Hypervalent hydridosilicates: Synthesis, structure and hydride bridging

Article abstract of DOI:10.1039/b713427d

A range of hydridosilicate anions has been prepared and characterised by spectroscopic, structural and computational methods. The general approach involved reaction of KH with a neutral silane precursor in the presence of [18]crown-6. In this manner, [K([

Full text of DOI:10.1039/b713427d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Hypervalent hydridosilicates: synthesis, structure and hydride bridging†‡§¶
Paul D. Prince,a Michael J. Bearpark,b G. Sean McGrady*c and Jonathan W. Steed*d
Received 31st August 2007, Accepted 4th October 2007
First published as an Advance Article on the web 23rd October 2007
DOI: 10.1039/b713427d
A range of hydridosilicate anions has been prepared and characterised by spectroscopic, structural and
computational methods. The general approach involved reaction of KH with a neutral silane precursor
in the presence of [18]crown-6. In this manner, [K([18]crown-6)]+ salts of [Ph3SiH2]− (1), [Ph3SiF2]− (9),
and [(p-FC6H4)3SiHF]−/[(p-FC6H4)3SiH2]− (12) were stabilised and characterized by NMR
spectroscopy and X-ray diffraction. In each case, the anion adopts a trigonal bipyramidal (TBP)
geometry with three equatorial phenyl groups eclipsing the axial Si–H/Si–F bonds. The Si–H · · · K
distances, along with DFT calculations on 1, indicate an electrostatic interaction that does not dictate
the geometry adopted by the anion. A [H2SiOiPr3]− salt (7) has also been crystallised in the same way;
X-ray diffraction shows in this case a distorted TBP array with axial hydride ligands, and both
Si–H · · · K and Si–O · · · K interactions. 1H NMR exchange experiments show 1 to undergo facile
hydride exchange with Ph3SiH. Compound 1 acts as a good hydride transfer reagent to a variety of
substrates, but its high reactivity often results in redistribution and other side reactions.
H · · · Li interactions are also known.20–22 An example of an
Introduction
Si–H · · · Na interaction is found in the crystal structure of a
sodium alcoholate, [Na8(O3C5H11)6][SiH3]2.23 Alkali metal silyls
of the form M[SiH(SitBu3)2] (M = Li, Na, K) have also been
prepared. The X-ray structure analysis of the benzene solvate
of K[SiH(SitBu3)2] shows a Si–H · · · K contact with an H · · · K
Hypervalent silicon compounds continue to generate a great deal
of interest.1 This activity stems from a desire to understand the
intermediacy of pentacoordinate silicon in nucleophilic substitu-
tions, and pentacoordination in general.2–7 Hypervalent silanes are
particularly topical in the context of intramolecular hypervalent
interactions in hypervalent silane transition metal complexes.8–16
While many pentacoordinate hypervalent fluorosilicates have been
studied and characterised, hydridosilicates are more reactive and
challenging experimentally, and reports have been limited largely
to descriptions of their preparation and reactivity. Few structural
studies have been reported, and those species that have been
characterised generally consist of zwitterionic systems, rather than
a discrete [X5Si]− (X = any atom) anion. Within the realm of
silane chemistry the Si–H · · · M hydride bridging interaction is
also of interest. Some reports have classed Si–H · · · M interactions
as agostic; Sekiguchi et al. have characterised a series of di(alkali
˚
distance of 3.01 A. This, at the outset of our study, was the only
example of an Si–H · · · K interaction in the Cambridge Structural
Database (CSD).24 We now report the synthesis, reactivity and
structural characterisation of a range of crown-ether stabilised
hypervalent silanes, which exhibit a range of interesting geometries
and Si–H · · · K interactions. Part of this work has been published
previously in preliminary form.25
Results and discussion
Synthesis and structure
=
metal), complexes [M(ligand)n]2[(HMe2Si)2C C(SiMe2H)2] (M =
We have undertaken the preparation and structural characteri-
sation of a range of hypervalent dihydrido-, fluorohydrido- and
difluorosilicates of the form [R3SiH2]−, [R3SiHF]− and [R3SiF2]−,
where R = an aryl or alkoxy group. The formation of these silicates
can be achieved either by reaction of an alkali metal with Ph2SiH2,
or by reaction of a triarylsilane with KH or KF. In each case, a
crown ether is included to complex the alkali metal counter cation,
affording greater solubility of the complex and stabilising the ionic
product. Hypervalent hydridosilicates of the form [(RO)3SiH2]−
and [(RO)4SiH]− have previously been prepared by the reaction
of (RO)3SiH with KH and KOR, respectively.26–28 We carried out
the reaction of KH with Ph3SiH in the presence of [18]crown-
6 in an attempt to produce [K([18]crown-6)][Ph3SiH2] (1), in the
manner used to synthesise the analogous alkoxyhydridosilicates;
however the only isolated product was (Ph3Si)2O·[18]crown-6 (2)
(see ESI¶), with the oxygen source apparently being the THF
solvent. However, reaction of Ph2SiH2 with potassium metal
resulted instead in a redistribution of phenyl ligands to produce
Li, Na, K, Rb, Cs), which exhibit such so-called Si–H · · · M
agostic interactions.17,18 Si–H · · · Li interactions have been es-
tablished by Goldfuss and co-workers in [M(Me2Si(H)NtBu)]
(M = Li, Mg0.5) systems.19 Further examples of similar Si–
aDepartment of Chemistry, King’s College London, Strand, London, UK
WC2R 2LS
bDepartment of Chemistry, Imperial College London, London, UK SW7
2AZ
cDepartment of Chemistry, University of New Brunswick, Fredericton, NB,
E3B 6E2, Canada. E-mail: smcgrady@unb.ca
dDepartment of Chemistry, Durham University, South Road, Durham, UK
DH1 3LE. E-mail: jon.steed@durham.ac.uk
† Dedicated to Professor Ken Wade on occasion of his 75th birthday.
‡ The HTML version of this article has been enhanced with colour images.
§ CCDC reference numbers 169646, 659284–659288 and 662558. For
crystallographic data in CIF or other electronic format see DOI:
10.1039/b713427d
¶ Electronic supplementary information (ESI) available: Selected NMR
spectra and figures. See DOI: 10.1039/b713427d
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 271–282 | 271
©

View Article Online
32
1 as the major product, in 84% yield based on Ph. The reaction
of PhSiH3 with potassium metal and [18]crown-6 also yielded 1,
in addition to other redistribution products.29 However, reaction
of (p-MeC6H4)2SiH2 (3) did not produce any hypervalent species
under the conditions in which 1 is produced, and the reaction
proceeded instead to give a silyl product.29 The 1H NMR spectrum
of 1 shows a singlet at d 5.95, with satellites from 29Si–1H coupling,
which is assigned to the SiH2 unit. The SiH resonance is shifted to
high frequency compared to that of Ph2SiH2 (d 3.9) and Ph3SiH
(d 5.6) and has a much reduced coupling constant of 131 Hz
(consistent with the hypercoordination at silicon) compared to
the coupling constants of 197 and 199 Hz measured for Ph2SiH2
and Ph3SiH, respectively, in the same solvent. The 29Si NMR
spectrum of 1 in THF-d8 reveals a triplet with 1J(29Si–1H) =
131 Hz, centred at d −74 ppm, consistent with both protons being
in equivalent environments or exchanging quickly on the NMR
timescale. The reaction was repeated using Ph2SiD2 (4) in place
of Ph2SiH2 to give [K([18]crown-6)][Ph3SiD2] (1D). Altering the
crown ether to [15]crown-5 gave [K([15]crown-5)2][Ph3SiH2] (5),
which was also isolated as a main product of the reaction between
PhSiH3, potassium metal and two equivalents of [15]crown-5.
Increasing the quantity of the silane in the reaction between
potassium metal, [18]crown-6 and Ph2SiH2, to greater than one
equivalent led to the formation of a mixture of [SiH3]−, Ph3SiH,
and SiH4, in addition to 1 in both THF and DME. Characterisa-
tion is based on the chemical shifts of silane species from 1H NMR
spectroscopic data. The broad nature of the resonances implies the
existence of a dynamic exchange process. The only comparable
species characterised by NMR spectroscopy in the literature are
Li[Ph3SiH2]30 and silicate anions of the form K[(OR)4SiH] or
K[(OR)3SiH2].31 Our NMR results for 1 are consistent with the
chemical shift observed for Li[Ph3SiH2]. The alkoxy-substituted
silicate anions appear quite different to 1, with dH ca. 4.5 and 1J(Si–
H) 190–220 Hz. Structural analyses of these systems have not been
reported, although their NMR spectra have been interpreted in
terms of [(OiPr)3SiH2]− and [(OtBu)3SiH2]− each possessing either
one axial (Si–Hax) and one equatorial Si–H bond (Si–Heq) about
the trigonal bipyramidal (TBP) silicon centre, or else two Si–Heq
bonds in a TBP array, depending on the nature of the counter-
cation and solvent.31
˚
˚
1.49(3) A in Ph3SiH). Similarly, the Si–C distances, at 1.928 A (b)
˚
˚
and 1.917–1.946 A (a; two independent distances), are 0.07–0.08 A
longer than those observed for Ph4Si.33 In the trigonal b form the
complex exhibits crystallographic threefold molecular symmetry
with a linear H–Si–H vector; Fig. 1. The K · · · H distance, based
on the position of the hydride substituent located in the difference
˚
˚
Fourier map, is 2.82 A (K · · · Si = 4.318 A), remarkably less than
the sum of the van der Waals radii34 and shorter than the K · · · H
distance in the binary hydride (2.854 A).35 In the monoclinic form
˚
there are two crystallographically independent half [K([18]crown-
6)]+ cations and a unique [Ph3SiH2]− anion. The interactions of
the anion with the two K+ centres are very different. One metal
ion is involved in a cation–p interaction36 to one of the phenyl
substituents, with the shortest K · · · C distances measured at 3.360
+
˚
and 3.375 A, while the other K ion engages in a K · · · H–Si
interaction similar to that observed in the trigonal form (K · · · H
˚
˚
distance = 3.108 A; Si · · · K distance = 4.709 A); Fig. 2. The
Fig. 1 Single-crystal X-ray structure of the trigonal (b) form of 1
showing the Si–H · · · K interaction as a dotted line. All C–H hydrogen
˚
atoms have been omitted for clarity. Selected interatomic distances (A):
K(1)–O(1) 2.794(2), K(1)–O(2) 2.799(2), Si(1)–C(3) 1.938(3), Si(1)–H(1S)
1.60, Si(1)–H(2S) 1.50, Si(1)–K(1) 4.32, K(1)–H(2) 2.82.
Highly air- and moisture-sensitive crystals of 1 and 5 were
isolated from the reaction mixture by decanting the reaction
solution, and were coated with a protective perfluoropolyalkyl
ether oil. The less concentrated solution of 1 was retained
and stored in a refrigerator, and after several days a further
crop of crystals with distinctly different habit formed. The a
form (mp 124 ◦C) deposited in far greater abundance from the
concentrated solution when DME or THF was used as the
solvent. Both forms (1a and 1b, respectively) were subjected to
single-crystal X-ray analysis. The geometry of the silane anion
in both polymorphs is essentially the same: a TBP array with the
hydrogen atoms occupying the axial positions, confirming the TBP
geometry of [Ph3SiH2]− inferred by Rot et al.30 The anion in the b
polymorph exhibits crystallographic D3h symmetry, whilst that in
the monoclinic a form has approximate D3h symmetry, owing to
canting of the phenyl group planes with respect to the H–Si–H axis.
The hypervalent nature of silicon is demonstrated by the reduced
Fig. 2 Single-crystal X-ray structure of the monoclinic (a) form of 1
showing the Si–H · · · K interaction as a dotted line. All C–H hydrogen
˚
atoms have been omitted for clarity. Selected interatomic distances (A):
K(1A)–O(av) 2.796, K(1B)–O(av) 2.780, Si(1)–C(av) 1.927, Si(1)–H(1S)
1.55, Si(1)–H(2S) 1.55, Si(1)–K(1B) 4.717(1), K(1B)–H(2S) 3.21,
K(1A)–C(16) 3.371(3), K(1A)–C(17) 3.358(3). Angle ∠H(1S)–Si(1)–H(2S)
177.8◦.
˚
bond order in the H–Si–H unit, with Si–H bond lengths of 1.64 A
˚
in the trigonal form and 1.55 A in the monoclinic polymorph (cf.
272 | Dalton Trans., 2008, 271–282
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
greater distance in the monoclinic form arises from the fact that
the anion is no longer situated directly above the cation, in order
to facilitate the K+ · · · p interactions.
The structure of [K([15]crown-5)2][Ph3SiH2] (5) exhibits consid-
erable disorder in the [K([15]crown-5)2]+ cations and a molecule
of DME solvent (see ESI¶). The silane anion in 5 is structurally
very similar to its counterpart in 1b, and exhibits a perfect TBP
arrangement with all three phenyl groups in equatorial sites. There
are however, no K · · · H–Si short contacts of the type observed
in 1. The potassium ion is sandwiched between two disordered
[15]crown-5 moieties, which isolate the metal centre and prohibit
interaction between the potassium ion and the silicon hydride
ligands. The results suggests that the TBP geometry of the anion
is not dictated by the K · · · H–Si interactions observed in 1. The
˚
average Si–C bond length is 1.948 A, and this is comparable with
those of 1. The plane of the phenyl rings is aligned with the H–Si–H
moiety, akin to both polymorphs of 1. The hypervalent anions in
the crystal structure of 5 form 2-dimensional hexagonal channels
within which the cations reside (see ESI¶). It is noteworthy that
even in the absence of cation–anion interactions, the preferred
geometry of the anion is TBP with axial hydride ligands.
In light of the novel Si–H · · · K interaction observed in the
X-ray structure determined for 1, a neutron diffraction study of
1D was attempted. After exhaustive attempts, only crystals of
the monoclinic a-form were obtained. Cooling to 20 K revealed
diffuse lines, but sharper diffraction maxima were observed at
120 K, hence the data were collected at this higher temperature.
Experimental problems meant that only a low precision data set
was obtained. While the structure confirms the positions of the
hydride substituents, the precision is low and the detailed structure
is not presented. Both Si–D bond lengths measured from the
Fig. 3 EXSY spectrum of a sample containing 1 and Ph3SiH in THF-d8.
The features on and near-diagonal correspond to exchange, and the eight
off-diagonal features intersecting at 5.5 and 6.0 with 7.5 and 8.2 ppm
correspond to through-space coupling.
hydride transfer from 1 to Ph3SiH; alternatively it may involve
“[K([18]crown-6)]+H−” as an intermediate. Another possible ex-
change mechanism is that previously proposed by Corriu et al.,
in which an anionic intermediate in the form of a bridged dimer
was assumed to mediate exchange in reactions observed between
K[HSi(OR)4] and phenylsilanes.38
˚
neutron diffraction study are of the order of 1.6 A (Si–H(1D)
˚
1.598(15) and Si–H(2D) 1.597(15) A); this gives a K(1B)–H(1D)
The exchange process was further investigated by variable
temperature H NMR spectroscopy in THF-d8. The resonances
1
˚
distance of 3.17 A. Assuming the trigonal form of [K([18]crown-
6)][Ph3SiH2] (1) has the same Si–H bond length as that determined
for the monoclinic form, the K · · · H distance for the trigonal form
corresponding to the Si–H protons of 1 and Ph3SiH broadened
upon heating, but the low boiling point of the solvent (56 ◦C) pre-
vented the observation of coalescence and the extraction of ther-
modynamic parameters for the process. In addition, the relatively
high temperatures had a deleterious effect on the mixture, with
the 1H NMR spectrum after heating showing a significant number
of new, unassignable resonances. On leaving the sample to stand
over 14 d, crystals were isolated and characterised by single crystal
X-ray diffraction as the known Ph4Si.33 Lineshape analysis for the
hydride exchange process between [K([18]crown-6)][Ph3SiH2] (1)
and Ph3SiH with the aid of the MEXICO39 program permitted
calculation of the activation energy (Ea) for the exchange process
(see ESI¶). Analysis of the data using the Arrhenius equation
yielded an Ea of 81.5(5) kJ mol−1; using the Eyring equation gave
average values of DH‡ = 74.3(5) kJ mol−1 and DS‡ = 25(2) J
K−1 mol−1, and an activation free energy barrier of 67 kJ mol−1.
The positive value of the entropy of activation deduced from this
experiment suggests that a dissociative mechanism is in operation.
The bridging (associative) model suggested by Corriu et al.31
for redistribution of OR ligands would presumably correspond
to a negative DS‡; accordingly we propose an exchange process
involving simple hydride transfer from 1 to Ph3SiH. However
the entropy values presented here must be treated with some
caution, as upon completion of the experiment the sample had
degraded, with many small, unassignable peaks appearing along
˚
would be approximately 2.72 A, very much shorter than the sum
of the van der Waals radii of the two atoms.
Dynamic NMR spectroscopic studies
Attempts to prepare 1 using an excess of Ph2SiH2, with respect to
potassium and [18]crown-6, resulted in a mixture of hydrosilane
products. The 1H NMR spectrum of this mixture shows, in
addition to a broadened resonance at ca. d 6 ppm associated with
1
1, a broad signal at approximately d 5.5 ppm with J(29Si–1H) =
200 Hz, assigned to Ph3SiH. A resonance at ca. d 1.2 ppm with
1J(29Si–1H) = 82 Hz was also evident. The breadth of the peaks
associated with 1 and Ph3SiH implies that a dynamic exchange
process may be occurring. An EXSY experiment confirmed this,
indicating proton exchange between 1 and Ph3SiH, the species
1
with a resonance at d 1.2. 29Si and 29Si{ H} NMR spectroscopy
carried out on the sample gave results that were consistent with
1 and Ph3SiH. The species responsible for the resonance in the
1H NMR spectrum at d 1.2 is consistent with the known NMR
data for a [SiH3]− with 1J(29Si–1H) ca. 80 Hz.37 Deliberate addition
of Ph3SiH (1 equivalent) to a solution of 1 and examination of
the mixture by EXSY (Fig. 3) confirmed SiH proton exchange
between these species. This exchange could arise from simple
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 271–282 | 273
©

View Article Online
with a broad feature at ca. d 5 ppm. This could correspond to
the formation of some Ph2SiH2, which implies that our proposed
exchange process is not necessarily the only one occurring. There
are a few studies of intermolecular hydride exchange reported
for this type of system,38,40,41 but the thermodynamics of these
processes have not been explored. However, similar studies have
been carried out for intramolecular fluoride site exchange based on
a pseudorotation mechanism in pentacoordinate fluorosilicates.
Typical activation energy values for such processes are around
40 kJ mol−1.42
NMR spectrum recorded after heating and subsequent cooling
to room temperature showed an increase in the ratio of the
[Ph3SiH2]− : Ph3SiH intensities. We conclude from this study
that the counter-cation of 1 must be involved in the exchange
mechanism. Sequestering the K+ ion between a pair of [15]crown-
5 ligands dramatically decreases the exchange rate.
Reactivity of [K([18]crown-6)][Ph3SiH2]
Several elementary reactions were carried out to assess the
reactivity of 1 as a hydride transfer reagent. In each case, these
were carried out in NMR tubes, and the conclusions are based on
subsequent 1H NMR analysis.
If the proposed mechanism involving hydride exchange is
correct, the reaction of KH and Ph3SiH in the presence of 18-
crown-6 should produce 1, although difficulties have been noted
previously with this procedure.40 The reaction was attempted in an
NMR tube, using 1.25 to 1.5 equivalents of hydride with respect
A recrystallised sample of 1, produces a colourless solution
in THF. Addition of 1 equivalent of Ph3GeH in THF to the
solution results in immediate effervescence and the formation of a
bright yellow solution. The same colour and effects are obtained
when Ph3GeH is reacted with potassium metal in the presence
1
to the silane and crown ether, and was followed by H NMR
spectroscopy at various time intervals (see ESI Fig. S6¶). It is
clear from the spectra that 1 is indeed formed from the reaction
between KH and Ph3SiH, but that this reaction is very slow (days).
The rate of this reaction, then, is likely to be affected by the
exchange process discussed above: as soon as 1 is formed, it is
free to exchange hydrides with the parent silane, decreasing the
number of sites available for nucleophilic attack of H− (by KH).
The overall effect will be to decrease the rate of formation of 1.
When a preparative scale reaction was conducted without stirring,
1 was isolated by quenching the reaction with toluene after 24 h.
This appears to be the effect of disturbing the exchange process,
as 1 is relatively insoluble in the toluene–THF solvent system,
whereas Ph3SiH remains in solution.
With a view to exploring the structure of 1 in solution, we carried
out a series of NOESY and COSY experiments. The results show
that the ortho protons couple through space to the Si–H protons
on the phenyl groups; there is also some evidence of through-
space coupling to the methylene protons on the crown ether,
indicating preservation of the solid-state structure in solution. Low
temperature 1H NMR experiments did not show any broadening
of the Si–H resonance down to 180 K, the lowest temperature
achievable with the solvent.
1
of [18]crown-6. Subsequent H NMR analysis of this reaction
mixture reveals the disappearance of resonances for both the Ge–
H and the Si–H moieties from 1, and the appearance of a resonance
associated with the Si–H proton of Ph3SiH. Upon addition of HCl
to this mixture the Ge–H resonance reappears. These observations
are consistent with Ph3GeH being deprotonated by 1 to produce
the anion [Ph3Ge]−, on account of the higher electronegativity
of Ge with respect to Si.34 From a consideration of the relative
electronegativities of C and Si, the reaction of Ph3CH with 1
would be expected to deprotonate the tertiary carbon in an
analogous manner to that inferred from the reaction of 1 with
Ph3GeH. However, addition of Ph3CH in THF to a solution of 1
in THF causes no such reaction. No effervescence is observed, and
1
the reaction mixture takes on a blood-red colour. The H-NMR
spectrum of this mixture is not easy to interpret, and addition of
HCl to the reaction does not regenerate the starting material.
Similar results are obtained for the reaction between PhCCH
1
and 1—the resulting H NMR spectrum contains a multitude
of unidentifiable resonances, and addition of HCl to the reaction
does not regenerate the starting material.
A mixture of Ph3SiF and 1 appears to undergo a redistribution
reaction, producing [Ph3SiF2]− and [Ph3SiFH]−. Although no
products were isolated, the details of the reaction are based on
1H and 19F NMR spectroscopic evidence. The 1H NMR spectrum
of the reaction products clearly shows the presence of Si–H with
both a chemical shift (d 5.4) and coupling constant (1J(29Si–1H) =
196 Hz) consistent with Ph3SiH, but there is no evidence of 1. The
19F NMR spectrum shows only a trace of Ph3SiF, and the major
peak (d −101.7, 1J(29Si–19F) = 255 Hz) is consistent with [Ph3SiF2]−
as reported for the complex [K([2.2.2]cryptand)][Ph3SiF2]
(d −102.26 and 1J(29Si–19F) = 259 Hz) by Yamaguchi et al.43
Influence of the crown ether on [Ph3SiH2]−
An analogous procedure to that employed in the synthesis of
1 from the reaction between potassium metal and Ph2SiH2,
was carried out using 2 equivalents of [15]crown-5 in place of
1
[18]crown-6 in order to produce 5. The H NMR spectrum of
the product in THF-d8 is similar to that of 1, with a1J(29Si–
1H) of 132 Hz. Crystals of 5 were easily obtained from the
reaction solution, using a similar procedure to that described
for 1, as noted above. Changing the crown ether in this case
does not appear to have an effect on the formation of the
hypervalent anion. However, the nature of the crown ether does
have a significant influence on the exchange process observed for
Studies of [H2SiO(iPr)3]−
1
1 and Ph3SiH. An H NMR study of the exchange between 5
The only isolated dihydridosilicates reported in the literature
prior to this research are of the form M[H2Si(OR)3].26,31,44,45 In
order to compare the structures of 1 and the silicate anions
[H2Si(OR)3]− we repeated the synthesis of [H2Si(OiPr)3]− by
reaction of KH with HSi(OiPr)3, both in the presence and
absence of [18]crown-6. Despite many attempts with crystals
that appeared visually to be diffraction quality, we could not
and Ph3SiH revealed no significant broadening of either of the
Si–H resonances up to 325 K. A lineshape analysis of the spectrum
was not possible on account of the very slow exchange, and it
appears a different mechanism is in operation; this manifests itself
in a reduction of the intensity of the resonance from the [Ph3SiH2]−
species with respect to Ph3SiH as the temperature increases. A 1H
274 | Dalton Trans., 2008, 271–282
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
solve the structure of K[H2Si(OiPr)3] (6). However, more success
was enjoyed when [K([18]crown-6)]+ was used as the counter-
ion. Although, obtaining good quality crystals proved difficult
using THF as the solvent, its replacement with Et2O produced
high quality crystals, which were used for an X-ray diffraction
study. The structure of [K([18]crown-6][H2Si(OR)3] (7) shows
the anion to adopt a distorted TBP array, with both hydride
substituents in axial positions (Fig. 4) The isopropoxy groups
occupy equatorial positions in a stereochemistry that appears
to be at odds with Bent’s Rule and known “apicophilicity”,46,47
i.e., that the most electronegative atoms should be in the axial
positions. The geometry determined by our study is also contrary
to that predicted by Corriu and co-workers based on NMR
spectroscopic data.31 The anion exhibits short contacts from both
an isopropoxy oxygen atom and one of the axial hydride ligands
to the K+ cation, in contrast to the single Si–H · · · K+ interaction
seen in 1. This bidentate cation–anion arrangement is similar to
those seen for hypervalent fluorosilicates. The H–Si–H moiety
shows significant distortion from a true TBP arrangement, with a
substantial deviation from linearity (<HSiH angle 160.3◦). This
appears to be brought about by the very short K+ · · · H contact at
Scheme 1 Addition of HSi(OiPr)3 to [K([18]crown-6)][H2SiPh3] results
in formation of an alkoxysilicate anion, and a proton exchange involving
[K([18]crown-6)][H2SiPh3] and Ph3SiH.
DFT Calculations on hypervalent anions [Ph3SiH2]− and
[H2Si(OiPr)3]−
DFT calculations were carried out on the anions [Ph3SiH2]−
and [(iPrO)3SiH2]− to assess the relative stability of the three
available stereoisomers; i.e., with Si–H bonds in axial–axial (aa),
axial–equatorial (ae) and equatorial–equatorial (ee) positions. The
results for calculated energy differences in the stereoisomers of the
two anions are summarised in Tables 1 and 2, respectively.
˚
2.49 A. The Si–H2S unit shows a similarly short interaction with
the tertiary C–H on the isopropoxy group, with H · · · H distances
˚
of 2.27 and 1.91 A, respectively; these are significantly shorter than
˚
the sum of the van der Waals radii (2.4 A).
A Mulliken charge analysis of aa-[Ph3SiH2]− ascribes the
following partial charges: Si +0.34, H −0.27, Ph −0.27. In ae-
[Ph3SiH2]− the corresponding values are Si +0.35, Hax −0.24, Heq
−0.15. Hence the negative charge in [Ph3SiH2]− is delocalised over
all five ligands, and there are no grounds for assuming that the
axial location of the Si–H bonds is a consequence of electrostatic
repulsion between the hydride ligands. The closest non-bonded
−
˚
C · · · C contact in aa-[Ph3SiH2] is 3.40 A, twice the van der Waals
radius of carbon.34 In ae-[Ph3SiH2] , this is reduced to 3.08 A.
−
˚
The results show repulsion between the three phenyl ligands to
be a more important factor in determining the stereochemistry at
Si than charge repulsion. The relatively equal apportionment of
negative charge between the H and Ph ligands is consistent with the
lengthening of the K · · · H–Si interaction to facilitate the K+ · · · p
interaction in the monoclinic a-form of 1, in contrast to the shorter,
linear K · · · H–Si interaction in the trigonal polymorph (b). These
results accord with a recent bond energy analysis of [Ph3SiH2]−
and related hypervalent silicate species,2 in which the axial hydride
Fig. 4 X-Ray structure of K([18]crown-6)[H2Si(OiPr)3] (7). Selected
bond lengths: K(1)–Si(1) 3.653(2), K(1)–O(3) 2.799(4), K(1)–H(1S) 2.83,
Si(1)–O(1) 1.697(4), Si(1)–O(2) 1.751(4), Si(1)–O(3) 1.716(4), Si(1)–H(1S)
Table 1 Relative energies of stereoisomers of [Ph3SiH2]−, calculated by
DFT25
˚
1.52, Si(1)–H(2S) 1.49 A.
Geometry
Energy/hartrees
DE/kJ mol−1
aa (D3h)
ae (Cs)
ee
−985.65324
0
+36
To explore the similarities and differences between 1 and 7, we
carried out some simple studies involving mixing 1 and HSi(OiPr)3,
and in a separate experiment mixing [K([18]crown-6)][H2Si(OiPr)3]
and Ph3SiH. The results are summarised in Scheme 1, and
indicate that 7 is formed upon combination of HSi(OiPr)3 and
1, presumably through hydride attack on the trialkoxysilane,
which also produces Ph3SiH. The Ph3SiH thus formed goes on
to exchange with unreacted 1, vide supra. In contrast, when 7 is
combined with Ph3SiH, there is no evidence for the formation of 1.
Corriu et al.38 have previously reported hydride exchange between
K[HSi(OEt)4], Ph2SiH2 and Ph3SiH.
−985.63951
Converged to aa isomer
Table 2 Relative energies of stereoisomers of [(iPrO)3SiH2]−, calculated
by DFT
Geometry
Energy/hartrees
DE/kJ mol−1
aa (D3h)
ae (Cs)
ee (C2v)
−872.12488
−872.11702
−872.11902
0
+21
+15
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 271–282 | 275
©

View Article Online
ligands were shown to be stabilized considerably by the presence
of three electron-withdrawing phenyl groups.
An extensive conformational analysis for [(iPrO)3SiH2]− to
include the isopropoxy groups was not attempted due to excessive
demands on computer time, and the calculations were constrained
on the crystal structure conformations for the isopropoxy groups.
The results give a good indication of the preferred isomer with
respect to substituents directly bonded to the silicon centre and
this certainly agrees with the observed aa geometry in the crystal
structure.
In conclusion, the DFT and X-ray results show that the
preferred sites of the hydride substituents in the TBP geometry
for both anions are axial. The axial/equatorial arrangement
suggested by NMR spectroscopic data may be attributed to the
formation of cation–anion pairs in solution, with the inequivalence
of the two Si–H substituents arising from the interaction of one
hydride with the K+ ion.
Fig. 5 The crystal structure of [K([18]crown-6)][Ph3SiF2] (9) obtained
from the reaction of Ph3SiF and KH in the presence of [18]crown-6.
Further studies of hypervalent hydrido- and fluorosilicates
The reaction of KH with Ph3SiH in the presence of crown ethers
led us to study the reactions of KH with other triarylsilanes and to
alter the alkali metal. The combination of Na metal and [15]crown-
5 with Ph2SiH2 or PhSiH3 gives [Na([15]crown-5)][SiH3], which
will be reported separately along with other [SiH3]− species.29
An inseparable mixture of both [Na([18]crown-6)][Ph3SiH2] (8)
and [Na([18]crown-6)][SiH3] was recovered from the reaction of
Na and [18]crown-6 with both PhSiH3 and Ph2SiH2. The NMR
spectroscopic data for 8 are in good agreement with those obtained
for 1 with the SiH resonance at d 5.93 ppm and J(29Si–1H) =
1
131 Hz (cf. 5.95 and 130 Hz, respectively, for 1). The 29Si NMR
signal, at d −73.9 (cf. −74.0 for 1) appears as a triplet that collapses
to a singlet upon proton decoupling.
Fig. 6 X-Ray crystal structure of [K([18]crown-6)]2[(p-FC6H4)3SiH2]·
[K([18]crown-6)][(p-FC6H4)3Si(F)H] (12).
The reaction of KH with Ph3SiF and [18]crown-6 in THF
produced [K([18]crown-6)][Ph3SiF2] (9) as the main product.
The mixed [Ph3SiHF]− was not isolated. The same product was
obtained from reaction of Ph3SiF and KF in the presence of
[18]crown-6, and has been previously reported.42 The crystal
structure has not previously been determined,43 although in our
hands the compound appeared to be reasonably stable in air. The
structure of a single-crystal of 9 was determined by X-ray diffrac-
tion, and is shown in Fig. 5. The fluorosilicate 9 may be compared
with 1 and also with [K([2.2.2]cryptand)][Ph3SiF2]43 (CSD code48
WOFJAC). The anions all adopt a trigonal bipyramidal geometry
with approximately linear X–Si–X units (X = H, F). The Si–C
distances in the hydrides (a- and b-1) are slightly longer than in
the fluorides, which may be correlated with the fact that the Si–F
distances are longer than Si–H.
the asymmetric unit occupy very different environments: one has
a silicate anion on either side, with one anion interacting with the
potassium ion through a H–Si–H/F · · · K interaction, whilst the
other anion interacts with potassium through a p · · · K contact.
The remaining cation is complexed with two molecules of THF
through an O · · · K contact (one per face of the crown), and the
whole cation complex is disordered over two sites. The fluoride in
the axial site was modelled with a site occupancy factor of 0.5.
1
The reaction between KH and 10 was studied by H NMR
spectroscopy, in order to investigate the fluoride redistribution
reaction. The reaction was followed as a function of time (see ESI
Fig. S7¶), the results clearly show that the first species to form is the
dihydridosilicate [K([18]crown-6)][(p-FC6H4)3SiH2], 11 followed
by the fluorohydridosilicate [K([18]crown-6)][(p-FC6H4)3Si(F)H]
(13). The resonances of both species in the aromatic region are
broad, implying that an exchange similar to that found between 1
and Ph3SiH is occurring. Also evident in the 1H NMR spectrum
are sharp resonances in the aromatic region, whose appearance
coincides with the formation of the fluorohydrido species 13;
these then must arise through fluoride abstraction from the aryl
substituents.
The reaction between KH, [18]crown-6 and (p-FC6H4)3SiH
1
(10) was also investigated using H NMR spectroscopy, and the
structure of the isolated product was determined by single-crystal
X-ray diffraction. The initial product of the reaction is the hyper-
valent dihydridosilicate [K([18]crown-6)][(p-FC6H4)3SiH2] (11),
1
which was characterised by H, 19F and 29Si NMR spectroscopy.
However, upon crystallisation the compound undergoes some
decomposition to release fluoride, resulting in the formation of
mixed crystals of the hydride [(p-FC6H4)3SiH2]− and the flu-
oro/hydrido anion [(p-FC6H4)3Si(F)H]−, i.e. [K([18]crown-6)]2[(p-
FC6H4)3SiH2][(p-FC6H4)3Si(F)H] (12); Fig. 6. The two cations in
The reaction of KH and [18]crown-6 with a number of
other silanes was attempted; namely (p-CF3C6H4)3SiH (14),49 (p-
MeOC6H4)3SiH (15), (p-MeC6H4)3SiH (16) and (m-F2C6H3)3SiH
(17); either by the direct reaction of KH with the silane in the
presence of the crown ether, or by reaction with 1. In all cases it
276 | Dalton Trans., 2008, 271–282
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
proved impossible to isolate analytically pure samples because of
the highly reactive nature of the compounds and complications
arising from side reactions; hence characterisation is based solely
upon NMR spectroscopic data. The reaction between KH and
14 in the presence of [18]crown-6 gave a hypervalent fluorohy-
dridosilicate product, [K([18]crown-6)][(p-CF3C6H4)3SiHF] (18),
presumably by fluoride abstraction from the -CF3 group. The
hypervalent nature of 18 is demonstrated in the 1H NMR spectrum
by a doublet assigned to SiH centred at d 5.74 with 1J(29Si–1H) =
125 Hz, comparable to 1 but smaller than for Ph2SiFH (19)50
deuteriated solvents were then decanted from the sodium and
˚
stored over fresh sodium or pre-dried 4A sieves; both methods
produced dry solvents. KH purchased from Aldrich as a 60%
suspension in mineral oil was transferred into a flask fitted with
Young’s valves, and under an argon atmosphere the mineral oil was
removed via cannula. The hydride was then washed, typically with
6 portions of pentane, removing each aliquot in turn by cannula
to remove all the mineral oil. The hydride was then dried in vacuo
and stored in a glove box. KH was always used in conjunction with
all-glass equipment. Crown ethers were dissolved in a minimum
1
19
1J(29Si–1H) = 231 Hz or the parent silane 14. The H{ F}, 19F
quantity of dry degassed Et2O and stored over pre-dried 4A sieves
˚
1
and 19F{ H} spectra show that the doublet arises from coupling
for at least 24 h. The solution was decanted from the sieves and
the solvent removed under reduced pressure. The crown ethers
were stored in a glove box. Potassium was cut into small chunks
in a glove box (after removal of the oxidised metal surface) and
used immediately. DME was dried over Na/benzophenone and
1
with fluorine, with J(19F–1H) = 38 Hz. This is certainly smaller
than the coupling constant measured for 19 (1J(19F–1H) = 54 Hz;
1
see ESI Fig. S8¶). The H and 19F and NMR spectra contain
other resonances which cannot be assigned to 18. Since there are
no other SiH resonances, this species may tentatively be assigned
as [p-CF3C6H4)3SiF2]−. The 19F NMR spectrum supports this
assignment, with a peak at d −101.7 and satellites corresponding
to 1J(29Si–19F) = 260 Hz.
˚
stored in ampoules over 4A molecular sieves. All other operations
were carried out in Schlenk tubes baked in an oven then treated
with Me3SiCl followed by three pump–purge steps. THF-d8 was
˚
dried over Na and stored in ampoules over 4A molecular sieves.
˚
The reaction of KH with (p-MeOC6H4)3SiH (15) was carried
out in THF-d8 and followed by 1H NMR spectroscopy. The
reaction produced a hypervalent product tentatively assigned as
[K([18]crown-6)][(p-MeOC6H4)3SiH2] (20) in very small amounts.
This assignment is based on the characteristic splitting of the
phenyl protons, resulting in the ortho resonances being shifted to
higher frequency and the observation of a signal at approximately
[18]crown-6 was dried by dissolving in dry diethyl ether over 4A
molecular sieves. Microanalyses for C, H and N were carried out at
the University of North London. NMR tubes used were standard
5 mm diameter borosilicate glass fitted with a Young’s NMR valve,
and treated by exactly the same process as the reaction glassware.
Samples were loaded into the NMR tube inside a glove box. Care
was taken to avoid contact with the PTFE valve of the NMR tube
to prevent reaction of the sample with the valve surface. THF-
d8 used for the NMR was dried over sodium metal for 14 d then
1
d 6 ppm with J(1H–29Si) = 127 Hz assigned to the hypervalent
H–Si–H unit. On standing for 72 h, the amount of hypervalent
product had increased and a further product had appeared,
which is possibly of the form (R3Si)2O or (R3Si)2 (where R =
(p-MeC6H4)).
˚
stored over baked 4A molecular sieves, and degassed using freeze–
pump–thaw cycles. The solvent was injected into the NMR tube
via a rubber septum using a dried glass syringe.
1H NMR spectroscopic studies of the combination of (p-
MeC6H4)3SiH (16) with KH/[18]crown-6 in THF showed no
evidence of reaction after 48 h. Reaction of the silane 16 with 1 in
THF showed evidence of both 16 and Ph3SiH in solution, along
with some resonances consistent with a hypervalent [R3SiH2]−
species. However, it is not clear if this is 1 or [K([18]crown-6)][(p-
MeC6H4)3SiH2]. This reaction requires further study.
X-Ray crystallography
Suitable single crystals were mounted using silicon grease on a thin
glass fibre. Crystallographic measurements were carried out using
a Nonius KappaCCD equipped with graphite monochromatic
Mo-Ka radiation (k = 0.71073 A). The standard data collection
˚
Combination of (m-F2C6H3)3SiH (17) with KH/[18]crown-6 as
with the other fluoroarylsilanes apparently resulted in fluoride
abstraction from the aryl ligand to give [K([18]crown-6)][(m-
F2C6H3)3SiHF] (21). This assignment is based on low coupling
temperature was 120 K, maintained using an open flow N2
Oxford Cryostream device. Integration was carried out using the
Denzo-SMN51 package. Data sets were corrected for Lorentz and
polarisation effects and for the effects of absorption. Structures
were solved using direct methods in SHELXS-9752 and developed
using conventional alternating cycles of least-squares refinement
with SHELXL-9753 and difference Fourier synthesis with the aid
of the graphical interface program XSeed.54 In all cases non-
hydrogen atoms were refined anisotropically. C–H hydrogen atoms
were fixed in idealized positions and allowed to ride on the atom
to which they were attached. Hydrogen atom thermal parameters
were tied to those of the atom to which they were attached. Si–H
hydrogen atoms were located experimentally and their positional
and displacement parameters refined wherever possible. Molecular
graphics were produced using the program POV-Ray.55
1
1
constants of J(1H–29Si) = 130 Hz and J(1H–19F) = 33.2 Hz;
comparable to those found for 13 and 18.
Experimental
Materials and methods
Molecular sieves purchased from either Lancaster or Aldrich and
◦
were stored in an oven at 80 C. Prior to use they were baked in
◦
vacuo at 250 C for a minimum of 3 h. The solvents Et2O, THF,
toluene, DME and Bu2O were dried by distillation from sodium,
using benzophenone as an indicator, and degassed by bubbling
argon via a cannula through the solvent for 15–20 min. For NMR
spectroscopy the solvents THF-d8 and toluene-d8, purchased from
Goss, were stored over sodium for a period of approximately 2–3
d and were then degassed by a freeze–pump–thaw method. The
DFT Calculations
Geometry optimisations for the anions [Ph3SiH2]− and
[(iPrO)3SiH2]− were performed at the B3LYP level for
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 271–282 | 277
©

View Article Online
[(iPrO)3SiH2]−, using a 6-31+G* basis set on Si and O and 6-
31G* on C and H atoms to allow for charge redistribution.
Frequency calculations were carried out to confirm the structures
obtained were minima. RMS Forces were converged, but not
the displacements, which is expected as the lowest vibrational
frequencies are between 10 and 15 cm−1. All calculations were
carried out using Gaussian.56 For [Ph3SiH2]−25 the optimisations
were carried out at the B3LYP/6-31G* level.
2.6 mmol) was added dropwise by syringe over a period of 5 min
and the solution took on a yellow/green tinge eventually turning
deep amber. After addition of Ph2SiH2 was complete the volume
was reduced to 20 mL by evaporation and placed in a refrigerator
at 5 ◦C for a period of 2 d where colourless crystals formed.
1H NMR d(THF-d8): 3.4 (s, 24H), 6.9–7.1 (m, 9H), 8.1 (dd,
1
6H),. 13C{ H} NMR d(THF-d8): 70.42 (CH2), 125.6 (meta), 126.1
1
(ortho), 135.3 (para), 158.3 (ipso). 29Si{ H} NMR d(THF-d8):
−74.9 (quintet, 1J(29Si–2H) = 20.5 Hz. 2H NMR d(THF-d8): 5.98
(s, 1J(29Si–2H) = 20.5 Hz). MS (FAB), m/z (relative intensity): 303
(1), 259 (0.06), 184 (0.08), 39 (0.5).
Syntheses
[K([18]crown-6)][Ph3SiH2] (1). [18]crown-6 (1.7 g, 6.4 mmol)
dissolved in DME (30 ml) was transferred into a Schlenk tube
containing chopped potassium metal (0.35 g, 6.4 mmol) whilst
stirring to produce a deep blue solution. To this solution Ph2SiH2
(1.7 g, 6.4 mmol) was added dropwise by syringe over a period
of 5 min; the solution took on a yellow/green tinge, eventually
turning deep amber. After addition of Ph2SiH2 was complete,
the volume was reduced to 20 mL by evaporation under reduced
pressure, and the solution placed in a refrigerator at 5 ◦C for a
period of 2 d, whereupon colourless crystals had formed. The
solution was transferred to a freezer at −20 ◦C, where more
crystals were obtained. The pale yellow solution was decanted,
and the colourless crystals were isolated. Stored under nitrogen the
crystals developed a brown coating that was removed by washing
(p-MeC6H4)2SiH2 (3). (p-MeC6H4)2SiCl2 (10 g, 36 mmol)
dissolved in Bu2O (60 mL) was placed in a pressure equalised
dropping funnel and added dropwise over a period of approxi-
mately 1 h to a slurry of LiAlH4 (1.4 g, 36 mmol) in Bu2O (20 mL)
stirred over an ice bath. The reaction mixture was allowed to
warm to room temperature and stirred for 24 h after which the
solvent was evaporated under reduced pressure. CHCl3 (40 mL)
was added whilst stirring the mixture over a ice bath, followed by
dropwise addition of HCl (approx. 1–2 M) (care!) until no more
effervescence was observed. The mixture was allowed to warm
to room temperature. The organic layer was separated and the
aqueous layer was washed with 2 × 20 mL portions of CHCl3. The
ethereal washings were combined dried over MgSO4 and filtered.
The final product was obtained by distillation 70–80 ◦C/approx.
1
with benzene. H NMR d(THF-d8): 3.14 (s, 24H), 5.81 (s, 2H
1
1J(29Si–1H) = 131 Hz), 6.8–6.9 (m, 9H), 7.98 (d, 6H). 13C{ H}
1
0.25 mbar. Yield: 6.2 g (83%). H NMR d(CDCl3): 2.34 (s, 3H),
NMR d(THF-d8): 68.48 (CH2), 123.79 (meta), 124.31 (ortho),
1
3
4.89 (s, J(29Si–1H) = 198 Hz, 2H), 7.17 (d, J(1H–1H) = 8 Hz,
1
135.25 (para), 156.24 (ipso). 29Si{ H} NMR d(THF-d8): −74.01
4H), 7.48 (d, J(1H–1H) = 8 Hz, 4H). 13C{ H} NMR d(CDCl3):
21.5 (Me), 128.0 (para), 128.9 (meta), 135.4 (ortho), 139.8 (ipso).
29Si NMR d(CDCl3): −34 (tt, 1J(29Si–1H) = 198 Hz, 3J(29Si–1H) =
3
1
(s). 29Si NMR d(THF-d8): −73.99 (t, J(29Si–1H) = 131 Hz). IR
◦
m/cm−1 (neat): m(Si–H) 1524(m). Mp 122–124 C. Anal. calc. for
C30H41O6SiK:C, 63.89;H, 7.32. Found:C, 61.34;H, 7.32%. Several
samples were tested all giving low readings for carbon possibly due
to decomposition.
5.5 Hz. 29Si{ H} NMR d(CDCl3): −34 (s). MS (EI), m/z (relative
1
intensity): 212 (M+, 0.89), 197 (M − CH3), 120 (C7H7SiH, 1), 105
(C6H4SiH, 0.021). IR m/cm−1 (neat): 2135 m(Si–H).
Crystal data for 1a. C30H41O6SiK, M = 564.82, colourless
block, 0.8 × 0.8 × 0.5 mm, monoclinic, space group P21/n
˚
−3
(No. 14), a = 9.3861(5), b = 20.2252(10), c = 16.5969(10) A,
Ph2SiD2 (4). Ph2SiCl2 (24 g, 95 mmol) was added dropwise
to a slurry of LiAlD4 (2 g, 50 mmol) in Et2O (100 mL). After
addition was complete, the mixture was refluxed for 24 h allowed
to cool to room temperature and filtered through a glass sinter,
to remove LiAlCl4, and the solvent reduced to half volume where
more LiAlCl4 precipitated. The remaining liquor was decanted
from the LiAlCl4 and distilled. One fraction was collected at 68–
◦
3
˚
b = 103.540(3) , V = 3063.1(3) A , Z = 4, Dc = 1.225 g cm ,
˚
F000 = 1208, KappaCCD, Mo-Ka radiation, k = 0.71073 A, T =
125(2) K, 2hmax = 55.0◦, 16817 reflections collected, 6802 unique
(Rint = 0.0810). Final GooF = 1.018, R1 = 0.0527, wR2 = 0.1039,
R indices based on 4249 reflections with I >2r(I) (refinement on
F2), 355 parameters, 0 restraints. Lp and absorption corrections
applied, l = 0.251 mm−1.
◦
1
72 C at 0.6 mbar. The H NMR spectrum revealed that Et2O
was also present in a silane : ether ratio of 3 : 1, this may
be due to the formation of a complex, with silicon acting as a
Lewis acid and the ether oxygen as the Lewis base. The crude
product was dissolved in Et2O (50 mL) and HCl (2 M) added
dropwise until effervescence ceased, this was followed by addition
of distilled water (10 mL). The ethereal layer was separated and
the aqueous layer was washed with 3 × 30 mL portions of
Et2O. All the organic extracts were combined, dried over MgSO4
and filtered. After the ether had been evaporated under reduced
pressure the product was obtained by reduced pressure distillation
68–72 ◦C/approx 0.6 mbar. Yield: 13.1 g (74%). 1H NMR
Crystal data for 1b. C30H41O6SiK, M = 564.82, very pale green
cubic blocks, 0.25 × 0.25 × 0.25 mm, trigonal, space group R3m:R
˚
(No. 160), a = b = c = 9.4377(12), a = b = c = 97.956(6) A, V =
3
−3
˚
813.80(18) A , Z = 1, Dc = 1.152 g cm , F000 = 302, KappaCCD,
Mo-Ka radiation, k = 0.71073 A, T = 100(2) K, 2hmax = 49.9◦,
˚
4637 reflections collected, 1043 unique (Rint = 0.0506). Final
GooF = 1.132, R1 = 0.0330, wR2 = 0.0697, R indices based on
1000 reflections with I > 2r(I) (refinement on F2 77 parameters, 1
restraint. Lp and absorption corrections applied, l = 0.237 mm−1.
Absolute structure parameter57 = 0.54(7).
[K([18]crown-6)][Ph3SiD2] (1D). Complex 1D was prepared in
a similar procedure to 1. 18-crown-6 (0.68 g, 2.6 mmol) dissolved in
1,2-DME (20 mL) was transferred into a Schlenk tube containing
chopped potassium metal (0.1 g, 2.6 mmol) whilst stirring to
produce a deep blue solution. To this solution Ph2SiD2 (0.48 g,
1
d(CDCl3): 7.57 (m, 6H), 7.59 (d, 4H). 13C{ H}NMR d(CDCl3):
128.1 (para), 129.9 (meta), 131.42 (ipso), 135.7 (ortho). 29Si NMR
1
d(CDCl3): −33.9 (quintet of multiplets, J(29Si–2H) = 30.5 Hz).
1
1
29Si{ H} NMR d(CDCl3): −33.9 (quin, J(29Si–2H) = 30.5 Hz).
278 | Dalton Trans., 2008, 271–282
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
2H NMR d(CHCl3): 5.1 (s, 1J(29Si–2H) = 30.5 Hz) MS (EI), m/z
(relative intensity): 186 (1.00), 107 (0.97), 105 (0.24), 79 (0.08). IR
m/cm−1 (neat): m(Si–D) 1549(s).
[Na([18]crown-6)][Ph3SiH2] (8) and [Na([18]crown-6)][SiH3]
(mixture). [18]crown-6 (1.93 g, 7.3 mmol) dissolved in THF
(40 mL) was transferred into a Schlenk tube containing freshly
chopped sodium metal (0.17 g, 7.4 mmol) whilst stirring to produce
an intense blue solution. To this solution PhSiH3 (0.85 g, 7.9 mmol)
was added dropwise by syringe over a period of 5 min and the
solution took on a gold colour, eventually obtaining a green tinge.
The mixture began to turn cloudy after 1 h, and after stirring for
24 h it had separated into an off-white precipitate and an amber
solution. The solution was removed and retained. The precipitate
was washed with pentane (20 mL) followed by Et2O (20 mL) and
these washings were combined with the reaction solution, which
produced a pale yellow crystalline product when placed in the
freezer. Warm THF (200 mL) dissolved most of the precipitate: on
cooling this too produced crystals. Both batches of crystals showed
[K([15]crown-5)2][Ph3SiH2] (5). [15]crown-5 (1.7 g, 6.4 mmol)
dissolved in DME (30 mL) was transferred into a Schlenk tube
containing freshly chopped potassium metal (0.35 g, 6.4 mmol)
whilst stirring to produce a deep blue solution. To this solution
Ph2SiH2 (1.7 g, 6.4 mmol) was added dropwise by syringe over
a period of 5 min and the solution took on a yellow/green tinge
eventually turning deep amber. After addition of Ph2SiH2 was
complete the volume was reduced to 20mL by evaporation and
placed in a refrigerator at 5 ◦C for a period of 2 d, whereupon
colourless crystals formed. The solution was transferred to a
freezer at −20 ◦C where more crystals were obtained. The solution
(now pale yellow) was decanted and the colourless crystals were
isolated. Stored under nitrogen the crystals developed a brown
1
identical H NMR spectra. Total yield of crystalline products =
1
1.8 g. H NMR d(THF-d8): 1.27 (s, 4H*, J(29Si–1H) = 78.5 Hz),
1
coating that was removed by washing with benzene. H NMR
3.51 (s, 38H), 5.93 (s, 2H J(29Si–1H) = 131 Hz), 6.9–7.2 (m, 9H),
8.1 (d, 6H). 13C NMR d(THF-d8): 70.15 (s, CH2), 125.59 (s, CH,
meta), 126.11 (s, CH, para), 137.08 (s, CH, ortho), 156.24 (s, CSi,
d(THF-d8): 3.44 (s, 40H), 5.96 (s, 2H 1J(29Si–1H) = 131 Hz),
1
6.9–7.1 (m, 9H), 8.11–8.14 (m, 6H). 13C{ H} NMR d(THF-d8):
68.79 (CH2), 125.57 (meta), 126.05 (para), 137.09 (ortho), 158.20
1
ipso). 29Si{ H} NMR d(THF-d8): −73.90 (s). 29Si NMR d(THF-
1
(ipso). 29Si{ H} NMR d(THF-d8): −74.01 (s). 29Si NMR d(THF-
d8): −73.86 (t, J(29Si–1H) = 131 Hz). *Given as 4 as the ratio of
d8): −73.99 (t, J(29Si–1H) = 131 Hz). Anal. calc. for C38H57O10SiK:
C, 61.59; H, 7.75. Found: C, 61.44; H, 7.39%.
[SiH3]−/[Ph3SiH2] protons.
[K([18]crown-6)][Ph3SiF2] (9). KF (0.15 g, 2.6 mmol) was
placed along with [18]crown-6 (0.68 g, 2.6 mmol) in a Schlenk
tube and suspended in THF (40 mL). To this was added Ph3SiF
(0.63 g, 2.3 mmol) in THF (30 mL) whilst stirring. The mixture
was left to stir for 24 h after which hexane was carefully added
to give a 3 : 1 THF–hexane layered system. Upon standing, white
crystals appeared. Yield. 1.13 g, 77%. 1H NMR d(THF-d8): 3.31
(s, 24H), 6.84–6.95 (m, 9H), 7.91–7.95 (m, 6H),. 13C NMR d(THF-
d8): 70.50 (s, C12H24O6), 125.70 (s, broad, meta), 135.41 (s, para),
137.94 (t, 3J(13C–19F) = 7.46 Hz), 152.10 (t, 2J(13C–19F) ≈ 40 Hz).
Crystal data for 5. C38H57O10SiK, M = 741.049, colourless
block, 0.70 × 0.50 × 0.40 mm, orthorhombic, space group
˚
Pnma (No. 62), a = 37.820(8), b = 12.712(3), c = 17.536(4) A,
V = 8431(3) A , Z = 8, Dc = 1.167 g cm−3, F000 = 2960,
3
˚
˚
KappaCCD, Mo-Ka radiation, k = 0.71073 A, T = 120(2) K,
2hmax = 50.0◦, 35476 reflections collected, 7703 unique (Rint
=
0.1724). Final GooF = 1.262, R1 = 0.1319, wR2 = 0.3374, R
indices based on 4149 reflections with I >2r(I) (refinement on
F2), 541 parameters, 2 restraints. Lp and absorption corrections
applied, l = 0.205 mm−1.
1
19F NMR d(THF-d8): −100.3 (s, 1J(29Si–19F) = 255 Hz). 19F{ H}
[K([18]crown-6][H2Si(OR)3]
(7). [18]crown-6
(1.02
g,
1
1
NMR d(THF-d8): −100.3 (s, J(29Si–19F) = 255 Hz). 29Si{ H}
3.86 mmol) was dissolved in Et2O (15 mL) and added to a
slurry of KH (0.16 g, 4.0 mmol) in Et2O (5 mL). (PriO)3SiH31
(0.76 g, 3.70 mmol) was added dropwise to the KH/[18]crown-6
mixture. Initially a very small amount of gas was evolved; possibly
SiH4. The reaction mixture was stirred for 2 h and placed in a
refrigerator, whereupon colourless crystals appeared which were
harvested by removing the supernatant liquor and washing with
2 (5 mL) portions of pentane. The remaining supernatant was
saved and pentane added to yield more crystals. Total yield 1.25 g
(66%). 1H NMR d(C6D6): 1.37 (d, 18H, 3J(1H–1H) = 6 Hz), 3.16
NMR d(THF-d8): −107.7 (t, J(29Si–19F) = 253 Hz). 29Si NMR
1
d(THF-d8): −107.7 (t, 1J(29Si–19F) = 255 Hz). Anal. calc. for
C30H39F2KO6Si: C, 59.97; H, 6.54. Found: C, 59.69; H, 6.80%.
Crystal data for [K([18]crown-6)][Ph3SiF2]. C30H39F2KO6Si,
M = 600.80, colourless block, 0.50 × 0.30 × 0.30 mm, monoclinic,
space group P21/c (No. 14), a = 9.5474(3), b = 37.429(3), c =
◦
3
˚
˚
17.6357(8) A, b = 103.016(16) , V = 6140.2(6) A , Z = 8, Dc =
1.300 g cm−3, F000 = 2544, KappaCCD, Mo-Ka radiation, k =
0.71073 A, T = 115(2) K, 2hmax = 55.0◦, 24097 reflections collected,
˚
11948 unique (Rint = 0.1088). Final GooF = 1.091, R1 = 0.0816,
wR2 = 0.1058, R indices based on 7304 reflections with I > 2r(I)
(refinement on F2), 722 parameters, 0 restraints. Lp and absorption
corrections applied, l = 0.264 mm−1.
1
(s, 24H), 4.42 (s, 2H, J(29Si–1H) = 222 Hz), 4.61 (septet, 3H,
1
3J(1H–1H) = 6 Hz)). 13C{ H} NMR d(C6D6): 27.84 (CH3), 62.68
1
(CH), 70.41 (CH2). 29Si{ H} NMR d(C6D6): −81.1 (s). 29Si NMR
d(THF-d8): −81.1 (t, J(29Si–1H) = 222 Hz).
Crystal data for 7. C21H47KO9Si, M = 510.78, colourless block,
0.30 × 0.20 × 0.20 mm, orthorhombic, space group P212121 (No.
(p-FC6H4)3SiH (10). Mg turnings (2.6 g, 107 mmol) were
placed in a round-bottom flask and stirred under an Ar atmo-
sphere until a reflective coating had been applied to the glass
surface, when Et2O (60 mLl) was introduced. To this was added
1,4-C6H4BrF (10.5 ml, 16.7 g, 95.4 mmol) dissolved in Et2O
(100 mL) dropwise at a rate to maintain steady reflux, the colour of
the mixture changing from colourless to green to deep brown over
this period. After addition the mixture was heated to maintain a
gentle reflux for 1 h and allowed to cool to room temperature.
HSiCl3 (3.0 ml, 4.0 g, 30.0 mmol) was dissolved in Et2O (100 mL)
˚
19), a = 9.2574(9), b = 15.1610(12), c = 20.5984(17) A, V =
3
−3
˚
2891.0(4) A , Z = 4, Dc = 1.174 g cm , F000 = 1112, Mo-Ka
radiation, k = 0.71070 A, T = 145(2) K, 2hmax = 50.0◦, 10561
˚
reflections collected, 4928 unique (Rint = 0.0809). Final GooF =
1.056, R1 = 0.0677, wR2 = 0.1453, R indices based on 3533
reflections with I >2r(I) (refinement on F2), 300 parameters, 0
restraints. Lp and absorption corrections applied, l = 0.266 mm−1.
Absolute structure parameter57 = 0.00(8).
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 271–282 | 279
©

View Article Online
and added dropwise to the brown mixture over 1 h. After addition
of the silane was complete, the mixture was stirred for 30 min
and dilute HCl was added dropwise to neutralise excess Grignard.
The ether layer was separated and washed with dilute HCl and
dried over MgSO4 to produce a green/yellow solution which when
filtered through charcoal to give a pale yellow solution. The Et2O
was removed under reduced pressure to give a pale yellow oil.
The final product was obtained by subliming at 300 ◦C at 0.6
mbar. The product was isolated as a colourless oil that crystallises
over a period of 2–3 d at room temperature. Yield 6.97 g (75%).
cool to room temperature. HSiCl3 (6.4 g (6.4 ml), 47.2 mmol) was
dissolved in Et2O (80 ml) and added dropwise to the brown mixture
over 1 h. After addition of the silane was complete, the mixture
was stirred for 30 min before pouring over ice–HCl. The ether layer
was separated, washed with dilute HCl and dried over MgSO4 to
produce a green/yellow solution. The Et2O was removed under
reduced pressur◦e to give the crude product. The product was
1
distilled at 295 C at approx. 0.5 mbar. Yield 13.6 g (82%). H
NMR d(CD2Cl2): 3.78 (s, 9H), 5.38 (s, 1J(29Si–1H) = 196 Hz, 1H),
3
3
6.91 (d, J(1H–1H) = 8.6 Hz, 6H), 7.46 (d, J(1H–1H) = 8.6 Hz,
6H). 13C NMR d(CDCl3): 54.62 (s, Me), 113.41 (s, meta), 124.46 (s,
ipso), 136.72 (s, ortho), 160.71 (s, para). 29Si NMR d(CDCl3): −19.0
1
1H NMR d(CDCl3): 5.52 (s, J(29Si–1H) = 201 Hz, 1H), 6.85–
6.95 (m, 6H), 7.30–7.35 (m, 6H). 13C NMR d(CD2Cl2): 115.8 (d,
2J(19F–13C) = 19.9 Hz, meta), 129.0 (d, 4J(19F–13C) = 3.6 Hz ipso),
(dm, 1J(29Si–1H) = 196 Hz). 29Si{ H} NMR d(CDCl3): −19.0 (s).
1
3
1
138.1, (d, J(19F–13C) = 7.7 Hz, ortho), 164.7 (d, J(19F–13C) =
MS (EI), m/z (relative intensity): 350 (1), 242 (0.97). IR m/cm−1
(HCBD mull): m(Si–H) 2138(s). Anal. calc. for (CH3OC6H4)3SiH
C, 71.96: H, 6.33. Found C, 71.71; H, 6.46%.
249.5 Hz, para). 19F{ H} NMR d(CD2Cl2): −111.0 (s). 19F NMR
1
d(CD2Cl2): −111.0 (m). 29Si{ H} NMR d(CDCl3): −19.1 (s). 29Si
1
NMR d(CDCl3): −19.1 (dt, 1J(29Si–1H) = 201 Hz, 3J(29Si–1H) =
5.5 Hz. MS (EI), m/z (relative intensity): 123 (0.18), 218 (1), 219
(0.68), 313 (0.35), 314 (0.34). IR m/cm−1 (HCBD mull): 2146 (s)
m(Si–H). Anal. calc. for (FC6H4)3SiH C, 68.77; H, 4.17. Found C,
68.67; H, 4.28%. Mp 42–44 ◦C.
(p-MeC6H4)3SiH (16). HSiCl3 (4.5 g, 33.2 mmol) was
dissolved in Et2O (10 mL) and added dropwise to a 4-
MeC6H4MgBr/Et2O solution (100 mL, 1 M (100 mmol of
reagent)) at room temperature whilst stirring. Near to complete
addition of the silane the formation of an off-white precipitate
was evident, the quantity of which increased as the addition was
completed. Excess Grignard reagent was neutralised by dropwise
addition of HCl (1–2 M). The organic layer was separated and the
aqueous layer was washed with 3 × 40 mL portions of Et2O. The
organic extracts were combined, dried over MgSO4 and filtered.
Et2O was removed under reduced pressure and a yellow oily
suspension resulted. The crude product was recrystallised from
hot petroleum spirit (bp 60–80 ◦C), then placed in a freezer at
−40 ◦C. Yield: 7.2 g (72%). 1H NMR d(CDCl3): 2.35 (s, 9H), 5.41
[K([18]crown-6)]2[(p-FC6H4)3SiH2]·[(FC6H4)3Si(F)H]
(12).
KH (0.060 g, 1.5 mmol) was suspended in THF (15 mL) and
stirred. [18]crown-6 (0.43 g, d 1.6 mmol) was dissolved in THF
(15 ml) and added to the KH suspension; to this was then added
(4-C6H4F)3SiH (0.51 g, 1.6 mmol) dissolved in THF (5 mL). The
mixture was stirred for 4 h, after which pentane was carefully
added to give a 3 : 1 THF–pentane layered system. Upon standing
for 72 h at approx. 5 ◦C the solution produced colourless crystals.
1
Yield 0.54 g, 54%. H NMR d(THF-d8): 3.46 (s, 24H), 5.92 (s,
1
1
(s, J(29Si–1H) = 198 Hz, 1H), 7.18 (d, J(1H–1H) = 8 Hz, 6H),
2H J(29Si–1H) = 132 Hz), 6.74 (m, 6H), 8.07 (m, 6H). 13C NMR
7.46 (d, J(1H–1H) = 8 Hz, 6H). 13C NMR d(CDCl3): 21.55 (s),
1
2
d(THF-d8): 70.49 (s), 112.06 (d, J(19F–13C) = 19.1 Hz meta),
128.8 (s), 130.1 (s), 135.8 (s), 139.6 (s). 29Si{ H} NMR d(CDCl3):
1
3
4
135.36 (d, J(19F–13C) = 6.3 Hz, ortho), 141.89 (d, J(19F–13C) =
−18.8 (s). 29Si NMR d(CDCl3): −18.8 (dm, 1J(29Si–1H) = 198 Hz,
3J(29Si–1H) = 5 Hz). MS (EI), m/z (relative intensity): 167 (1), 210
(0.25), 91 (0.15), 302 (0.06). IR m/cm−1 (Nujol mull): 2116 m(Si–H).
Mp = 80–82 ◦C. Anal. calc. for (CH3C6H4)3SiH C, 83.38: H, 7.33.
Found C, 83.46; H, 7.39%.
3.6 Hz, ipso), 161.15 (d, J(13C–19F) = 241 Hz, para). 19F{ H}
NMR d(THF-d8): −120.0 (s). 19F NMR d(THF-d8): −120.0 (m).
1
1
29Si{ H} NMR d(THF-d8): −75.1 (s). 29Si NMR d(THF-d8):
1
−75.1 (t, broad, 1J(29Si–1H) = 135 Hz).
Crystal data for 12 [K([18]crown-6)]2[(FC6H4)3SiH2]·
[(FC6H4)3Si(F)H]·ca.2THF. C70.5H91F6.5K2O14Si2, M =1420.32,
colourless block, 0.30 × 0.20 × 0.20 mm, monoclinic, space group
(3,5-F2C6H3)3SiH (17). Mg turnings (1.26 g, 51.8 mmol)
were placed in a round-bottom flask and stirred under an Ar
atmosphere until a reflective coating had been applied to the glass
surface, when Et2O (40 mL) was slowly introduced. To this was
added 1-Br, 3,5-C6H2F2 (9.16 g, 47.5 mmol) dissolved in Et2O
(60 mL) dropwise at a rate to maintain steady reflux, the colour of
the mixture changing from colourless to green to brown over this
period. After the addition the mixture was heated to maintain a
gentle reflux for 1 h and allowed to cool to room temperature.
HSiCl3 (0.76 mL, 2.12 g, 15.6 mmol) was dissolved in Et2O
(60 mL) and added dropwise to the brown mixture over 1 h. After
addition of the silane was complete, the mixture was stirred for
30 min and dilute HCl was added dropwise to neutralise excess
Grignard. The ether layer was separated, washed with dilute HCl
and dried over MgSO4 to produce a green/yellow solution which
when filtered through charcoal gave a pale yellow solution. The
Et2O was removed under reduced pressure to give a pale yellow
oil. The final product was obtained by distillation at 300 ◦C at
˚
−3
C2/c (No. 15), a = 43.5927(18), b = 9.6984(5), c = 40.2731(18) A,
◦
3
˚
b = 120.188(2) , V = 14717.5(12) A , Z = 8, Dc = 1.282 g cm ,
˚
F000 = 6004, KappaCCD, Mo-Ka radiation, k = 0.71073 A, T =
120(2) K, 2hmax = 50.0◦, 27159 reflections collected, 12445 unique
(Rint = 0.0734). Final GooF = 1.121, R1 = 0.0935, wR2 = 0.1490,
R indices based on 8745 reflections with I >2r(I) (refinement on
F2), 840 parameters, 0 restraints. Lp and absorption corrections
applied, l = 0.238 mm−1. The Si–F fluorine atom was modelled at
0.5 occupancy because of apparent H/F disorder.
(p-MeOC6H4)3SiH (15). Mg turnings (3.8 g, 156 mmol)
were placed in a round-bottom flask and stirred under an Ar
atmosphere until a reflective coating had been applied to the glass
surface. To this was added dropwise 1,4-MeOC6H4Br (28.04 g,
18.8 mL, 150 mmol) dissolved in Et2O (80 mL) over a period of
1 h, the colour of the mixture changing from colourless to green
to deep brown over this period. After the addition the mixture
was heated to maintain a gentle reflux for 4 h and allowed to
1
approximately 0.6 mbar. Yield 2.2 g (73%). H NMR d(CDCl3):
1
5.44 (s, J(29Si–1H) = 213 Hz, 1H), 6.92–6.98 (m, 3H), 7.03–7.1
280 | Dalton Trans., 2008, 271–282
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
(m, 6H). 13C NMR d(CDCl3): 106.7 (t, 2J(19F–13C) = 26 Hz, para),
118.3 (m, ortho), 135.7, (t, 3J(19F–13C) = 5.4 Hz, ipso), 163.6 (dd,
11 S. Lachaize and S. Sabo-Etienne, Eur. J. Inorg. Chem., 2006, 2115.
12 A. L. Osipov, S. F. Vyboishchikov, K. Y. Dorogov, L. G. Kuzmina,
J. A. K. Howard, D. A. Lemenovskii and G. I. Nikonov, Chem.
Commun., 2005, 3349.
3
1
1J(19F–13C) = 252 Hz, J(19F–13C) 11 Hz, meta). 19F{ H} NMR
1
d(CDCl3): −109.65 (s). 19F NMR d(CDCl3): −109.65 (m). 29Si{ H}
13 S. K. Ignatov, N. H. Rees, B. R. Tyrrell, S. R. Dubberley, A. G.
Razuvaev, P. Mountford and G. I. Nikonov, Chem.–Eur. J., 2004, 10,
4991.
14 I. Atheaux, F. Delpech, B. Donnadieu, S. Sabo-Etienne, B. Chaudret,
K. Hussein, J. C. Barthelat, T. Braun, S. B. Duckett and R. N. Perutz,
Organometallics, 2002, 21, 5347.
NMR d(CDCl3): −17.4 (m). 29Si NMR d(CDCl3): −17.4 (d of
1
multiplets, J(29Si–1H) = 212 Hz). Mr = 368.348 g mol−1. MS
(EI), m/z (relative intensity): 368 (0.58), 254 (1).
15 G. I. Nikonov, L. G. Kuzmina and J. A. K. Howard, J. Chem. Soc.,
Dalton Trans., 2002, 3037.
16 W. Scherer, G. Eickerling, M. Tafipolsky, G. S. McGrady, P. Sirsch and
N. P. Chatterton, Chem. Commun., 2006, 2986.
Conclusions
We have shown that although reactive, hypervalent hydridosilicates
of the form [R3SiHX]− (where X = H or F) are amenable to study
and are isolable; hence they can no longer be thought to exist
only as transitory intermediates in reactions. The hypervalent H–
Si–H unit does not seem to engage in intermolecular hydrogen
bonding, but prefers instead to interact in an electrostatic manner
with a metal cation forming a hydride bridge; when a suitable
cation is not available for this type of interaction a TBP ar-
rangement with axial hydrides persists. Qualititative observations
suggest that the sequestered cation of 5 leads to a more air-
and moisture-stable compound. Accessibility of the cation also
appears to enhance the proclivity of [Ph3SiH2]− to engage hydride
exchange reactions. We have also shown that although silicon has
a propensity to become hypervalent, this appears to be somewhat
dependent on the substituents attached to the silicon centre.
NMR spectroscopic data for the hypervalent hydridosilicates and
fluorosilicates reported here are remarkably consistent across all
of the systems, suggesting a similar TBP arrangement with axial
H–Si–H and H–Si–F units in each case. In all of the systems ligand
redistribution reactions of hydride, fluoride and phenyl groups are
highly prevalent.
17 M. Ichinohe, A. Sekiguchi, M. Takahashi and H. Sakurai, Bull. Chem.
Soc. Jpn., 1999, 72, 1905.
18 A. Sekiguchi, M. Ichinohe, M. Takahashi, C. Kabuto and H. Sakurai,
Angew. Chem., Int. Ed. Engl., 1997, 36, 1533.
19 B. Goldfuss, P. V. Schleyer, S. Handschuh, F. Hampel and W. Bauer,
Organometallics, 1997, 16, 5999.
20 K. Junge, E. Popowski, R. Kempe and W. Baumann, Z. Anorg. Allg.
Chem., 1998, 624, 1369.
21 J. Schneider, E. Popowski, K. Junge and H. Reinke, Z. Anorg. Allg.
Chem., 2001, 627, 2680.
22 T. Matsuo, H. Watanabe, M. Ichinohe and A. Sekiguchi, Inorg. Chem.
Commun., 1999, 2, 510.
23 H. Pritzkow, T. Lobreyer, W. Sundermeyer, N. Hommes and P.
Vonragueschleyer, Angew. Chem., Int. Ed. Engl., 1994, 33, 216.
24 N. Wiberg, W. Niedermayer, H. Noth and M. Warchhold, J. Organomet.
Chem., 2001, 628, 46.
25 M. J. Bearpark, G. S. McGrady, P. D. Prince and J. W. Steed, J. Am.
Chem. Soc., 2001, 123, 7736.
26 R. J. P. Corriu, C. Guerin, B. Henner and Q. J. Wang, Organometallics,
1991, 10, 2297.
27 R. Corriu, C. Guerin, B. Henner and Q. Wang, J. Organomet. Chem.,
1989, 365, C7.
28 B. Becker, R. Corriu, C. Guerin, B. Henner and Q. Wang, J. Organomet.
Chem., 1989, 359, C33.
29 P. D. Prince, Ph.D. Thesis, King’s College London, 2003.
30 N. Rot, T. Nijbacker, R. Kroon, F. J. J. de Kanter, F. Bickelhaupt, M.
Lutz and A. L. Spek, Organometallics, 2000, 19, 1319.
31 R. J. P. Corriu, C. Guerin, B. J. L. Henner and Q. J. Wang,
Organometallics, 1991, 10, 3574.
32 J. Allemand and R. Gerdil, Cryst. Struct. Commun., 1979, 8, 927.
33 C. Glidewell and G. M. Sheldrick, J. Chem. Soc. A, 1971, 3127.
34 A. Bondi, J. Phys. Chem., 1964, 68, 441.
35 S. R. Gunn and L. G. Green, J. Am. Chem. Soc., 1958, 80, 4782.
36 J. C. Ma and D. A. Dougherty, Chem. Rev., 1997, 97, 1303.
37 E. Weiss, G. Hencken and H. Kuhr, Chem. Ber., 1970, 103, 2868.
38 R. J. P. Corriu, C. Guerin, B. J. L. Henner and Q. Wang, J. Organomet.
Chem., 1992, 439, C1.
Acknowledgements
We thank the EPSRC for a studentship (to PDP), Dr Matthias
Gutmann (Rutherford Appleton Lab.) for his efforts with the
neutron structure of 1D, the Royal Society for provision of a
microscope facility, and Prof. Alex Bain (McMaster University,
Ontario, Canada) for help with analysis of the NMR experiments
and for the MEXICO program.
39 A. D. Bain, MEXICO 3.0, 2002, http://www.chemistry.mcmaster.
ca/∼bain/mexmanc.html.
40 J. L. Brefort, R. Corriu, C. Guerin and B. Henner, J. Organomet. Chem.,
1989, 370, 9.
41 B. Becker, R. J. P. Corriu, C. Guerin and B. J. L. Henner, J. Organomet.
Chem., 1989, 369, 147.
Notes and references
42 R. Damrauer, B. Oconnell, S. E. Danahey and R. Simon,
Organometallics, 1989, 8, 1167.
43 S. Yamaguchi, S. Akiyama and K. Tamao, Organometallics, 1999, 18,
2851.
44 R. J. P. Corriu and J. C. Young, Hypervalent Silicon Compounds, in
The Silicon Heteroatom Bond, ed. Z. Rappoport and S. Patai, John
Wiley & Sons, Chichester, 1991, p. 1.
45 B. Becker, R. J. P. Corriu, C. Guerin, B. Henner and Q. Wang,
J. Organomet. Chem., 1989, 368, C25.
46 H. A. Bent, Chem. Rev., 1961, 61, 275.
1 G. S. McGrady and J. W. Steed, Hypervalent Compounds, in Encyclo-
pedia of Inorganic Chemistry 2, ed. R. B. King, Wiley, Chichester, 2005,
p. 1938.
2 E. P. A. Couzijn, A. W. Ehlers, M. Schakel and K. Lammertsma, J. Am.
Chem. Soc., 2006, 128, 13634.
3 R. Z. Mu, Z. G. Liu, Z. Q. Liu, L. Yang, L. M. Wu and Z. L. Liu,
J. Chem. Res. (S), 2005, 469.
4 M. Nakash, D. Gut and M. Goldvaser, Inorg. Chem., 2005, 44, 1023.
5 A. K. Sahoo, T. Oda, Y. Nakao and T. Hiyama, Adv. Synth. Catal.,
2004, 346, 1715.
47 R. G. Cavell, J. A. Gibson and K. I. The, J. Am. Chem. Soc., 1977, 99,
6 M. Nakash, M. Goldvaser and I. Goldberg, Inorg. Chem., 2004, 43,
5792.
7841.
48 F. H. Allen, Acta Crystallogr., Sect. B, 2002, 58, 380.
49 R. A. Benkeser and F. J. Riel, J. Am. Chem. Soc., 1951, 73, 3472.
50 H. Ohmura, H. Matsuhashi, M. Tanaka, M. Kuroboshi, T. Hiyama, Y.
Hatanaka and K. Goda, J. Organomet. Chem., 1995, 499, 167.
51 Z. Otwinowski and W. Minor, Methods Enzymol., 1997, 276,
307.
7 L. N. Parshina, L. A. Oparina, M. Y. Khil’ko and B. A. Trofimov,
J. Organomet. Chem., 2003, 665, 246.
8 G. I. Nikonov, Adv. Organomet. Chem., 2005, 53, 217.
9 M. J. Paterson, N. P. Chatterton and G. S. McGrady, New J. Chem.,
2004, 28, 1434.
10 S. F. Vyboishchikov and G. I. Nikonov, Chem.–Eur. J., 2006, 12, 8518.
This journal is
The Royal Society of Chemistry 2008
Dalton Trans., 2008, 271–282 | 281
©

View Article Online
52 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467.
53 G. M. Sheldrick, in SHELXL-97, University of Go¨ttingen, Germany,
1997.
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 01,
Development Version (Revision B.01), Gaussian, Inc., Pittsburgh, PA,
2001.
54 L. J. Barbour, J. Supramol. Chem., 2001, 1, 189.
55 C. J. Cason, POV-Ray, 2002, http://www.povray.org/.
56 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, J. J. A. Montgomery, T. Vreven, K. N. Kudin,
J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X.
Li, J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo,
57 H. D. Flack, Acta Crystallogr., Sect. A, 1983, 39, 876.
282 | Dalton Trans., 2008, 271–282
This journal is
The Royal Society of Chemistry 2008
©