Niobocene hydrides with functionalised silyl ligands. Tuning of interligand hypervalent interactions (IHI) M-H → Si-X

Article abstract of DOI:10.1039/b110998g

This paper presents syntheses and X-ray structure determinations of a variety of mono- and bis-(silyl) niobocene hydride complexes with functionalised silyl ligands. Complexes of the type [NbCp₂(SiR₃)H₂] (SiR₃ = SiMe₂Cl (3), Si(OEt)₃, SiMe₂OEt (4)) can be prepared conveniently by heating the precursor [NbCp₂H₃] in the presence of the corresponding silane HSiR₃. As the electrondonating ability of the R group increases, the corresponding mono(silyl) derivatives [NbCp₂(SiR₃)H₂] are either unstable (SiR₃ = SiMe₃) or do not exist at all (SiR₃ = SiEt₃, Si(OCH₂CH₂)₃N). The previously described bis(silyl) complex [NbCp₂(SiMe₂H)₂H] easily reacts with half an equivalent of iodine, giving the functionalised derivative [NbCp₂(SiMe₂I)₂H] (10). A labelling experiment for the analogous reaction with bromine establishes that an electrophile attacks the more accessible Si-H bond rather than the more electron-rich Nb-H bond. Bis(silyl) complexes [NbCp₂(SiMe₂X)₂H] (X= Br (8), I (10), OTfl (11)) can be conveniently prepared by the reaction of [NbCp₂(SiMe₂Cl)₂H] (2) with the corresponding silane XSiMe₃. Reactions of 2 with fluorinating reagents such as [CPh₃][BF₄], BF₃·OEt₂ and AgBF₄ do not work well for the synthesis of the compound [NbCp₂(SiMe₂F)₂H]. Nucleophilic chlorine substitution in 2 under the action of lithium amides and alkoxides does not occur, whereas reaction with soft reagents such as MeLi, LiPPh₂ and LiSPh leads to the corresponding derivatives [NbCp₂(SiMe₃)₂H] and [NbCp₂(SiMe₂EPh_n)₂H] (E = S, n = 1 (14); E = P, n = 2 (15)). Methanolysis of 10 gives high yields of the derivative [NbCp₂(SiMe₂OMe)₂H] (18). The mono(silyl) precursor [NbCp₂(SiMe₂Cl)H₂] (3) reacts with LiOEt affording a mixture of [NbCp₂(SiMe₂OEt)H₂] and [NbCp₂H₃]. Reaction of 3 with LiPPh₂ gives high yields of the compound [NbCp₂(SiMe₂PPh₂)H₂] (17). Reduction of 3 with LiAlH₄ results in the dimeric compound [NbCp₂H₂Al(μ-H)₂]₂ (13). Carrying out this reaction in the presence of an amine leads to the formation of the complex [NbCp₂(SiMe₂H)H₂]. X-Ray structure determinations were performed for the compounds (10), (13), (17), and (18). Complex (10) has 5 centre-6 electron interligand hypervalent interactions (IHIs) I-Si ... H ... Si-I; the analogous IHI O-Si ... H ... Si-O in 18 is significantly weaker, whereas 17 has a classical structure. Complex 13 has a cenrosymmetric dimeric structure, with two five-co-ordinate aluminium centres bridged by hydride ligands.

Full text of DOI:10.1039/b110998g

View Article Online / Journal Homepage / Table of Contents for this issue
Niobocene hydrides with functionalised silyl ligands. Tuning of
interligand hypervalent interactions (IHI) M–H Si–X
Georgii I. Nikonov,*a Lyudmila G. Kuzminab and Judith A. K. Howardc
a Chemistry Department, Moscow State University, Vorob’evy Gory, 119899, Moscow, Russia
b Institute of General and Inorganic Chemistry, RAS, Leninsky Prosp. 31, 117907, Moscow,
Russia
c Chemistry Department, University of Durham, South Road, Durham, UK DH1 3LE
Received 3rd December 2001, Accepted 27th May 2002
First published as an Advance Article on the web 4th July 2002
This paper presents syntheses and X-ray structure determinations of a variety of mono- and bis-(silyl) niobocene
hydride complexes with functionalised silyl ligands. Complexes of the type [NbCp2(SiR3)H2] (SiR3 = SiMe2Cl (3),
Si(OEt)3, SiMe2OEt (4)) can be prepared conveniently by heating the precursor [NbCp2H3] in the presence of the
corresponding silane HSiR3. As the electrondonating ability of the R group increases, the corresponding mono(silyl)
derivatives [NbCp2(SiR3)H2] are either unstable (SiR3 = SiMe3) or do not exist at all (SiR3 = SiEt3, Si(OCH2CH2)3N).
The previously described bis(silyl) complex [NbCp2(SiMe2H)2H] easily reacts with half an equivalent of iodine, giving
the functionalised derivative [NbCp2(SiMe2I)2H] (10). A labelling experiment for the analogous reaction with
bromine establishes that an electrophile attacks the more accessible Si–H bond rather than the more electron-rich
Nb–H bond. Bis(silyl) complexes [NbCp2(SiMe2X)2H] (X = Br (8), I (10), OTfl (11)) can be conveniently prepared
by the reaction of [NbCp2(SiMe2Cl)2H] (2) with the corresponding silane XSiMe3. Reactions of 2 with fluorinating
reagents such as [CPh3][BF4], BF3ؒOEt2 and AgBF4 do not work well for the synthesis of the compound
[NbCp2(SiMe2F)2H]. Nucleophilic chlorine substitution in 2 under the action of lithium amides and alkoxides
does not occur, whereas reaction with soft reagents such as MeLi, LiPPh2 and LiSPh leads to the corresponding
derivatives [NbCp2(SiMe3)2H] and [NbCp2(SiMe2EPhn)2H] (E = S, n = 1 (14); E = P, n = 2 (15)). Methanolysis of 10
gives high yields of the derivative [NbCp2(SiMe2OMe)2H] (18). The mono(silyl) precursor [NbCp2(SiMe2Cl)H2] (3)
reacts with LiOEt affording a mixture of [NbCp2(SiMe2OEt)H2] and [NbCp2H3]. Reaction of 3 with LiPPh2 gives
high yields of the compound [NbCp2(SiMe2PPh2)H2] (17). Reduction of 3 with LiAlH4 results in the dimeric
compound [NbCp2H2Al(µ-H)2]2 (13). Carrying out this reaction in the presence of an amine leads to the formation
of the complex [NbCp2(SiMe2H)H2]. X-Ray structure determinations were performed for the compounds (10), (13),
(17), and (18). Complex (10) has 5 centre–6 electron interligand hypervalent interactions (IHIs) I–Si ؒ ؒ ؒ H ؒ ؒ ؒ Si–I;
the analogous IHI O–Si ؒ ؒ ؒ H ؒ ؒ ؒ Si–O in 18 is significantly weaker, whereas 17 has a classical structure. Complex 13
has a cenrosymmetric dimeric structure, with two five-co-ordinate aluminium centres bridged by hydride ligands.
groups and present further examples of IHI in these systems.
Si–H bond oxidative addition, metathetical (salt elimination)
and electrophilic and nucleophilic substitution reactions at the
co-ordinated silyl groups were tested and compared.
Introduction
Transition metal silyl complexes [LnM–SiR3] are of significant
interest1 as intermediates in metal mediated transformations,
such as different silylation reactions2 and dehydropolymeris-
ation of silanes.3 The majority of silyl complexes studied bear
Results and discussion
chemically inert alkyl and aryl substituents at silicon, whereas
complexes with functionalised silyl groups are usually restricted
to trichloro, trifluoro and trialkoxy derivatives.4 However, one
can expect that other functionalised silyl complexes can be
valuable reagents for the synthesis of new silicon-based
materials. Apart from this practical aspect we were encouraged
by the discovery that transition metal silyl complexes of the
type [LnM(H)(SiR2X)], bearing basic hydride ligands, can pos-
sess non-classical interligand hypervalent interactions (IHIs)
between the hydride and functionalised silyl groups.5 Com-
plexes [NbCp2(SiMe2X)2H] with X = F, Cl and Br were thor-
oughly studied by X-ray, neutron diffraction, NMR relaxation
and DFT calculation techniques.5a–d
To explore further the phenomenon of IHI, we were inter-
ested in extending these studies to other functional substituents
X at silicon. Synthetic approaches to complexes with function-
alised silyl groups are scarce and usually not generally applic-
able. Here we report our search for convenient routes to the
mono- and bis-(silyl) niobocene hydrides bearing SiMe2X
1. Oxidative addition route
Oxidative addition of Si–H bonds is one of the most frequently
used approaches for the synthesis of transition metal silyls
and a variety of Group 5 hydridosilyl derivatives were made by
thermolysis of [NbCp2H3] (1) in the presence of silanes
HSiR3.6–10 However, in case of electron-donating R groups on
the silicon atom the corresponding silyls are unstable. Thus
heating 1 with HSiEt3 results in bis(niobocene) [Nb(η5-
Cp)(µ,η5:η5-C5H4)(H)]2, although intermediate complexation
of the silane to niobium was inferred from a labelling experi-
ment.6 Our attempts to prepare [NbCp2H2(SiEt3)] under milder
conditions5c by addition of HSiEt3 to a transient [NbCp2H]
generated from [NbCp2(BH4)] under the action of NEt3 again
produced only bis(niobocene). Thus we can conclude that
[NbCp2H2(SiEt3)] is intrinsically unstable towards silane elim-
ination even at room temperature. The very similar trialkyl-
silane [NbCp2H2(SiMe3)], appears to be the borderline case: it
DOI: 10.1039/b110998g
J. Chem. Soc., Dalton Trans., 2002, 3037–3046
This journal is © The Royal Society of Chemistry 2002
3037

View Article Online
can be made in 35% yield by addition of silane HSiMe3 to 1 at
35 ЊC but decomposes when excess silane is pumped off.9 In an
analogous fashion, the preparation of [NbCp2H2{Si(OCH2-
CH2)3N}] via addition of HSi(OCH2CH2)3N to 1 failed.
Although the target compound [NbCp2H2{Si(OCH2CH2)3N}]
would be closely related to the known (and stable) tris(alkoxy)-
derivative [NbCp2H2{Si(OEt)3}],5c the essential difference
between them is that the former has an intramolecular dative
Si–N bond, making the silyl centre more electron-rich. Appar-
ently, the stability of these mono(silyl) systems correlates with
the electron density on the silicon atom. Thus the richer the
silicon, the less stable is the complex towards silane elimination.
Heating the compounds 1 or [NbCp2H2{Si(OEt)3}] with
HSi(OEt)3 at 100 ЊC gives bis(niobocene) as the sole product
rather than the target bis(silyl) derivative [NbCp2H(Si(OEt)3)2].
This is in contrast to the reactivity of [NbCp2(SiMe2Cl)H2] (3)
which eliminates dihydrogen under similar conditions and oxi-
datively adds another equivalent of silane HSiMe2Cl, affording
the bis(silyl) [NbCp2(SiMe2Cl)2H].5c The essential difference
between [NbCp2H(Si(OEt)3)2] and 3 is that the latter exists as a
mixture of two isomers, differing in the position of the silyl
ligand, with the isomer 3b having the silyl group in the lateral
position in the niobocene bisectral plane and the two hydrides
cis to each other. By way of contrast [NbCp2H(Si(OEt)3)2]
exists only in the form of the central isomer in which the
hydrides are separated by the silyl ligand, and the lateral isomer
apparently is not available even at high temperature. Therefore,
the nature of the functional substituent X at silicon plays a
crucial role in this reaction. Previous DFT calculations showed
that in the mono(silyl) niobocene dihydrides the interligand
hypervalent interactions (IHI) of the type H ؒ ؒ ؒ S–X are
stronger for the lateral isomers and, in principle, can outweigh
the increased steric repulsion therein.5c With the aim of testing
the performance of the OR substituent in the SiMe2OR group
we prepared the compound [NbCp2H2(SiMe2OEt)] (4) using
thermolysis of 1 in the presence of HSiMe2OEt at 55 ЊC. Only
recrystallisation, failed. Apparently the presence of reactive
functional substituents at silicon is not quite compatible with
the highly reducing conditions of the metathetical route.
(1)
3. Electrophilic functionalisation of co-ordinated silyls
An elegant way of synthesising halogenosilyl complexes by
electrophilic activation of the Si–H bond in silyl derivatives was
suggested by Schubert et al.14 We previously used this approach
for the preparation of complexes [NbCp2(SiXMe2)2H] (X = F,
Br (8)) by reacting the compound [NbCp2(SiHMe2)2H] (9) with
CPh3BF4 and [Br2ؒdiox] (diox = 1,4-dioxane), respectively.5c
Using two equivs of [Br2ؒdiox] led then to only 25% yield of 8.
Wishing to improve the performance we analysed the mechan-
ism of this reaction. In principle, the Si–X bond could arise
either from the attack of an electrophile on the Si–H bond or
on the Nb–H bond with further rearrangement. In total three
routes are possible (Scheme 1). This point has been elucidated
by means of labelling experiments. The labelled compound
[NbCp2(SiDMe2)2H] (9D) was prepared in 90% yield anal-
ogously to 9 by the reaction of 2 with LiAlD4. 9D is sufficiently
stable for some days, but on keeping at room temperature a slow
redistribution of the Nb–H and Si–D atoms occurs. After
2 weeks of reaction 33% deuterium incorporation into the
hydride position was achieved. In principle, the H–D exchange
could proceed via HDSiMe2 elimination followed by oxidative
addition into the Si–D bond. However, an intramolecular route
cannot be excluded taking into account Berry’s report that
isomerisation of the lateral isomer of [TaCp2H2(SiR3)] into the
central one occurs without silane elimination.7b In our experi-
ments a reaction of the complex 9D with an equivalent of [Br2ؒ
diox] gave only the compound 8 in high yield. Careful integra-
1
the central isomer 4a was observed by H NMR spectroscopy,
suggesting that any interligand interaction of the type H–
Si ؒ ؒ ؒ OEt is rather weak. Structural data for the related bis-
(silyl) derivative (vide infra) support this conclusion. As for the
Si(OEt)3 derivative mentioned above, refluxing 4a in toluene in
the presence of silane, gives only bis(niobocene).
1
tion of the hydride region of the H NMR spectrum revealed
Finally, we report that heating [NbCp2H(η2-C2H3Ph)] with
the silane HSiMe2Cl in addition to the previously described
compound [NbCp2H(SiMe2Cl)2] (2, 60%)5a also affords [Nb-
Cp2(SiMe2Cl) (η2-C2H3Ph)] (5), isolated as a mixture of
exo- and endo-isomers in 15% yield and characterised by NMR
spectroscopy. This by-product most likely arises from styrene
elimination from the starting material [NbCp2H(η2-C2H3Ph)],
as was originally postulated by Green et al. for the related tin
chemistry,11 and its recombination with the intermediate com-
plex [NbCp2(SiMe2Cl)].
no significant deuterium incorporation. It is noteworthy that
some small amount of the mono(brominated) complex
[NbCp2(SiDMe2)(H)(SiMe2Br)] also observed in the NMR
spectrum of the reaction mixture did not exhibit any Si–H
hydrogen signal. Among the three mechanisms shown in
Scheme 1 only route c does not allow any deuterium incorp-
oration into the hydride position upon the reaction of 9D with
[Br2ؒdiox]. This route involves the interaction of an electrophile
with the Si–D bond, affording the Si–X derivative, followed by
the attack on the second Si–D bond by DX thus released, even-
tually giving the disubstituted product. Therefore routes a and b
do not contribute significantly to the formation of 8 and all
reaction steps occur at the silicon–hydrogen bond. This result
is surprising, taking into account the greater basicity of the
Nb–H bond, and is most likely caused by its steric protection by
the two bulky silyl groups. Another conclusion of this study is
that the acids HX are very effective halogenating agents in this
reaction and therefore only one equivalent of halide X2 is
required. This conclusion is supported by the preparation in
70% isolated yield of the bromo derivative 8 by reacting 9 with
exactly one equivalent of [Br2ؒdiox] and also by the synthesis of
the new iodo derivative [NbCp2(SiIMe2)2H] (10). Reaction of 9
with 2 equivs of I2 gives 10 in only 17% yield. The lower yield of
10 in comparison with 8 by using this stoichiometry is most
likely caused by the greater decomposition of the product under
the action of excess iodine and the more acidic HI. However,
with an exact 1 : 1 stoichiometry the compound 10 was
obtained in 43% isolated yield (Scheme 2). The identity of 10
2. Metathetical approach
Green and Hughes found that addition of Me3SiCl to
[NbCp2H2]Li12 results in a slow reaction leading to a mixture of
[NbCp2H2(SiMe3)], [NbCp2H(SiMe3)2] and bis(niobocene).9 We
independently observed the same result. However, we also
found that reaction of [NbCp2H2]Li with Cl2SiPh2 in THF gives
the expected product [NbCp2H2(SiPh2Cl)] (6) isolated as a mix-
ture of two isomers (central 6a and lateral 6b) albeit in a very
low yield (eqn. (1)). An improved preparation of this material
has been developed.13 Surprisingly, when this reaction was
attempted in hexanes, the main product in the reaction mixture
was the complex [NbCp2H2(SiPh2H)] (7). This compound
exhibits a hydride signal as a singlet at Ϫ4.44 ppm and a Si–H
signal as a singlet at 5.93 ppm. The origin of the Si–H bond
remains unknown. Unfortunately, 7 was contaminated by a
significant amount of 6a,b and its isolation in a pure form, by
3038
J. Chem. Soc., Dalton Trans., 2002, 3037–3046

View Article Online
Scheme 1
Scheme 2
was established by IR and NMR spectroscopies and confirmed
[NbCp2(SiClMe2)H(SiMe2Br)] after ten days. However, heating
by X-ray diffraction study (vide infra).
at 80 ЊC for some days allows for the preparation of 8 in 70%
yield (Scheme 2). The triflate derivative [NbCp2H(SiMe2OTfl)2]
(11) was prepared analogously to 10 by Me3SiOTfl addition to 2
at room temperature (68% yield). No reaction, even after heat-
ing at 60 ЊC, was observed between 2 and Me3SiN3. Attempted
reaction of [NbCp2(SiClMe2)H2] (3) with Me3SiI resulted in
fast decomposition.
The most convenient route to complexes 8 and 10 consists of
the reaction of the readily available bis(chlorosilyl) compound
[NbCp2H(SiClMe2)] (2) with halogenosilanes Me3SiX (X = Br,
I) (Scheme 1). Thus, 2 reacts readily with Me3SiI at room temp-
erature, giving a high yield of 10. Reaction of 2 with Me3SiBr is
slow under these conditions, affording a 10 : 1 mixture of 2 and
J. Chem. Soc., Dalton Trans., 2002, 3037–3046
3039

View Article Online
Chloride appears to be a poorer leaving group than hydride
in electrophilic substitution. Thus no chloride abstraction
occurs under the action of the trityl reagent [CPh3][BF4] or
[BF3ؒOEt2] on 2. Reaction with AgBF4 proceeds slowly and
results in a mixture of starting compound 2, [NbCp2(SiCl-
(2)
1
Me2)(H)(SiFMe2)] and [NbCp2(H)(SiFMe)2], identified by H
NMR spectroscopy. This reaction is much less convenient for
the synthesis of [NbCp2(H)(SiFMe)2] than the previously
described reaction of 2 with [CPh3][BF4].5c Fast decomposition
and formation of a silver mirror is observed when the dihydride
mono(silyl) 3 is mixed with AgBF4.
4. Nucleophilic functionalisation of co-ordinated silyls
Bis(silyl) complex [NbCp2(SiMe2Cl)2H] (2) and mono(silyl)
complex [NbCp2(SiMe2Cl)H2] (3) present themselves as
“natural” precursors for nucleophilic chloride substitution by
the desired functional groups X. However, their interaction
with lithium alkoxy and amide reagents was not successful. No
reaction of 2 with LiOMe, LiOEt, LiNHtBu, LiNHPh or LiNEt2
was observed. In contrast, the mono(silyl) compound 3 reacts
with these reagents, giving mixtures containing noticeable
amounts of the trihydride [NbCp2H3] (1). Reaction of 3 with
LiOEt, however, does give high yields of the desired product
[NbCp2(SiMe2OEt)H2] along with a small amount of 1. Forma-
tion of 1 is most likely caused by the niobocene anion
[NbCp2H2]Ϫ rather than chloride substitution at the silicon
centre. This transition metal anion displacement has prece-
dents, for example, in the reaction of [FeCp(CO)2(SiHMe2)]
with P(OMe)3.15
Attempted synthesis of the compound [NbCp2H2(SiHMe2)]
(12) by the method used for preparing 9, namely by chloride for
hydride substitution, also met with little success. Thus, reaction
of NaBH4 with 3 gives [NbCp2H3] and the known borohydride
complex [NbCp2BH4] whereas treatment with LiAlH4 gives the
related hydride [Cp2Nb(µ-H)2Al(µ-H)H]2 (13). The molecular
structure of the latter long-known compound16 has been even-
tually established by us and is reported here. In contrast to the
mononuclear boron congener, it is a hydride-bridged dimer,
which accounts for its low solubility in aromatic solvents and
diethyl ether. Formation of these Group 3 substituted products
could be again caused by [NbCp2H2]Ϫ displacement reactions,
or alternatively they may result from the reactions of the Nb–Si
bonds of the initially formed complex 12 with the Lewis acids
EH3 (E = B, Al), emerging in the reaction mixture. The latter
possibility appears to be more likely because carrying out the
reaction of 3 with LiAlH4 in the presence of a Lewis base NEt3
does produce 12 as a mixture with 13 (eqn. (2)). It is interesting
that 12 also emerges as the result of thermal decomposition of
[NbCp2(SiHMe2)2H] (9). Partial conversion to 12 was observed
after heating a C6D6 solution of 9 at 60 ЊC in a sealed NMR
tube. The silane product of this formal silylene extrusion
process is the disilane (HSiMe2)2 identified by its Si–H signal
(septet at 3.92 ppm, JH–H = 4.2 Hz) and Me-group signal (doub-
let at Ϫ0.4 ppm, JH–H = 4.2 Hz). Longer thermolysis at 100 ЊC
proceeds further to give eventually the bis(niobocene). Anal-
ogously, thermolysis of deuterated compound 9D at 100 ЊC for
2.5 h gives a mixture of 9D and 12D, in which deuterium is
distributed among the niobium and silicon positions, and par-
tially (about 50%) deuterated disilane (H{D}SiMe2)2. Appar-
ently, deuterium/proton scrambling among the Nb–H and Si–H
positions in 9D occurs prior to disilane elimination which does
not allow us to make any conclusion about the mechanism of
this transformation. Unexpectedly, compound 12 is also formed
by the reaction of LiI with complex [NbCp2(SiMe2Cl)H2] (3),
the other product being bis(niobocene).
formed after two days reaction of 2 with MgBr2 in ether. No
Si–H bond formation products were observed in these trans-
formations. These alkali metal salt reactions are of no synthetic
significance since a considerable decomposition accompanied
by the formation of a toluene insoluble material is observed.
No reaction of 2 with NaN3 in THF occurs.
Although hard nucleophilic reagents are not effective in the
chloride substitution of 2 and 3, we found that soft reagents
exhibit a better performance. Thus MeLi easily reacts with
2 giving the known compound [NbCp2H(SiMe3)2].9 LiSPh
(2 equivs) does not react with 2 in ether because of the relative
insolubility of the lithium thiolate, but nevertheless smoothly
reacts in THF giving the disubstituted compound [NbCp2-
(SiMe2SPh)2H] (14) in almost quantitative yield (Scheme 2).
The analogous reaction with LiPPh2 proceeds even in ether and
affords the compound [NbCp2(SiMe2PPh2)2H] (15). The course
of this reaction was monitored by NMR spectroscopy and
intermediate formation of the mono(substituted) product
1
[NbCp2(SiClMe2)H(SiMe2PPh2)] (16) was identified. In the H
NMR spectrum of the reaction mixture two different methyl
groups of 16 resonate closely to the corresponding signals of
2 and 15 (singlet at 0.60 ppm and doublet at 0.47 ppm, 3JH–P
=
5.0 Hz) but the hydride signal at Ϫ4.33 ppm lies in between
those of 2 and 15 (Ϫ5.15 ppm and Ϫ3.45 ppm, respectively).
Unfortunately, 16 cannot be made by the reaction of exactly
one equivalent of LiPPh2 with 2 as a mixture of 15, 16 and
unreacted 2 is formed. The monosilyl 3 also reacts with LiPPh2
giving the substituted product [NbCp2H2(SiMe2PPh2)] (17). An
X-ray quality crystal of this compound was grown and the
molecular structure determination performed (vide infra).
We have already mentioned that complex 2 is surprisingly
stable to water and alcohols.5c Stirring 2 in wet THF under
argon requires some days until a brown precipitate of decom-
position products forms. No new compound was identified in
the THF soluble part by NMR. Reaction of 2 with EtOH in the
presence of amine proceeds very slowly and only 15% conver-
sion into [NbCp2H(SiMe2OEt)2] was observed after two weeks.
Keeping the reaction mixture for longer results in a mixture of
other as yet unidentified products. Complex 10, having a much
better leaving group such as iodide, is more reactive. In the
presence of an amine it readily reacts with a slight excess of
MeOH giving the substitution product [NbCp2H(SiMe2OMe)2]
(18) in almost quantitative yield (Scheme 2). The structure of
this compound was confirmed by X-ray diffraction analysis (see
next section).
5. X-Ray structure determinations
The molecular structure of [NbCp2(SiMe2I)2H] (10) is shown in
Fig. 1. To the best of our knowledge this is the first transition
metal iodosilyl derivative characterised by X-ray structure
analysis. Determination of the structure of this compound
completes our study of the whole series of symmetrical halogen
Chloride displacement by iodide was attempted also on 2.
Reaction with LiI in ether proceeds slowly affording a mixture
of the mono(substituted) product [NbCp2(SiClMe2)H(SiMe2I)]
and 10. A mixture of [NbCp2(SiClMe2)H(SiMe2Br)] and 8 is
3040
J. Chem. Soc., Dalton Trans., 2002, 3037–3046

View Article Online
Table 1 Selected molecular parameters for complexes [NbCp2(SiXMe2)2H] (X = F, Cl (2), Br (8), I (10))
Compound
Nb–Si/Å
Si–X/Å
Si–Nb–Si/Њ
Si ؒ ؒ ؒ H/Å
[NbCp2(SiFMe2)2H]
2.618(1)
2.622(1)
2.597(1)
2.604(2)
2.595(3)
1.652(3)
1.644(3)
2.163(1)
2.349(2)
2.590(3)
105.57(4)
1.98
[NbCp2(SiClMe2)2H]
[NbCp2(SiBrMe2)2H]
[NbCp2(SiIMe2)2H]
104.27(5)
103.37(7)
104.4(1)
2.06
2.05
2.07
Table 2 Comparison of the silicon–halogen bond lengths in silanes and in the niobocene silylhydrido complexes
Si–X bonds
in silanes/Å
Lengthening of the
Si–X bond in complexes/Å
Lengthening of the
Si–X bond in complexes (%)
Compound
[NbCp2(SiFMe2)2H]
[NbCp2(SiClMe2)2H]
[NbCp2(SiClMe2)H2]
[NbCp2(SiBrMe2)2H
[NbCp2(SiIMe2)2H]
1.550
2.020
2.020
2.235a
2.537b
0.102
0.143
0.150
0.114
0.053
6.6
7.1
7.4
5.1
2.1
a Si–Br bond length in Me3SiBr;17 average Si–Br distance for 4 structures from CCDC18 is 2.250 Å. b Average for 5 structures from CCDC.18
Fig. 1 Molecular structure of 10. Hydrogen atoms apart from
hydrides are omitted for clarity.
substituted bis(silyl) niobocenes hydrides of the type [NbCp2-
(SiMe2X)2H] (X = F, Cl, Br, I). The structures of the fluoro-,
chloro- and bromo-analogues have been reported elsewhere.5c
The most important molecular parameters for the complexes
[NbCp2(SiMe2X)2H] are collected in Table 1. The Nb–Si bond
Fig. 2 Dependence of the Si–X and Nb–Si bond length on the halogen
X. (a) Relative lengthening of the Si–X bond in complexes
length in 10 is close to that in the chloro- and bromo-
substituted analogues, whereas the Si–Nb–Si bond angle and
the value of the Si–H contact does not differ much for all com-
plexes of this series. The observed Si–I bond (2.590(3) Å) is
longer than the structurally characterised Si–I bond in tetra-
co-ordinate silanes (range 2.487–2.574 Å, mean 2.537 Å for five
structures). The data for the Si–X bond length in complexes
[NbCp2(SiMe2X)2H] and in corresponding tetra-co-ordinate
silanes are compared in Table 2. It can be seen that lengthening
of the Si–X bond in silyl complexes relative to the parent silanes
XSiMe3 forms an inverted V-shaped function on going down
the halogen group (Fig. 2a). This trend is opposite to the change
of the Nb–Si bond lengths in the series X = F, Cl, Br, where the
Nb–Si bond lengths form a V-shaped function with respect to
the halogen position in the Periodic Table (Fig. 2b). This
irregularity was discussed previously in terms of the super-
position of two factors responsible for the shortening of the
Nb–Si and relative lengthening of the Si–X bonds, namely due
to the operation of Bent’s rule and interligand hypervalent
interactions (IHI).5c The importance and contributions of these
factors to the Nb–Si and Si–X parameters change in opposite
[NbCp2(SiMe2X)2H] in comparison with silanes X–SiR3 versus halogen
position in the Periodic Table. (b) Nb–Si bond length in complexes
[NbCp2(SiMe2X)2H] versus halogen position in the Periodic Table.
directions on going down the halogen group. Summation of
two monotonic functions with opposite slopes creates this
V-shaped curve (or inverted V-shaped curve in the case of the
Si–X distance). The Si–I bond length follows the trend shown in
Fig. 2a, while the Nb–Si bond length is less than expected (Fig.
2b). This can be due to crystal packing effects, which for the
large iodine group could shorten both the Nb–Si and Si–I
bonds.
The structure of the related methoxy substituted derivative
18 has several different features (Fig. 3 and Table 3), signifying a
substantial decrease in interligand interactions. Compared with
the halogenosilyl analogues, in this compound there is a notice-
able elongation of the Nb–Si bonds to 2.624(1) and 2.629(1) Å,
which are however still shorter than the range usually observed
for alkyl and aryl substituted silyl derivatives of niobo- and
tantalo-cenes (2.646–2.669(1) Å)6–8,19 and close to the Ta–Si
bond length in [Cp2Ta(SiHMe2)2H] (2.624(2) and 2.633(2) Å).20
J. Chem. Soc., Dalton Trans., 2002, 3037–3046
3041

View Article Online
Table 3 Selected molecular parameters (bonds in Å, bond angles in Њ)
for the compound [NbCp2(SiMe2OMe)2H] 18
Table 4 Selected molecular parameters (bonds in Å, bond angles in Њ)
for the compound [NbCp2H2(SiMe2PPh2)] (17)
Nb(1)–Si(1)
Si(1)–O(1)
O(1)–C(15)
Nb(1)–H(1)
2.6240(9)
1.696(3)
1.414(5)
1.88(5)
Nb(1)–Si(2)
Si(2)–O(2)
O(2)–C(16)
2.6285(9)
1.691(2)
1.419(5)
Nb(1)–Si(1)
Nb(1)–H(1Ј)
P(1)–C(11)
2.6491(4)
1.65(2)
1.842(2)
P(1)–Si(1)
Nb(1)–H(2Ј)
P(1)–C(17)
2.3065(5)
1.70(2)
1.849(2)
H(1Ј)–Nb(1)–H(2Ј) 118.1(10)
Si(1)–Nb(1)–H(1Ј) 54.1(7)
P(1)–Si(1)–Nb(1)
Si(1)–Nb(1)–H(2Ј)
111.08(2)
64.2(7)
Si(1)–Nb(1)–Si(2)
Si(1)–Nb(1)–H(1)
O(1)–Si(1)–Nb(1)
108.20(3)
55.4(15)
109.83(9)
Si(2)–Nb(1)–H(1)
O(2)–Si(2)–Nb(1)
52.8(15)
108.79(9)
Table 5 Selected molecular parameters (bonds in Å, bond angles in Њ)
for the compound [NbCp2(AlH4)]2 (13)
Nb(1)–Al(1)
Nb(1)–H(3)
Al(1)–H(1)
Al(1)–H(3)
2.621(1)
1.59(3)
1.64(2)
1.98(3)
Al(1)–Al(1Ј)
Nb(1)–H(4)
Al(1)–H(2)
Al(1)–H(4)
2.683(2)
1.69(3)
1.48(3)
1.99(3)
Al(1)–Nb(1)–H(3)
H(3)–Nb(1)–H(4)
Nb(1)–Al(1)–H(1)
Nb(1)–Al(1)–H(2)
H(1)–Al(1)–H(2)
Al(1Ј)–Al(1)–H(3)
H(2)–Al(1)–H(3)
Al(1Ј)–Al(1)–H(4)
H(2)–Al(1)–H(4)
49.0(12)
98.4(17)
118.4(9)
128.4(11)
104.0(14)
110.0(10)
127.9(15)
132.1(10)
111.5(15)
Al(1)–Nb(1)–(H)4
Nb(1)–Al(1)–Al(1Ј) 130.66(5)
49.4(12)
Al(1)–Al(1Ј)–H(1)
Al(1Ј)–Al(1)–H(2)
Nb(1)–Al(1)–H(3)
H(1)–Al(1)–H(3)
Nb(1)–Al(1)–(H4)
H(1)–Al(1)–H(4)
H(3)–Al(1)–H(4)
40.0(8)
100.6(11)
37.2(10)
126.6(13)
40.0(10)
96.5(13)
77.2(14)
(2.27 Å). In another transition metal silyl derivative with a
terminal phosphido substituent at silicon, the compound
[Cp(CO)2Fe{SiMeBut(PH{SiF(2,4,6-Pri3C6H3)2})}], the Si–P
bond is almost the same (2.297 Å).24 In the complex
[Cp(CO)2W(η2-SiMe2PPh{N(SiMe3)2})], where the silicon-
bound phosphido group is also co-ordinated to the metal via
the phosphorus lone pair, the Si–P bond is 2.197(3) Å.25 For
comparison, in the phosphido substituted silylene complex the
direct Si–P single bond is 2.294(1) Å,26 whereas in the phos-
phino stabilised silylene complex the donor–acceptor Si–P
bond is elongated to 2.364(1) Å.27
Fig. 3 Molecular structure of 18. Hydrogen atoms apart from hydride
are omitted for clarity.
The increased Si–Nb–Si bond angle of 108.20(3)Њ lies between
the values found for [NbCp2(SiMe2X)2H] (103–106Њ, see Table
1) and classical compounds [NbCp2(SiMe2Ph)2H] (110.81(6)Њ)
and [TaCp2(SiMe2H)2H] (109.90(7)Њ), which also suggests
diminished IHI. This conclusion allows us to understand the
absence of the lateral isomer in the case of the alkoxy derivative
[NbCp2H2(SiMe2OEt)] and suggests that the electronegativity
of the functional group at silicon is not the only prerequisite for
the occurrence of a strong interligand hypervalent interaction.
The ability of the substituent to serve as a good leaving group is
also important.21 The Si–O bond of 1.696(3) Å in 18 is less than
that in five-co-ordinate silicon compounds having an alkoxy
group in the apical position (mean 1.785 Å from the range
1.656–1.855 Å for 20 structures from the CCDC).18 However, it
is still elongated in comparison with the normal silicon–oxygen
covalent bond (1.640 Å, average for 1237 structures with
Si–OC(sp3) bonds in tetra-coordinated organosilanes;18 ∆ =
0.056 Å, 3.4%) and the Si–O bonds in two other structurally
characterised compounds with SiR2OR fragments, namely the
complexes [Cp(CO)2Fe{SiMe(OMe)(C6H4{CH2NHMe2}-2)}]-
[PF6] (1.671 Å)22 and [Ir(acac)(PCy3)2(η2-SiPh2OCH-
(OCH3)C(CHCO2CH3))] (1.656(4) Å).23 This elongation can
be either due to the stronger electron donating ability of the
[NbCp2] fragment (i.e. manifestation of Bent’s rule) or the
presence of a weak H ؒ ؒ ؒ Si–O interaction, or more likely is
due to the simultaneous operation of both effects.
The molecular structure of the binuclear complex [Cp2Nb-
(µ-H)2AlH(µ-H)]2 (13) is shown in Fig. 5 and selected molecular
parameters are given in Table 5. For a long time complex 13 was
believed to be monomeric in analogy with the structurally char-
acterised complex [NbCp2(µ-H)2(BH2)].28 Its dimeric structure
is unequivocally established in this study. Two aluminium
hydride units are associated via two hydride bridges, with each
unique aluminium centre being five-coordinate. The co-
ordination environment of aluminium can be described as
pseudo-square-pyramidal. This structural motif is abundant29
in Group 4 metallocene chemistry30 and in other complexes31
but for the Group 5 metallocenes the only relevant example is
the ditantalum complex [TaCp2(µ-H)2AlH(µ-OBu)]2 where the
two units are bridged by two OBun groups rather than
hydrides.32 A related alkoxy-bridged species is the compound
[Ta(dmpe)2(µ-H)2Al(OC2H4OMe)(µ-OC2H4OMe)]2.33 In 13 all
hydride atoms were located from difference Fourier maps and
refined isotropically. The Nb–H distances are 1.59(3) and
1.69(3) Å, whereas the Al–H bond to the bridging hydrides are
1.98(3) and 1.99(3) Å to the niobium bridges and 1.64(2) Å to
30b
the aluminium bridge. In [TiCp*2(µ-H)2AlH(µ-H)]2 the cor-
responding Al–H bond values are 1.67(1) Å (average of two
distances) and 1.56(1) Å and similar parameters were observed
for [ZrCp*2(H)(µ-H)2AlH(µ-H)]2 30c (1.65(7) Å for the Al–H(Zr)
bond and 1.75(6) Å for the Al–H(Al) bond). In contrast, the
Al–H bond to the terminal hydride in 13 is much shorter
(1.48(3) Å) and can be compared to the terminal Al–H bonds in
[TaCp2(µ-H)2AlH(µ-OBu)]2 (1.39 Å), [TiCp*2(µ-H)2AlH(µ-H)]2
(1.51(1) Å) and [ZrCp*2(H)(µ-H)2AlH(µ-H)]2 (1.52(6) Å). The
Nb–Al contact of 2.621(1) Å is somewhat shorter than in the
tantalum product (2.671(4) Å), probably due to the different
substituents at the aluminium.
The molecular structure of [NbCp2H2(SiMe2PPh2)] (17) is
shown in Fig. 4 and selected molecular parameters are collected
in Table 4. This is a typical trisubstituted niobocene structure
with a centrally positioned main-group element substituent.
The phosphorus centre is pyramidal, with the sum of the bond
angles being equal to 312.35Њ. The Nb–Si bond length
(2.6491(4) Å) is normal for an organosilicon group having only
electron-donor substituents at the silicon atom. The P–Si bond
length of 2.3065(5) Å is close to the sum of the covalent radii
3042
J. Chem. Soc., Dalton Trans., 2002, 3037–3046

View Article Online
Fig. 4 Molecular structure of 17. Hydrogen atoms apart from hydrides are omitted for clarity.
Fig. 5 Molecular structure of 13. Hydrogen atoms apart from hydrides are omitted for clarity.
Significant I–Si–H(Nb)–Si–I interligand hypervalent inter-
Conclusions
actions (IHI) were found by X-ray structure determination for
the iodine derivative [NbCp2(SiMe2I)2H] but not for [NbCp2-
(SiMe2OMe)2H]. The latter compound is intermediate between
the non-classical halogeno-substituted complexes [NbCp2(Si-
Me2X)2H] and classical alkyl substituted compounds [NbCp2-
(SiR3)2H]. The phosphorus derivative [NbCp2(SiMe2PPh2)H2]
is classical. Summarising this stage of study of IHI, we can
conclude that this type of interligand interaction is more
pronounced for functional substituents that are both highly
electronegative and good leaving groups.
Transition metal complexes with functionalised silyl groups can
be made by a variety of methods, the success of which depends
on the choice of the functionalised substituent. The bis(silyl)
niobocene complexes [NbCp2(SiMe2X)2H can be equally well
made either by electrophilic or nucleophilic functionalisation
reactions. The most efficient approach to the derivatives with
X = F, Br, I is electrophilic substitution at the Si–H bond in the
precursor compound [NbCp2(SiMe2H)2H]. Complexes with
X = Br, I, OTfl can also be made by chlorine substitution in
[NbCp2(SiMe2H)2H] by XSiMe3. Sulfur and phosphorus sub-
stituted complexes [NbCp2(SiMe2EPhn)2H] (E = S, n = 1; E = P,
n = 2) and also [NbCp2(SiMe3)2H] can be made by nucleophilic
chlorine displacement in the compound [NbCp2(SiMe2Cl)2H].
We found that chlorine substitution does not work for hard
nucleophiles such as amides and alkoxides. Also this reaction
does not work well with alcohol for making the alkoxy deriv-
atives [NbCp2(SiMe2OR)2H], but analogous substitution of
iodine to give [NbCp2(SiMe2OMe)2H] readily occurs under the
action of methanol. Synthesis of mono(silyl) dihydrides
[NbCp2(SiMe2X)H2] turned out to be more difficult. Attempted
electrophilic substitutions lead to decomposition, whereas
nucleophilic substitution can be used only for soft reagents such
as phosphides, alkyls, etc. In this case, oxidative addition of
silanes can be helpful provided the silane HSiR2X is available.
For example, silyls [NbCp2(SiMe2OEt)H2] and [NbCp2(Si-
{OEt}3)H2] were made by this method. Approaches based on
the transition metal anion [Cp2NbH2]Ϫ met with little success
due to concurrent reactions.
Experimental
General methods
All manipulations were carried out using conventional Schlenk
high vacuum and argon techniques. Solvents were dried
over sodium or sodium benzophenone ketyl and distilled into
the reaction vessel by high vacuum gas phase transfer. NMR
spectra were recorded on a Varian VXR-400 spectrometer (1H,
400 MHz; 13C, 100,4 MHz). IR spectra were obtained in Nujol
with a FTIR Perkin-Elmer 1600 series spectrometer. Silanes
were purchased from Aldrich. [Cp2NbH3] was prepared accord-
ing to a literature method.16
Preparation of [NbCp2(C2H3Ph)(SiMe2Cl)]
A solution of [NbCp2H3] (1.67 g, 1.42 mmol) in toluene was
treated with a five-fold excess styrene. Next day all volatile
J. Chem. Soc., Dalton Trans., 2002, 3037–3046
3043

View Article Online
materials were pumped off and the residue was redissolved in
50 mL of toluene. A four-fold excess of the silane HSiMe2Cl
was added and the mixture heated at 65–70 ЊC for 3 h. All
volatiles were removed in vacuo and the residue extracted with
ether, filtered and dried. The 1H NMR spectrum in CDCl3
Preparation of [NbCp2H2(SiMe2PPh2)] (17)
A supension of 0.971 g (2.95 mmol) of Ph2PLiؒ2THF in 30 mL
of Et2O was added to 0.664 g (2.08 mmol) of [NbCp2H2-
(SiMe2Cl)] in 20 mL of Et2O. The mixture was stirred over-
night. Filtration and solvent removal gave a brown oily com-
pound. This was redissolved in 40 mL of ether and cooled
to Ϫ30 ЊC. Light beige crystals were formed over 2 days. The
cold solution was filtered and the residue dried. Yield: 0.406 g
showed
a mixture of [NbCp2(SiMe2Cl)2H] and [NbCp2-
(C2H3Ph)(SiMe2Cl)] in the ratio 5 : 2. The mixture was dis-
solved in ether (≈ 100 mL) and left at Ϫ25 ЊC. The cold solution
was filtered from the precipitate and dried to give 0.97 g of
(0.867 mmol, 42%). IR (Nujol): νNb–H = 1751 cmϪ1. H NMR
1
a
mixture of [NbCp2(SiMe2Cl)2H] and [NbCp2(C2H3Ph)-
(C6D6): 7.83 (t, JH–H = 6.9 Hz, 4, Ph), 7.06 (m, Ph), 4.49 (s, 10,
Cp), 0.61 (d, JP–H = 6.0 Hz, 6, Me), Ϫ4.44 (s, 2, Nb–H).
13C-{1H} NMR (C6D6): 135.0 (d, JC–P = 22.1 Hz, Ph), 126.5
(Ph), 88.0 (Cp), 9.4 (d, JC–P = 20.1 Hz, Me). 31P-{1H} NMR
(C6D6): Ϫ27.3. C24H28NbPSi: calcd C 59.23, H 6.02; found:
C 58.79, H 5.83%.
(SiMe2Cl)] in the ratio 10 : 1. Yield of the precipitate was
0.974 g. According to the 1H NMR spectrum in CDCl3 it con-
sisted of [NbCp2(SiMe2Cl)2H] and [NbCp2(C2H3Ph)(SiMe2Cl)]
in the ratio 7 : 6. Complete isolation of [NbCp2(C2H3Ph)-
(SiMe2Cl)] from [NbCp2(SiMe2Cl)2H] was not achieved. 1H
NMR (C6D6): 4.46 (s, 5, Cp), 4.20 (s, 5, Cp), 2.39 (pseudo-
triplet, JH–H = 12.2 Hz, 1, PhCHCH2), 1.08 (dd, JH–H = 6.8 Hz,
JH–H = 12.6 Hz, 1, PhCHCH2), 0.96 (s, 3, SiCH3), 0.90 (s, 3,
SiCH3), 0.80 (dd, JH–H = 6.8 Hz, JH–H = 13.0 Hz, 1, PhCHCH2).
Preparation of [NbCp2H(SiMe2Br)2] (8)
Improved method (a). 0.31 g (1.23 mmol) of [Br2ؒdiox] in
25 mL of ether was added to 0.42 g (1. 23 mmol) of [NbCp2H-
(SiMe2D)2]. A light crimson solution and a precipitate are
formed. The mixture was stirred overnight at room tempera-
ture. The solution was filtered and stripped of volatiles. Yield:
0.43 g (0.86 mmol, 70%).
Preparation of [NbCp2H2(SiMe2OEt)] (5)
A solution of 0.30 mL (2.18 mmol) of HSiMe2OEt in 5 mL of
toluene was added to 10 mL of a toluene solution of
[NbCp2H3] (0.321 g, 1.42 mmol). The mixture was heated for
5 h at 55 ЊC. The solution was filtered and volatiles removed
leaving a yellow product. Yield of the crude material: 0.371 g.
This was recrystallised from ether giving 0.27 g (.83 mmol, 58%)
of 5. 1H NMR (C6D6): 4.66 (s, 10, Cp), 3.79 (q, JH–H = 6.4 Hz, 2,
OCH2), 1.31 (t, JH–H = 6.4 Hz, 3, CH2CH3), 0.66 (s, 6, SiCH3),
Ϫ5.38 (s, 2, Nb–H). C14H23NbOSi: calcd C 51.22, H 7.06;
found: C 50.89, H 7.00%.
New method (b). 0.296 g (0.72 mmol) of [NbCp2H(SiMe2Cl)2]
and 1.5 mL of BrSiMe3 were heated at 80 ЊC in 10 mL of
toluene for 1 week. The solution was filtered and stripped of
volatiles. The residue was recrystallised from ether to give
1
0.260 g (0.52 mmol, 72%). H NMR (C6D6): 4.70 (s, 10, Cp),
0.71 (s, 12, SiCH3), Ϫ5.16 (s, 1, Nb–H). 13C-{1H} NMR (C6D6):
93.76 (Cp), 14.06 (SiMe2). For full characterization data see
ref. 5c.
Thermolysis of 5 with HSiMe2OEt
Preparation of [NbCp2H(SiMe2I)2] (10)
A mixture of 0.20 g (0.61 mmol) of [NbCp2H2(SiMe2OEt)] and
0.23 mL (1.67 mmol) of HSiMe2OEt were refluxed in 10 mL of
toluene for 3 h. The solution was filtered and volatiles removed
in vacuo. The 1H NMR spectrum taken from the residue
revealed the formation of bis(niobocene).
Method (a). Iodine (0.813 g, 3.20 mmol) was added to 0.55 g
(1.61 mmol) of [NbCp2H(SiMe2H)2] in 30 mL of ether. A
purple solution and a dark red precipitate developed. The
solution was filtered and the precipitate dried in vacuo. Yield:
0.474 g. Volatiles were removed in vacuo from the solution,
leaving a red substance. This was recrystallised from ether.
Yield: 0.16 g (0.269 mmol, 17%).
Reactions of [NbCp2H2(SiMe2Cl)] with LiAlH4
Method (a). 0.342 g (1.073 mmol) of [NbCp2H2(SiMe2Cl)]
were treated with an excess of LiAlH4 in 15 mL of ether. A red-
orange solution is formed. The volatiles were removed in vacuo,
and the residue extracted with 10 mL of toluene, filtered and
dried to give a light orange powder. The NMR spectrum
showed the formation of a mixture of the known compound
1316 (major) and [NbCp2H3]. Crystals of 13 suitable for X-ray
diffraction study were obtained from a dilute pentane solution.
13: NMR (C6D6): 5.4 (br s, 1, Al–H), 4.71 (s, 10, Cp), 3.6 (br s,
1, Al–H–Al), Ϫ9.79 (s, 2, Nb–H–Al). 13C-{1H} NMR (C6D6):
85.40 (Cp).
Method (b). 1.1 mL (8.1 mmol) of ISiMe3 was added to
30 mL of a toluene solution of 1.15 g (2.8 mmol) of [NbCp2H-
(SiMe2Cl)2]. The mixture was stirred for 4 days at room
temperature. The solution was filtered and stripped of volatiles.
Yield: 1.13 g (1.90 mmol, 68%). IR (Nujol): νNb–H = 1713 cmϪ1
.
1H NMR (C6D6): 4.71 (s, 10, Cp), 0.870 (s, 12, SiCH3), Ϫ4.91 (s,
1, Nb–H). 13C-{1H} NMR (C6D6): 95.69 (Cp), 14.65 (SiMe2).
C14H23NbSi2I2: calcd C 28.30, H 3.90; found C 28.02, H 3.75%.
Preparation of [NbCp2H(SiMe2OMe)2] (18)
0.06 mL (1.48 mmol) of HOMe and 1 mL of NEt3 were added
by microsyringe to 30 mL of an ether suspension of 0.316 g
(0.53 mmol) of [NbCp2NbH(SiMe2I)2]. The mixture was stirred
for 2 days at room temperature. The solution was filtered, the
residue washed with 10 mL of ether and the volatiles were
removed from the combined fractions. The yield was quantita-
tive. IR (Nujol): νNb–H = 1735 cmϪ1. 1H NMR (C6D6): 4.60 (s, 10,
Cp), 3.42 (s, 6, OCH3), 0.50 (s, 12, SiCH3), Ϫ4.49 (s, 1, Nb–H).
13C-{1H} NMR (C6D6): 89.08 (Cp), 50.86 (OMe), 8.44 (SiMe2).
C16H29NbO2Si2: calcd C 47.75; H 7.26; found C 47.43, H 7.10%.
Method (b). 0.299 g (0.938 mmol) of [NbCp2H2(SiMe2Cl)]
were treated with an excess of LiAlH4 in a solution of 3 mL of
amine NEt3 in 15 mL of ether. A red-orange solution is formed.
The volatiles were removed in vacuo, and the residue extracted
with 10 mL of toluene, filtered and dried to give a light orange
1
powder, which according to H NMR is a mixture of 13 and
[Cp2NbH2(SiMe2H)] (12) in the ratio 3 : 1. This material was
treated with 10 mL of pentane. Filtration followed by volatiles
removal in vacuo gave an orange oil of 12. 12: 1H NMR (C6D6):
5.06 (sept, JH–H = 3.6 Hz, 1, Si–H), 4.58 (s, 10, Cp), 0.69 (s,
JH–H = 3.6 Hz, 6, SiCH3), Ϫ5.05 (s, 2, Nb–H). 13C-{1H} NMR
(C6D6): 95.69 (Cp), 8.09 (SiMe2).
Preparation of [NbCp2H(SiMe2OTfl)2] (11)
0.6 mL (3.31 mmol) of Me3SiOTfl was added to 20 mL of an
ether solution of 0.44 g (1.06 mmol) of [NbCp2H(SiMe2Cl)2].
The mixture was stirred for 24 h at room temperature. The
The analogous tantalum compound [NbCp2H2(SiMe2H)] has
been previously characterised.7b
3044
J. Chem. Soc., Dalton Trans., 2002, 3037–3046

View Article Online
Table 6 Crystal data and structure refinement for 10, 18, 17 and 13
10
18
17
13
Empirical formula
Formula weight
C16H23NbI2Si2
594.21
C16H29NbO2Si2
402.48
C24H28NbPSi
468.43
C10H14AlNb
254.10
Crystal size/mm
Crystal system, space group
0.44 × 0.36 × 0.26
Tetragonal, I4/m
0.66 × 0.18 × 0.08
Triclinic, P1
0.44 × 0.42 × 0.20
Monoclinic, P21/c
0.30 × 0.10 × 0.10
Monoclinic, P21/c
¯
a/Å
b/Å
18.996(3)
18.996(3)
7.7729(1)
10.5018(2)
13.3692(2)
12.8331(1)
5.894(2)
12.499(2)
c/Å
α/Њ
13.031(3)
90
13.0324(2)
70.100(1)
12.6511(2)
90
14.256(4)
90.00
β/Њ
90
90
4702.2(1)
72.739(1)
73.936(1)
936.79(3)
92.848(1)
90
2167.85(5)
101.01(2)
90.00
1031.0(5)
γ/Њ
V/Å3
Z
8
2
4
4
µ/mmϪ1
3.230
0.772
0.691
1.196
T /K
173(2)
1417
1122
1.115
153.0(2)
6066
3648
1.111
R1 = 0.0361, wR2 = 0.0843
R1 = 0.0442, wR2 = 0.0959
153.0(2)
5007
4991
1.091
R1 = 0.0209, wR2 = 0.0497
R1 = 0.0256, wR2 = 0.0552
150.0(2)
2138
2031
1.152
R1 = 0.0251, wR2 = 0.0491
R1 = 0.0342, wR2 = 0.0541
Reflections collected
Independent reflections
Goodness-of-fit on F 2
Final R indices [I > 2σ(I )]
R indices (all data)
R1 = 0.0560, wR2 = 0.1658
R1 = 0.0736, wR2 = 0.1758
solution was cooled to Ϫ26 ЊC for some days. Tiny yellow crys-
formed over 2 days. The cold solution was filtered and the
tals were formed. The solution was slowly concentrated while
residue dried. Yield: 0.12 g. IR (Nujol): νNb–H = 1759 cmϪ1
.
1
cold and filtered. Yield: 0.48 g (0.75 mmol, 71%). H NMR
1H NMR (C6D6): 7.76 (t, JH–H = 7.0 Hz, 8, Ph), 7.52 (m, Ph),
7.03 (m, Ph), 4.59 (s, 10, Cp), 0.49 (d, 3JP–H = 5.0 Hz, 12, SiCH3),
Ϫ3.33 (s, 1, Nb–H). 13C-{1H} NMR (C6D6): 90.3 (Cp), 7.2
(C6D6): 4.70 (s, 10, Cp), 0.33 (s, 12, SiCH3), Ϫ6.76 (s, 1, Nb–H).
13C-{1H} NMR (C6D6): 91.88 (Cp), 11.02 (SiMe2). C16H23F6-
NbO6Si2S2: calcd C 30.10, H 3.63, found C 29.79, H 3.31%.
2
(d, JP–C = 14.7 Hz, SiMe2). 31P-{1H} NMR (C6D6): Ϫ27.2 (s).
Satisfactory elemental analysis could not be obtained.
Reaction of [NbCp2H(SiMe2Cl)2] with MgBr2
Crystal structure determinations
Fresh MgBr2 was prepared by reaction of 0.22 g (9.05 mmol) of
magnesium with dibromethane. MgBr2 was suspended in 25 mL
of THF and added to 20 mL of a THF solution of [NbCp2H-
(SiMe2Cl)2] (0.628 g, 1.53 mmol). The mixture was stirred for
2 days at room temperature to give a light green solution. THF
was removed in vacuo and the residue extracted with 25 mL
of toluene. The solution was filtered and stripped of volatiles.
Pink crystals of [NbCp2H(SiMe2I)2] (10), orange crystals of
[NbCp2H2Al(µ-H)2]2 (13), beige crystals of [NbCp2H2-
(SiMe2PPh2)] (17) and colourless crystals of [NbCp2H(SiMe2-
OMe)2] (12) were grown from ether by cooling the solutions to
Ϫ25 to Ϫ30 ЊC. The crystals were covered by polyperfluoro oil.
Crystals of 13, 17, and 18 were mounted directly on the Bruker
Smart three-circle diffractometer with CCD area detector at
Ϫ120 ЊC (Ϫ123 ЊC for 13) and a crystal of 10 was mounted on
an Enraf-Nonius CAD-4 diffractometer at Ϫ100 ЊC. The crys-
tallographic data and characteristics of structure solution and
refinement are given in Table 6. The structure factor amplitudes
for all independent reflections were obtained after the Lorentz
and polarization corrections. An empirical absorption correc-
tion based on measurements of several equivalent reflections
was applied for 10 and 13. The structures were solved by heavy-
atom methods34 and refined by full-matrix least squares pro-
1
Yield: 0.71 g. The H NMR spectrum revealed a mixture of
[NbCp2H(SiMe2Cl)2], [NbCp2H(SiMe2Br)(SiMe2Cl)] and [Nb-
Cp2H(SiMe2Br)2]. NMR data for [NbCp2H(SiMe2Br)(SiMe2-
1
Cl)]: H NMR (C6D6): 4.69 (s, 10, Cp), 0.74 (s, 12, BrSiCH3),
0.58 (s, 12, ClSiCH3). 13C-{1H} NMR (C6D6): 93.11 (Cp).
Preparation of [NbCp2H(SiMe2SPh)2] (14)
2.61 mmol of PhSLi, prepared by the reaction of PhSH with
MeLi, in 10 mL of THF was added to 15 ml of a THF solution
of 0.536 g (1.30 mmol) of [NbCp2H(SiMe2Cl)2]. The mixture
was stirred for 5 days affording a crimson solution. Volatiles
were removed in vacuo and the residue extracted with 35 mL of
ether. Filtration and solvent removal gave 0.689 g of crude
product which contained 14 as the only Cp-containing com-
2
cedures, using ω(|Fo | Ϫ |Fc2|)2 as the refined function.34 All
hydrogen atoms were found from the difference map. In the
final cycles of refinement, all the non-hydrogen atoms were
refined with anisotropic temperature parameters. The hydride
ligands were refined isotropically, other hydrogen atoms were
refined using the riding scheme. The largest residuals in the final
difference Fourier maps were small (1.086 and Ϫ0.768 e ÅϪ3 for
10, 0.563 and Ϫ0.472 e ÅϪ3 for 13, 0.374 and Ϫ0.329 e ÅϪ3 for
17, 0.559 and Ϫ0.915 e ÅϪ3 for 18), location and magnitude of
the residual electron density was of no chemical significance.
The largest residuals in the final difference Fourier maps of
[NbCp2H(SiMe2I)2] were more significant and could be ration-
alised either as some partial co-crystallisation of an ether
molecule or polyiodide.
1
pound. In addition to signals of 14 in the H NMR spectrum
this compound also displays THF signals (3.50 and 1.26 ppm)
probably due to an admixture of (LiCl)ؒxTHF. An analytically
pure white microcrystalline material was obtained by cooling
a saturated ether solution at Ϫ30 ЊC. IR (Nujol): νNb–H
=
1725 cmϪ1. 1H NMR (C6D6): 7.59 (d, JH–H = 1.8 Hz, 4, Ph), 7.08
(m, Ph), 4.85 (s, 10, Cp), 0.46 (s, 12, SiCH3), Ϫ4.29 (s, 1, Nb–H).
13C-{1H} NMR (C6D6): 135.88, 135.01, 120.46 (Ph),
91.85(Cp), 9.60 (SiMe2). C26H33NbSi2S2: calcd C 55.89, H 5.95;
found C 55.50, H 5.80%.
CCDC reference numbers 175461–175464.
See http://www.rsc.org/suppdata/dt/b1/b110998g/ for crystal-
lographic data in CIF or other electronic format.
Preparation of [NbCp2H(SiMe2PPh2)2] (15)
A suspension of 0.818 g (2.43 mmol) of Ph2PLiؒ2THF in 30 mL
of Et2O was added to 0.500 g (1.22 mmol) of [NbCp2H-
(SiMe2Cl)2]. The mixture was stirred for 2 days. Filtration and
solvent removal gave a brown oily compound. This was dis-
solved in ether and cooled to Ϫ30 ЊC. Light beige crystals were
Acknowledgements
G.I.N. and L.G.K. thank The Russian Fund for Basic Research
for the support of this work (project numbers 00–03–32850
J. Chem. Soc., Dalton Trans., 2002, 3037–3046
3045

View Article Online
13 K. Yu. Dorogov and G. I. Nikonov, unpublished results.
14 U. Schubert, B. Wörle and P. Jandik, Angew. Chem., Int. Ed. Engl.,
1981, 20, 695.
15 H. Kobayashi, K. Ueno and H. Ojino, Chem. Lett., 1999, 119–120.
16 J. A. Labinger and K. S. Wong, J. Organomet. Chem., 1979, 170,
373.
and 01–03–32474). J.A.K.H. and L.G.K. also thank The Royal
Society for support through a joint research project and the
EPSRC for a Senior Research Fellowship (J.A.K.H.)
References and notes
17 M. D. Harmony and M. R. Strand, J. Mol. Spectrosc., 1980, 81,
1 (a) T. D. Tilley, in The Chemistry of Organosilicon Compounds, ed.
S. Patai and Z. Rappoport, Wiley, New York, 1989, ch. 24; (b)
J. Y. Corey and J. Braddock-Wilking, Chem. Rev., 1999, 99, 175.
2 I. Ojima, in The Chemistry of Organosilicon Compounds, ed. S. Patai
and Z. Rappoport, Wiley, New York, 1989, p. 1479.
308.
18 F. H. Allen and O. Kennard, Chem. Des. Automat. News, 1993, 8,
31.
19 J. Arnold, T. D. Tilley, A. L. Rheingold and S. J. Geib,
Organometallics, 1987, 6, 473.
20 I. Tanaka, T. Ohhara, N. Niimura, Y. Ohashi, Q. Jiang, D. H. Berry
and R. Bau, J. Chem. Res., 1999, 14.
3 T. D. Tilley, Acc. Chem. Res., 1993, 26, 22.
4 Functionalised silyls were systematically studied by Malisch et al.,
see for example: (a) W. Malisch, H. Jehle, S. Möller, G. Thum,
J. Reising, A. Gbureck, V. Nagel, C. Fickert, W. Kiefer and
M. Nieger, Eur. J. Inorg. Chem., 1999, 1597–1605; (b) W. Malisch,
H. Jehle, S. Möller, K. Saha-Möller and W. Adam, Eur. J. Inorg.
Chem., 1998, 1585–1587; (c) G. Thum and W. Malisch,
J. Organomet. Chem., 1984, 264, C5–C9; (d ) H.-U. Wekel and
W. Malisch, J. Organomet. Chem., 1984, 264, C10–C13.
5 (a) G. I. Nikonov, L. G. Kuzmina, D. A. Lemenovskii and
V. V. Ko t ov, J. Am. Chem. Soc., 1995, 117, 10133; (b) G. I. Nikonov,
L. G. Kuzmina, D. A. Lemenovskii and V. V. Kotov, J. Am. Chem.
Soc., 1996, 118, 6333; (c) G. I. Nikonov, L. G. Kuzmina, S. F.
Vyboishchikov, D. A. Lemenovskii and J. A. K. Howard, Chem. Eur.
J., 1999, 5, 2497; (d ) V. I. Bakhmutov, J. A. K. Howard, D. A. Keen,
L. G. Kuzmina, M. A. Leech, G. I. Nikonov, E. V. Vorontsov and
C. C. Wilson, J. Chem. Soc., Dalton Trans., 2000, 1631; (e) S. B.
Duckett, L. G. Kuzmina and G. I. Nikonov, Inorg. Chem. Commun.,
2000, 3/3, 126; ( f ) G. I. Nikonov, P. Mountford, S. K. Ignatov, J. C.
Green, P. A. Cooke, M. A. Leech, L. G. Kuzmina, A. G.
Razuvaev, N. H. Rees, A. J. Blake, J. A. K. Howard and D. A.
Lemenovskii, J. Chem. Soc., Dalton Trans., 2001, 2903.
6 M. D. Curtis, L. G. Bell and W. M. Butler, Organometallics, 1985,
4, 701.
7 (a) D. H. Berry, T. S. Koloski and P. J. Carroll, Organometallics,
1990, 9, 2952; (b) Q. Jiang, P. J. Carroll and D. H. Berry,
Organometallics, 1991, 10, 3648.
8 A. Antiñolo, F. Carrillo, M. Fajardo, A. Otero, M. Lanfranchi and
M. A. Pellinghelli, Organometallics, 1995, 14, 1518.
9 M. L. H. Green and A. K. Hughes, J. Organomet. Chem., 1996, 506,
221.
10 Complex 4 was obtained as a mixture of central (silyl in the central
position in the niobocene plane 4a) and lateral isomers (silyl in the
lateral position 4b).
21 R. J. P. Corriu and J. C. Young, in The Chemistry of Organosilicon
Compounds, ed. S. Patai and Z. Rappoport, Wiley, 1989, p. 1245.
22 H. Kobayashi, K. Ueno and H. Ogino, Organometallics, 1995, 14,
5490.
23 M. A. Esteruelas, F. J. Lahoz, E. Oñate, L. A. Oro and L. Rodríguez,
Organometallics, 1995, 14, 263.
24 M. Driess, H. Pritzkow and U. Winkler, J. Organomet. Chem., 1997,
529, 313.
25 H.-U. Reisacher, E. N. Duesler and R. T. Paine, J. Organomet.
Chem., 1997, 539, 37.
26 M. Weinmann, G. Rheinwald, L. Zsolnai, O. Walter, M. Buchner,
B. Schiemenz, G. Huttner and H. Lang, Organometallics, 1998, 17,
3299.
27 K. Kawamura, H. Nakazawa and K. Miyoshi, Organometallics,
1999, 18, 1517.
28 N. I. Kirilova, A. I. Gusev and Yu. T. Struchkov, Zh. Strukt. Khim.,
1974, 15, 718.
29 For a review see: B. M. Bulychev, Polyhedron, 1990, 9, 387.
30 (a) V. K. Belskii, A. I. Sizov, B. M. Bulychev and G. L. Soloveichek,
Koord. Khim., 1985, 11, 1003; (b) V. K. Belskii, A. I. Sizov, B. M.
Bulychev and G. L. Soloveichik, J. Organomet. Chem., 1985, 280, 67;
(c) N. Etkin, A. J. Hoskin and D. W. Stephan, J. Am. Chem. Soc.,
1997, 119, 11420.
31 (a) G. S. Girolami, G. Wilkinson, M. Thornton-Pett and M. B.
Hursthouse, J. Am. Chem. Soc., 1978, 105, 6752; (b) C. G. Howard,
G. S. Girolami, G. Wilkinson, M. Thornton-Pett and M. B.
Hursthouse, J. Chem. Soc., Dalton Trans., 1985, 921; (c) A. R.
Barron, D. Lyons, G. Wilkinson, M. Motevalli, A. J. Howes and
M. B. Hursthouse, J. Chem. Soc., Dalton Trans., 1986, 279.
32 M. D. Fryzuk, G. K. B. Clentsmith and S. J. Rettig, Inorg. Chim.
Acta, 1997, 259, 51.
33 T. J. McNeese, S. S. Wreford and B. M. Foxman, J. Chem. Soc.,
Chem. Commun., 1978, 500.
34 G. M. Sheldrick, SHELXTL 5.04/VMS, An integrated system for
solving, refining and displaying crystal structures from diffraction
data, Siemens Analytical X-ray Instrument Inc., Madison, WI,
1995.
11 M. L. H. Green, A. K. Hughes and P. Mountford, J. Chem. Soc.,
Dalton Trans., 1991, 1407.
12 D. A. Lemenovskii, I. E. Nifant’ev, E. G. Perevalova, T. V.
Timofeeva, Yu. L. Slovokhotov and Yu. T. Struchkov, J. Organomet.
Chem., 1986, 342, 31.
3046
J. Chem. Soc., Dalton Trans., 2002, 3037–3046