Bent vs. linear imido ligation at the octahedral molybdenum(VI) dithiocarbamate stabilised centre

Article abstract of DOI:10.1039/a907382e

Molybdenum(VI) bis(imido) complexes [Mo(NR)₂(S₂CNEt₂)₂] 3a-3p have been prepared via two synthetic routes and their structures investigated. Thermolysis of [MoO₂(S₂CNEt₂)₂] 1 with 2,6-disubstituted aryl isocyanates affords air-stable bis(imido) complexes 3i-3m in high yields after column chromatography, a route which fails for sterically less demanding isocyanates. A more general preparation involves room temperature addition of two equivalents of [S₂CNEt₂]^- to the labile bis(imido) complexes [Mo(NR)₂Cl₂(dme)] 2a-2p (dme = 1,2-dimethoxyethane) to afford 3a-3p in high yields. In this manner related complexes [Mo(NR)₂(S₂CNMe₂)₂] 4 (R = Ph or Bu^t) and [Mo(NPh)₂(S₂PPh₂)₂] 5a have also been prepared. Structural and spectroscopic studies have been carried out in order to determine the degree of bending of the imido ligands. Six complexes have been characterised by X-ray crystallography. In all cases one of the imido ligands is linear (Mo-N-C ≥ 160°), while the angle at the second is dependent upon the nature of R and ranges from 180 to 139°. Solid-state ¹³C CPMAS spectroscopy reveals that the difference in bond angles at the two nitrogen centres (Δ°) correlates closely with the difference in chemical shifts of the ipso-carbon atoms (Δδ). This then has been used to probe the relative bond angles at nitrogen for non-crystallographically characterised complexes. In solution even at low temperatures both imido ligands are equivalent indicating a rapid interconversion between linear and bent bonding modes. The Royal Society of Chemistry 1999.

Full text of DOI:10.1039/a907382e

View Article Online / Journal Homepage / Table of Contents for this issue
Bent vs. linear imido ligation at the octahedral molybdenum(VI)
dithiocarbamate stabilised centre
Patrick Barrie, Thérèse A. Coffey, Glyn D. Forster and Graeme Hogarth*
Chemistry Department, University College London, 20 Gordon Street, London,
UK WC1H 0AJ. E-mail: g.hogarth@ucl.ac.uk
Received 13th September 1999, Accepted 29th October 1999
Molybdenum() bis(imido) complexes [Mo(NR)2(S2CNEt2)2] 3a–3p have been prepared via two synthetic routes and
their structures investigated. Thermolysis of [MoO2(S2CNEt2)2] 1 with 2,6-disubstituted aryl isocyanates affords air-
stable bis(imido) complexes 3i–3m in high yields after column chromatography, a route which fails for sterically less
demanding isocyanates. A more general preparation involves room temperature addition of two equivalents of
[S2CNEt2]Ϫ to the labile bis(imido) complexes [Mo(NR)2Cl2(dme)] 2a–2p (dme = 1,2-dimethoxyethane) to afford 3a–
3p in high yields. In this manner related complexes [Mo(NR)2(S2CNMe2)2] 4 (R = Ph or But) and [Mo(NPh)2(S2PPh2)2]
5a have also been prepared. Structural and spectroscopic studies have been carried out in order to determine the
degree of bending of the imido ligands. Six complexes have been characterised by X-ray crystallography. In all cases
one of the imido ligands is linear (Mo–N–C ≥ 160Њ), while the angle at the second is dependent upon the nature of R
and ranges from 180 to 139Њ. Solid-state 13C CPMAS spectroscopy reveals that the difference in bond angles at the
two nitrogen centres (∆Њ) correlates closely with the difference in chemical shifts of the ipso-carbon atoms (∆δ). This
then has been used to probe the relative bond angles at nitrogen for non-crystallographically characterised complexes.
In solution even at low temperatures both imido ligands are equivalent indicating a rapid interconversion between
linear and bent bonding modes.
etry.3 Indeed, simple well characterised examples of bent imido
Introduction
ligands remain rare,14–17 although complexes are also known in
Considerable current interest surrounds transition metal imido
which the imido ligand is forced to bend by virtue of being part
complexes, primarily since they are anticipated to be potential
of a chelate ring,18–24 and transient bent imido complexes have
aziridination and amination reagents,1 but also as a result of
also been proposed.25–27 In 1979, Haymore et al.14 gave details
their application as catalysts for ring opening metathesis poly-
of the first bent imido complex, reporting the synthesis and
crystal structure of [Mo(NPh)2(S2CNEt2)2]. This molyb-
denum() bis(imido) complex was shown to contain inequiv-
alent imido ligands in the solid state, being characterised by
merisation (ROMP).2 The imido moiety is a strong π-donor
ligand, and as such is generally found bound to metals in high
oxidation states, that is with d0–d2 electronic configurations.3
When binding to a single metal centre, two extreme co-
bent and linear Mo–N–C angles of 139.4(4) and 169.9(2)Њ
ordination geometries can be identified, being classified as
respectively at Ϫ80 ЊC. Such a geometry is easily understood in
linear when the M–N–C angle is between 180 and 160Њ and bent
terms of the valence bond descriptions of linear and bent imido
ligation previously alluded to. Thus, the molybdenum bis-
(dithiocarbamate) moiety is a 12-electron fragment and the
further 6 electrons required for an 18-electron configuration at
when between 150 and 130Њ.3 These can easily be conceptual-
ised in terms of valence bond theory, such that in the bent case
a lone pair of electrons is localised on nitrogen (A), while for
the metal centre are provided by 2- and 4-electron bent and
linear imido ligands respectively.
During attempts to develop applications of imido complexes
as nitrene transfer reagents, high-valent imido complexes con-
taining only linear imido ligands have generally shown poor
reactivity towards a range of unsaturated organics and thus are
unsuitable in this capacity.3 We reasoned that complexes con-
linear co-ordination the two electrons are delocalised between
nitrogen and the metal centre (B). A further linear form (C)
results when symmetry restrictions do not allow for lone-pair
donation, giving a linear imido ligand with a formal metal–
nitrogen double bond.3,4 If we consider the electron donor
capacity of the imido moiety in terms of a radical formulation,
then we would consider the linear (B) and bent (A) forms to
be 4- and 2-electron donor ligands respectively. For M–N–C
angles of between 160 and 150Њ the term “semi-bent” has
appeared in the literature,4–12 although it is now generally con-
sidered that this falls at the low end of linear co-ordination.3,12
During the thirty years since the report of the first imido
containing complex, [OsO3(NtBu)],13 a large number of such
complexes have been prepared and structurally characterised
and in by far the vast majority the imido adopts a linear geom-
taining the bent ligand were likely to be far more reactive, since
sterically the nitrogen centre was more accessible, the localis-
ation of charge should induce higher reactivity, while the
formal metal–nitrogen double bond should be less thermo-
dynamically stable than linear triply bonded counterparts
and thus more easily be cleaved. In light of the success of the
analogous bis(oxo) complexes towards a range of oxo transfer
reactions,28 molybdenum() bis(imido) complexes appeared
to be particularly attractive as nitrene transfer reagents. During
the transfer of the multiply bonded ligand, the metal centre
formally undergoes a 2-electron reduction and thus ancillary
ligands should be capable of stabilising the metal centre in a
range of oxidation states. Dithiocarbamate ligands have this
ability, for example they are known to stabilise molybdenum in
oxidation states from ϩ6 to Ϫ1,29 and thus bis(imido) com-
J. Chem. Soc., Dalton Trans., 1999, 4519–4528
This journal is © The Royal Society of Chemistry 1999
4519

View Article Online
Chart 1
plexes of the type [Mo(NR)2(S2CNEt2)2] appear to be poten-
deactivated alumina. In all cases, however, the major reaction
tially attractive nitrene transfer reagents.
product was the molybdenum() imidodisulfur complex
[Mo(NR)(S2)(S2CNEt2)2] formed as a result of a double sulfur–
carbon bond cleavage of a dithiocarbamate ligand,32 while
small amounts of dimeric molybdenum() complexes were
also isolated.33 With less sterically demanding aryl isocyanates
significant amounts of the bis(imido) complexes were not
isolated, but rather yields of the co-products detailed above
increased. The only exception to this was the reaction with
ortho-tolyl isocyanate in which 3b was prepared in moderate
amounts, however since it did not survive chromatography
separation from co-products proved impossible. Reaction
with the bulky tertiary butyl isocyanate did not afford 3n. We
attribute the latter to the reduced reactivity of alkyl versus
aryl isocyanates in oxo substitution reactions, a result of the
electron-releasing alkyl substituent reducing the dipole
moment at carbon. The reason for the lack of formation of
the relevant bis(imido) complexes from reactions with steric-
ally undemanding aryl isocyanates was initially thought to be
the thermal instability of the products. However, prolonged
thermolysis of d8-toluene solutions of 3a and 3b in sealed
NMR tubes revealed no evidence of degradation products.
In contrast, however, heating the mixed oxo–imido complex
[MoO(NPh)(S2CNEt2)2]35 under the same conditions resulted
in slow formation of the imidodisulfur complex, [Mo(NPh)(S2)-
(S2CNEt2)2].32 Thus, we conclude that, while in all cases the
first oxo substitution readily occurs at 110 ЊC, for sterically
undemanding isocyanates the product formed is unstable at
these elevated temperatures and dithiocarbamate degradation
must be more rapid than the second substitution. In the case
of the 2,6-disubstituted aryl isocyanates, the second substitu-
tion process must be in competition with the ancillary ligand
degradation and thus significant amounts of the bis(imido)
complexes are produced. We have not been able to ascertain
whether this is attributable to an increase in the rate of
the second substitution process or a decrease in the rate of the
dithiocarbamate degradation. On steric grounds the former
seems improbable, and since both electron-releasing and
-withdrawing substituents favour bis(imido) formation it is
difficult to see how electronic factors influence the rate of the
second substitution. We thus favour a decrease in the rate of
dithiocarbamate degradation upon introduction of 2,6-
substituents on the aryl ring. Since this process involves addi-
tion of the sulfur–carbon bonds to the metal centre, then an
increase in the co-ordination number is anticipated in the
transition state, the energy of which will be increased upon
Two examples of this class of complex have previously been
prepared, namely the phenyl and para-tolyl derivatives and on
the basis of the crystallographic structure of the former they were
assumed to adopt a similar structure, that is with one linear
and one bent imido ligand.14 The synthetic route employed by
Maatta and Wentworth involves oxidative addition of two
equivalents of an aryl azide to the molybdenum() dicarbonyl
complex [Mo(CO)2(S2CNEt2)2]. Reasoning that if such com-
plexes were to act as nitrene transfer reagents, then a range of
aryl- and alkyl-substituted analogues would be required, but
wanting to avoid use of organic azides we sought alternative
synthetic routes to such complexes. Herein we describe the
development of two of these, the first being quite general, while
the second is only applicable for aryl complexes with ortho-
substituents. Investigation of their solid-state structural proper-
ties by both room temperature X-ray crystallographic and 13C
CPMAS NMR spectroscopy reveals that the difference in bond
angles at the two imido ligands depends critically upon the
nature of the organic substituent, which in turn affects the
reactivity of the complexes with respect to their stability in air
and their ability to act as nitrene transfer reagents. An aspect of
this work has been previously communicated.30
Results and discussion
Synthesis
Imido ligands used in this work are shown in Chart 1 together
with the two synthetic pathways utilised. As previously alluded
to, Maatta and Wentworth14 prepared the bis(imido) complexes
3a and 3d via addition of two equivalents of the respective aryl
azide to [Mo(CO)2(S2CNEt2)2].14 We have now developed two
new synthetic routes to 3. The substitution of oxo for imido
ligands upon addition of isocyanates to metal oxides has widely
been utilised in order to prepare a range of new imido com-
plexes.31–33 This strategy benefits from the ready availability of
a wide range of metal oxides and organic isocyanates, but a
major disadvantage is the high temperatures generally required.
Since the required molybdenum() bis(oxo) complex
[MoO2(S2CNEt2)2] 1 is easily prepared on a large scale,34 this
appeared to be a potentially excellent route to 3. Indeed,
thermolysis of toluene solutions of 1 with a slight excess of 2,6
disubstituted aryl isocyanates afforded the required bis(imido)
complexes 3i–3m in yields of 14–27% after chromatography on
4520
J. Chem. Soc., Dalton Trans., 1999, 4519–4528

View Article Online
introduction of sterically demanding ligands within the co-
ordination sphere.
observations, and specifically the behaviour of 3a, to the rapid
interconversion of bent and linear imido functionalities in
solution on the NMR timescale (Scheme 1).
Since the isocyanate route is limited to arylimido complexes
containing 2,6-substituents, we sought a more general syn-
thetic strategy towards 3. Schrock and co-workers36 have
developed a very simple and generally applicable route to
molybdenum() bis(imido) complexes of the type [Mo-
(NR)2Cl2(dme)] 2 (dme = 1,2-dimethoxyethane), which has
been extended and elaborated on by Gibson and his group.5,11
This method is particularly beneficial since the source of the
imido ligands is widely available primary amines, while sodium
molybdate is cheap and plentiful. Addition of two equivalents
of [NH4][S2CNEt2] or dehydrated Na[S2CNEt2] to diethyl ether
solutions of [Mo(NR)2Cl2(dme)] 2a–2p at room temperature
resulted over approximately 12 h in the slow formation of
[Mo(NR)2(S2CNEt2)2] 3a–3p in yields of 20–96% after removal
of volatiles and extraction into benzene. In all cases the
reaction was clean and appreciable amounts of other metal-
containing products were not formed. The extraction pro-
cedure proved necessary to remove salts and the isolated yields
of the bis(imido) complexes were to some extent a function of
Scheme 1
1
While the aromatic region of the H NMR spectra does not
change significantly upon varying the temperature from Ϫ80 to
ϩ100 ЊC, substantial changes were seen in other regions. At
elevated temperatures the methyl and ethyl resonances of the
dithiocarbamate ligands appeared as a sharp triplet and quartet
respectively; upon lowering the temperature these broadened
and then reappeared as pairs of triplets and quartets. We
attribute these changes to the well established restricted
rotation about the carbon–nitrogen bond of dithiocarb-
amates,29 a result of the resonance hybrid in which positive
charge is localised on nitrogen. For the 2,6-diisopropyl-
1
their solubility; examination of the reaction mixtures by H
NMR spectroscopy showed that all reactions were essentially
quantitative.
1
substituted complex, 3j, other features of the H NMR spec-
trum were also found to be temperature dependent. At 100 ЊC
the methyls of the isopropyl group appear as a single triplet,
which collapses upon cooling into two equal intensity triplets at
Ϫ20 ЊC, and we associate these changes with the geared rotation
of the aryl groups about the carbon–nitrogen bond and in sup-
port of this supposition have previously noted the restricted
rotation of other sterically demanding arylimido ligands.33
All aryl complexes were brown or red-brown solids, while
alkyl complexes were orange. The stability of the complexes in
air was found to be highly dependent upon the nature of the
organic substituent. Thus, solutions of the 2,6-disubstituted
complexes 3i–3m proved to be indefinitely stable in air and
could easily be recrystallised upon slow diffusion of methanol
into dichloromethane solutions. The tertiary butyl complex 3n
and aryl complexes containing a single ortho-substituent 3b and
3e were air-stable in the solid state, but decomposed over a
period of hours when solutions were exposed to air. The
phenylimido complex 3a and other complexes which did not
have substituents in the ortho positions were found to be highly
air-sensitive in solution, decomposing over a period of minutes,
while even in the solid state reaction with moisture resulted in
the slow formation of the bis(oxo) complex and primary
amines. Recrystallisation of the more air-sensitive complexes
was best effected upon cooling concentrated diethyl ether
solutions to Ϫ40 ЊC.
Solid-state structures
In order to probe the changes in bond angles at nitrogen upon
varying the nature of the organoimido ligands, room temper-
ature X-ray crystallographic studies were carried out on five of
the new bis(imido) complexes, namely 3b, 3i–3k and 3n. Since
the original solid-state studies of 3a were carried out at
Ϫ130 ЊC, for the sake of direct comparison, the structure of the
latter was redetermined at room temperature and only slight
differences were noted. The results of these studies are summar-
ised in Fig. 1 and Table 1, which shows a comparison of the
essential structural features of the six bis(imido) complexes at
room temperature and 3a at Ϫ130 ЊC.
Characterisation and solution fluxionality
All show the same gross structural features and consist of
discrete molecular units which contain a central, approximately
octahedral, molybdenum centre. The imido ligands adopt a
relative cis conformation with the N–Mo–N angle varying
between 101.4(3) and 107.3(3)Њ. This variation appears not to
be a function of steric effects since it is the bulky 2,6-Cl2 aryl
complex 3k for which the closest approach is found. The struc-
ture of 3a is widely quoted in the literature as a rare example of
the coexistence of both linear and bent imido moieties at the
same metal centre,14 the difference in bond angles at nitrogen
being 29.5(8)Њ. Inspection of the other five structures reveals
that the angles at nitrogen are highly sensitive to the nature of
the organic substituent. In the case of the bulky 2,6-disubsti-
tuted aryl complexes 3i–3k, the two bond angles are the same
within experimental error (in 3j the imido ligands are symmetry
related) while differences of 8.3(7) and 10.7(1.1)Њ are found for
3b and 3n respectively. In all the five new structures, however,
the most acute angle is 162Њ and both imido ligands can only be
considered as being linear. Thus, while the bond angle at nitro-
gen is quite soft in these complexes, varying between 162.0(6)
and 175.1(4)Њ, they differ with respect to 3a in best being classi-
fied as linear–linear rather than linear–bent species.
Characterisation of bis(imido) complexes 3 proved straight-
forward. The IR spectra were relatively uninformative showing
the anticipated absorptions associated with the dithio-
carbamate ligands, the absence of metal–oxo stretches and the
appearance of absorptions between 1300 and 1200 cmϪ1 tenta-
tively assigned to the molybdenum–nitrogen moieties.3,37 EI
Mass spectra showed molecular ions for all complexes, however
it is noteworthy that FAB spectra consistently showed ions that
were too heavy for the mononuclear formulation. At room tem-
perature, 1H NMR spectra confirmed that the ratio of imido to
dithiocarbamate ligands was 1:1, the latter appearing as a
quartet and triplet respectively at room temperature. The sim-
plicity of these signals suggests that the two imido ligands are
equivalent under these conditions, and indeed for all complexes
a single imido environment was observed. While this is to be
expected if both of the imido ligands are linear, if as is the case
with 3a, one is bent and the other linear, then two separate
environments are expected. For the latter, Maatta and Went-
worth14 previously reported the observation of a complex
multiplet (no further details were given) in the aromatic region
of the spectrum. In contrast to this, we find that at all temper-
atures this region of the spectrum appears simply as a doublet
and two triplets as would be expected for two equivalent phenyl
groups. We also find analogous behaviour for all other com-
plexes down to temperatures of Ϫ100 ЊC. We attribute these
It is well established that the metal–nitrogen bonds in a wide
range of imido complexes show only a small variation in
length.3 The linear–bent complex 3a is characterised by short
[1.745(5) Å] and long [1.774(5) Å] molybdenum–nitrogen
J. Chem. Soc., Dalton Trans., 1999, 4519–4528
4521

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ)
Complex
Mo–Na
Mo–Sb
Mo–Sc
N–Ca
Mo–N–Ca
N–Mo–N
S–Mo–Sd
3a
3b
3i
1.774(5)
1.745(5)
1.776(5)
1.757(6)
1.770(6)
1.756(6)
1.776(2)
1.760(7)
1.769(7)
1.753(7)
1.721(8)
1.789(4)
1.754(4)
2.762(2)
2.605(2)
2.682(2)
2.652(2)
2.685(2)
2.684(2)
2.672(1)
2.648(3)
2.651(3)
2.719(2)
2.694(3)
2.755(2)
2.602(2)
2.464(2)
2.454(2)
2.462(2)
2.458(2)
2.462(2)
2.462(2)
2.464(1)
2.452(3)
2.449(3)
2.469(3)
2.468(3)
2.461(2)
2.453(2)
1.402(9)
1.383(9)
1.352(9)
1.373(8)
1.374(9)
1.398(9)
1.376(3)
1.368(11)
1.346(11)
1.418(10)
1.429(15)
1.392(6)
1.386(6)
141.4(5)
170.9(6)
166.8(6)
175.1(4)
162.0(6)
162.8(6)
169.9(2)
162.2(7)
162.9(7)
162.5(7)
173.2(8)
139.4(4)
169.4(4)
103.2(2)
105.7(3)
102.5(3)
67.3(1)
69.2(1)
68.6(1)
68.3(1)
68.2(1)
68.0(1)
68.6(1)
68.6(1)
68.8(1)
67.4(1)
67.6(1)
69.51(4)
67.60(5)
3je
3k
104.7(1)
101.4(3)
3n
107.3(3)
103.5(2)
3a (143 K)f
a First entry corresponds to the most bent and the second the least bent imido ligand. b First entry is that trans to the most bent and the second is trans
to the least bent imido ligand. c cis to the imido ligands: first entry is associated with the same dithiocarbamate ligand as that trans to the most bent
imido ligand, etc. d Associated with the dithiocarbamate ligands represented by the distances in the rows. e The molecule has a crystallographic plane
of symmetry. f Ref. 14.
Fig. 1 Molecular structures of bis(imido) complexes 3.
vectors respectively. For the four new arylimido complexes, the
molybdenum–nitrogen bond lengths vary between 1.776(5)
and 1.756(6) Å, with the largest difference of 0.019(7) Å being
found in the ortho-tolyl complex 3b, which also exhibits the
largest bond angle difference of 8.3(7)Њ. For the alkyl-
substituted complex 3n a difference in bond lengths of
0.032(10) Å is found, which is close to that of 0.029(7) Å seen in
linear–bent 3a. The former is, however, only associated with a
10.7Њ difference in bond angles at nitrogen. In the three com-
plexes where there is a significant difference in bond angles at
nitrogen, namely 3a–3b and 3n, it is noteworthy that the shorter
molybdenum–nitrogen bond is associated with the more linear
imido ligand and vice versa. Thus, while in general metal–
nitrogen bond lengths may be somewhat insensitive to the
degree of bending at nitrogen, the small differences observed
may be significant.
As has previously been noted for dithiocarbamate-stabilised
molybdenum imido complexes,14 the nature of molybdenum–
sulfur interactions are highly dependent upon the relative
orientation of the chalcogen with respect to the strong π-donor
ligand. Thus, for all complexes, the molybdenum–sulfur vectors
cis to the imido ligands are short and essentially invariant
[2.449(3)–2.469(3) Å], while those trans to imido ligands are
elongated significantly [2.762(2)–2.605(2) Å], the precise value
being highly dependent upon the nature of the imido ligand it
lies trans to, long distances being associated with bending at
nitrogen and vice versa. Thus, the trans molybdenum–sulfur
bond length of 2.762(2) Å in 3a is the only one associated with
a truly bent, 2-electron donor imido ligand, the next longest
distance being 2.719(2) Å found in the tertiary-butyl complex
3n and associated with a bond angle at nitrogen of 162.5(7)Њ.
For the linear imido ligand in 3a, formally a 4-electron donor,
4522
J. Chem. Soc., Dalton Trans., 1999, 4519–4528

View Article Online
Table 2 13C CPMAS data for the ipso-carbons (δ) and Mo–N–C bond
angle differences (Њ) for crystallographically characterised complexes
(first block) and extrapolated values from the NMR data (second
block)
the trans molybdenum–sulfur bond length of 2.605(2) Å is
significantly shorter than for the other bis(imido) complexes.
In the latter, the range of trans molybdenum–sulfur bond
lengths is far more limited varying only between 2.651(3) and
2.694(3) Å, while the biggest difference within any one complex
is 0.030(3) Å seen in 3b, which is considerably smaller than the
0.157(3) Å difference found in 3a. We believe that this is further
evidence that in the five new complexes the essentially linear
imido ligands are best considered as acting as net 3-electron
donors to the molybdenum centre.
Complex
ipso-C
∆δ
∆Њ
3a
3b
3i
161.5 156.6
158.6 157.3
157.4
153.9
152.1
71.7 69.7
155.5 151.7
146.2
5.0
1.3
0
0
0
2.0
3.8
0
29.5
8.3
0.8
0
0.7
10.7
23
0
3j
3k
3na
3d
3e
3f
Other structural features are generally unremarkable. The
small bite-angle dithiocarbamate ligands subtend angles of
approximately 68Њ at molybdenum, while no obvious trends are
apparent in the N–Cipso distances. One further significant point
of note, however, is that in all structures the most bent imido
ligand bends in towards the second and not away from it. At
first sight this appears counter intuitive, especially for the
relatively bulky 2,6-disubstituted complexes 3i–3k, since the
organoimido ligands would generally be considered as more
sterically demanding than dithiocarbamates. The bending of
imido ligands towards each other has previously been observed
in [OsO2(NR)2].5 Given the lack of steric hindrance here it
suggests that the origin of the bending is electronic in nature.
For both complexes 3i and 3j, the origin of the different
155.5
0
0
3g
3h
3l
158.2 155.4
160.4 156.8
153.2
71.1
160.3
72.8 69.9
156.7 155.1
2.8
3.6
0
0
0
2.9
1.6
17
22
0
0
0
15
9
3oa
4a
4n
5a
a Values refer to the α-carbons of the tBu groups.
1
alkyl signals in the low-temperature limiting H NMR spectra
is easily explained in light of the solid-state structures, being
attributed to the restricted rotation about the N–Cipso vector.
Thus, for each imido ligand one of the substituents lies over
the aryl ring of the second imido ligand, while the other lies
towards the ethyl substituents of the dithiocarbamate ligand.
With respect to this, a comparison of the structure of the ortho-
tolyl complex 3b with that of the dimethyl substituted 3i reveals
that the preferred orientation is that in which the methyl group
lies over the aryl ring of the second since the second rotamer
was not observed either in solution or the solid state.
13C CPMAS NMR Spectroscopy
From the crystallographic analyses described above, it is appar-
ent that the metal–nitrogen–carbon angle in the bis(imido)
complexes 3 is relatively soft, varying over approximately 40Њ.
As the study developed it became obvious that, short of carry-
ing out a single crystal X-ray diffraction study on each new
complex, another method of characterising the degree of bend-
ing of the M–N–C angle was needed. Since even in 3a, which
has very different imido environments in the solid state, imido
ligands interconvert in solution via a process which we have not
been able to freeze out even at low temperatures, then a solid-
state technique was required. To the best of our knowledge,
solid-state NMR data have not previously been reported for
imido complexes. This is, however, potentially an excellent way
of characterising differences in bond angles at nitrogen with
bis- and poly-imido complexes, since the chemical shift of the
carbon directly bound to nitrogen is likely to be highly depend-
ent upon the degree of bending at the latter.
Fig. 2 13C CPMAS spectra of the aryl region of bis(imido) complexes
3.
In order to test this hypothesis, the 13C CPMAS spectra of
complexes 3 were recorded at room temperature. The results are
given in Table 2 while key spectra are shown in Fig. 2. While the
general features of the spectrum are as expected, the main
feature is the appearance of inequivalent ipso-carbon reson-
ances at δ 161.5 and 156.5, that is a ∆δ of 5.0 ppm. This con-
firms the results of the crystallographic studies, that is the two
imido environments are quite different. In contrast, but again in
accord with the structural study, the spectrum of 3j shows only
a single ipso-carbon resonance (∆δ 0 ppm) as expected for
equivalent imido environments. Similar spectra for the other
crystallographically characterised complexes revealed ∆δ to be
a function of the difference in bond angles (∆Њ) at nitrogens, 3b
(1.3 ppm, 8.3Њ); 3i/3k (0 ppm, 0Њ); 3n (2.0 ppm, 10.7Њ), the latter
representing the difference in chemical shift of the α-carbons of
the t-butyl groups. These data are presented graphically in Fig.
3, the main feature of which is the relatively linear nature of the
relationship, although it is acknowledged that the plot com-
prises only four points. Regardless of the physical basis of this
relationship, it provides a quick and easy method of screening
new complexes. Hence from the measurement of the chemical
shift difference (∆δ), the difference in bond angles at nitrogen
(∆Њ) can easily be determined. The results of this analysis for a
number of complexes 3 are displayed in Table 2. It is clear from
these data that complexes which are substituted in one or both
ortho positions always adopt the linear–linear conformation.
The meta- and para-substituted complexes, however, adopt a
more distorted co-ordination geometry. Thus, ∆Њ for the tolyl
derivatives vary in the order para (23.1Њ) 3d, para-meta (17.0Њ)
J. Chem. Soc., Dalton Trans., 1999, 4519–4528
4523

View Article Online
ther, a 15N CPMAS TOSS (total sideband suppression) experi-
ment on 3b run over 5 d showed a single strong dithiocarbamate
resonance at δ Ϫ206.0 and two very weak broad resonances at
δ 67 and 55 tentatively assigned to the inequivalent imido
environments (∆Њ 8.3Њ). The observation of only a single dithio-
carbamate resonance, however, suggests that interpretation of
15N CPMAS data in terms of imido bending requires further
consideration.
Other ancillary ligands
A further approach adopted was to explore briefly the effect of
changing the ancillary ligand on the structure adopted. For this
[Mo(NPh)2(S2CNMe2)2] 4a and [Mo(NPh)2(S2PPh2)2] 5 were
prepared in 40 and 52% yield respectively in a manner analo-
gous to the synthesis of 3a. At room temperature both showed
Fig. 3 Relationship between ∆δ and ∆Њ for bis(imido) complexes 3.
1
a single imido-phenyl environment in the H NMR spectrum
3g (a sufficiently pure sample of 3c was not obtained), ortho
(8.3Њ) 3b. Since the electronic properties of these substituents is
anticipated to be essentially invariant, we surmise that it is
steric factors which are primarily responsible for these differ-
ences. A further series of dimethyl-substituted complexes 3f–3i
also supports this hypothesis, since it is only those that are not
ortho substituted which show any significant difference in bond
angles. That this is not a result of the electron-donating nature
of the methyl group is confirmed by the observation that 2,6-
dichloro 3k and 2,6-difluoro 3l derivatives both adopt equiv-
alent imido environments.
This analysis only provides information as to the relative
bending of ligands and cannot give absolute bond angles at
nitrogen. We have tried to correlate the chemical shifts (cor-
rected for substituent effects) with bond angles, however no
meaningful relationship could be discerned. If we assume that,
as is the case with all structurally characterised complexes, one
of the imido ligands is linear (160–180Њ), then for complexes
with a measurable ∆δ we can place upper and lower limits on
the degree of bending at the second ligand. For example, the
para-tolyl complex 3d has a ∆δ of 3.8 ppm, that is a ∆Њ of 23.1Њ
placing limits on the more bent imido ligand of 157–137Њ. Per-
haps the most noteworthy feature of the solid-state 13C NMR
spectra is the realisation that the parent complex 3a, originally
prepared by Maatta and Wentworth,14 is significantly different
to all other complexes of this type reported here. Hence, it is the
only complex that can unambiguously be described as contain-
ing both linear and bent imido moieties and appears to repre-
sent an extreme case.
In light of the sensitivity of the chemical shift of the ipso-
carbon to variations in bond angle, solid-state 15N NMR spec-
troscopy should be a useful tool in determining the degree of
bending of imido ligands. Bradley et al.38 have previously
recorded natural abundance 15N NMR spectra for a range of
imido complexes in solution. Notably in this study the spectrum
of 3a was shown to contain only a single resonance, further
evidence for the interconversion of imido environments. As far
as we are aware, 15N CPMAS spectra have not previously been
reported for imido complexes, although dithiocarbamate com-
plexes have been studied.39 For 3 a major drawback is the low
amounts of natural abundance 15N, which makes accumulation
times very long. In order to test the viability of this technique,
the 15N CPMAS spectrum of 3j (in which imido sites are
crystallographically equivalent) was recorded over 24 h.
Two nitrogen environments are seen, the high-field signal at
δ Ϫ210.82 being assigned to the dithiocarbamate ligands, while
that to lower field at δ 77.37 is due to the equivalent imido
ligands. For 3n at room temperature, two dithiocarbamate
resonances are seen at δ Ϫ214.5 and Ϫ209.0. This is as expected
from the crystal structure in which the two imido ligands are
inequivalent (∆Њ 10.7Њ) and suggests that 15N CPMAS may be a
sensitive probe to bending. However, no resonances could be
observed for the imido nitrogens at room temperature or
100 ЊC, an observation which we are at a loss to explain. Fur-
and as we were unable to grow suitable single crystals of either,
13C CPMAS spectra were recorded. Somewhat surprisingly in
view of the bent–linear structure of 3a only a single ipso-carbon
resonance was observed for 4a (δ 160.3) suggesting that both
imido ligands are equivalent. For 5 two ipso resonances were
seen at δ 156.7 and 155.1, the difference (∆δ 1.6 ppm) being
comparable to a 9Њ difference in bond angles. In order further to
probe 5, a 31P CPMAS spectrum was run and showed two dis-
tinct resonances at δ 68.6 and 65.5, while in solution a single
peak was observed (δ 64.5). The ∆δ here of 3.1 ppm has no
direct comparison but does suggest that 31P NMR may be a
sensitive tool for probing imido environments in appropriate
systems. While these results should be treated with caution as
they are not supported by crystallographic data, they do sug-
gest that the difference in bond angles at the imido ligands (i.e.
the conformation adopted) may be very sensitive to the nature
of the ancillary ligands. Indeed we have also run the 13CPMAS
of [Mo(NPh)2Cl2(dme)] 2a and this shows a single ipso-carbon
resonance as do almost all the other dme complexes 2.
In a further experiment [Mo(NBut)2(S2CNMe2)2] 4n was pre-
pared in 41% yield. In solution a single tertiary butyl resonance
at room temperature was indicative of a single imido environ-
ment, however, two Cα resonances were seen in the 13C CPMAS
spectrum at δ 72.8 and 69.9 (∆δ 2.9) suggesting a bond angle
difference (∆Њ) of around 15Њ. We do not have crystallographic
verification of this but again it suggests that ancillary ligand
choice can play a significant role in the adopted bis(imido)
configuration.
Conclusions
This work has shown that the bond angle at nitrogen in molyb-
denum() bis(imido) complexes depends strongly upon the
nature of the substituents on nitrogen and also on the ancillary
ligand. The precise reason for the former dependence is not
totally clear. Extended Hückel calculations carried out by Nik
Kaltsoyannis and Woo-Sung Kim of our department40 have
revealed that the total energy change upon deformation of one
M–N–C angle in [Mo(NPh)2(S2CNEt2)2] 3a between 180 and
140Њ is essentially zero suggesting that there is no electronic
preference for the adoption of a bent–linear over a linear–linear
conformation. This in turn suggests that the bent–linear struc-
ture found for 3a may be a result of crystal packing effects or
other undetected intermolecular interactions (although no such
4524
J. Chem. Soc., Dalton Trans., 1999, 4519–4528

View Article Online
interactions are seen in the crystal structure). This is further
supported by our solid-state NMR data for [Mo(NPh)2-
(S2CNMe2)2] 4a which suggest that the two imido ligands are
equivalent and as such are both linear. Clearly further work
is needed to determine the precise role of the ancilliary
ligand, and we note that in [Mo(NPh)2Cl2(dme)] 2a again only a
single ipso resonance is seen in the 13C solid-state NMR spec-
trum. Further calculations on [Mo(NBut)2(S2CNEt2)2] 3n and
[Mo(N-C6H3Cl2-2,6)2(S2CNEt2)2] 3k40 show that while the total
energy varies little upon deformation from 180 to 160Њ, at
around 155Њ it rises steeply. This indicates that a major factor
favouring the adoption of linear–linear co-ordination for ortho-
substituted aryl and t-butyl complexes is the adverse steric
interactions between these substituents upon bending.
red 2j (90%); brown 2k (92%); red 2l (65%); brown 2m (88%);
yellow 2n (85%); yellow 2o (90%); red 2p (62%); red 2a-d6
(70%).
[Mo(NR)2(S2CNEt2)2] 3. Solid [Mo(NR)2Cl2(dme)] 2 (4.56
mmol) and [NH4][S2CNEt2] (9.20 mmol) were loaded into a
Schlenk tube. To this was added diethyl ether (40 cm3) and
the mixture stirred for 16 h at room temperature. This solution
was filtered into another Schlenk tube, the remaining residues
were washed with diethyl ether (2 × 25 cm3) and added to the
original solution. The solution was concentrated to ca. 20 cm3
under reduced pressure, cooled to Ϫ20 ЊC to give crystalline 3,
which was isolated after filtration, washed with light petroleum
(10 cm3) and dried under vacuum. Red 3a (53%); brown 3b
(45%); brown 3c (23%); brown 3d (20%); red 3e (42%); brown 3f
(55%); brown 3g (58%); purple 3h (48%); brown 3i (80%);
brown 3j (85%); brown 3k (75%); red 3l (38%); brown 3m (96%);
orange 3n (60%); orange 3o (70%); red 3p (28%); red 3a-d10
(46%).
Experimental
General
All reactions were carried out under an atmosphere of dinitro-
gen using standard vacuum line and glove-box techniques.
Liquids were transferred via stainless-steel cannulae using a
dinitrogen pressure and filtrations by using cannulae modified
to accept a glass micro-fibre filter at one end. Solvents were
dried over suitable drying agents under a dinitrogen atmos-
phere, stored in ampoules over 4 Å molecular sieves or
sodium–potassium alloy and degassed thoroughly before use.
Chromatography was performed on columns of deactivated
alumina (6% w/w water) wet packed with light petroleum (bp
40–60 ЊC). Solutions were absorbed onto approximately 5 g of
alumina before chromatography. Chemicals were purchased
from Aldrich, Fluka, Lancaster Synthesis, Avocado and
Janssen and used as supplied. All amines were distilled prior to
use as was Me3SiCl. Dithiocarbamate salts were dried under
vacuum prior to use, while sodium salts were dehydrated in
vacuo whilst heating at 80 ЊC. Both [MoO2(S2CNEt2)2] 134 and
Na[S2PPh2] were prepared by standard literature methods. 1,2-
Dimethoxyethane (dme) complexes [Mo(NR)2Cl2(dme)] 2 were
prepared via the method of Schrock and co-workers.36 Many
have previously been described, but some here for the first time.
Since these were merely precursors to dithiocarbamate com-
plexes full characterisation was not attempted. Microanalyses
were performed in-house and IR spectra were recorded on a
Nicolet 205-FT-IR spectrometer as KBr discs. Solution NMR
spectra were recorded on a Varian VXR400 spectrometer at
20 ЊC in C6D6 unless otherwise stated and referenced internally
to residual solvent. Air-sensitive samples were prepared in a
glove-box in tubes fitted with a Young’s valve or sealed under
vacuum. Solid-state 13C and 15N NMR spectra were recorded
on a Bruker MSL300 spectrometer as part of the University of
London Intercollegiate Research Service (ULIRS) and mass
spectra were obtained in-house.
The 2,6-disubstituted aryl complexes 3i–3k were also pre-
pared upon thermolysis of toluene solutions of [MoO2(S2-
CNEt2)2] 1 and the appropriate isocyanate. For 3i, [MoO2(S2-
CNEt2)2] 1 (1.34 g, 3.16 mmol) was dissolved in toluene (100
cm3) and 2,6-dimethylphenyl isocyanate (0.50 ml, 3.42 mmol)
added via a degassed syringe. The resulting solution was
refluxed for 1 d, after which time the volatiles were removed
under reduced pressure. The resulting solid was absorbed onto
alumina and passed down a chromatography column. A brown
band [Mo(2,6-Me2C6H3N)2(S2CNEt2)2] 3i (0.60 g, 26%) was
obtained upon elution with 20% dichloromethane in light
petroleum. Complexes 3j (27%) and 3k (14%) were prepared in
an analogous fashion. Crystals of 3i–3k were obtained by slow
diffusion of methanol into dichloromethane solutions.
3a.14 IR: 1501m, 1468m, 1431m, 1383m, 1357w, 1266s,
1207w, 1146m, 1095m, 1068m, 1020m, 803w, 755w, 686m and
670w cmϪ1. Mass spectrum (EI): m/z 575 (Mϩ), 486 (M Ϫ
NPh), 429 (M Ϫ dtc), 397 (M Ϫ 2NPh) and 369 (M Ϫ NPh Ϫ
1
dtc); H NMR: δ 7.47 (d, 4 H, J 7.3, H2,6), 6.97 (t, 4 H, J 7.5,
H3,5), 6.70 (t, 2 H, J 7.4, H4), 3.25 (q, 8 H, J 7.0, CH2) and 0.74
(t, 12 H, J 7.0 Hz, Me). 13C-{1H} NMR: δ 202.8 (C᎐N), 158.7
᎐
(Cipso), 128.6, 125.3, 123.4, 46.1 (CH2) and 12.1 (Me). 13C
CPMAS TOSS: δ 200.7, 161.5 (Cipso), 156.5 (Cipso), 130.1, 128.6,
126.7, 47.5, 44.5, 13.8 and 12.6. Found (calc.) for C22H30MoN4-
S4: C, 45.56 (45.98); H, 5.28 (5.26); N, 9.63 (9.75); S, 21.66
(22.30)%.
3b. IR: 1633w, 1497s, 1432s, 1295m, 1273s, 1206m, 1146m,
1076m and 759m cmϪ1. Mass spectrum (EI): m/z 604 (Mϩ), 499
(M Ϫ NR), 457 (M Ϫ dtc) and 352 (M Ϫ NR Ϫ dtc). 1H NMR
(C6D5CD3): δ (333 K) 7.36 (d, 2 H, J 7.9, H6), 6.83 (dd, 4 H,
J 7.9, 7.4, H3,5), 6.62 (t, 2 H, J 7.4, H4), 3.33 (q, 8 H, J 7.1, CH2),
2.44 (s, 6 H, Me) and 0.84 (t, 12 H, J 7.1 Hz, Me); (213K): 7.55
(d, 2 H, J 7.6, H6), 6.80 (t, 2 H, J 7.6, H5), 6.72 (d, 2 H, J 7.5,
H3), 6.61 (t, 2 H, J 7.3, H4), 3.15 (br, 8 H, CH2), 2.53 (s, 6 H,
Me) and 0.57 (d, 12 H, J 6.2 Hz, Me). 13C-{1H} NMR
Syntheses
[Mo(NR)2Cl2(dme)] 2. White anhydrous sodium molybdate
(2.00 g, 9.72 mmol) was thoroughly degassed and 1,2-
dimethoxyethane (40 cm3) added to give a suspension. Triethyl-
amine (5.4 cm3, 38.98 mmol), trimethylsilyl chloride (10.0 cm3,
78 mmol) and two equivalents of the amine (19.44 mmol) were
added via a degassed syringe to the stirred suspension. This
solution was left to stir at 65 ЊC for 18 h during which time it
became highly coloured (red, brown or yellow). After cooling to
room temperature, filtration and removal of volatiles under
reduced pressure gave solid 2 which was washed with light
petroleum (2 × 20 cm3) and dried under vacuum. The solid was
redissolved in the minimum volume of diethyl ether and fil-
tered. Cooling to Ϫ20 ЊC gave crystalline samples which were
dried and transferred into the glove-box for storage. Red 2a
(72%); red 2b (65%); brown 2c (58%); brown 2d (60%); red 2e
(73%); red 2f (80%); red 2g (82%); red 2h (85%); red 2i (94%);
(C D CD ). δ (333 K) 203.2 (C᎐N), 158.2 (Cipso), 130.2, 129.9,
᎐
6 5 3
126.2, 45.8 (CH2), 18.9 (Me) and 12.3 (Me); (193 K) 200.9
(C᎐N), 157.8 (Cipso), 130.5, 129.2, 126.3, 45.2 (CH2), 45.0 (CH2),
᎐
19.8 (Me), 12.3 (Me) and 11.6 (Me). 13C CPMAS TOSS:
δ 202.4, 199.0, 158.6 (Cipso), 157.3 (Cipso), 134.3, 131.6, 130.1,
127.0 (br), 122.6, 49.1, 45.9, 23.1, 21.2, 19.8, 14.1 and 12.8. 15
N
CPMAS TOSS: δ 67, 55 and Ϫ206. Found (calc.) for
C24H34MoN4S4: C, 47.77 (47.82); H, 5.70 (5.69); N, 9.18 (9.29);
S, 21.33 (21.28)%.
3c. IR: 1527s, 1458s, 1438s, 1402s, 1356m, 1278m, 1205w,
1153m, 1094m, 1075m and 793w cmϪ1. Mass spectrum (EI):
m/z 604 (Mϩ), 499 (M Ϫ NR), 457 (M Ϫ dtc) and 352 (M Ϫ
1
NR Ϫ dtc). H NMR: δ 7.01 (t, J 7.6, 2 H, H5), 6.56 (t, 2 H,
J 7.6, H4,6), 6.49 (s, 2 H, H2) 6.47 (d, J 7.9, 2 H, H4,6), 3.92–3.79
(m, 8 H, CH2), 2.24 (s, 6 H, Me) and 1.46 (t, 12 H, J 7.0 Hz,
Me). Satisfactory elemental analysis could not be obtained.
J. Chem. Soc., Dalton Trans., 1999, 4519–4528
4525

View Article Online
3d.14 1H NMR: δ 7.50 (d, 4 H, J 7.9, H2,6), 6.79 (d, 4 H, J 7.9,
H3,5), 3.27 (q, 8 H, J 7.0, CH2), 1.90 (s, 6 H, Me) and 0.75 (t,
(S2CN), 153.8 (ipso-C), 141.6 (o-C of C6H3), 124.0 (p-C of
C6H3), 122.1 (m-C of C6H3), 46.0, 45.4 (CH2), 28.1 (CH), 24.5,
23.6 ((CH3)2CH), 12.8 and 12.1 (CH3CH2). 13C CPMAS TOSS:
δ 200.4, 182.0, 153.9 (Cipso), 145.0, 142.2, 126.4, 123.0, 45.8,
27.6, 27.2, 25.1, 21.0, 14.7 and 13.9. 13C CPMAS NQSTOSS:
δ 200.4, 153.9, 145.0, 142.2, 27.6, 27.2, 25.1, 21.0, 14.7 and 13.9.
12 H, J 7.0 Hz, Me). 13C CPMAS: δ 203.0 (C᎐N), 155.5 (Cipso),
᎐
151.7 (Cipso), 46.8 (CH2), 22.0 (Me) and 13.4 (Me).
3e. Mass spectrum (EI): m/z 612 (Mϩ), 503 (M Ϫ NR) and
465 (M Ϫ dtc). 1H NMR: δ 7.54 (t, 2 H, J 7.9, H6), 6.68 (t, 2 H,
J 8.0, H4,5), 6.64 (t, 2 H, J 7.7, H4,5), 6.43 (q, J 6.1, 2 H, H3), 3.21
(m, 8 H, CH2) and 0.73 (t, 12 H, J 6.9 Hz, Me). 13C CPMAS
TOSS: δ 201.2, 199.4, 151.0 (C–F), 146.2 (Cipso), 132–116 (br),
46.1 and 12.8. 13C CPMAS NQS (non-quaternary suppres-
sion): δ 201.2, 199.4, 150.9, 146.1, 13.4 and 12.8. Found (calc.)
for C22H28F2MoN4S4: C, 42.16 (43.26); H, 4.68 (4.62); N, 8.75
(9.17); S, 20.07 (21.00)%.
15N CPMAS:
δ 77.4 and Ϫ210.8. Found (calc.) for
C34H54MoN4S4: C, 54.54 (54.96); H, 7.32 (7.32); N, 7.31 (7.54);
S, 17.50 (17.26)%.
3k; IR: 1508s, 1432s, 1311m, 1275m, 1205w, 1070w, 960m
and 790m cmϪ1 1H NMR (CDCl3): δ 7.12 (d, 4 H, J 7.5,
.
H3,5), 6.68 (t, 2 H, J 7.2 Hz, H4), 3.82 (br, 8 H, CH2) and
1.46 (br, 12 H, CH3). 13C CPMAS TOSS: δ 199.9, 152.1 (Cipso),
134.0, 128.9, 125.8, 122.6, 47.6, 46.3 and 11.4. 13C CPMAS
NQSTOSS: δ 199.8, 152.0, 137.4, 130.0, 126.9, 123.2 and 11.4.
Found (calc.) for C22H26Cl4MoN4S4: C, 37.48 (37.08); H, 3.94
(3.68); Cl, 19.90 (19.72); N, 7.63 (7.87); S, 18.07 (17.96)%.
3l. IR: 1508s, 1468s, 1436m, 1384s, 1275m, 1265m, 1236w,
1208w, 1019w, 1001m, 965w and 649w cmϪ1. Mass spectrum
(EI): m/z 647 (Mϩ), 520 (M Ϫ NR) and 500 (M Ϫ dtc). 1H
NMR: δ 6.37 (dd, 4 H, J 7.6, H3,5), 6.12 (m, 2 H, H4), 3.19 (q, 8
H, J 6.7, CH2) and 0.72 (t, 12 H, J 6.7 Hz, Me). 13C-{1H} NMR:
3f. IR: 1497s, 1474m, 1456m, 1429m, 1357m, 1300m, 1274s,
1208w, 1148m, 1092w, 1073w and 814m cmϪ1. Mass spectrum
(EI): m/z 633 (Mϩ), 513 (M Ϫ NR), 483 (M Ϫ dtc) and 361
(M Ϫ 2NR). 1H NMR: δ 7.60 (d, 4 H, J 7.9, H6), 6.74 (d, 2 H,
J 8.1, H5), 6.71 (s, 2 H, H3), 3.30 (q, 8 H, J 6.1, CH2), 2.65 (s,
6 H, Me), 2.03 (s, 6 H, Me) and 0.79 (t, 12 H, J 7.0 Hz, Me).
13C-{1H} NMR: δ 203.1 (C᎐N), 156.0 (Cipso), 134.9, 130.7,
᎐
130.6, 127.0, 125.6, 45.7 (CH2), 21.1 (Me), 19.0 (Me) and 12.3
(Me). 13C CPMAS TOSS: δ 202.0, 155.5, 138.5, 135.6, 133.9,
132.4, 130.4, 127.6, 126.8, 50.7, 45.7, 23.9, 23.3, 21.6, 19.6, 16.8,
13.3 and 11.5. 13C CPMAS NQSTOSS: δ 202.0, 155.5, 138.6,
135.6, 134.0, 132.4, 23.3, 19.6, 16.8, 13.3 and 11.4. Found (calc.)
for C26H38MoN4S4: C, 48.62 (49.50); H, 5.93 (6.07); N, 8.57
(8.88); S, 19.57 (20.33)%.
δ 202.6 (C᎐N), 157.3 (Cipso), 154.8, 128.6, 127.4, 127.2, 124.0,
᎐
111.1, 110.9, 45.6 (CH2), 21.1 (Me) and 12.2 (Me). 13C CPMAS:
δ 199.2, 195.5, 153.2 (Cipso), 136.1, 127.8, 121.7, 111.6, 46.8 and
13.2. 13C CPMAS NQS: δ 201.5, 199.5, 136.2 and 13.0. Found
(calc.) for C22H26F4MoN4S4: C, 39.99 (40.87); H, 4.10 (4.02);
N, 8.45 (8.67); S, 20.21 (19.81)%.
3g. IR: 1497s, 1456m, 1429m, 1384s, 1336m, 1306m, 1275s,
1208m, 1147m, 1022m, 845m and 813m cmϪ1. Mass spectrum
(EI): m/z 632 (Mϩ), 513 (M Ϫ NR), 485 (M Ϫ dtc), 397
(M Ϫ 2NR) and 366 (M Ϫ dtc Ϫ NR). 1H NMR: δ 7.42 (d, 2 H,
J 6.1, H6), 7.41 (s, 2 H, H2), 6.81 (d, 2 H, J 6.1, H5), 3.29 (q, 8 H,
J 7.0, CH2), 1.91 (s, 6 H, Me), 1.84 (s, 6H, Me) and 0.77 (t, 12 H,
3m. IR: 1494s, 1457s, 1420s, 1390m, 1356m, 1319s, 1273s,
1247s, 1208m, 1147m, 1089w, 953m, 868w, 858w and 817s cmϪ1
.
Mass spectrum (EI): m/z 788 (Mϩ) and 591 (M Ϫ NR); 1H
NMR: δ 7.04 (s, 4 H, H3,5), 3.84 (br, 8 H, CH2), 2.30 (s, 12 H,
Me) and 1.30 (t, 12 H, J 7.1 Hz, Me). Found (calc.) for
C26H36Br2MoN4S4: C, 40.11 (39.59); H, 4.68 (4.57); N, 7.12
(7.11)%.
J 7.2 Hz, Me). 13C-{1H} NMR: δ 203.2 (C᎐N), 157.8 (Cipso),
᎐
136.3, 134.1, 129.7, 125.2, 121.8, 40.5 (CH2), 19.6 (Me), 19.4
(Me) and 12.3 (Me). 13C CPMAS TOSS: δ 202.1, 158.2 (Cipso),
155.4 (Cipso), 136.1, 130.4, 124.6, 122.9, 48.0, 21.2 and 14.8.
Found: C, 48.71; H, 6.20; N, 8.69; S, 20.01%.
3n. IR: 1497s, 1429m, 1263s, 1209m, 1097s, 1022s and 802s
cmϪ1. Mass spectrum (EI): m/z 536 (Mϩ), 465 (M Ϫ NBut) and
1
316 (M Ϫ NBut Ϫ dtc). H NMR: δ 3.36 (q, 8 H, J 7.0, CH2),
3h. IR: 1629m, 1613m, 1596m, 1529s, 1501m, 1456m, 1438m,
1382m, 1354m, 1323w, 1274s, 1263s, 1206m, 1146m, 1095s,
1078s, 1021s, 845w and 803m cmϪ1. Mass spectrum (EI): m/z
633 (Mϩ), 514 (M Ϫ NR), 485 (M Ϫ dtc), 397 (M Ϫ 2NR) and
1.57 (s, 18 H, Me) and 0.82 (t, 12 H, J 7.1 Hz, Me). 13C-{1H}
NMR: δ 203.9 (C᎐N), 71.2 (C ), 45.8 (CH ), 31.3 (CMe ) and
᎐
α
2
3
12.4 (Me). 13C CPMAS: δ 203.3, 201.4, 71.7 (Cα), 69.7 (Cα),
47.2, 44.4, 32.1, 14.3 and 13.0. 15N CPMAS: δ (373 K) Ϫ214.1
and Ϫ209.1; (293 K) Ϫ214.5 and Ϫ209.0. Found (calc.) for
C18H38MoN4S4: C, 39.85 (40.43); H, 6.82 (7.16); N, 10.22
(10.48); S, 24.06 (23.99)%.
1
366 (M Ϫ dtc Ϫ NR). H NMR: δ 7.21 (s, 4 H, H2,6), 6.44 (s,
2 H, H4), 3.29 (q, 8 H, J 7.0, CH2), 1.96 (s, 12 H, Me) and 0.77
(t, 12 H, J 7.1 Hz, Me). 13C-{1H} NMR: δ 203.1 (C᎐N), 158.1
᎐
(Cipso), 137.8, 127.3, 121.4, 45.8 (CH2), 21.1 (Me) and 12.3 (Me).
13C CPMAS TOSS: δ 201.4, 160.4 (Cipso), 156.8 (Cipso), 139.1,
137.7, 130.7, 124.9, 118.1, 47.9, 43.9, 25.2, 23.7, 22.5, 15.5, 14.7
and 12.1. Found: C, 48.51; H, 6.11; N, 8.40%.
3o. IR: 1490s, 1449m, 1426s, 1301m, 1270s, 1211m, 1144m
and 1095m cmϪ1. Mass spectrum (EI): m/z 692 (Mϩ) and 543
(M Ϫ NRm). 1H NMR: δ 3.40 (q, 8 H, J 7.0, CH2), 2.35 (s, 12 H,
Hβ), 2.01 (s, 6 H, Hγ), 1.59 (d, 6 H, J 11.8, Hδ), 1.49 (d, 6 H,
J 11.8, Hδ) and 0.86 (t, 12 H, J 7.0 Hz, Me). 13C-{1H} NMR:
3i. IR: 1504s, 1457w, 1440w, 1426m, 1352m, 1274s, 1245w,
1
1149w, 1092w, 957w, 846w, 778w and 572w cmϪ1. H NMR
δ 204.2 (C᎐N), 71.1 (C ), 45.8 (C ), 45.0 (CH ), 36.7 (CH ), 30.2
᎐
α
β
2
δ
(CDCl3); δ (313 K) 6.90 (d, 4 H, J 7.2, H3,5), 6.73 (t, 2 H, J 7.2
Hz, H4), 3.83 (br, 8 H, CH2) 2.35 (s, 12 H, CH3C6H3) and 1.28
(s, 12 H, CH2CH3); (213K) 6.90 (d, 4H, J 7.2, H3,5), 6.73 (t, 2 H,
J 7.2, H4), 3.96–4.05 (m, 2 H, CH2), 3.78–3.82 (m, 4 H, CH2),
3.60–3.69 (m, 2 H, CH2), 1.34 (s, 12 H, CH3C6H3), 1.26 (t, 6 H,
J 7.2, CH2CH3) and 1.23 (t, 6 H, J 7.2 Hz, CH2CH3). 13C
CPMAS TOSS: 199.9, 157.4 (Cipso), 133.3, 128.2, 125.7, 47.5,
46.3, 20.3, 18.9 and 11.4. Found: C, 49.37; H, 5.63; N, 8.68; S,
20.14%.
(Cγ) and 12.5 (Me). 13C CPMAS: δ 202.6, 71.7 (Cα), 45.8, 45.6,
37.4, 30.7, 30.4, 15.4 and 13.1. Found (calc.) for C30H50MoN4S4:
C, 51.97 (52.15); H, 7.31 (7.29); N, 8.03 (8.11); S, 18.72
(18.56)%.
3p. Mass spectrum (EI): m/z 756 (Mϩ), 608 (M Ϫ dtc), 575
(M Ϫ NR), 460 (M Ϫ 2dtc) and 427 (M Ϫ NR Ϫ dtc). 1H
NMR: δ 3.26 (q, 8 H, J 7.0 Hz, CH2) and 2.32 (s, 12 H, CH2).
13C-{1H} NMR: δ 201.3 (C᎐N), 142–114 (m, Ph), 45.7 (CH )
᎐
2
and 12.1 (Me). 13C CPMAS: δ 200.9, 137.3, 133.8, 49.9, 46.5,
14.9 and 11.8. 13C CPMAS TOSS: δ 200.9, 150–130 (br), 133.7,
49.8, 14.9 and 11.7. 13C CPMAS NQSTOSS: δ 200.9, 133.8,
14.9 and 11.7. Satisfactory elemental analysis could not be
obtained.
3j. IR: 1267s, 1210m, 1149m, 1095w, 1075m, 962m, 847w and
755s cmϪ1. 1H NMR (C6D5CD3): δ (373 K) 6.95 (d, 4 H, J 7.6,
H3,5), 6.80 (t, 2 H, J 7.6, H4), 4.08 (septet, 4 H, J 6.8, CHMe2),
3.46 (q, 8 H, J 7.0, CHMe2), 1.19 (d, 24 H, J 6.8, CHMe2) and
0.96 (t, 12 H, J 7.0 Hz, CH2Me); (253 K) 6.96 (d, 4 H, J 7.6,
H3,5), 6.84 (t, 2 H, J 7.6, H4), 4.17 (septet, 4 H, J 6.8, CHMe2),
3.47 (septet, 2 H, J 6.8, CH2Me), 3.14 (m, 4 H, CH2Me), 2.93
(septet, 2 H, CH2Me), 1.28 (d, 12 H, J 6.8, CHMe2), 1.16 (d,
12 H, J 6.8, CHMe2), 0.81 (t, 6 H, J 6.8, CH2Me) and 0.74 (t, 6 H,
J 7.0 Hz, CH2Me). 13C-{1H} NMR (CDCl3, 253 K): δ 201.4
3a-d10. Mass spectrum (EI): m/z 584 (Mϩ), 490 (M Ϫ NPh)
1
and 439 (M Ϫ dtc). H NMR: δ 3.28 (q, 8 H, J 7.1, CH2) and
0.77 (t, 12 H, J 7.1 Hz, Me). 13C-{1H} NMR: δ 202.8 (C᎐N),
᎐
158.5 (Cipso), 124.9, 124.7, 124.5, 123.2, 122.9, 122.6, 45.8 (CH2)
and 12.3 (Me). 13C CPMAS TOSS: δ 200.7, 161.2, 156.3, 129.6,
125.9, 120.2, 47.5, 44.6, 13.7 and 12.6. Found (calc.) for
4526
J. Chem. Soc., Dalton Trans., 1999, 4519–4528

View Article Online
Table 3 Crystallographic data for complexes 3a, 3b, 3i–3k and 3n at 293 K
3a
3b
3i
3j
3k
3n
Formula
Formula weight
C22H30MoN4S4
574.74
C24H34MoN4S4
602.80
C26H38MoN4S4
630.86
C34H54MoN4S4
742.99
C22H26Cl4MoN4S4
712.50
C18H38MoN4S4
534.78
Crystal system
Space group
Monoclinic
P21/a
Triclinic
P1
Orthorhombic
P212121
Monoclinic
C2/c
Orthorhombic
P212121
Monoclinic
P21/c
¯
a/Å
b/Å
c/Å
α/Њ
16.973(4)
10.293(1)
17.369(3)
8.467(2)
10.486(3)
16.388(4)
73.32(2)
86.18(2)
89.13(2)
1390.6
2
7.68
5442
5140
298
10.753(2)
10.761(1)
26.972(5)
16.336(2)
12.977(2)
19.230(2)
10.690(4)
10.701(3)
26.728(5)
17.014(7)
9.193(4)
19.064(7)
β/Њ
116.01(2)
111.70(2)
115.09(3)
γ/Њ
V/Å3
2727.1
4
7.79
5402
5231
280
3120.9
4
6.87
3677
3622
316
3788.7
4
5.76
4593
4556
195
3057.3
4
10.54
3078
3029
316
2700.4
4
7.82
5463
5061
244
Z
µ(Mo-Kα)/cmϪ1
Data measured
Unique data used
No. parameters
R
0.048
0.042
0.050
0.054
0.040
0.049
0.031
0.035
0.040
0.043
0.049
0.050
RЈ
C22H20D10MoN4S4ؒ0.5CH2Cl2: C, 43.44 (43.09); H, 3.17 (3.35);
N, 9.14 (8.94)%.
using an automated four-circle diffractometer (Nicolet R3mV)
equipped with Mo-Kα radiation (λ = 0.71073 Å). Important
crystallographic parameters are summarised in Table 3. The
lattice vectors were identified by application of the automatic
indexing routine of the diffractometer to the positions of a
number of reflections taken from a rotation photograph and
centred by the diffractometer. The ω–2θ technique was used to
measure reflections in the range 5 ≤ 2θ ≤ 50Њ and three standard
reflections (remeasured every 97 scans) showed no significant
loss in intensity during data collection. The data were corrected
for Lorentz-polarisation effects, and empirically for absorption.
The unique data with I ≥ 3.0σ(I) were used to solve and refine
the structures, being solved by direct methods and developed
using alternating cycles of least-squares refinement and
Fourier-difference synthesis. The non-hydrogen atoms were
refined anisotropically while hydrogens were placed in idealised
positions (C–H 0.96 Å) and assigned a common isotropic
thermal parameter (U = 0.08 Å2). Structure solution used either
the SHELXTL41 or SHELXTL PLUS42 program packages
on a microVax or IBM PC respectively.
[Mo(NPh)2(S2CNMe2)2] 4a. A diethyl ether solution (50
cm3) of [Mo(NPh)2Cl2(dme)] 2a (2.00 g, 4.56 mmol) and
Na[S2CNMe2] (1.32 g, 9.20 mmol) was stirred for 16 h at room
temperature. Filtration gave a brown solution and residue. The
latter was extracted with dichloromethane (3 × 30 cm3) and
added to the dme solution. Removal of volatiles under reduced
pressure gave a black solid which was washed with light petrol-
eum. This was then redissolved in the minimum amount of
dichloromethane and light petroleum added. Cooling to
Ϫ20 ЊC gave complex 4a (0.94 g, 40%) as a dark brown solid.
Orange 4n (41%) was prepared in a similar fashion.
4a. Mass spectrum (EI): m/z 521 (Mϩ), 430 (M Ϫ NPh) and
398 (M Ϫ dtc). 1H NMR: δ 7.50 (d, 4 H, J 8.1, H2,6), 6.98 (t, 4
H, J 7.5, H3,5), 6.71 (t, 2 H, J 7.5 Hz, H4) and 2.48 (s, 12 H, Me).
13C CPMAS TOSS: δ 199.8, 190.6, 189.4, 184.0, 177.6, 160.3
(Cipso), 129.5, 128.2, 124.4, 123.8, 116.6 and 39.7. 13C CPMAS
NQSTOSS: δ 199.9, 160.3, 128.8 and 39.7. Found (calc.) for
C18H22MoN4S4ؒ0.5CH2Cl2: C, 39.00 (39.47); H, 4.22 (4.09); N,
10.41 (9.96)%.
CCDC reference number 186/1716.
4n. Mass spectrum (EI): m/z 480 (Mϩ) and 409 (M Ϫ NBut).
1H NMR: δ 2.61 (s, 12 H, Me) and 1.58 (s, 18 H, But). 13C-{1H}
NMR: δ 204.9 (C᎐N), 71.1 (C ), 40.3 (Me) and 31.4 (But). 13C
Acknowledgements
We thank the EPSRC for funding of this work in the form of
studentships to T. A. C. and G. D. F. and for the provision of a
glove-box. Dr Simon P. Redmond is thanked for the synthesis
of complexes 3m and 5a and Mr Kayodi Ajaya for 4a. We also
thank Dr Abil Aliev and Mr David Butler of the ULIRS solid-
state NMR service for their help with this work, Drs Andrea
Sella and Nik Kaltsoyannis for their interest and contributions
and Dr Derek A. Tocher for help with crystallography.
᎐
α
CPMAS TOSS: δ 204.1, 203.3, 72.8 (Cα), 69.9 (Cα), 44.0, 42.4,
41.9, 32.4 and 31.8. Found (calc.) for C14H30MoN4S4ؒCH2Cl2:
C, 31.78 (31.86); H, 6.26 (5.66); N, 10.65 (9.91)%.
[Mo(NPh)2(S2PPh2)2] 5a. A diethyl ether solution (50 cm3) of
[Mo(NPh)2Cl2(dme)] 2a (2.00 g, 4.56 mmol) and Na[S2PPh2]
(2.50 g, 9.20 mmol) was stirred for 16 h at room temperature.
Filtration gave a red solution which after removal of volatiles
under reduced pressure gave a bright red solid. This was washed
with light petroleum then redissolved in the minimum volume
of diethyl ether. Cooling to Ϫ20 ЊC gave complex 5a (1.84 g,
52%) as a dark red microcrystalline solid. 1H NMR: δ 8.05–7.69
(m, 6 H, PPh2 ϩ H2,6), 6.92 (m, 6 H, PPh2), 6.84 (t, 4 H, J 7.6,
H3,5) and 6.68 (t, 2 H, J 7.3 Hz, H4). 13C-{1H} NMR: δ 158.2
(Cipso) and 137.9–131.3 (m, Ph). 13C CPMAS: δ 156.7 (Cipso),
155.1 (Cipso), 139.0, 132.3, 128.1 and 124.5. 31P NMR: δ 64.5 (s).
References
1 P. J. Walsh, A. M. Baranger and R. G. Bergman, J. Am. Chem. Soc.,
1992, 114, 1708; D. Mansuy, J.-P. Mahy, A. Dureault, G. Bedi and
P. Battioni, J. Chem. Soc., Chem. Commun., 1984, 1161; J. T. Groves
and T. Takahashi, J. Am. Chem. Soc., 1983, 105, 2073.
2 R. R. Schrock, J. S. Murdzek, G. C. Bazan, J. Robbins, M. DiMare
and M. B. O’Regen, J. Am. Chem. Soc., 1990, 112, 3875; J. Robbins,
G. C. Bazan, J. S. Murdzek, M. B. O’Regen and R. R. Schrock,
Organometallics, 1991, 10, 2902; R. R. Schrock, R. T. DePue,
J. Feldman, C. J. Schaverien, J. C. Dewan and A. H. Liu, J. Am.
Chem. Soc., 1988, 110, 1423; R. R. Schrock, R. T. DePue,
J. Feldman, K. B. Yap, D. C. Yang, W. M. Davis, L. Y. Park,
M. DiMare, M. Schofield, J. Anhaus, E. Walborsky, E. Evitt, C.
Kruger and P. Betz, Organometallics, 1990, 9, 2262.
31P CPMAS:
δ
68.6 and 65.5. Found (calc.) for
C36H30MoN4P2S4: C, 54.86 (55.66); H, 3.82 (3.89); N, 3.18
(3.61); P, 8.17 (7.97); S, 16.41 (16.51)%.
X-Ray crystallography
3 D. E. Wigley, Prog. Inorg. Chem., 1994, 42, 239; W. A. Nugent and
B. L. Haymore, Coord. Chem. Rev., 1980, 31, 123; W. A. Nugent and
J. M. Mayer, Metal–Ligand Multiple Bonds, Wiley, New York, 1988.
Suitable single crystals were mounted on a glass fibre and all
geometric and intensity data were taken from these samples
J. Chem. Soc., Dalton Trans., 1999, 4519–4528
4527

View Article Online
4 G. Parkin, A. van Asselt, D. J. Leahy, L. R. Whinnery, N. G. Hua,
R. W. Quan, L. M. Henling, W. P. Schaefer, B. D. Santarsiero and
J. E. Bercaw, Inorg. Chem., 1992, 31, 82.
5 A. Bell, W. Clegg, P. W. Dyer, M. R. Elsegood, V. C. Gibson and
E. L. Marshall, J. Chem. Soc., Chem. Commun., 1994, 2247.
6 W. A. Nugent, R. L. Harlow and R. J. McKinney, J. Am. Chem.
Soc., 1979, 101, 7265.
7 D. W. H. Rankin, H. E. Robertson, A. A. Danopoulos, P. D. Lyne,
D. M. P. Mingos and G. Wilkinson, J. Chem. Soc., Dalton Trans.,
1994, 1563.
23 O. D. Sloan and P. Thornton, Polyhedron, 1988, 7, 329.
24 R. R. Schrock, S. W. Seidel, N. C. Mosch-Zanetti, D. A. Dobbs,
K. Y. Shih and W. M. Davis, Organometallics, 1997, 16, 5195.
25 B. A. Arndtsen, H. F. Sleiman, A. K. Chang and L. McElwee-
White, J. Am. Chem. Soc., 1991, 113, 4871.
26 A. Saravanamuthu, D. M. Ho, M. E. Kerr, C. Fitzgerald, M. R. M.
Bruce and A. E. Bruce, Inorg. Chem., 1993, 32, 2202.
27 Z. Li, R. W. Quan and E. N. Jacobsen, J. Am. Chem. Soc., 1995, 117,
5889.
28 R. H. Holm, Chem. Rev., 1987, 87, 1401.
8 G. V. Goeden and B. L. Haymore, Inorg. Chem., 1983, 22, 157.
9 W. A. Nugent and R. L. Harlow, J. Chem. Soc., Chem. Commun.,
1979, 1105.
10 G. K. Lahiri, S. Goswami, L. R. Favello and A. Chakraworty,
Inorg. Chem., 1987, 26, 3365.
29 A. M. Bond and R. L. Martin, Coord. Chem. Rev., 1984, 54, 23;
D. Coucouvanis, Prog. Inorg. Chem., 1970, 11, 233.
30 T. A. Coffey, G. D. Forster, G. Hogarth and A. Sella, Polyhedron,
1993, 12, 2741.
31 I. S. Kolomnikov, Yu. D. Koreshkov, T. S. Lobeeva and M. E.
Volpin, J. Chem. Soc., Chem.Commun., 1970, 1432; E. A. Maatta,
Inorg. Chem., 1984, 23, 2650; A. J. Nielson, Inorg. Synth., 1986, 24,
194; M. L. H. Green, G. Hogarth, P. C. Konidaris and P.
Mountford, J. Chem. Soc., Dalton Trans., 1990, 3781.
32 T. A. Coffey, G. D. Forster and G. Hogarth, J. Chem. Soc., Dalton
Trans., 1996, 183; J. Chem. Soc., Chem. Commun., 1993, 1524.
33 T. A. Coffey, G. D. Forster and G. Hogarth, J. Chem. Soc., Dalton
Trans., 1995, 2337.
34 F. W. Moore and M. L. Larson, Inorg. Chem., 1967, 6, 998.
35 D. D. Devore and E. A. Maatta, Inorg. Chem., 1985, 24, 2846.
36 H. H. Fox, K. B. Yap, J. Robbins, S. Cai and R. R. Schrock, Inorg.
Chem., 1992, 31, 2287.
37 K. Dehnicke and J. Strahle, Angew. Chem., Int. Ed. Engl., 1981, 20,
413; W. P. Griffith, A. J. Nielson and M. J. Taylor, J. Chem. Soc.,
Dalton Trans., 1988, 647.
38 D. C. Bradley, S. R. Hodge, J. D. Runnacles, M. Hughes, J. Mason
and R. L. Richards, J. Chem. Soc., Dalton Trans., 1992, 1663.
39 A. V. Ivanov, T. Rodyna and O. N. Antzutkin, Polyhedron, 1998, 17,
3101.
11 R. C. B. Copley, P. W. Dyer, V. C. Gibson, J. A. K. Howard,
E. L. Marshall, W. Wang and B. Whittle, Polyhedron, 1996, 15, 3001.
12 V. C. Gibson, C. Redshaw, W. Clegg, M. R. Elsegood, U. Siemeling
and T. Türk, J. Chem. Soc., Dalton Trans., 1996, 4513.
13 A. F. Clifford and C. S. Kobayashi, Inorg. Synth., 1960, 6, 204;
N. A. Milas and I. M. Iliopulos, J. Am. Chem. Soc., 1959, 81, 6089.
14 B. L. Haymore, E. A. Maatta and R. A. D. Wentworth, J. Am.
Chem. Soc., 1979, 101, 2063; E. A. Maatta and R. A. D. Wentworth,
Inorg. Chem., 1979, 18, 2409.
15 V. C. Gibson, E. L. Marshall, C. Redshaw, W. Clegg and M. R.
Elsegood, J. Chem. Soc., Dalton Trans., 1996, 4197.
16 A. A. Danopoulos, G. Wilkinson and D. J. Williams, J. Chem. Soc.,
Chem. Commun., 1991, 181.
17 A. A. Danopoulos, G. Wilkinson, T. K. N. Sweet and M. B.
Hursthouse, J. Chem. Soc., Chem. Commun., 1993, 495; J. Chem.
Soc., Dalton Trans., 1994, 1037; 1995, 205; 937.
18 T. I. Gountchev and T. D. Tilley, J. Am. Chem. Soc., 1997, 119,
12831.
19 M. Minelli, R. L. Kuhlman, S. J. Shaffer and M. Y. Chiang,
Inorg. Chem., 1992, 31, 3891.
20 M. Minelli, M. R. Carson, D. W. Whisenhunt, Jr., W. Imhof and
G. Huttner, Inorg. Chem., 1990, 29, 4801.
21 K. J. Weller, S. D. Gray, P. M. Briggs and D. E. Wigley,
Organometallics, 1995, 14, 5588.
40 N. Kaltsoyannis and W.-S. Kim, unpublished results.
41 G. M. Sheldrick, SHELXTL, University of Göttingen, 1985.
42 G. M. Sheldrick, SHELXTL PLUS, Program package for structure
solution and refinement, version 4.2, Siemens Analytical
Instruments Inc., Madison, WI, 1990.
22 S. Cai, R. R. Schrock, R. R. Kempe and W. M. Davis, Polyhedron,
1998, 17, 747.
Paper 9/07382E
4528
J. Chem. Soc., Dalton Trans., 1999, 4519–4528