bond construction processes may be within reach. Although an
aggregate of these reports might suggest that the mediation of
these processes is primarily the preserve of mid or late metals
from the transition series, there is a growing awareness that
main group elements selected from the s- and p-blocks of the
periodic table may also provide comparable levels of small
molecule reactivity.7
application in a wide range of molecular and macromolecular
homogeneous catalytic processes.8 The hydrosilylation of C]X
multiple bonds (X ¼ C, N, O), for example, may be catalysed by a
diverse array of transition and main group element species in
which the mode of activation of the Si–H bond is determined by
the nature of the catalytic metal centre.9 Whereas oxidative
addition pathways have been found to predominate for activa-
tion of the Si–H function by variable oxidation state transition
metal compounds,10 reactivity derived by the intermediacy of
more electropositive d0 main group or lanthanide species
generally proceeds via polarized s-bond metathesis pathways to
provide metal hydrides together with a silicon-containing by-
product.11
This reactivity may be elaborated to a catalytic regime
through the introduction of a protic coupling partner E0–H,
resulting in a net dehydrocoupling reaction and the formation
of an E0–Si bond. In cases where E0–H represents an amine
function (Scheme 1) this reactivity will expedite the formation of
silylamines, valuable molecules which may be utilised as sily-
lating agents,12 bases13 and ligands as well as polymer and
ceramic precursors.14 The synthesis of Si–N bonds is most
commonly achieved through the aminolysis of chlorosilanes.15
Consequently the dehydrogenative coupling of silanes and
amines, obviating the production of HCl, has attracted signi-
cant attention. Heterogeneous catalysis has been achieved with
several supported palladium systems16 while a range of
Department of Chemistry, University of Bath, Claverton Down, Bath, BA2 7AY, UK.
E-mail: msh27@bath.ac.uk
† Electronic supplementary information (ESI) available: Experimental procedures
and characterisation data for all new compounds. CCDC 945785–945789. For ESI
and crystallographic data in CIF or other electronic format see DOI:
10.1039/c3sc51797g
4212 | Chem. Sci., 2013, 4, 4212–4222
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
Scheme 1
Scheme 2 Proposed mechanism for Si–N bond formation mediated by III.
titanium(IV),17 chromium(III),18 copper(I),19 rhodium(I),20 ruthe- a self-consistent series of readily available heavier alkaline
nium(0),21 ytterbium(II),22 samarium(II)22 and uranium(IV)23 silylamide derivatives, [M{N(SiMe3)2}2]2 (M ¼ Mg (1), Ca (2) and
species have been identied for the homogeneous catalysis of Sr (3)).29 Through a combination of stoichiometric and catalytic
this process. Although complexes of the heavier alkaline earth studies we also provide a provisional mechanistic rationale for
elements have an established pedigree as reagents for multiple variations observed in activity and selectivity of these processes
bond heterofunctionalisation24 and have recently been applied in relation to alkaline earth cation size, charge density and
quite broadly in amine–borane25 and to a lesser extent group 15 polarisation effects.
dehydrocoupling catalysis,26 reports of group 2-mediated
dehydrogenative silylamine synthesis are rather more limited.
The rst report of such an alkaline earth-promoted catalysis was
Results and discussion
Catalytic scope
provided by Harder and co-workers who,27 inspired by the
ytterbium ketamine derivative (I),22a,b utilised the isostructural
calcium species (II) along with a dibenzyl calcium complex for
the coupling of a variety of primary or secondary amines with
Ph3SiH or Ph(Me)SiH2. In a similar manner Sadow has more
An initial NMR investigation was undertaken into potential
substrate scope for reactions catalysed by compounds 1–3
between the range of organosilanes and organic amines shown
in Table 1. On addition of the pre-catalyst to a mixture of the
recently
demonstrated
that
the
tris(oxazolinyl)bor-
amine and silane in C6D6 a pronounced bubbling was observed
for many of the reactions. In general, hydrogen evolution
occurred on mixing at room temperature, although some reac-
tions required heating to 60 ꢀC for the observation of appre-
ciable product formation. While the available literature data
and conditions applied in previous studies are somewhat
disparate,16–23,27,28 the requisite conditions for this group 2-cat-
alysed dehydrogenative coupling are, at worst, commensurate
with the best of the previously reported catalytic systems, for
example Cui's recently described NHC-supported Yb(II) pre-
catalyst,22c and are considerably milder than many of the late
transition metal-based processes.
atomagnesium complex (III) may also act as a potent pre-cata-
lyst for the dehydrocoupling of silane and amine (including
ammonia and hydrazine) reaction partners.28 The calcium-
centred reactivity was proposed to occur through a sequential
metathesis-based mechanism analogous to the process depic-
ted in Scheme 1, while kinetic analysis of the dehydrocoupling
reaction between Ph(Me)SiH2 and tBuNH2 catalysed by III
yielded the rate law shown as eqn (1).
Rate ¼ k[amine]0[silane]1[catalyst]1
(1)
A large, negative entropy of activation, DS‡, was supportive of
a highly ordered transition state while kinetic isotope (kH/kD
A number of notable general trends could be identied. Each
of the pre-catalysts was found to be active for a wide range of
¼
1.0(2)) and Hammett analyses supported only limited Si–H both amine and silane coupling partners. Qualitatively, the
bond cleavage during the rate determining step of the reaction. calcium and strontium pre-catalysts, 2 and 3, appeared to
These observations led Sadow to postulate the model presented provide superior dehydrocoupling activity than their magne-
in Scheme 2 in which nucleophilic attack of the magnesium sium counterpart, 1, while the reactivity survey indicated a clear
amide on silane is followed by a rapid and energetically insig- kinetic dependence upon the relative steric demands of the
nicant hydride transfer from the transient hypervalent silicon amine and silane coupling partners. Triphenylsilane was
centre to yield the product neutral silazane complexed to almost entirely inactive with magnesium (Table 1, entries 13–
magnesium.28
18), providing only modest yields of the silazane products with
In this contribution we demonstrate that this catalytic the least bulky amines, n-benzylamine (Table 1, entry 13) and
silane–amine dehydrocoupling reactivity may be generalised to pyrrolidine (Table 1, entry 16). The dehydrocoupling of primary
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 4212–4222 | 4213

View Article Online
Chemical Science
Edge Article
Table 1 The scope of silicon–nitrogen dehydrocoupling catalysed by complexes 1–3
Entry
Catalystc
Silane
Amine
Product
Conversion/%
1
2
3
4
5
6
7
8
1
PhSiH3
BnNH2
Intractable mixture
PhSiH2NHtBu + PhSiH(NHtBu)2
PhSiH2NHDipp + PhSiH(NHDipp)2
PhSiH2(N(CH2)4) + PhSiH(N(CH2)4)2
PhSiH2NEt2 + PhSiH(NEt2)2
PhSiH2N(SiMe3)2
99a
tBuNH2
DippNH2
99 (86 : 14)a
99 (39 : 61)a
d
(CH2)4NH
EtNH2
99 (29 : 71)a
99 (99 : trace)a
(Me3Si)2NH
BnNH2
71a
Ph2SiH2
Ph3SiH
PhSiH3
Ph2SiHNHBn + Ph2Si(NHBn)2
99 (97 : 3)a
tBuNH2
Ph2SiHNHtBu
95b
99a
d
9
DippNH2
(CH2)4NH
Et2NH
Ph2SiHNHDipp
Ph2SiHN(CH2)4
Ph2SiHNEt2
No reaction
Ph3SiNHBn
No reaction
No reaction
Ph3SiN(CH2)4
10
11
12
13
14
15
16
17
18
19
20
99a
99a
(Me3Si)2NH
BnNH2
0b
25b
tBuNH2
0b
d
DippNH2
(CH2)4NH
Et2NH
0b
7b
No reaction
No reaction
0b
(Me3Si)2NH
BnNH2
0b
2
Intractable mixture
PhSiH2NHtBu + PhSiH(NHtBu)2 +
(PhSiH2)2NtBu
PhSiH2NHDipp + PhSiH(NHDipp)2 +
(PhSiH2)2NDipp
PhSiH(N(CH2)4)2
PhSiH2NEt2 + PhSiH(NEt2)2
PhSiH2N(SiMe3)3
99a
tBuNH2
99 (78 : 10 : 12)a
99 (12 : 20 : 68)a
d
21
DippNH2
22
23
24
25
(CH2)4NH
EtNH2
(Me3Si)2NH
BnNH2
99a
99 (87 : 13)a
99a
Ph2SiH2
Ph2SiHNHBn + Ph2Si(NHBn)2 +
(PhSiH)2NBn
99 (80 : 12 : 8)a
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
tBuNH2
DippNH2
(CH2)4NH
EtNH2
Ph2SiHNHtBu
99a
d
Ph2SiHNHDipp
Ph2SiHN(CH2)4
Ph2SiHNEt2
No reaction
99a
94 (98 : 2)a
88a
(Me3Si)2NH
BnNH2
0a
Ph3SiH
PhSiH3
Ph3SiNHBn
99a
tBuNH2
Ph3SiNHtBu
7b
0b
d
DippNH2
(CH2)4NH
Et2NH
No reaction
Ph3SiN(CH2)4
Ph3SiNEt2
No reaction
Intractable mixture
PhSiH2NHtBu + PhSiH(NHtBu)2
PhSiH2NHDipp (PhSiH2)2NDipp
PhSiH2(N(CH2)4) + PhSiH(N(CH2)4)2
PhSiH2NEt2 + PhSiH(NEt2)2
Ph2SiHNHBn + Ph2Si(NHBn)2 +
(PhSiH)2NBn
99b
54b
(Me3Si)2NH
BnNH2
0b
3
99a
tBuNH2
99 (58 : 42)a
99 (31 : 33)a,e
99 (15 : 85)a
96 (89 : 11)a
99 (5 : 35 : 55)a,e
d
DippNH2
(CH2)4NH
EtNH2
Ph2SiH2
Ph3SiH
BnNH2
99 (77 : 23)a
d
43
44
45
46
47
48
49
DippNH2
Ph2SiHNHDipp + (Ph2SiH)2NDipp
Ph3SiNHBn
BnNH2
99a
54b
0b
tBuNH2
Ph3SiNHtBu
d
DippNH2
(CH2)4NH
Et2NH
No reaction
Ph3SiN(CH2)4
Ph3SiNEt2
No reaction
91a
83b
0b
(Me3Si)2NH
a
Reaction performed at room temperature. b Reaction performed at 60 ꢀC. c Reactions performed at 5 mol% catalyst loading. d Dipp ¼ 2,6-di-iso-
e
propylphenyl. Higher oligomers constitute the remainder of the conversion.
4214 | Chem. Sci., 2013, 4, 4212–4222
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
amines of low to moderate steric demands with all three silane singly dehydrocoupled PhSiH2(N(CH2)4) when this amine was
partners was readily catalysed by each of the group 2 species combined with phenylsilane for all three pre-catalysts. Although
(Table 1, entries 1–3, 7–9, 19–21, 25–27, 31, 37–39, 42–44) with very bulky secondary amines such as hexamethyldisilazane
the exception of the aforementioned issues with magnesium. provided some limited reactivity with the less sterically con-
The bulkiest amine, 2,6-di-iso-propylaniline, however, could not gested phenylsilane (Table 1, entries 6, 24), the more
be dehydrocoupled with the most sterically encumbered silane, substituted di- and triphenylsilane provided no reaction
Ph3SiH, by any of the group 2 pre-catalysts (Table 1, entries 15, (Table 1, entries 12, 18, 30, 36, 49). Consistently high conver-
33, 46). This sterically demanding aniline did, however, provide sions were otherwise observed, indicating a set of broadly active
high activity with the less sterically congested phenylsilane (vide pre-catalysts with good tolerance for a variety of substrates and
infra), yielding doubly dehydrocoupled products and higher reasonable activity under mild conditions.
oligomers in preference to single dehydrocoupling irrespective
Finally, multiple dehydrocoupling to yield bis- and tris-
of the reaction stoichiometry (Table 1, entries 3, 21, 39). The aminosilanes or bis-silylamines was observed particularly for
sterically demanding primary aliphatic amine, tert-butylamine, phenylsilane (Table 1, entries 1–5, 19–21, 23, 37–41). The least
also provided only poor yields in dehydrocoupling reactions sterically congested pairing of phenylsilane and benzylamine
with Ph3SiH catalysed by 2 and 3 (Table 1, entries 32, 45). These rapidly and consistently yielded intractable mixtures of higher
observations suggest that both the steric demands and elec- oligomers suggesting that subsequent dehydrocoupling of the
tronic character (nucleophilicity/basicity) of the amine initially formed silazane, PhSiH2N(H)Bz is very rapid for all
substrate exert a signicant inuence on reactivity with, in metals with these substrates (Table 1, entries 1, 19, 37).
contrast to dehydrocoupling reactions mediated by III,28 the far Furthermore, the strontium pre-catalyst 3 yielded higher olig-
more acidic aniline coupling more readily than its aliphatic omers with phenylsilane and the highly active 2,6-di-iso-
partners where the bulk of the silane coupling partner is propylaniline (Table 1, entry 39) and also through the combi-
insignicant.
nation of benzylamine with diphenylsilane (Table 1, entry 42).
Secondary amines of low steric demands were readily dehy- Pre-catalyst 3 also provided multiple silane substitution in all
drocoupled, with pyrrolidine providing more facile reactivity dehydrocoupling reactions involving either phenyl- or diphe-
than its less basic acyclic analogue diethylamine with phenyl- nylsilane (Table 1, entries 37–43). The rational synthesis of
silane (Table 1, entries 4, 5, 23, 24, 40, 41). This enhanced more substituted silazane species was, thus, investigated by
reactivity, however, resulted in lower selectivity with the doubly NMR spectroscopy utilising amine to silane ratios reective of
dehydrocoupled PhSiH(N(CH2)4)2 forming in preference to the the stoichiometries of the desired heavier aminosilane
Table 2 Group 2-catalysed silicon–nitrogen dehydrocoupling to yield higher silazanes
Entry
Catalystc
Silane
Amine
Ratio
Product
Conversion/%
1
2
3
4
5
6
7
8
1
PhSiH3
BnNH2
1 : 3
1 : 2
3 : 1
1 : 2
1 : 2
2 : 1
1 : 2
2 : 1
1 : 2
1 : 3
1 : 3
1 : 2
2 : 1
3 : 1
1 : 2
1 : 2
2 : 1
1 : 2
2 : 1
1 : 2
1 : 3
1 : 3
1 : 2
3 : 1
1 : 2
2 : 1
(BnNH)3SiPh + (BnNH)2SiHPh
PhSiH2NHtBu + PhSiH(NHtBu)2
PhSiH2(N(CH2)4) + PhSiH(N(CH2)4)2
99 (91 : 9)a
92 (77 : 23)a
66 (100 : 0)b
99 (84 : 16)a
90 (30 : 70)a
99 (99 : trace)a
64b
tBuNH2
(CH2)4NH
Ph2SiH2
BnNH2
tBuNH2
Ph2SiHNHBn + Ph2Si(NHBn)2
Ph2SiHNHtBu
99b
9
(CH2)4NH
BnNH2
Ph2SiHN(CH2)4 + Ph2Si(N(CH2)4)2
Intractable mixture
55 (83 : 17)b
99a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
2
tBuNH2
PhSiH(NHtBu)2 + (PhSiH2)2NtBu
66 (100 : 0)b,d
87 (93 : 7)a
99 (0 : 100)a
99 (0 : 100)b
99 (100 : 0)a
99 (0 : 100 : 0)a
99 (52 : 9 : 39)a
99 (65 : 35)b
99 (100 : 0)b
95 (12 : 88)a
99a
(CH2)4NH
BnNH2
PhSiH(N(CH2)4)2 + PhSi(N(CH2)4)3
Ph2SiH2
Ph2SiHNHBn + Ph2Si(NHBn)2 +
(Ph2SiH)2NBn
tBuNH2
Ph2SiHNHtBu + Ph2Si(NHtBu)2
(CH2)4NH
nBnNH2
tBuNH2
Ph2SiHN(CH2)4 + Ph2Si(N(CH2)4)2
Intractable mixture
3
PhSiH3
PhSiH(NHtBu)2
66a,d
99a,d
(CH2)4NH
BnNH2
PhSi(N(CH2)4)3
Ph2Si(NHBn)2 + (Ph2SiH)2NBn
99b
Ph2SiH2
99 (88 : 12)a
99 (100 : 0)a
a
Reaction performed at room temperature. b Reaction performed at 60 ꢀC. c Reactions performed at 5 mol% catalyst loading. d Higher oligomers
constitute the remainder of the conversion.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 4212–4222 | 4215

View Article Online
Chemical Science
Edge Article
oligomers. The information provided in Table 2 summarises the HN(SiMe3)2 (vide infra) the alkyl reagents, commercially avail-
outcome of these studies. able di-butylmagnesium, 4, and the calcium and strontium
Poly-substituted silazanes were found to be generally acces- dialkyl species [M{CH(SiMe3)2}2(THF)2] (5, M ¼ Ca; 6, M ¼ Sr),30
sible, although conversion of the mono-substituted products were employed in place of the amide pre-catalysts 1–3 utilised in
appeared to occur less readily than the initial dehydrocoupling. the earlier catalytic study.
Given the bulk-sensitive nature of these reactions and the effect
An initial room temperature reaction between a 1 : 1 : 2
of aminosilane formation on the hydridicity of silicon– mixture of di-butylmagnesium, phenylsilane and 2,6-di-iso-
hydrogen bonds this is unsurprising. Of particular note was the propylaniline in heptanes yielded the doubly dehydrocoupled
reaction of benzylamine with phenylsilane in a 3 : 1 ratio cata- species DippNHSiPh(H)NHDipp, 7, (Dipp ¼ 2,6-di-iso-propyl-
lysed by the magnesium pre-catalyst 3, which yielded well- phenyl) as the only isolable compound aer separation of the
dened tri-and di-substituted products in contrast to the results reaction mixture from a nely dispersed colourless solid by
of the 1 : 1 reaction (Table 2, entry 1). The heavier congener pre- ltration. Crystals of 7 suitable for analysis by single crystal
catalysts 2 and 3, however, again yielded intractable products X-ray diffraction were grown direct from the ltered solution
(Table 2, entries 10, 21). Reactions performed at a 1 : 3 phe- and the results of this analysis are shown in Fig. 1. Details of the
nylsilane : amine ratio provided access to the tris-aminated analysis and selected bond length and angle data are provided
products for amines with lower overall steric demands (Table 2, in Tables 3–5 respectively.
entries 1, 14, 24; cf. entries 3, 11, 22). Although bis(amino)-
In contrast, a repeat of this reaction performed in THF
silanes were accessible from use of initial 1 : 2 silane : amine provided, by fractional crystallisation from toluene at reduced
ratios, these products only formed cleanly in a limited number temperature, both the magnesium-anilide–silylanilide 8 and
of cases (Table 2, entries 11, 15, 26) and only for pre-catalysts 2 the magnesium bis(amido)silane 9. Compounds 8 and 9 were
and 3 derived from the heavier alkaline earth metals, Ca and Sr. identied by mechanical separation of the two types of crystals
With bulkier amines the singly dehydrocoupled product was and subsequent X-ray diffraction analysis. The structures of
generally favoured (Table 2, entries 2, 4, 9, 15, 23), irrespective these species are shown in Fig. 2 and 3 respectively while details
of the reaction stoichiometry. Aside from a few notable excep- of the X-ray analysis and selected bond length and angle data
tions (Table 2, entries 13, 17, 25), where they were observed to are summarised in Tables 3–5.
form as a small proportion of the product mixture, disilyl-
amines, targeted from reactions performed with 2 : 1 silane : competitively and we, thus, surmise that the production of
amine ratios, were generally inaccessible. these magnesium species along with the conjugate acid 7 may
The formation of compounds 8 and 9 evidently occurs
This study demonstrated that amine–silane dehydrocoupling be accounted for by the stepwise series of Si–H and N–H
is readily catalysed by group 2 amides under mild conditions metathesis reactions depicted in Scheme 3. In this case an
giving moderate to excellent conversions for a broad range of initially formed, but unobserved, magnesium bis-anilide
coupling partners. A number of trends may also be discerned.
Both the steric demands and the electronic character of both
coupling partners have a signicant inuence on reactivity. The
degree to which steric bulk inuences the reaction is profoundly
inuenced by the size of the metal centre; effects are smallest on
strontium and largest on magnesium. This has an effect both on
the reactions of bulky amines and silanes and also multiple
dehydrocouplings to yield bis(amino)silanes. In contrast to the
observations of Sadow and co-workers,28 more acidic anilines
were generally found to undergo dehydrocoupling more readily
than aliphatic amines, suggestive of a contrasting mechanism.
These empirical studies also indicated that strontium and
calcium appear to provide signicantly higher activity than
magnesium as evidenced by both their shorter reaction times
and milder conditions, but notably reduced selectivity.
Stoichiometric studies
The catalytic formation of the product compounds listed in
Tables 1 and 2 is predicated upon a sequence of N–H/M–H
dihydrogen elimination and Si–H/M–N silicon–nitrogen bond
forming reactions (Scheme 1). We, thus, undertook a series of
stoichiometric reactions in an attempt to provide direct
evidence of such metal-centred reactivity through the isolation
Fig. 1 ORTEP representation of compound 7. Thermal ellipsoids at 30% prob-
of possible group 2-containing intermediate species. In order to
ability and hydrogen atoms, except those attached to N(1), Si(1) and N(2), omitted
avoid complications resulting from the presence of liberated for clarity.
4216 | Chem. Sci., 2013, 4, 4212–4222
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
Table 3 Crystallographic analyses of compounds 7–11
7
8
9
10
11
Molecular formula
Formula weight (g molꢁ1
Crystal system
C30H42N2Si
458.75
Monoclinic
P21/a
15.1354(4)
9.2400(3)
20.1280(6)
90
101.781(2)
90
C38H58MgN2O2Si
627.26
Triclinic
ꢀ
P1
C44H68MgN2O3.5Si
733.40
Monoclinic
P21/n
10.0310(1)
30.0170(5)
16.5560(3)
90
104.903(1)
90
C49H72CaN2O3Si
805.26
Monoclinic
P21/n
10.2642(2)
22.7621(6)
20.7835(5)
90
98.249(2)
90
C46H71N2O4SiSr
831.76
Monoclinic
P21
10.3700(3)
18.8520(8)
12.6320(5)
90
113.60(2)
90
)
Space group
˚
a (A)
11.2940(4)
12.1680(4)
13.6040(4)
96.759(2)
90.610(2)
90.961(2)
1856.13(10)
2
0.113
1.122
3.53 to 27.49
0.0612, 0.1597
0.989, 0.1843
˚
b (A)
˚
c (A)
a (deg)
b (deg)
g (deg)
3
˚
V (A )
2755.63(14)
4
0.105
4817.34(13)
4
0.098
4805.5(2)
4
0.195
2262.83(15)
2
1.260
Z
m (mmꢁ1
)
)
r (g cmꢁ3
1.106
1.011
1.113
1.221
q range (ꢀ)
3.81 to 27.48
0.0578, 0.1372
0.0757, 0.1484
3.62 to 27.44
0.0815, 0.2128
0.1607, 0.2434
5.04 to 25.03
0.0756, 0.1990
0.1379, 0.2270
3.68 to 27.52
0.0640, 0.119
0.1090, 0.1238
R1, wR2 [I > 2s(I)]
R1, wR2 (all data)
Measured/independent reections/Rint 35 403/6254/0.0772 25 602/8357/0.0590 10 746/10 746/0.0000 64 099/8386/0.1702 37 584/10 228/0.1217
compound formation, these observations provide clear
evidence of the viability of the individual reaction steps illus-
trated by Scheme 1 and a stepwise mechanism for the magne-
sium-catalysed silane/amine dehydrocoupling reactions
compiled in Tables 1 and 2.
In a similar manner, reactions between a 1 : 1 : 2 mixture of
either of the heavier alkaline earth alkyl compounds, 5 or 6, with
phenylsilane and 2,6-di-iso-propylaniline in THF produced a
rapid exotherm and immediate foaming of the solution
(Scheme 4). In both cases, evaporation and low temperature
crystallisation of the resulting solids from toluene solution
yielded crystals of compounds 10 and 11 suitable for single
crystal X-ray diffraction analysis. The results of the structural
Fig. 2 ORTEP representation of compound 8. Thermal ellipsoids at 30% prob-
ability. Minor disordered fractional atoms and hydrogen atoms, except those
attached to N(1) and Si(1), omitted for clarity.
undergoes Si–H metathesis with formation of the silylated
aniline DippN(H)SiH2Ph and an anildomagnesium hydride.
These latter species have not been isolated but can either
immediately undergo self-protonolysis to form compound 8
and dihydrogen or engage in further Si–H/Mg–N metathesis to
produce compound 7 and MgH2, which precipitates for reac-
tions performed in heptanes but which may themselves elimi-
nate dihydrogen to provide compound 9 when performed in the
more solubilising medium of THF (route A in Scheme 3).
Alternatively, compound 8 may itself engage in intramolecular
Si–H/Mg–N metathesis with subsequent H2 elimination via the
resultant magnesium hydride (route B). Although we have not
fully identied the conditions which may be employed to
maximise the production of either the magnesium compounds
8 and 9 or compound 7, and irrespective of the precise order of
Fig. 3 ORTEP representation of compound 9. Thermal ellipsoids at 30% prob-
ability. Minor disordered fractional atoms and hydrogen atoms, except for that
attached to Si(1), omitted for clarity.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 4212–4222 | 4217

View Article Online
Chemical Science
Edge Article
˚
Table 4 Selected bond lengths (A) for compounds 7–11
alkaline earth coordination number from ve for compounds 9
and 10 to six for compound 11.
7
8a
9a
10b
11c
Mechanistic considerations
M(1)–N(1)
M(1)–N(2)
M(1)–O(1)
M(1)–O(2)
M(1)–O(3)
M(1)–O(4)
N(1)–Si(1)
N(2)–Si(1)
—
—
—
—
—
—
1.9795(19) 2.044(2) 2.281(3) 2.489(4)
2.0073(18) 2.108(3) 2.342(3) 2.494(4)
The dehydrocoupling reaction of diethylamine with diphe-
nylsilane catalysed by the metal silylamides 1–3 (Scheme 5) was
selected for more detailed analysis due to its appropriate reac-
tion kinetics and well-dened products which provide readily
identiable resonances in 1H NMR spectra. NMR solutions were
prepared in the glovebox and immediately cooled to 193 K on
removal before being warmed to 298 K within the spectrometer.
Concentrations of substrate and product were monitored by
integration over three half-lives in comparison to a tetrakis-
(trimethylsilyl)silane standard, added at the end of the run.
Consistent with the expectation provided by the earlier
observations of silane–amine dehydrocoupling catalysis
(Tables 1 and 2) the calcium-based pre-catalyst 2 provided a
turnover frequency (TOF) that was more than an order of
magnitude higher than either the analogous magnesium or
strontium species (Table 6).
Catalysis of the reactions shown in Scheme 4 with both the
magnesium and calcium silylamides, 1 and 2, was found to
proceed with rst order rst order kinetics over a 5–10 mol%
range of [1] and [2], while further pseudo-rst order experi-
ments, in which tenfold excesses of silane were reacted with
single equivalents of amine and vice versa, yielded zero and rst
order dependences on [Ph2SiH2] and [Et2NH] respectively. The
rate laws for both the magnesium- and calcium-catalysed
reactions were thus deduced to be of the form
2.0341(1)
2.157(2) 2.335(3) 2.591(3)
2.0194(16) 2.057(2) 2.377(3) 2.597(3)
—
—
—
2.075(2) 2.388(3) 2.645(3)
2.561(4)
1.699(3) 1.699(3) 1.704(4)
—
—
1.7119(16)
1.7281(17) 1.6908(18) 1.696(2) 1.693(3) 1.693(4)
a
c
M ¼ Mg. b M ¼ Ca. M ¼ Sr.
determinations are illustrated in Fig. 4 and 5 and reveal
compounds 10 and 11 to be the respective calcium and stron-
tium analogues of the magnesium bis(amido)silane, compound
9. Although the data collected from crystals of compound 10
was of low quality [R(int) ¼ 0.1702], the connectivity was
unambiguous and details of both analyses and selected bond
length and angle data are, thus, included for purposes of
comparison in Tables 3–5 respectively.
While the bond lengths and angles in the group 2 amide
complexes 8–11 are generally unremarkable and comparable to
previously reported amide,31 silylamide29,32 and bis(amido)-
silane33 derivatives of Mg, Ca or Sr, the constitutions of
compound 7 and the dianionic ligand formed by its deproto-
nation in compounds 9–11 are unprecedented for any metal
complex in containing a bridging Si–H function between the
two nitrogen centres. The increase of M2+ radius with group 2
atomic weight is reected by the M–N bond lengths of
˚
d½amineꢂ
compounds 9 (M ¼ Mg, 2.044(2), 2.108(3) A), 10 (M ¼ Ca,
1
1
0
ꢁ
ð1; 2Þ ¼ k½catꢂ ½amineꢂ ½silaneꢂ
(2)
˚
˚
2.281(3), 2.342(3) A), and 11 (M ¼ 2.489(4), 2.494(4) A) as well as
dt
the number of molecules of coordinated THF which raise the
Scheme 3
Scheme 4
4218 | Chem. Sci., 2013, 4, 4212–4222
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
Scheme 5
Assuming that eqn (2) reects an elementary reaction step
and correlates with molecularity in the catalysis, these rate laws
are suggestive of a turnover-limiting process involving one
molecule of amine and a single magnesium or calcium catalytic
centre. In contrast, a zero order dependence on [silane] appears
to indicate that Ph2SiH2 plays no role in the rate-determining
process. Similar experiments performed with compound 3
revealed that the strontium-based catalysis displayed divergent
behaviour, providing second order rate constants as a conse-
quence of rst order dependences on both [Et2NH] and
[Ph2SiH2]. A second order dependence on [3] was also deduced
indicating the rate law shown as eqn (3) for the strontium
catalysis.
Fig. 4 ORTEP representation of compound 10. Thermal ellipsoids at 30%
probability. Minor disordered fractional atoms and hydrogen atoms, except for
that attached to Si(1), omitted for clarity.
d½amineꢂ
2
1
1
ꢁ
ð3Þ ¼ ½catꢂ ½amineꢂ ½silaneꢂ
(3)
dt
Additional insight into the course of these reactions was
provided by monitoring of 1H NMR spectra obtained from a
further series of reactions between a 1 : 1 ratio of diethylamine
and diphenylsilane in the presence of 5 mol% 1–3. Whereas
only ca. 50% of the Mg–N(SiMe3)2 environments of 1 were
consumed during the catalytic reaction, with production of
Ph2SiHN(SiMe3)2 and a smaller quantity (ca. 5% by ratio) of
HN(SiMe3)2, both of the heavier congeners, compounds 2 and 3,
were observed to undergo complete consumption of any
calcium- or strontium-bound hexamethyldisilazide residues. In
these latter cases, catalyst initiation occurred with complete
conversion either to the product of Si–H/M–N metathesis,
Ph2SiHN(SiMe3)2, or N–H/M–N protonolysis, HN(SiMe3)2, for 2
and 3 respectively. In the case of 2, this exclusive, irreversible
metathesis-based activation yields a brief induction period
which initially complicated kinetic analysis. As a result of this,
premixing of the precatalyst, 2, and diphenylsilane was utilised
to deconvolute the kinetics, rendering them amenable to
analysis.
We have previously reported that the efficacy of group 2-
catalysed alkene hydroamination is heavily inuenced by the
ability of the amine substrate to enter the catalytic manifold and
Fig. 5 ORTEP representation of compound 11. Thermal ellipsoids at 30%
probability. Minor disordered fractional atoms and hydrogen atoms, except for
that attached to Si(1), omitted for clarity.
Table 5 Selected bond angles (ꢀ) for compounds 7–11
7
8a
9a
77.15(10)
96.05(10)
164.22(10)
111.61(10)
92.75(11)
90.60(11)
99.43(12)
10b
69.46(12)
109.10(12)
114.58(13)
160.41(13)
95.34(15)
93.31(15)
101.88(16)
11c
65.04(12)
107.00(12)
107.25(12)
87.95(12)
94.77(15)
94.90(16)
104.10(19)
N(1)–M(1)–N(2)
N(1)–M(1)–O(1)
N(2)–M(1)–O(1)
N(2)–M(1)–O(2)
M(1)–N(1)–Si(1)
M(1)–N(2)–Si(1)
N(2)–Si(1)–N(2)
—
—
—
—
—
—
114.07(8)
112.31(8)
110.04(8)
106.86(7)
—
123.18(10)
—
104.15(8)
a
c
M ¼ Mg. b M ¼ Ca. M ¼ Sr.
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 4212–4222 | 4219

View Article Online
Chemical Science
Edge Article
Table 6 Turnover frequencies (298 K) and activation parameters for the catalytic
while it would appear that a single bis(trimethylsilyl)amide
remains bound to Mg during catalysis with compound 1, this in
itself is insufficient to replicate the behaviour dictated by the
tris(oxazolinyl)borate ancillary ligand of III. Notably, a similar
ligand-dependent behaviour has only very recently been estab-
lished in the more widely considered area of group 2-based
alkene hydroamination catalysis.36 We have previously hypoth-
esised that similar variations in the rates of reactions of these
latter catalytic systems are a consequence of the differing charge
densities of the M2+ cations and their resultant ability to
mediate metathesis and C]C insertion.35 A similar rationale
dehydrocoupling of Et2NH and Ph2SiH2
Ea
DH‡
DS‡
DG‡298
(kcal
TOF
(hꢁ1
(kcal
(kcal
(cal
Catalyst
)
molꢁ1
)
molꢁ1
)
molꢁ1 Kꢁ1
)
molꢁ1
)
1
2
3
57.1(4)
2822.5(54)
125.5(22)
14.6(2)
7.4(1)
11.2(23)
13.9(2)
6.8(1)
10.6(23)
ꢁ29.8(6)
ꢁ46.8(1)
ꢁ41.2(81)
22.8
20.7
23.5
the basicity and resultant ability of the group 2 amide or alkyl has been proposed to account for differences in the ability of
pre-catalyst to effect deprotonation of primary or secondary group 2 cations to engage in amine–borane dehydrocoupling25
amine substrates.34,35 In the current instance it appears that the and comparable effects may indeed be the origin of the varia-
relative ease of Si–H/M–N or N–H/M–N metathesis is a further tions in the Ea and DH‡ values of Table 6. The similarity of the
signicant consideration affected by the identity of the specic free energy values for the reactions catalysed by 1–3 are a
group 2 pre-catalyst and co-ligand employed. Furthermore, the consequence of the negative entropy of activation (ꢁ46.8(1) cal
persistence of HN(SiMe3)2 in the reaction medium during molꢁ1 Kꢁ1) associated with the calcium-based catalysis. We
catalysis using compounds 1 and 3 is, unlike the Ph2SiHN- have previously reported that an entropic component can play a
(SiMe3)2 produced during initiation of 2, likely to have an major role in deterministic kinetic analyses of group 2-based
inhibitory effect on the reaction and may account, at least in inter- and intramolecular alkene hydroamination and intra-
part, for the increased activity of calcium relative to the other molecular alkyne hydroalkoxylation catalysis.35 In the current
congeners in these reactions.
case, however, it is likely that catalyst solution molecularity,
While these observations and the rst order rate dependence mode of activation and a potential for the establishment of
on [1] (vide supra) suggest that a monomeric constitution may transaminative pre-equilibria between the amine substrate and
be maintained through persistent hexamethyldisilazide ligation HN(SiMe3)2 play an equally signicant role in the relative effi-
of magnesium, the origin of a similar rate dependence for cacy of each individual catalysis.
calcium but second order behaviour with respect to [3] is less
Although it would be imprudent to attempt to present a
readily ascribed. We have earlier suggested that a second order denitive interpretation of these variations, with this caveat in
rate dependence on [catalyst] in group 2-based processes is place it is possible to draw some tentative conclusions
suggestive of a dimeric active catalytic centre,34b,35 which regarding the transition in mechanism of the individual reac-
generally results in the case of the larger heavier alkaline earth tions. For magnesium and calcium, the deduced rate laws (eqn
dications. Although the apparently differing modes of pre- (2)) suggest that protonolysis of the amine by the hydride
catalyst activation are probably signicant, in the absence of formed in the Si–H/M–N s-bond metathesis step is rate deter-
further corroborative evidence, we surmise that the current data mining. While the energetics of this process should indeed be
are a result of similar variations in solution molecularity.
inuenced by the charge density of the M2+ cation in question,
Arrhenius and Eyring analyses provided activation parame- the likely origin of this variation, viz. the increased radius of the
ters (Table 6) for the reactions depicted in Scheme 4 which are Ca2+ centre, also has marked consequences for both the mode
commensurate with the superior activity provided by of pre-catalyst activation and the ordering of the protonolysis
compound 2. Although the calcium-based process enjoys an transition state. Furthermore, previous computational studies
advantage of some 4–8 kcal molꢁ1 in its calculated activation on group 2-mediated hydroamination have noted a marked
energy and enthalpy of activation, the resultant DG2‡98 values for decrease in the barrier to protonolysis down the group.35a
the three group 2 systems are less indicative of such signicant Meaningful comparison is further complicated for the stron-
discrimination in reaction rates. Similar activation parameters tium system described by the rate law in eqn (3), in which case
as those derived here for catalysis by 2 and a deduced zero order the kinetic data evidence mediation of the reaction via a cata-
dependence upon [amine] (eqn (1)) led Sadow to propose that lytic dimer. The rst order dependences in [amine] and [silane]
dehydrocoupling catalysis of tBuNH2 and PhMeSiH2 by III, for catalysis by 3 are redolent of equal involvement of Et2NH
occurs via rate-determining nucleophilic attack by an interme- and Ph2SiH2 in the rate determining step and suggest that the
diate magnesium amide on silicon to form a ve-coordinate Si–H/Mg–N s-bond metathesis step to yield a metal hydride is
silicon centre (Scheme 2).28 While the rst order dependence on rate determining. While rapid amine protonolysis is also
[amine] depicted in eqn (2) and (3) does not allow us to discount consistent with the apparent mode of activation of compound 3
a similar mechanism in reactions catalysed by the simple during catalysis, the equally signicant rate dependence on
homoleptic bis(trimethyl)amide derivatives, 1–3, our observa- [amine] requires further consideration. We and others have
tions suggest that different mechanisms are likely to be active previously suggested that rate determining C]C insertion in
across the various catalytic systems; an observation that is group 2-mediated hydroamination catalysis is assisted by the
further corroborated by the observed divergence in activity for presence of further coordinated equivalents of the protic
aniline dehydrocoupling between III and 1–3. For example, substrate within the coordination sphere of the catalytic
4220 | Chem. Sci., 2013, 4, 4212–4222
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
centre.35,37 The lower charge density and larger radius of the Sr2+ plausible manner. Hence PLATON SQUEEZE was employed to
centre may, thus, also be signicant in its ability to provide a treat these regions. Based on pre-PLATON electron density and
less constrained and labile platform for reaction.
the results for the SQUEEZE analysis, guest solvent was
In conclusion, we have shown that the readily available included in the asymmetric unit as ½ of a THF molecule.
magnesium, calcium and strontium amides, [M{N(SiMe3)2}2]2, Although the data quality for compound 10 was poor, the crystal
are active catalysts for the dehydrogenative coupling of silanes structure was unambiguous. The asymmetric unit contained
and amines. Stoichiometric studies have provided evidence for one molecule of toluene solvent in addition to one molecule of
amide intermediates consistent with a stepwise mechanism the complex. H(1) was located and freely rened. Some THF
predicated upon a sequence of metal-centred amine deproto- carbons disordered in 55 : 45 ratio [C(36), C(37), C(39), C(40),
nation and/or silane-based metathesis steps. While all three C(41) and C(42)]. ADP restraints were applied to fractional
pre-catalysts studied displayed broad reactivity with a wide occupancy atoms and distance restraints applied in disordered
variety of amine and silane coupling partners, in most cases the regions to assist convergence. For compound 11, H(1), attached
calcium pre-catalyst, 2, provided signicantly superior activity to Si(1), was located and freely rened. C(40) was disordered
to either of its magnesium- or strontium-based congeners. over two sites in a 70 : 30 ratio, while C(43)–C(46) were disor-
These empirical observations were borne out by a comparative dered in 50 : 50 ratio. Some ADP, C–O and C–C restraints used
kinetic study of the catalysed reaction between Et2NH and in disordered regions to assist convergence.
Ph2SiH2 which also highlighted distinct and divergent catalytic
behaviour between all three amide pre-catalysts studied herein
as well as the recently reported borato-magnesium species, III.28
Acknowledgements
While our interpretation of these data is inconclusive it is clear We thank the EPSRC and the University of Bath for funding.
that the intrinsic character (coordinative unsaturation and
molecularity, ancillary ligation, bond polarity and charge
density) of the group 2 centre employed in the catalysis has a
Notes and references
profound effect upon not only the efficacy but also the mode
and mechanism of reaction. Distinct differences in reactivity are
now emerging across the entire series of alkaline earths. It is
clear, therefore, that compounds of these elements must always
be judged on their own merits and are predisposed to display a
more diverse chemistry than their historical treatment would
imply. We are continuing in our attempts to identify the causes
of these variations and to deconvolute their impacts upon
chemical reactivity and catalysis.
1 (a) D. W. Stephan, Angew. Chem., Int. Ed., 2000, 39, 314; (b)
T. J. Clark, K. Lee and I. Manners, Chem.–Eur. J., 2006, 12,
8634; (c) R. Waterman, Curr. Org. Chem., 2008, 12, 1322; (d)
R. Waterman, Dalton Trans., 2009, 18.
2 (a) L. D. Schwartz and P. C. Keller, J. Am. Chem. Soc., 1972, 94,
3015; (b) P. C. Keller, J. Am. Chem. Soc., 1974, 96, 3078; (c)
P. C. Keller, Inorg. Chem., 1975, 14, 438; (d) P. C. Keller,
Inorg. Chem., 1975, 14, 440.
3 (a) C. W. Hamilton, R. T. Baker, A. Staubitz and I. Manners,
Chem. Soc. Rev., 2009, 38, 279; (b) A. Staubitz,
A. P. M. Robertson, M. E. Sloan and I. Manners, Chem.
Rev., 2010, 110, 4023.
X-ray crystallography
Data for 7–11 were collected at 150 K on a Nonius Kappa CCD
diffractometer equipped with an Oxford Cryosystem low
temperature device, using graphite monochromated MoKa
4 J. Y. Corey, Adv. Organomet. Chem., 2004, 51, 1.
5 S. Greenberg and D. W. Stephan, Chem. Soc. Rev., 2008, 37,
1482.
˚
radiation (l ¼ 0.71073 A). Data were processed using the Nonius
6 (a) E. W. Corcoran, Jr and L. G. Sneddon, Inorg. Chem., 1983,
22, 182; (b) E. W. Corcoran, Jr and L. G. Sneddon, J. Am.
Chem. Soc., 1984, 106, 7793; (c) E. W. Corcoran, Jr and
L. G. Sneddon, J. Am. Chem. Soc., 1985, 107, 7446; (d)
H. Braunschweig and F. Guethlein, Angew. Chem., Int. Ed.,
2011, 50, 12613; (e) H. Braunschweig, C. Claes and
F. Guethlein, J. Organomet. Chem., 2012, 706–707, 144; (f)
H. Braunschweig, P. Brenner, R. D. Dewhurst, F. Guethlein,
Soware.38 Structure solution, followed by full-matrix least
squares renement was performed using the programme suite
X-SEED.39 For compound 7, H(1) and H(2), attached to N(1) and
N(2) respectively, were located and rened at a distance of 0.98
˚
A from the relevant parent atoms. H(1A), attached to Si(1), was
also located and rened without restraints. For compound 8,
the hydrogen atoms attached to Si(1) were located and rened
subject to being similar distances from the parent. H(1)
attached to N(1) was similarly located and rened at distance of
¨
J. O. C. Jimenez-Halla, K. Radacki, J. Wolf and L. Zollner,
Chem.–Eur. J., 2012, 18, 8605; (g) H. C. Johnson,
C. L. McMullin, S. D. Pike, S. A. Macgregor and
A. S. Weller, Angew. Chem., Int. Ed., 2013, DOI: 10.1002/
anie.201304382, in press.
7 (a) P. P. Power, Nature, 2010, 463, 171; (b) T. Chivers and
J. Konu, Comments Inorg. Chem., 2009, 30, 131.
8 See, for example; (a) T. Hayashi, Acc. Chem. Res., 2000, 33, 354;
(b) B.-H. Kim, M.-Y. Cho and H.-G. Woo, Synlett, 2004, 761.
9 Modern Reduction Methods, ed. P. G. Andersson and I. J.
Munslow, Wiley-VCH, Weinheim, 2008.
˚
0.98 A from the nitrogen. C(36) and C(37) were found to be
disordered in a 50 : 50 ratio over two proximate sites. The C–C
distances were, thus, rened with distance restraints in the
associated THF molecule to assist convergence. For compound
9, H(1a), attached to Si(1) was located and rened freely. Some
disorder was present. In particular, C(36)/C(37) were disordered
over 2 sites in a 50 : 50 ratio, while C(41) was similarly frag-
mented but in a 70 : 30 ratio. The unit cell was seen to contain 2
areas of diffuse solvent which could not be modelled in any
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 4212–4222 | 4221

View Article Online
Chemical Science
Edge Article
10 Organotransition Metal Chemistry: from bonding to catalysis, ed.
dehydrocoupling catalysis by main group elements, see, (b)
R. J. Less, R. L. Melen and D. S. Wright, RSC Adv., 2012, 2, 2191.
J. F. Hartwig, University Science Books, Sausalito, Ca, 2010.
11 G. A. Molander and J. A. C. Romero, Chem. Rev., 2002, 102, 2161. 27 F. Buch and S. Harder, Organometallics, 2007, 26, 5132.
12 (a) C. A. Roth, Ind. Eng. Chem. Prod. Res. Dev., 1972, 11, 134; (b) 28 J. F. Dunne, S. R. Neal, J. Engelkemier, A. Ellern and
Y. Tanabe, M. Murakami, K. Kitaichi and Y. Yoshida,
A. D. Sadow, J. Am. Chem. Soc., 2011, 133, 16782.
Tetrahedron Lett., 1994, 35, 8409; (c) Y. Tanabe, T. Misaki, 29 (a) P. B. Hitchcock, M. F. Lappert, G. A. Lawless and B. Royo,
M. Kurihara, A. Iida and Y. Nishii, Chem. Commun., 2002, 1628.
13 A. P. Smith, J. J. S. Lamba and C. L. Fraser, Org. Synth., 2002,
78, 82.
14 Y. D. Blum, K. B. Schwartz and R. M. Laine, J. Mater. Sci.,
1989, 24, 1707.
J. Am. Chem. Soc., 1990, 30, 96; (b) M. Westerhausen, Inorg.
Chem., 1991, 30, 96.
30 M. R. Crimmin, A. G. M. Barrett, M. S. Hill, D. MacDougall,
M. F. Mahon and P. A. Procopiou, Chem.–Eur. J., 2008, 14,
11292.
15 (a) C. Eaborn, Organosilicon Compounds, Butterworths, 31 See, for example; (a) A. Xia, J. Heeg and C. H. Winter,
London, 1960; (b) R. Fessenden and J. S. Fessenden, Chem.
Rev., 1961, 61, 361; (c) V. Passarelli, G. Carta, G. Rossetto
and P. Zanella, Dalton Trans., 2003, 413.
Organometallics, 2002, 21, 4718; (b) M. M. Olmstead,
W. J. Grigsby, D. R. Chacon, T. Hascall and P. P. Power,
Inorg. Chim. Acta, 1996, 251, 273; (c) D. R. Armstrong,
W. Clegg, R. E. Mulvey and R. E. Rowlings, J. Chem. Soc.,
Dalton Trans., 2001, 409; (d) A. G. M. Barrett, I. J. Casely,
M. R. Crimmin, M. S. Hill, J. R. Lachs, M. F. Mahon and
P. A. Procopiou, Inorg. Chem., 2009, 48, 4445.
16 L. H. Sommer and J. D. Citron, J. Org. Chem., 1967, 32, 2470.
17 H. Q. Liu and J. F. Harrod, Organometallics, 1992, 11, 822.
18 E. Matarasso-Tchiroukhine, J. Chem. Soc., Chem. Commun.,
1990, 681.
19 H. Q. Liu and J. F. Harrod, Can. J. Chem., 1992, 70, 107.
20 (a) J. A. Reichl and D. H. Berry, Adv. Organomet. Chem., 1998,
197; (b) W. D. Wang and R. Eisenberg, Organometallics, 1991,
10, 2222.
32 (a) W. Vargas, U. English and K. Ruhlandt-Senge, Inorg.
Chem., 2002, 41, 5602; (b) L. T. Wendell, J. Bender, X. He,
B. C. Noll and K. W. Henderson, Organometallics, 2006, 25,
¨
4953; (c) M. S. Hill, G. Kociok-Kohn and D. J. MacDougall,
21 (a) Y. Blum and R. M. Laine, Organometallics, 1986, 5, 2081;
Inorg. Chem., 2011, 50, 5234.
(b) C. Biran, Y. D. Blum, R. Glaser, D. S. Tse, K. A. Youngdahl 33 (a) M. Veith, W. Frank, F. Tollner and H. Lange, J. Organomet.
and R. M. Laine, J. Mol. Catal., 1988, 48, 183.
Chem., 1987, 326, 315; (b) C. Pi, L. Wan, Y. Gu, H. Wu,
C. Wang, W. Zheng, L. Weng, Z. Chen, X. Yang and L. Wu,
Organometallics, 2009, 28, 5281; (c) D. Yang, Y. Ding,
H. Wu and W. Zheng, Inorg. Chem., 2011, 50, 7698; (d)
V. L. Blair, W. Clegg, A. R. Kennedy, Z. Livingstone,
L. Russo and E. Hevia, Angew. Chem., Int. Ed., 2011, 50, 9857.
34 (a) M. R. Crimmin, M. Arrowsmith, A. G. M. Barrett,
I. J. Casely, M. S. Hill and P. A. Procopiou, J. Am. Chem.
Soc., 2009, 131, 9670; (b) M. Arrowsmith, M. R. Crimmin,
A. G. M. Barrett, M. S. Hill, G. Kociok-Kohn and
P. A. Procopiou, Organometallics, 2011, 30, 1493; (c)
A. G. Avent, M. R. Crimmin, M. S. Hill and P. B. Hitchcock,
Dalton Trans., 2005, 278.
22 (a) K. Takaki, T. Kamata, Y. Miura, T. Shishido and
K. Takehira, J. Org. Chem., 1999, 64, 3891; (b) K. Takaki,
K. Komeyama and K. Takehira, Tetrahedron, 2003, 59,
10381; (c) W. Xie, H. Hu and C. Cui, Angew. Chem., Int. Ed.,
2012, 51, 11141. See also, (d) Y. Chen, H. Song and C. Cui,
Angew. Chem., Int. Ed., 2010, 49, 8958.
23 J. X. Wang, A. K. Dash, J. C. Berthet, M. Ephritikhine and
M. S. Eisen, J. Organomet. Chem., 2000, 610, 49.
24 For reviews, see; (a) A. G. M. Barrett, M. R. Crimmin,
M. S. Hill and P. A. Procopiou, Proc. R. Soc. A, 2010, 466,
927; (b) S. Harder, Chem. Rev., 2010, 110, 3852.
25 (a) M. R. Crimmin, A. G. M. Barrett, M. S. Hill,
P. B. Hitchcock and P. A. Procopiou, Organometallics, 2007, 35 (a) A. G. M. Barrett, C. Brinkmann, M. R. Crimmin, M. S. Hill,
24, 10410; (b) D. J. Liptrot, M. S. Hill, M. F. Mahon and
D. J. MacDougall, Chem.–Eur. J., 2010, 16, 8508; (c)
M. S. Hill, M. F. Mahon and T. P. Robinson, Chem.
Commun., 2010, 46, 2498; (d) M. S. Hill, M. Hodgson,
D. J. Liprot and M. F. Mahon, Dalton Trans., 2011, 40,
P. Hunt and P. A. Procopiou, J. Am. Chem. Soc., 2009, 131,
12906; (b) C. Brinkmann, A. G. M. Barrett, M. S. Hill and
P. A. Procopiou, J. Am. Chem. Soc., 2012, 134, 2193; (c)
C. Brinkmann, A. G. M. Barrett, M. S. Hill, P. A. Procopiou
and S. Reid, Organometallics, 2012, 31, 7287.
¨
7783; (e) J. Spielmann, G. Jansen, H. Bandmann and 36 (a) M. Arrowsmith, M. S. Hill and G. Kociok-Kohn,
S. Harder, Angew. Chem., Int. Ed., 2008, 47, 6290; (f)
J. Spielmann and S. Harder, J. Am. Chem. Soc., 2009, 131,
5064; (g) J. Spielmann, M. Bolte and S. Harder, Chem.
Organometallics, 2011, 30, 1291; (b) B. Liu, T. Roisnel,
J.-F. Carpentier and Y. Sarazin, Angew. Chem., Int. Ed.,
2012, 51, 4943.
Commun., 2009, 6934; (h) J. Spielmann, D. F.-J. Piesik and 37 J. F. Dunne, D. B. Fulton, A. Ellern and A. D. Sadow, J. Am.
S. Harder, Chem.–Eur. J., 2010, 16, 8307; (i) P. Bellham, Chem. Soc., 2010, 132, 17680.
M. S. Hill, G. Kociok-Kohn and D. J. Liptrot, Dalton Trans., 38 DENZO-SCALEPACKZ, Otwinowski and W. Minor,
¨
2013, 42, 737; (j) P. Bellham, M. S. Hill, D. J. Liptrot,
D. J. MacDougall and M. F. Mahon, Chem. Commun., 2011,
Processing of X-ray Diffraction Data Collected in
Oscillation Mode, Methods in Enzymology, ed. C. W. Carter,
Jr and R. M. Sweet, Academic Press, 1997, vol. 276:
Macromolecular Crystallography, part A, pp. 307–326.
¨
47, 9060; (k) P. Bellham, M. S. Hill, G. Kociok-Kohn and
D. J. Liptrot, Chem. Commun., 2013, 49, 1960.
26 (a) M. S. Hill, M. F. Mahon and T. P. Robinson, Chem. 39 L. J. Barbour, X-Seed – A soware tool for supramolecular
Commun., 2010, 46, 2498; For
a
general review of
crystallography, J. Supramol. Chem., 2001, 1, 189–191.
4222 | Chem. Sci., 2013, 4, 4212–4222
This journal is ª The Royal Society of Chemistry 2013