Cerium(III) dialkyl dithiocarbamates from [Ce{N(SiMe3)2}3] and tetraalkylthiuram disulfides, and [Ce(κ2-S2 CNEt2)4] from the CeIII precursor; TbIII and NdIII analogues

Article abstract of DOI:10.1039/b311397c

The synthesis and characterisation of the first neutral cerium dialkyl dithiocarbamate complexes, using a novel oxidative displacement of the amido ligands of [Ce{N (SiMe₃)₂}₃] by tetraalkylthiuram disulfides [R₂NC(S)S]₂ (R = Me, Et) in thf solution, are reported. In the absence of other donors, the complexes [Ce(κ²-S₂CNMe₂)₃ (thf)₂] 2 and Ce(κ²-S₂CNEt₂)₃ 3 were obtained. The addition of a polypyridyl ligand allowed easy access to a range of complexes of general formula [Ce(κ²-S₂ CNR₂)₃(L∩L)][R = Me and L∩L = 2,2′-bipy (4), or 4,7-diphenyl-1,10-phenanthroline (6); or R = Et and L∩L = 2,2′-bipy (5)]. Brief exposure of the Ce(III) dithiocarbamate 3 to oxygen gas afforded in high yield the diamagnetic, crystalline Ce(IV) dithiocarbamate [Ce(κ²-S₂CNEt₂)₄] 7. The neodymium (8) and terbium (10) complexes, isoleptic with 2, were prepared from the appropriate 4f metal (Ln) bis(trimethylsilyl)amide [Ln{N(SiMe₃)₂}₃] [Ln = Nd or Tb (9)] and [Me₂NC(S)S]₂. The structures of the crystalline complexes 2, 4, 6, 7, 9 and 10 have been determined by X-ray crystallography. Some evidence has been obtained for the formation of the cerium(IV) complex Ce{N(SiMe₃)₂}₂ (κ²-S₂CNMe₂)₂. The cerium(IV) complex 7 has the metal coordinated to eight sulfur atoms of four planar chelating S₂CNC₂ moieties and its geometry is intermediate between dodecahedral and square prismatic; the mean Ce-S bond length of 2.803 A in 7 compares with the 2.950 A in the Ce(III) complex 2.

Full text of DOI:10.1039/b311397c

View Article Online / Journal Homepage / Table of Contents for this issue
Cerium(III) dialkyl dithiocarbamates from [Ce{N(SiMe ) } ] and
3
2 3
2
III
tetraalkylthiuram disulfides, and [Ce(ꢀ -S CNEt ) ] from the Ce
2
2 4
III
III
precursor; Tb and Nd analogues
Peter B. Hitchcock, Alexander G. Hulkes, Michael F. Lappert* and Zhengning Li
The Chemistry Laboratory, University of Sussex, Brighton, UK BN1 9QJ.
E-mail: m.f.lappert@sussex.ac.uk
Received 17th September 2003, Accepted 30th October 2003
First published as an Advance Article on the web 18th November 2003
The synthesis and characterisation of the first neutral cerium dialkyl dithiocarbamate complexes, using a novel
oxidative displacement of the amido ligands of [Ce{N(SiMe ) } ] by tetraalkylthiuram disulfides [R NC(S)S] (R =
3
2
3
2
2
2
Me, Et) in thf solution, are reported. In the absence of other donors, the complexes [Ce(κ -S CNMe ) (thf ) ] 2 and
Ce(κ -S CNEt ) 3 were obtained. The addition of a polypyridyl ligand allowed easy access to a range of complexes
of general formula [Ce(κ -S CNR ) (L L)] [R = Me and L L = 2,2Ј-bipy (4), or 4,7-diphenyl-1,10-phenanthroline (6);
or R = Et and L L = 2,2Ј-bipy (5)]. Brief exposure of the Ce() dithiocarbamate 3 to oxygen gas afforded in high
2
2
3
2
2
2
2 3
2
∩
∩
2
2 3
∩
2
yield the diamagnetic, crystalline Ce() dithiocarbamate [Ce(κ -S CNEt ) ] 7. The neodymium (8) and terbium (10)
2
2 4
complexes, isoleptic with 2, were prepared from the appropriate 4f metal (Ln) bis(trimethylsilyl)amide
Ln{N(SiMe ) } ] [Ln = Nd or Tb (9)] and [Me NC(S)S] . The structures of the crystalline complexes 2, 4, 6, 7, 9
[
3
2
3
2
2
and 10 have been determined by X-ray crystallography. Some evidence has been obtained for the formation of the
2
cerium() complex Ce{N(SiMe ) } (κ -S CNMe ) . The cerium() complex 7 has the metal coordinated to eight
3
2
2
2
2 2
sulfur atoms of four planar chelating S CNC moities and its geometry is intermediate between dodecahedral and
2
2
square prismatic; the mean Ce–S bond length of 2.803 Å in 7 compares with the 2.950 Å in the Ce() complex 2.
Introduction
We have an ongoing programme of research into non-
classical lanthanide compounds, including into early 4f-meta5l
There are few areas of chemistry in which, when it comes to
the development of genuinely useful new technologies, the lan-
thanides can compete as equals with the main group and transi-
tion metals, but in electro- and photo-luminescence research
they have achieved some eminence. Indeed, they are in some
respects superior choices for the development and fabrication
of thin film displays and other related technologies, as they are
cheap and readily available, exhibit relatively straightforward
reaction chemistry, have well defined and understood electronic
properties largely unaffected by the ligand environment and,
most crucially, present an almost unbroken series which allows
a fine tuning of their chemistry by systematic variation of the
attached ligands.
There are, however, gaps in what is otherwise an uninter-
rupted spectrum of behaviour. Cerium has a relatively stable
tetravalent state as far as inorganic compounds, such as CeO2,
are concerned and so the synthesis and handling of its trivalent
organic compounds is often considerably – sometimes pro-
hibitively – more demanding than is the case for any of the
other lanthanides. This would be a minor problem were it not
for the fact that cerium has been shown to be an excellent
compounds in the formerly unknown ϩ2 oxidation states.
Herein we report results which in part represent a comple-
mentary investigation. The objective was to extend the bound-
aries relating to organic Ce() compounds. The isolation of the
crystalline [CeCl{N(SiMe ) } ] represented our first steps
3
2 3
6
in that direction, and this was followed by isolation of the
corresponding bromide. We then turned to tetraalkylthiuram
7
disulfides as oxidants. This led to the main subject of the paper,
namely the synthesis and characterisation of the first neutral
cerium() dithiocarbamato complexes, and of a homoleptic
cerium() dithiocarbamate.
0
Well authenticated organic compounds of f Ce() are
rare and their reactions little studied. Until recently, such
compounds were largely restricted to alkoxides or related
t
t
8
O-centered compounds, such as [Ce(OBu ) (NO ) (HOBu ) ],
2
3
2
2
t
t
t
8
t
[
(
(
Ce(OBu ) (µ -OBu ) (µ -OBu ) Na(dme)] , [{Ce(OCBu ) } -
2 2 2 3 2 2 3 3 2
9 10 5
µ-OC H O-1,4)], [Ce{c-C H ) Si O } (py) ], [Ce(η -C H ) -
6 4 6 11 8 8 13 2 3 5 5 3
1 1 t 11 i 12
OR )] (R = Bu , Pr ). These were obtained from the
inorganic Ce() precursor [NH ] [Ce(NO ) ]. The latter has an
4
2
3 6
13
important role as an oxidant in organic chemistry, and has the
acronym CAN.
1
dopant in the synthesis of luminescent SrS thin films, and so
there is a demand for new (preferably oxygen-free) molecular
Breakthroughs came with reports that highly reactive Ce()
amides can be made. The first, showed that the tetrapodal (tris-
2
precursors. Similar requirements are commonly satisfied by
dithiocarbamato complexes (from here on, “dithiocarbamato”
refers specifically to a dialkyl dithiocarbamato ligand), but this
has not been an option in the case of cerium, as there are to the
best of our knowledge no known neutral homo- or heteroleptic
cerium dithiocarbamates. Several 4f metal diethyl dithio-
carbamates [Ln(S CNEt ) ] (but not for Ln = Ce) have been
III
1
(
amido)amine)cerium() complex [Ce (NN )] was oxidised
3
IV 1 14
by I to give [Ce (NN )I], whereas Br and Cl gave the
2
3
2
2
1
mixed Ce()/Ce() valence compounds [{Ce(NN )} (µ-X)]
3
2
1
t
14
[
[
NN 3 = N(CH CH NSiMe Bu ) , X = Cl or Br]. By contrast,
2 2 2 3
14 7
Ce{N(SiMe ) } ] A was unreactive towards Cl , Br2 or
3
2
3
2
7
I . The second, at Sussex, revealed that [Ce{N(SiMe ) } ] could
be converted into [CeCl{N(SiMe ) } ] by using TeCl4 as
2
3 2 3
2
2 3
6
3
2 3
prepared from the appropriate LnBr and 3Na[S CNEt ] in
ethanol (Ln = Pr, Nd, Eu, Ho, Er or Yb); the X-ray structure
3
2
2
oxidant.
3
(
no details) of the Nd() compound was mentioned. The only
notable advance in this area was made almost twenty-five years
Results and discussion
ago with the synthesis of Na[Ce(S CNEt ) ], as part of a
2
2 4
study of the entire (expect Pm) 4f metal series of complexes
Na[Ln(S CNEt ) ], including the X-ray structure of the La
compound.
The homoleptic cerium() complex [Ce{N(SiMe ) } ] (A)
3 2 3
combines a readily oxidisable metal centre with three reactive
amido ligands, and so it might have been expected that it should
2
2 4
4
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6
129

View Article Online
Scheme 1 Synthesis of the new cerium dithiocarbamates 1–7. Reagents and conditions: i, 6[Me NC(S)S] , thf, 20 ЊC; ii, 3[Me NC(S)S] , thf, 20 ЊC;
2
2
2
2
iii, 3[Et NC(S)S] , Et O, 20 ЊC; iv, 3[NH Et ][S CNEt ], Et O, 20 ЊC; v, 3[Me NC(S)S] , bipy, thf, 20 ЊC; vi, 3[Et NC(S)S] , bipy, thf, 20 ЊC;
2
2
2
2
2
2
2
2
2
2
2
2
vii, 3[Me NC(S)S] , dipphen (dipphen = 4,7-diphenyl-1,10-phenanthroline), thf, 20 ЊC; viii, O , PhMe, 20 ЊC, 1 min.
2
2
2
2
engage in quite a varied oxidative chemistry; for our initial
characteristic of κ -S CNMe at δ 202.7, along with the remain-
2 2
6,7
forays, see above. We now chose to examine one potential
such reaction, namely that of A with a tetraalkylthiuram
disulfide. This appeared to be particularly favourable as two-
electron reduction of the disulfide bond in [R NC(S)S] creates
ing dithiocarbamato (δ 40.9) and trimethylsilyl (δ 5.6) carbon
resonances. No further reliable indicators of composition were
obtained, as the EI mass spectrum contained no readily assign-
able peaks (except for those corresponding to fragmentation of
the amido ligand). Taken together, the evidence gathered for the
formation of 1 is at present equivocal.
2
2
two dithiocarbamato ligands, and with them a significant
degree of possible stabilisation for a newly formed – and Lewis
acidic – metal centre. Somewhat perversely, thiuram disulfides
themselves can be formed from dithiocarbamato salts by the
In both the previous examples of the isolation of a Ce()
derivative of A, namely [CeX{N(SiMe ) } ] (X = Cl or Br ),
6
7
3
2 3
application of oxidants as weak as I , which is incapable of
oxidising A, although with a different amido substrate (see
the greater part of the starting material was consumed in side-
reactions, limiting the yield. This aspect of its reactivity may be
exploited to considerable advantage when tetraalkylthiuram
disulfides are used as oxidants. Thus, A with three equivalents
of [R NC(S)S] (R = Me, Et) in thf gave the appropriate tris-
2
7
above) the nominally large Ce()/() couple can be over-
14
come. A recent paper has shown that a catalytic cycle based on
CeCl with molecular oxygen effects the direct hydroxylation of
3
2
2
15
β-dicarbonyls to yield acyloins.
(dialkyl dithiocarbamato)cerium() complex 2 (R = Me) or 3
(R = Et) (ii and iii in Scheme 1). It is important to note
that (i) there was no colour change, which might have been
Our experiments based on reactions of the cerium() amide
A with tetraalkylthiuram disulfides [R NC(S)S] (R = Me, Et)
2
2
IV
are summarised in Scheme 1. Compound A was first reported in
indicative of the formation of 1 or another Ce species in
16
1
973, although a yield was not cited. Its preparation, from
solution, and (ii) the precise stoichiometry was of paramount
importance.
CeCl and three equivalents of lithium bis(trimethylsilyl)amide,
3
was repeated more recently (17% yield), in the context of its use
An alternative, essentially classical, method for 3 was the
for the preparation of thin films of SrS:Ce by atomic layer
epitaxy, and its X-ray structure was recorded. We prepared A,
reaction of A with [NH Et ][S CNEt ] (iv in Scheme 1). Com-
2
2
2
2
2
pound 2 crystallised from the filtered solution over the course
of several days as the bis-thf adduct in moderate yield; but
when R = Et, the product was isolated (after the addition of
toluene and the evaporation of the majority of the remaining
as shown in eqn. (1), using anhydrous Ce() triflate as pre-
cursor. The use of lanthanoid metal (Ln) triflates as precursors
to not only [Ln{N(SiMe ) } ] but also Ln() aryloxides or
3
2 3
17
2
cyclopentadienyls was first noted at Sussex; and more recently
was reported for Ln = La, Nd, Sm and Er in yields ranging from
thf ) as the solvent-free Ce(κ -S CNEt ) 3. The absence of
2
2 3
coordinated thf in 3 reflects the increase in steric crowding
around the metal ion in the final product, and this (along with
enhanced ligand basicity) greatly increases the sensitivity of the
complex towards even traces of oxygen gas, at least in solution;
in the solid state, both 2 and 3 were surprisingly resistant to
atmospheric oxygen for up to 1 min, due in no small part to
their high degree of crystallinity. Both were high-melting,
bright yellow solids. The methyl derivative 2 lost its coordinated
thf upon heating above 100 ЊC.
Compound 2 was insoluble in common solvents, and 3 was
only slightly soluble even in thf or hot toluene, rendering NMR
spectral characterisation impractical. The solid state structure
of 3 is, at present, unknown, but the presumed coordinative
unsaturation at the metal centre, combined with the observ-
ation of peaks ascribed to formation of dimers in the EI mass
spectrum, suggests the possibility of higher nuclearity species.
The formulation of the methyl derivative 2 rests on mass spec-
tral data (although no peaks retaining the coordinated thf were
observed) and its microanalysis, and, more conclusively, on an
X-ray crystallographic study performed on single crystals of 2
as deposited from the reaction solution (Fig. 1 and Table 1).
18
5
4 to 68%.
(
1)
When ca. 6 to 8 equivalents of tetramethylthiuram disulfide
were held in suspension with a thf solution of A for ca. 3 h, a
deep purple coloration developed. In contrast to the ease with
which this reaction was performed, isolation of the chromo-
phore proved to be exceptionally difficult. On the one occasion
that crystalline material was isolated it was not of sufficient
1
13
quality for an X-ray crystallographic study, but H and
C
NMR spectroscopic evidence suggested the formation of the
mixed amido/dithiocarbamato complex 1 (i in Scheme 1). There
was no obvious paramagnetic shift in either spectrum, the
dithiocarbamato and trimethylsilyl protons resonating at δ 2.66
and 0.72, respectively in C D solution, while the ratio of their
6
6
integrated intensities (12 : 36) unquestionably indicated that the
molecule 1 contained an equal number of dithiocarbamato
13
and amido ligands. The C NMR spectrum showed a signal
1
30
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6

View Article Online
Table 1 Selected bond lengths (Å) and angles (Њ) for 2
Table 2 Selected bond lengths (Å) and angles (Њ) for 4
Ce–S(1)
Ce–S(3)
S(1)–C(1)
S(3)–C(3)
2.9487(5)
2.9825(6)
1.7186(16)
1.719(2)
Ce–S(2)
2.9208(5)
2.5051(14)
1.718(2)
Ce–S(3)
Ce–S(5)
Ce–N(4)
S(1)–C(1)
S(3)–C(4)
S(5)–C(7)
C(1)–N(1)
2.9258(10)
2.9408(11)
2.655(3)
1.720(3)
1.720(4)
1.718(4)
1.328(4)
Ce–S(4)
Ce–S(6)
Ce–N(5)
S(2)–C(1)
S(4)–C(4)
S(6)–C(7)
N(2)–C(4)
2.9927(11)
2.9064(10)
2.676(3)
1.727(3)
1.710(4)
1.730(4)
1.324(5)
Ce–O(1)
S(2)–C(3)
N(1)–C(1)
1.341(4)
S(1)–Ce–S(1)Ј
S(2)–Ce–S(2)Ј
O(1)–Ce–O(1)Ј
O(1)–Ce–S(2)
S(2)–C(3)–S(3)
60.31(2)
159.95(2)
155.86(8)
91.45(4)
S(2)–Ce–S(3)
S(3)–Ce–S(3)Ј
O(1)–Ce–S(1)
O(1)–Ce–S(3)
60.18(2)
79.92(2)
72.65(4)
77.75(4)
S(1)–Ce–S(2)
S(5)–Ce–(6)
S(1)–C(1)–S(2)
S(5)–C(7)–S(6)
S(1)–Ce–S(5)
N(4)–Ce–S(5)
61.24(2)
61.33(3)
118.6(2)
119.7(2)
153.49(3)
78.91(7)
S(3)–Ce–S(4)
N(4)–Ce–N(5)
S(3)–C(4)–S(4)
S(2)–Ce–S(6)
N(5)–Ce–S(5)
N(5)–Ce–S(4)
59.84(3)
61.40(9)
118.8(2)
71.69(3)
85.27(6)
80.25(7)
118.91(12)
Symmetry transformations used to generate equivalent atoms: ЈϪx, y,
Ϫz ϩ 1/2.
By treatment of the Ce() amide A with three equivalents of
tetramethylthiuram disulfide and 2,2Ј-bipyridine in thf, well
formed, deep red–brown octahedra of up to 4 mm in length
were deposited from the reaction mixture over the course of
2
several days (v in Scheme 1). The product [Ce(κ -S CNMe ) -
2
2 3
(
2,2Ј-bipy)] 4 has been identified only by microanalysis and
a single crystal X-ray study, as it was too insoluble for NMR
solution spectroscopy and its mass spectrum showed only peaks
derived from the bipy ligand. Analogues are known for some
20
other lanthanides (La, Pr, Nd and Sm–Lu), and comparison
of the solid state structure of 4 (Fig. 2 and Table 2) with that of
20
its erbium congener shows the two to be isostructural. There
is a slightly greater variation in the Ce–S bond lengths in 4 than
for 2, but even so the total range is no greater than 0.1 Å, which
(
when the absence of orbital effects for the lanthanides is borne
in mind) again suggests that they have little or no chemical
significance, the apparent differences being simply manifest-
ations of both packing effects and the great precision of the
C–S bonds suggests. A dithiolene-type resonance form may be
the most accurate description of the bonding between the metal
ion and the S2C units.
Fig. 1 Molecular structure of compound 2
The cerium ion in crystalline 2 is eight-coordinate (6S and 2O
atom donor sites) and the molecule resides on a two-fold
crystallographic axis which passes through Ce, C(1) and N(1).
As a consequence, there are just two unique dithiocarbamato
ligands, both of which are planar; in one of these, the two Ce–S
distances are equal, Ce–S(1) 2.9487(5) Å. In the other, there
is a slight asymmetry in coordination, resulting in bonds of
length 2.9208(5) and 2.9825(6) Å for Ce–S(2) and Ce–S(3),
respectively. All three associated S–C bond lengths are the
same, indicating delocalisation within each S C fragment and
2
ascribing little chemical significance to the marginal variations
in the observed Ce–S separations. The geometry at the metal
does not conform to any reasonably idealised description as a
direct result of the small bite angles of the dithiocarbamato
ligands.
Further development of compounds similar to 2 and 3 by
modification of the alkyl groups on the dithiocarbamato
ligands – specifically in order to lower their melting points and
increase vapour pressure with a view to using them as dopant
precursors – may be straightforward as the required tetra-
alkylthiuram disulfides are readily accessible.
Fig. 2 Molecular structure of compound 4
There are numerous examples in the literature where the
combination of a lanthanide ion, three chelating bidentate
ligands and a polypyridyl ligand (L L) (acting both to activate
While the molecular structure of 4 is not particularly surpris-
ing, its mode of synthesis is novel and warranted further exten-
sion of such an oxidative cleavage method towards other useful
complexes. Two routes were followed: that of increasing the
lipophilicity of the dithiocarbamato ligands which led to
∩
and saturate the inner coordination sphere of the metal) often
results in the formation of potentially useful electroluminescent
2
2
complexes. The target molecules have normally contained a
the synthesis (vi in Scheme 1) of [Ce(κ -S CNEt ) (bipy)] 5, and
2
2 3
substituted β-diketonato ligand, and are available in high yield
via straightforward three-component reactions under aerobic
the utilisation of 4,7-diphenyl-1,10-phenanthroline (dipphen)
as an activating and volatilising ligand which afforded (vii in
2
conditions, e.g., [Eu(DTP) (dipphen)] (DTP = 1,3-dithienyl-
Scheme 1) [Ce(κ -S CNMe ) (dipphen)] 6.
3
2
2 3
1
9
diketonate, dipphen = 4,7-Ph -1,10-phenanthroline). As noted
The ethyl derivative 5 proved to be more readily soluble than
4 (and hence a more complicated work-up procedure was
required than for 4) and also had a lower melting point, below
its decomposition temperature. This is a highly desirable
characteristic, and clearly shows that it should in future be
possible systematically to lower the melting point further by the
2
earlier, such straightforward syntheses are not applicable to
cerium dithiocarbamates. However, we now report that similar
results may be achieved simply by adding the required donor to
a mixture of A and the thiuram disulfide in thf solution, (v, vi
and vii in Scheme 1).
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6
131

View Article Online
Table 3 Selected bond lengths (Å) and angles (Њ) for 6
Table 4 S–Ce–SЈ bond angles (Њ) for 7
Ce–S(1)
Ce–S(3)
Ce–S(5)
Ce–N(4)
S(1)–C(1)
S(3)–C(4)
S(5)–C(7)
N(1)–C(1)
2.927(3)
2.932(4)
2.930(3)
2.665(10)
1.677(12)
1.741(14)
1.724(14)
1.346(14)
Ce–S(2)
Ce–S(4)
Ce–S(6)
Ce–N(5)
S(2)–C(1)
S(4)–C(4)
S(6)–C(7)
N(2)–C(4)
2.931(3)
2.936(4)
2.876(4)
2.664(11)
1.733(13)
1.690(14)
1.703(13)
1.354(17)
S(1)–Ce–S(1)Ј
S(1)–Ce–S(4)Ј
S(1)–Ce–S(4)
S(4)Ј–Ce–S(4)
S(1)–Ce–S(3)Ј
S(4)–Ce–S(3)Ј
S(1)–Ce–S(3)
S(4)–Ce–S(3)
89.66(4)
131.09(2)
84.95(3)
132.69(4)
165.46(2)
85.26(3)
88.71(3)
63.11(2)
S(3)Ј–Ce–S(3)
S(3)Ј–Ce–S(2)Ј
S(4)–Ce–S(2)Ј
S(3)–Ce–S(2)Ј
S(1)–Ce–S(2)
S(4)–Ce–S(2)
S(3)–Ce–S(2)
S(2)Ј–Ce–S(2)
96.40(4)
81.03(3)
68.15(2)
130.85(2)
62.99(2)
135.83(2)
85.36(3)
129.06(4)
Symmetry transformations used to generate equivalent atoms: ЈϪx ϩ 1,
y, Ϫz ϩ 1/2.
S(1)–Ce–S(2)
S(5)–Ce–S(6)
S(1)–C(1)–S(2)
S(5)–C(7)–S(6)
60.73(10)
61.50(11)
120.6(7)
S(3)–Ce–(4)
N(4)–Ce–(5)
S(3)–C(4)–S(4)
60.75(11)
61.5(3)
119.7(8)
120.1(8)
The crystalline, toluene-soluble cerium() dithiocarbamate
2
[
Ce(κ -S CNEt ) ] 7 was obtained (viii in Scheme 1) by briefly
2
2 4
presence of even more bulky alkyl groups at the periphery of
the coordination sphere. The associated enhancement in vola-
bubbling oxygen gas through a suspension of the yellow
2
Ce(κ -S CNEt ) 3 in toluene at ambient temperature. Reaction
2
2 3
tility was also evident in the EI mass spectrum, which showed
was rapid, as evident from the almost immediate colour change
to the black 7. The latter was isolated by filtration and crystal-
lisation from the filtrate, in high yield, indicating that its form-
ation is appropriately represented by eqn. (2); the coproduct
was not identified, but is tentatively suggested to have been
Ce O or CeO .
ϩ
[
2M Ϫ dtc] as its heaviest fragment ion. Again, analogues of 5
are known for all of the other lanthanides, along with an X-ray
crystal structure of the europium complex, which is mono-
19
nuclear; thus, the observation of dinuclear fragments in the EI
mass spectrum of 5 may be a poor guide to its nuclearity in the
solid state, and this may also be the case for the unsaturated
compound 3. As a final observation on the properties of
this complex, it should be noted that the steric saturation
brought about by the bipy ligand renders the eight-coordinate
cerium() complex 5 to be considerably more resistant towards
traces of oxygen gas than the apparently six-coordinate
2
3
2
(2)
The formulation of 7 is based on its cerium analysis (C, H
2
and N data were unsatisfactory, attributed to its air-sensitivity),
complex Ce(κ -S CNEt ) 3.
2
2 3
1
13
diamagnetism, appropriate H and C NMR spectra, and its
X-ray structure, Fig. 4.
Once again, as in the case of the dipphen complex 6, defini-
tive characterising data were hard to compile, and in this
respect the most important information came from X-ray
crystallography. The coordination about the metal (Fig. 3 and
Table 3) is very similar to that encountered in the structure of
the bipy complex 4. The planes of the dithiocarbamato ligands
in 6, as defined by the atoms S(3), C(4), S(4) and S(5), C(7), S(6)
are deflected away from the cerium atom. In the case of the
former this may be a result of close contact with one of the
occluded solvent molecules, but this explanation does not apply
to the other inclined ligand. It is clear that the dipphen in 6
extends far beyond the rather compact space enclosing the
‘
Ce(dtc) ’ core, no doubt a significant factor in forcing the unit
3
cell to include two thf molecules in the otherwise vacant space
above the plane of the polypyridyl. The loss in the lattice energy
which this entails may be the cause of the small, but noticeable,
compression of the Ce–S bonds, which are 0.03 to 0.04 Å shorter
than their counterparts in 4. This is associated both with a wider
variation in C–S bond lengths within each dithiocarbamato lig-
Fig. 4 Molecular structure of compound 7
2
and and a slight lengthening of the C(sp )–N bonds, observations
which are indicative of increased electron-localisation within the
The molecule 7 lies on a two-fold rotation axis, in which the
S2CN fragment. The metric differences between the structures of
cerium is bonded to eight sulfur atoms from four chelating
2
4
and 6 are, however, of marginal significance.
dithiocarbamato ligands. It is isomorphous with [Th(κ -S -
2
CNEt ) ], which was the first example of a metal complex
2
4
21
coordinated to eight sulfur atoms; 7 is a also isoleptic with
2
22
2
4
[
Ti(κ -S CNEt ) ], and the anions of Na[Ce(κ -S CNEt ) ]
2 2 4 2 2 4
2 3
and [NEt ][Np(κ -S CNEt ) ]. We note that the dodecahedral
4
2
2 4
2
Ce() and Th() compounds [M{κ -(OC(CH Ph)) CH} ] are
2
2
4
23
also isomorphous. The cerium atom environment in 7 (Fig. 4)
is intermediate between the ideal dodecahedral and square
antiprismatic arrangements. The angles between the Ce–S
bonds and the two-fold axis fall into two groups, with θ = 44.8
A
and 48.2Њ and θ = 64.5 and 66.3Њ, respectively. These may be
B
compared with Hoard’s values for the most favourable dodeca-
hedron (θ = 35.2Њ, θ = 73.5Њ) and square antiprism, θ = θ =
A
B
A
B
24
5
7.3Њ.
The major geometric parameters for 7 are summarised in
Fig. 5 and Table 4; the data are very similar to those for [Th-
2
21
(
κ -S CNEt ) ], except that the mean Th–S bond length
2 2 4
of 2.87 Å is ca. 0.07 Å longer than its Ce–S counterpart in 7.
Fig. 3 Molecular structure of compound 6
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6
The data for 7 are closely similar to those for 2, as far as the
1
32

View Article Online
2
Fig. 5 Bond lengths (Å) and angles (Њ) (see also Fig. 4 and Table 4) in [Ce(κ -S CNEt ) ] 7
2
2 4
Table 5 Selected bond lengths (Å) and angles (Њ) for 9
Table 6 Selected bond lengths (Å) and angles (Њ) for 10.
Tb–N
2.233(12)
Si–N
1.722(7)
Tb–S(1)
Tb–S(3)
S(1)–C(1)
S(3)–C(3)
2.8651(8)
2.8854(9)
1.724(2)
1.714(3)
Tb–S(2)
2.8187(8)
2.4138(19)
1.716(3)
Tb–O(1)
S(2)–C(3)
N(1)–C(1)
N–Tb–NЈ
Si–N–Tb
112.96(8)
129.8(4)
Si–N–Si
SiЉ–N–Tb
123.1(8)
106.1(3)
1.313(6)
Symmetry transformations used to generate equivalent atoms: ЈϪx ϩ y
ϩ 1, Ϫx ϩ 1, z, ЉϪx ϩ y ϩ 1, y, Ϫz ϩ 1/2.
S(1)Ј–Tb–S(1)
S(2)Ј–Tb–S(2)
O(1)–Tb–O(1)Ј
O(1)–Tb–S(2)
S(2)–Tb–S(3)
61.73(3)
158.54(4)
154.42(12)
90.99(5)
S(2)–Tb–S(3)
S(3)Ј–Tb–S(3)
O(1)–Tb–S(1)
O(1)–Tb–S(3)
62.07(3)
77.43(4)
72.53(6)
77.29(6)
2
Ce(κ -S CNR ) fragment is concerned (R = Me 2, R = Et 7),
2
2
62.07(3)
except that the average Ce–S bond length in 7 is 2.803 Å,
whereas in 2 it is 2.950 Å. The mean of five of the six Ce–S
Symmetry element used to generate equivalent atoms: ЈϪx, y, 0.5 Ϫ z.
2
distances in [Ce(κ -S CNMe ) (dipphen)]ؒ2thf 6 is 2.931 Å, the
2
2 3
remaining Ce–S bond being of 2.876 Å in length. The Ce()
complexes 2 and 6, like 7, have an eight-coordinate Ce environ-
terbium atom is 0.60 Å out of the plane of the adjacent nitro-
gen atoms and the Tb–N bond length is 2.233(12) Å [cf. ref. 27
for the isoleptic Er compound, the metal atom is 0.58 Å out of
ment (6S ϩ 2N in 2 and 6); 2 and 7 are both C symmetric. The
2
significantly shorter Ce–S distance in 7 than in 2 or 6 is
appropriate for 7 being in the higher oxidation state.
the N plane and the Er–N bond length is 2.21(1) Å].
3
2
The crystalline compounds [Ln(κ -S CNMe ) (thf ) ] (Ln =
2
2
3
2
The discovery of the remarkably ready conversion of cerium
tris(trimethylsilyl)amide A into various Ce() dithiocarb-
amates (Scheme 1), led us to examine the possible generality of
this procedure for other of the lanthanoid elements. The choice
fell on the neodymium and terbium analogues of A; the latter
seemed to be a particularly promising candidate as a potential
Nd, 8; Ln = Tb 10) were readily prepared under identical condi-
tions to those which had been used for the isoleptic cerium
compound 2, eqn. (4). They were identified by microanalysis,
mass spectra and for 10 the X-ray structure (Fig. 7 and Table 6).
The crystalline complex 10 is isomorphous with the isoleptic
Ce() complex 2, and as such it lies on a crystallographic two-
fold axis and its structure is unexceptional. As required by the
lanthanide contraction the mean Ln–S and Ln–O bond lengths
are 0.094 and 0.087 Å, respectively, shorter in the Tb compound
7
precursor to the unknown Tb() (f ) oxidation state (this
aspect is still under investigation).
The 4f metal {Ln{N(SiMe ) } ] series was last reviewed in
3
2 3
25
1
996. The neodymium compound was originally prepared
from NdCl and the lithium amide, and more recently from
Nd(OTf ) and the sodium amide. The terbium amide 9, was
1
0 than in the Ce analogue 2.
16
3
1
8
26
3
here prepared as shown in eqn. (3). After sublimation and then
(
3)
crystallisation, it was obtained as colourless needles. It was
25
characterised by its X-ray structure, which is unexceptional.
Thus the molecule 9 (Fig. 6 and Table 5) has point group C3,
with the Tb atom disordered above and below that plane.. The
Fig. 7 Molecular structure of compound 10
Initially, we attributed the ready formation of the Ce()
dithiocarbamates 2–6 from the cerium() amide A on the
availability of the Ce() oxidation state, and it was therefore a
surprise to find that similar results were obtained with the Nd
and Tb analogues of A, since the Ln() state had not hitherto
proved to be normally accessible for lanthanoids other than Ce.
While the homoleptic dithiocarbamates of the other lan-
thanides are conveniently prepared using the more traditional
method of ligand transfer between M(S CNR ) (M = Li, Na,
2
2
K) and LnCl or Ln(OTf ) , we do not envisage that this new
3
3
route would usefully be applied to a 4f element other than
2
cerium. However, the successful synthesis of the [Ln(κ -S -
Fig. 6 Molecular structure of compound 9, showing only one of the
two alternative Tb sites
2
CNMe
) (thf )
3
] compounds 8 (Nd) and 10 (Tb) puts conjecture
2
2
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6
133

View Article Online
(
4)
about the reaction mechanism on a firmer foundation, as it is
for a brief period under vacuum, yielding 2 (Found: C, 31.3;
unlikely that a Ln() intermediate such as B (R = SiMe ) was
first formed, except in the case of Ce; cf., compound 1. A more
H, 5.4; N, 6.7. C H CeN O S requires C, 31.7; H, 5.31;
3
17 34
3
2 6
N, 6.51%), mp > 250 ЊC (lost thf at > 100 ЊC); m/z 500
ϩ ϩ
plausible mechanism for Nd and Tb is of the S 2-type at Ln, C.
([M Ϫ 2thf] , 6%), 485 ([M Ϫ 2thf Ϫ Me] , 1%), 380 ([Ce(S -
2
ϩ ϩ ϩ
H
The fate of the supposedly eliminated aminyl radical, ؒNR has
CNMe ) ] , 9%), 121 ([HS CNMe ] , 62%), 88 ([SCNMe ] ,
2
2
2
2
2
2
still to be established, but the volatile disilazane HNR is the
100%).
2
likely candidate.
2
Ce(ꢀ -S CNEt ) 3. (a) A solution of tetraethylthiuram
2
2 3
3
disulfide (3.75 g, 12.7 mmol) in diethyl ether (100 cm ) was
added to a solution of the amide A (2.62 g, 4.22 mmol) in
diethyl ether (50 cm ). A yellow–brown solid gradually formed;
3
the colour of the solution changed from yellow to red–brown.
Filtration afforded the yellow solid 3 (1.32 g, 53%).
(
b). A solution of the amide A (1.76 g, 2.83 mmol) in diethyl
3
Experimental
ether (50 cm ) was added to a suspension of diethylammonium
diethyl dithiocarbamate (1.88 g, 7.42 mmol) in diethyl ether
50 cm ). The mixture was stirred overnight. A part of the
volatiles was removed in vacuo and the residue (ca. 50 cm ) was
filtered yielding yellow needles of 3 (1.12 g, 72%) (Found: C,
All manipulations were carried out under argon using Schlenk
techniques. Solvents were dried and distilled over sodium–
potassium alloy under nitrogen gas prior to use. The tetra-
alkylthiuram disulfides, 2,2Ј-bipyridine and 4,7-diphenyl-1,10-
phenanthroline were purchased from Aldrich Chemical Co. and
used as received. Microanalyses were carried out by Medac Ltd.
3
(
3
3
7
2
1
0.0; H, 5.2; N, 7.0. C H CeN S requires C, 30.8; H, 5.17; N,
15 30 3 6
1
.18%), mp > 250 ЊC; H NMR (300 MHz, CDCl ): δ 3.36 (s, br,
3
13
1
H), 0.48 (s, 3H); C{ H} NMR (75 MHz, CDCl ): δ 48.20,
1.44; m/z 1168 ([2M] , 1%), 1020 ([Ce (S CNEt ) ] , 8%), 584
([M] , 62%), 436 ([Ce(S CNEt ) ] , 100%), 149 ([HS CNEt ] ,
3
ϩ
(
Brunel University). The NMR spectra were recorded using the
ϩ
2
2
2 5
DPX Bruker instrument, calibrated internally to the residual
ϩ
ϩ
ϩ
1
13
29
2 2 2 2 2
solvent for H and C, or externally using SiMe for Si. EI
4
4
9%).
Ce(ꢀ -S CNMe ) (2,2Ј-bipy)] 4. A mixture of the amide A
mass spectra were obtained from solid samples using a Kratos
MS 80 RF instrument; m/z values are reported for the most
common isotope, and observed envelopes were compared with
calculated patterns. Melting points were taken in sealed capil-
laries and are uncorrected.
2
[
2
2 3
(
1.15 g, 1.85 mmol), 2,2Ј-bipyridine (0.29 g, 1.86 mmol) and
3
tetramethylthiuram disulfide (1.3 g, 5.42 mmol) in thf (40 cm )
was stirred for 1 h. The resulting red suspension was filtered.
The filtrate was collected in a fresh Schlenk vessel and set aside
for 6 d. The pale supernatant liquor was decanted off from the
large (up to 4 mm in length), deep red–brown octahedral crys-
tals of 4 which had formed. These were dried under vacuum,
yielding 4 (0.70 g, 58%) (Found: C, 34.7; H, 4.2; N, 10.6.
C H CeN S requires C, 34.7; H, 3.99; N, 10.66%), mp
Preparations
[
Ce{N(SiMe ) } ] A. A solution of sodium bis(trimethyl-
3 2 3
3
silyl)amide (9.33 g, 50.9 mmol) in thf (120 cm ) was added to
cerium triflate (9.46 g, 16.4 mmol). The mixture was refluxed; a
homogeneous yellow solution formed in ca. 1 h, and refluxing
was continued for a further 3 h. After cooling to room temper-
ature, the solvent was removed in vacuo. The residual yellow
19
26
5 6
ϩ
ϩ
(decomp.) > 230 ЊC; m/z 156 ([bipy] , 100%), 78 ([py] , 28%).
3
2
solid was extracted with pentane (2 × 50 cm ) and the mixture
was filtered. The filtrate was evaporated to dryness in vacuo.
Sublimation of the residue at 150 ЊC/10 mmHg provided the
Ce(ꢀ -S CNEt ) (2,2Ј-bipy) 5. The amide A (1.05 g, 1.7
mmol), 2,2Ј-bipyridine (0.26 g, 1.7 mmol.) and tetraethyl-
thiuram disulfide (1.5 g, 5.06 mmol.) in diethyl ether (40 cm )
2
2 3
Ϫ4
3
1
golden solid A (8.43 g, 82%). H NMR (300 MHz, C D ):
were stirred together. A salmon-pink solid precipitated from the
initially formed red solution after ca. 5 min, and stirring was
continued for a further 1 h, whereafter the solid 5 was isolated
by filtration and washed with diethyl ether. The residue was
dissolved in a hot toluene–thf mixture, and the filtered solution
was set aside at room temperature to furnish the brown crystal-
line compound 5 (1.02 g, 81%) (Found: C, 40.5; H, 5.3; N,
9.4. C H CeN S requires C, 40.5; H, 5.17; N, 9.45%), mp
6
6
13
1
29
1
δ Ϫ3.45; C{ H} NMR (75 MHz, C D ): δ 2.5; Si{ H} NMR
6
6
(
99.45 MHz, C D ): δ Ϫ17.5.
6 6
2
Ce{N(SiMe ) ) (ꢀ -S CNMe )
2 2
1.
Tetramethylthiuram
3
2
2
2
disulfide (1.83 g, 7.63 mmol) was added to a stirred solution of
the amide A (0.80 g, 1.29 mmol) in thf (30 cm ). Stirring was
continued for 3 h; the colour of the mixture changed from yel-
low to brown and then deep purple. The mixture was filtered
and the filtrate was concentrated in vacuo to ca. 3 cm . After the
residue had been at Ϫ30 ЊC for a few days, some solid, tenta-
tively formulated as 1, formed; this was isolated by filtration,
3
25
38
5 6
ϩ
200–202 ЊC; m/z 1177 ([Ce (S CNEt ) (bipy)] , 1%), 1020
2 2 2 5
ϩ ϩ
3
([Ce (S CNEt ) ] , 11%), 584 ([M Ϫ bipy Ϫ H] , 65%), 436
2 2 2 5
ϩ
([Ce(S CNEt ) ] , 100%).
2 2 2
1
2
and dried in vacuo. H NMR (300 MHz, C D ): δ 2.66 (s, 1 H),
[Ce(ꢀ -S CNMe ) (dipphen)]ؒ2thf 6. The amide A (0.41 g,
2 2 3
6
6
0
.72 (s, 3 H).
0.66 mmol), 4,7-diphenyl-1,10-phenanthroline (0.22 g, 0.66
mmol) and tetramethylthiuram disulfide (0.47 g, 1.96 mmol) in
thf (5 cm ) were stirred together for 30 min, whereafter the red
2
3
[
Ce(ꢀ -S CNMe ) (thf) ] 2. The amide A (1.0 g, 1.61 mmol)
2
2
3
2
and tetramethylthiuram disulfide (1.16 g, 4.83 mmol) were
stirred together while thf (40–50 cm ) sufficient to dissolve all
mixture was filtered and the filtrate was reduced slightly in
volume. Crystallisation was initiated by cooling and was con-
tinued at room temperature, giving pale red–brown crystals of 6
(0.35 g, 54%) (Found: C, 49.7; H, 5.0; N, 7.4. C H CeN O S
3
the solid had been added. Stirring was continued for 20 min.
The mixture was filtered and the filtrate was transferred into a
fresh Schlenk vessel. After the solution had been set aside at
room temperature for 5 d, large golden yellow crystals had
formed, which were isolated by decanting the solvent, and dried
41
50
5
2 6
requires C, 50.4; H, 5.16; N, 7.17%), mp > 250 ЊC; m/z 332
ϩ
ϩ
ϩ
([dipphen] , 97%), 255 ([dipphen Ϫ Ph] , 45%), 77 ([Ph] ,
100%).
1
34
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6

View Article Online
2
[
Ce(ꢀ -S CNEt ) ] 7. Oxygen gas was bubbled through a
2 2 4
2
mixture of Ce(κ -S CNEt ) 3 (1.32 g, 2.27 mmol) in toluene
50 cm ) for 1 min. The mixture was filtered and the filtrate was
2
2 3
3
(
transferred to a fresh Schlenk flask. The filtrate was set aside
at room temperature for a few days, yielding black crystals
of compound 7 (0.93 g, 74%) (Found: C, 30.7; H, 5.0; N, 5.0,
Ce, 19.0. C H CeN S requires C, 32.8; H, 5.50; N, 7.64;
2
0
40
4 8
1
Ce, 19.11%); H NMR (300 MHz, CDCl ): δ 3.99 (q, J = 7.01
3
1
13
Hz, 2 H), 1.21 (t, J = 7.01 Hz, 3 H); C{ H} NMR (75 MHz,
ϩ
CDCl ): δ 201.06, 46.43, 12.42; m/z 265 ([Ce(S CNMe )] ,
3
2
2
ϩ
1
8%), 88 ([SCNMe ] , 100%).
2
[
Tb{N(SiMe ) } ] 9. Sodium bis(trimethylsilyl)amide (3.67 g,
3 2 3
1
9.6 mmol) was added to a stirred suspension of terbium tri-
3
chloride (1.73 g, 6.53 mmol) in thf (30 cm ). The mixture was
refluxed overnight. After cooling to room temperature, the sol-
vent was removed in vacuo and the residue was twice sublimed
Ϫ3
at 160 ЊC/10 mmHg, affording the colourless solid terbium
amide 9 (2.01 g, 48%). Colourless, needles of single crystal
quality were obtained from hexane solution.
2
[
Tb(ꢀ -S CNMe ) (thf) ] 10. Tetramethylthiuram disulfide
2
2
3
2
(
0.44 g, 1.83 mmol) was added to a stirred solution of the ter-
3
bium amide 9 (0.39 g, 0.61 mmol) in thf (20 cm ). The mixture
was stirred for 1 h, then filtered and the filtrate was transferred
to a fresh Schlenk vessel. The residue was washed with thf
3
(
20 cm ). The combined filtrate and washings were set aside at
room temperature for 3 d, yielding the colourless compound 10
0.29 g, 71%) (Found: C, 30.2; H, 5.30; N, 6.38; Tb, 23.58.
C H N O S Tb requires C, 30.8; H, 5.16; N, 6.33, Tb,
(
17
34
3
2 6
2
2
3
3.94%), as needles of single crystal quality; m/z 519 ([M Ϫ
ϩ
ϩ
thf] , 70%), 399 ([M Ϫ 2thf Ϫ S CNMe ] , 100%), 367 (14%),
2
2
21 (58%), 280 (M Ϫ 2thf Ϫ 2S CNMe ϩ 1, 10%), 121
2
2
ϩ
ϩ
(
[HS CNMe ] , 6%), 73 ([thf ϩ 1] , 12%).
2 2
2
[
Nd(ꢀ -S CNMe ) (thf) ] 8. This compound was prepared
2 2 3 2
using a similar method as that for the Ce analogue 2, from
[
Nd{N(SiMe ) } ] (1.6 g, 2.58 mmol) and tetramethylthiuram
3
2 3
disulfide (3 eq.). Crystallisation, induced by storage at Ϫ25 ЊC,
afforded the blue microcrystalline powder of compound 8
(
Found: C, 30.5; H, 5.3; N, 6.4. C H N NdO S requires C,
17 34 3 2 6
3
2
1.5; H, 5.28; N, 6.47%), mp > 250 ЊC (lost thf at > 100 ЊC); m/z
ϩ ϩ
65 ([Nd(S CNMe )] , 18%), 88 ([SCNMe ] , 100%).
2
2
2
Crystal data and refinement details for 2, 4, 6, 7, 9 and 10
Diffraction data for each of 2, 4, 6, 7, 9 and 10 were collected on
an Enraf-Nonius Kappa CCD diffractometer using mono-
chromated Mo-Kα radiation, λ = 0.71073 Å at 173(2) K. Crys-
tals were coated in oil and then directly mounted on the
diffractometer under a stream of cold nitrogen gas. The struc-
2
28
tures were refined on all F using SHELXL-97. Non-hydrogen
atoms were refined anisotropically; hydrogen atoms were
included in the riding mode. Crystals of 6 and 9 gave very weak
diffractions. For 6, two molecules of disordered solvent were
both included as carbon atoms with H atoms omitted. The Tb
atom in 9 was disordered over two positions related by the two-
Ϫ3
fold rotation axis. For 10, there was a residual peak of 6 e Å
lying on the two-fold axis at a distance of 2–3 Å from the Tb
atom. Further details are found in Table 7. All figures show
atoms with 50% ellipsoids.
CCDC reference numbers 220010–220015.
See http://www.rsc.org/suppdata/dt/b3/b311397c/ for crystal-
lographic data in CIF or other electronic format.
Acknowledgements
For the award of fellowships, we thank the EPSRC (A. G. H.)
and the Royal Society (Z. L.).
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6
135

View Article Online
1
4 C. Morton, N. W. Alcock, M. R. Lees, I. J. Munslow, C. J. Sanders
and P. Scott, J. Am. Chem. Soc., 1999, 121, 11255.
Notes and references
1
2
3
4
5
I. Tanaka, Y. Izumi, K. Tanaka, Y. Inoue and S. Okamoto,
J. Lumin., 2000, 87–89, 1189.
15 J. Christoffers, T. Werner, S. Unger and W. Frey, Eur. J. Org. Chem.,
2003, 425.
16 D. C. Bradley, J. S. Ghotra and F. A. Hart, J. Chem. Soc.,
Dalton Trans., 1973, 1021.
17 S. A. Holmes, D. Phil Thesis, University of Sussex, 1992, cited in:
A. D. Frankland, P. B. Hitchcock, M. F. Lappert and G. A. Lawless,
J. Chem. Soc., Chem. Commun., 1994, 2435.
18 S. A. Schuetz, V. W. Day, R. D. Sommer, A. L. Rheingold and
J. A. Belot, Inorg. Chem., 2001, 40, 5292.
W. S. Rees, Jr., O. Just and D. S. Van Derveer, J. Mater. Chem., 1999,
9
, 249.
D. Brown, D. G. Holah and C. E. F. Rickard, J. Chem. Soc. A, 1970,
86.
7
M. Ciampolini, N. Nardi, P. Colamarino and P. Orioli, J. Chem.
Soc., Dalton Trans., 1977, 379.
M. C. Cassani, Yu. K. Gun’ko, P. B. Hitchcock, A. G. Hulkes, A. V.
Khvostov, M. F. Lappert and A. V. Protchenko, J. Organomet.
Chem., 2002, 647, 71.
19 K. Okada, Y.-F. Wang, T.-M. Chen, M. Kitamura, T. Nakaya and
H. Inoue, J. Mater. Chem., 1999, 9, 3023.
6
7
8
9
O. Eisenstein, P. B. Hitchcock, A. G. Hulkes, M. F. Lappert and
L. Maron, Chem. Commun., 2001, 1560.
P. B. HitchcockA. G. Hulkes and M. F. Lappert, Inorg. Chem., in
press.
20 C. Su, N. Tang, M. Tan and K. Yu, Polyhedron, 1996, 15, 233.
21 D. Brown, D. G. Holah and C. E. F. Rickard, J. Chem. Soc. A, 1970,
423.
22 M. Colapietro, A. Vaciago, D. C. Bradley, M. B. Hurthouse and
I. F. Rendall, J. Chem. Soc., Dalton Trans., 1972, 1052.
23 L. Wolf and H. Bärnighausen, Acta Crystallogr., 1960, 10, 605.
24 J. L. Hoard and J. V. Silverton, Inorg. Chem., 1963, 2, 235.
25 R. Anwander, in Organolanthanoid Chemistry: Synthesis, Structure,
Catalysis, ed. W. A. Herrmann, Springer, Berlin, 1996, ch. 2
(Lanthanide Amides), pp. 52–64.
26 M. Allen, H. C. Aspinall, S. R. Moore, M. B. Hursthouse and
A. I. Karvalov, Polyhedron, 1992, 11, 409.
27 W. A. Herrmann, R. Anwander, F. C. Munck, W. Scherer,
V. Dufaud, N. W. Huber and G. R. J. Artus, Z. Naturforsch., Teil B,
1994, 49, 1789.
W. J. Evans, T. J. Deming, J. M. Olofson and J. W. Ziller,
Inorg. Chem., 1989, 28, 4027.
A. Sen, H. A. Stecher and A. L. Rheingold, Inorg. Chem., 1992, 31,
4
73.
1
1
1
0 Yu. K. Gun’ko, R. Reilly, F. T. Edelmann and H.-G. Schmidt,
Angew. Chem., Int. Ed., 2001, 40, 1279.
1 W. J. Evans, T. J. Deming and J. W. Ziller, Organometallics, 1989, 8,
1
581.
2 A. Gulino, M. Casarin, V. P. Conticello, J. G. Gaudiello,
H. Mauermann, J. Fragalá and T. J. Marks, Organometallics, 1988,
7
, 2360.
3 V. Nair, J. Mathew and J. Prabhaken, Chem. Soc. Rev., 1997, 26,
27.
1
28 G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Refinements, University of Göttingen, 1997.
1
1
36
D a l t o n T r a n s . , 2 0 0 4 , 1 2 9 – 1 3 6