Recent advances in the synthesis of N-heteroatom substituted imido complexes containing a nitrido bridge [M=N-E] (M = group 4, 5 and 6 metal, E = B, Si, Ge, P, S)

Article abstract of DOI:10.1002/zaac.200390137

The focus of the current report lies on recent developments of synthetic methods applied to the synthesis of some high-valent complexes containing the nitrido functionality [N]^3- as a link between a group 4, 5 or 6 transition metal and a main group element E (E = B, Si, Ge, P, S). Emphasis is put on results, that have been obtained within the Schwerpunktprogramm Nitridobruecken funded by the Deutsche Forschungsgemeinschaft. The synthetic methods include condensation reactions of reactive chloro and oxo complexes (M = V, Nb, Ta, Cr, Mo, W) with silylamines, sulfonylamides, with N-silyl and N-lithio iminophosphoranes, furthermore methatesis reactions of oxo complexes with N-sulfonyl sulfinyl amides (M = V, Cr, Mo, W), the oxidative addition of element azides to d² metal centers (M = V, W), and finally transamination reactions of N-H iminophosphoranes with amido complexes (M = Ti, Sm).

Full text of DOI:10.1002/zaac.200390137

Recent Advances in the Synthesis of N-Heteroatom Substituted Imido
Complexes Containing a Nitrido Bridge [M؍N؊E]
(M ؍ Group 4, 5 and 6 Metal, E ؍ B, Si, Ge, P, S)
Klaus Weber, Klaus Korn, Andrea Schorm, Jennifer Kipke, Martin Lemke, Alexander Khvorost, Klaus Harms,
and Jörg Sundermeyer*
Marburg, Fachbereich Chemie der Philipps-Universität Marburg, Hans-Meerwein-Straße
Received December 18th, 2002.
Abstract. The focus of the current report lies on recent develop-
ments of synthetic methods applied to the synthesis of some high-
valent complexes containing the nitrido functionality [N]3Ϫ as a
link between a group 4, 5 or 6 transition metal and a main group
element E (E ϭ B, Si, Ge, P, S). Emphasis is put on results, that
have been obtained within the “Schwerpunktprogramm “Nitri-
dobrücken” funded by the Deutsche Forschungsgemeinschaft. The
synthetic methods include condensation reactions of reactive
chloro and oxo complexes (M ϭ V, Nb, Ta, Cr, Mo, W) with silyl-
amines, sulfonylamides, with N-silyl and N-lithio iminophos-
phoranes, furthermore methatesis reactions of oxo complexes with
N-sulfonyl sulfinyl amides (M ϭ V, Cr, Mo, W), the oxidative ad-
dition of element azides to d2 metal centers (M ϭ V, W), and finally
transamination reactions of N-H iminophosphoranes with amido
complexes (M ϭ Ti, Sm).
Keywords: Nitrido bridges; Imido complexes
Jüngste Fortschritte in der Synthese von N-Heteroatom-substituierten Imidokomplexen
mit Nitridobrücken [M؍N؊E] (M ؍ Metall der Gruppe 4, 5 und 6, E ؍ B, Si, Ge, P, S)
Inhaltsübersicht. Der Fokus des vorliegenden Berichts liegt auf der
Entwicklung synthetischer Methoden für die Synthese höhervalen-
ter Komplexe, die eine Nitridobaugruppe [N]3Ϫ als Brückenfunk-
tion zwischen einem Übergangsmetall der Gruppe 4, 5 oder 6 und
einem Hauptgruppenelement E (E ϭ B, Si, Ge, P, S) enthalten.
Besonders gewichtet werden Ergebnisse, die im Rahmen des durch
die Deutsche Forschungsgemeinschaft geförderten Schwerpunkt-
programms „Nitridobrücken“ erhalten wurden. Die synthetischen
Methoden umfassen Kondensationsreaktionen reaktiver Chloro-
und Oxokomplexe (M ϭ V, Nb, Ta, Cr, Mo, W) mit Silylaminen,
Sulfonylamiden, mit N-Silyl- und N-Lithio-Iminophosphoranen,
weiterhin Methatese Reaktionen von Oxokomplexen mit N-Sulfo-
nyl-sulfinylamiden (M ϭ V, Cr, Mo, W), die Oxidative Addition
von Elementaziden an d2 Metalkomplexen (M ϭ V, W), und
schließlich Transaminierungsreaktionen von N-H-funktionellen
Iminophosphoranen mit Amidokomplexen (M ϭ Ti, Sm).
Introduction
complexes. Prominent examples for organoimido catalysts
include vanadium [6] and chromium [7] complexes in olefin
polymerizations, osmium imido complexes in the asymmet-
ric syn-aminohydroxylation [8], molybdenum imido com-
plexes in catalytic olefin methatesis reactions [9] and
titanium imido complexes as the reactive intermediates in
the amination of alkynes [10]. The donor strength and cone
angle of the organoimido ligand can be controlled by the
steric and electronic properties of the organic substituent
within the backbone [MϭNϪR]. Variation of the organic
radical R allows the fine-tuning of the catalyst, e.g. the first
chiral, C2-symmetrical bis-imido complexes, isolobal with
ansa metallocenes have been described [11]. Much less
investigated are effects induced by the variation of a
heteroatom subtituent E within the heteroimido donor
function [MϭNϪE]. By considering the different bonding
interactions between nitrogen and carbon compared to ni-
trogen and any other main group element E, it becomes
clear, that heteroatom substitution will have a much greater
influence on the donor properties of the ligand [MϭNϪE]
than variations within the pattern of organic substituents R
at E. The two largest classes of N-heteroatom substituted
The chemistry of N-organoimido complexes has experi-
enced a considerable development during the last two dec-
ades [1]. The organoimido ligand [NR]2Ϫ (R ϭ alkyl, aryl)
coordinates to a metal by one σ and up to two π orbital
interactions, similar to the cyclopentadienyl ligand [2]. This
so called cyclopentadienyl-imido-ligand-analogy [3, 4, 5] is
the basis of the isolobal relationship of group 4 metallocene
fragments [Ti(η5-C5R5)2] and their corresponding imido
complex fragments of group 5 and 6, [V(NR)(η5-C5R5)] and
[Cr(NR)2]. With respect to the fundamental importance of
group 4 metallocenes as catalysts in olefin polymerizations
and other catalytic transformations, the cyclopentadienyl-
imido-ligand-analogy has become a strategic tool for the
development of new classes of catalytically active imido
* Prof. Dr. J. Sundermeyer
Fachbereich Chemie der Universität
D-35032 Marburg
Fax: ϩ49-(0)6421-2828917
E-mail: jsu@chemie.uni-marburg.de
744
 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 0044Ϫ2313/03/629/744Ϫ754 $ 20.00ϩ.50/0 Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754

Synthesis of N-Heteroatom Substituted Imido Complexes Containing a Nitrido Bridge [MϭNϪE]
imido complexes are those with electron withdrawing N-
phosphonio ϪPR3 substituents, so called phosphoranimi-
M(NSiMe3)(η1-C5H5)] (2a,b) are readily synthesized by re-
action of [M(NSiMe3)Cl3(py)2] with NaCp.
ϩ
nato complexes, and those with isoelectronically related N-
Complexes with three or more σ,2π donor ligands are
π bond saturated [22]; there are more ligand orbitals of π
symmetry available for π interaction than empty metal d
orbitals. Therefore electron density has to remain at ligand
centered non-bonding orbitals. As a consequence, both, the
solid state structure of 2b as well as dynamic nmr spectra
of solutions of 2a and 2b reveal two η5- and one η1-cyclo-
pentadienyl ligands. Our study proved, that a trimethylsilyl
imido ligand can push a cyclopentadienyl ligand out of π
coordination (Fig. 1). The σ,π bonding mode is also favored
by the sterically congested situation.
silyl ϪSiR3 substituents. Due to the positive formal charge
ϩ
of the strongly electron withdrawing ϪPR3 substituent,
phosphoraniminato ligands are mono-anionic, while the re-
lated N-silylimido ligands are dianionic. Both ligands are
less basic than organoimido ligands, therefore they may well
be suitable for the design of catalytically active metal
Lewis acids.
The mono-anionic ligand [NPR3]Ϫ is bonded via one σ
and up to two π bonds, it can replace a cyclopentadienyl
ligand [C5R5]Ϫ, both are considered to be isolobally related
[12]. Recently, phosphoraniminato complexes of nickel [13]
and titanium [14] have proven their ability to act as catalysts
in olefin polymerizations and those of samarium turned out
to be catalysts in the ring opening polymerization of lac-
tones [15Ϫ18]. Nearly all of the early work with phosphor-
animinato ligands focused on P-phenyl derivatives [19].
Only recently it turned out, that an even larger cone angle
seems to be a precondition for the activity in olefin poly-
merizations: [CpTi(NPPh3)Cl2] and [Ti(NPPh3)2Cl2] with
less sterically demanding phenyl substituents were among
the first d-electron poor complexes known with this ligand
system [20], but they are not catalytically active in olefin
polymerizations. Recently however, Stephan et al. demon-
strated, that the sterically more demanding tert-butyl de-
rivatives [CpTi(NPtBu3)Cl2] and [Ti(NPtBu3)2Cl2] are
highly active catalysts in MAO mediated α-olefin polymeri-
zations [14]. Alkyl substituents at phosphorus also proved
to generate the highly active nickel complexes in norbor-
nene polymerization reactions [13]. Finally it has been re-
ported, that silylimido complexes may become active cata-
lysts, when sterically demanding tert-butyl groups in [Cp*
Ta(NSitBu3)Cl2] are attached at the N-heteroatom [21].
These trends demonstrate, that there is a need for de-
veloping synthetic methods for the synthesis of N-heteroim-
ido complexes. The following report is not organized in
chapters describing chemistry of a certain metal or a certain
N-heteroimido ligand but in chapters describing a certain
synthetic strategy of introducing a sterically demanding
linkage [MϭNϪE] with an electron withdrawing substitu-
ent E to group 4Ϫ6 transition metals.
In order to directly compare the σ,π-donor capability of
[Me3Si-N]2Ϫ and [Me3C-N]2Ϫ ligands, two related com-
plexes [(η5-C5H5)2Ta(N-EMe3)(η1-C5H5)] (E ϭ Si 2b, C 2c)
have been prepared. Both were investigated by X-ray crys-
tallography and vt-NMR measurements. The stronger do-
nor character of the tert-butyl imido ligand is in accord
with a shorter Ta-NC bond 176.1(5) pm in 2c versus Ta-
NSi 179.6(7) pm in 2b [22]. The higher degree of RNǞM
π-donation and the greater sterical demand of the tert-
butylimido versus silylimido ligand is responsible for a
higher barrier of rotation ∆G϶(TC) about the Ta-Cipso bond
of the η1-C5H5 ligand in 2c compared to 2b. On the other
hand, the lower coalescence temperature for the σ,π-ex-
change of silylimido complex 2b suggest, that the stronger
donor [NCMe3]2Ϫ promotes the ring slippage into a η3- and
finally the η1-cyclopentadienyl bonding mode more readily
than the weaker donor [NSiMe3]2Ϫ. Our study is one of the
very few directly comparing the bonding capability of
related [N-EMe3]2Ϫ ligands (E ϭ C, Si) and adding experi-
mental proof, that tert-butylimido ligands are the better
donors than silylimido ligands, and that both are better do-
nors towards a d0 metal atom than a η5-C5H5 ligand.
Attempts to prepare halfsandwich complexes by treat-
ment of [M(NSiMe3)Cl3(py)2] with an equimolar amount of
NaCp or LiCp* failed. However, the synthesis of [(η5-
C5Me5)M(NSiMe3)Cl2] (3a,b) has been accomplished by re-
action of [(η5-C5Me5)MCl4] with LiN(SiMe3)2, while re-
lated hydridotrispyrazolylborato complexes [Tp*M-
(NSiMe3)Cl2] (4a, b; Tp*ϭ κ3-HB(3,5-Me2pz)3) have been
prepared from the pyridine adducts [22]. The result of the
crystal structure analysis on 4b is shown in Figure 2. For
both complex types the isoelectronic organoimido counter-
parts [(η5-C5Me5)M(NCMe3)Cl2] (M ϭ Nb, Ta) [24] and
[Tp*M(NCMe3)Cl2] (M ϭ Nb, Ta) [25] are known.
Results
Condensation reactions with bistrimethylsilyl amine
and derivatives
Sulfonylimido complexes via arylsulfonyl
sulfinylamide methatesis
Trimethylsilylimido niobium and tantalum complexes of
metallocene and halfsandwich type have been obtained via
condensation reactions as synthetic strategy [22]. The key
complexes [M(NSiMe3)Cl3(py)2] (M ϭ Nb, Ta) are readily
prepared from the metal pentachlorides and (Me3Si)2NH in
the presence of pyridine [23]. We found that bent metallo-
cenes [(η5-C5H5)2M(NSiMe3)Cl] (1a,b) and [(η5-C5H5)2-
Sulfonyl groups ϪSO2R belong to the strongest electron
withdrawing substituents known. Taking the higher acidity
of aryl sulfonylamides ArSO2NH2 compared to water into
consideration, the donor capability of an aryl sulfonylimido
ligand should be lower than that of an oxo ligand. The high
reactivity of [MϭNϪSO2Ar] groups is reflected by the pro-
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754
745

K. Weber, K. Korn, A. Schorm, J. Kipke, M. Lemke, A. Khvorost, K. Harms, J. Sundermeyer
Fig. 1 Molecular structure of [(η5-C5H5)2Ta(NSiMe3)(η1-C5H5)]
(2b) (CCDC 132475). Selected bond distances/pm and angles/°:
Ta-N 179.6(7), N-Si 170.1(8), Ta-N-Si 171.9(5).
Fig. 2 Molecular structure of [Tp*Ta(NSiMe3)Cl2] (4b) (CCDC
132477). Selected bond distances/pm and angles/°:
Ta-N(1) 176.7(6), Si-N(1) 174.2(6), Ta-N(1)-Si 164.8(4).
nounced sensitivity towards hydrolysis and the tendency to
undergo oxidative nitren transfer. We set out to develop a
new strategy for the synthesis of electron-poor and highly
reactive aryl sulfonylimido complexes containing the N-het-
eroatom link [MϭNϪS] between metals such as V, Cr, Mo
and W and a six-valent sulfur atom. One of the synthetic
strategies used in early work is the oxidative addition of N-
Cl bonds of N,N-dichloro aryl sulfonylamides to tungsten
hexacarbonyl [26, 27]. The coordination polymer 5 dissolves
in MeCN to yield structurally characterized 6.
Protolysis of [WCl6] with tosylamide offers direct access
to [W(NSO2Tol)2Cl2]n (7a) [28]. The by-product of amino-
lysis is HCl. In presence of donor ligands coordination
polymer 7a can be converted into molecular complexes, e.g.
[W(NTs)2Cl2(dme)], [W(NTs)2Cl2(PMe3)2] and [W(NTs)2-
Cl2(4,4Ј-Me2bipy)]. For a long time, we were not able to
synthesize the corresponding less stable molybdenum(VI)
746
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754

Synthesis of N-Heteroatom Substituted Imido Complexes Containing a Nitrido Bridge [MϭNϪE]
Fig. 3 Molecular structure of [W(NSO2Tol)Cl2(PMe3)3] (9) (CCDC
132478). Selected bond distances/pm) and angles/°:
Mo-N 147.9(6), N-S 163.3(6), Mo-N-S 177.6(4).
bis-phenylsulfonylimido complex by either oxidative N-Cl
addition or protolysis strategies Ϫ “MoCl6” is unknown,
until the reaction with sulfinylamines RNϭSϭO was stud-
ied. Originally, we developed the sulfinylamine methatesis
protocol, in order to introduce electron-withdrawing
haloarylimido ligands via a desoximidation at terminal oxo
complexes [29, 30]. This strategy worked well, when com-
mon strategies such as condensation reactions with silyl-
amines failed. We learned, that methatesis of [Mo(O)2Cl2]n
with sulfinyl sulfonylamides ArSO2-NSO offers a straight
access to highly reactive complexes of the type [Mo-
(NSO2Ar)2Cl2]n (8a؊c) [30].
The synthetic potential of 8a؊c lies in their ability to
transfer an electron-poor 6 VE nitren [NSO2Ar] to electron-
rich substrates. An attempt to synthesize a molybdenum
phosphine complex analogous tungsten congener
[W(NTs)2Cl2(PMe3)2] led to oxidative nitren transfer to the
phosphine. Me3PϭNTs was formed and the d2 molyb-
denum complex 9 was isolated and characterized by a crys-
tal structure analysis (Fig. 3) [28]. Interestingly, a similar
nitren abstraction is inhibited in the corresponding tungsten
phosphine complex while the bipy complex is reacting with
PMe3 under nitren abstraction to yield [W(NTs)Cl2-
(4,4ЈϪMe2bipy)(PMe3)2]. This observation is in accord with
a nitren transfer induced via direct attack of the phosphine
HOMO into the π* orbital of the MϭNS bond (LUMO)
[28].
In collaboration with the group of Hanack, another inter-
esting reactivity pattern of [W(NSO2Ph)2Cl2]n (6) has been
investigated. The nitrido chlorides [MoNCl3] and [WNCl3]
are good starting materials for the synthesis of novel
phthalocyanine metal nitrides [PcM(N)] (10a,b) [31]. Better
yields of 10b are obtained, when 6 is chosen as molecular
precursor for the central [WN] building block. This may
well be due to the known tendency of homolytic or hetero-
lytic cleavage of the NϪS bond within the backbone [Wϭ
N-SO2Ph] under the reaction conditions of 220 °C. The
reactivity of corresponding molybdenum tosylimido
complexes is currently investigated.
We finally investigated the methatesis of [VOCl3] and
[CrO2Cl2] with arylsulfonyl sulfinylamides ArSO2-NSO.
While [VOCl3] is transformed without reduction into the
purple crystalline complexes [V(NSO2Ar)Cl3] (11a,b) [30],
[CrO2Cl2] is reduced. The paramagnetic product has the
composition [{Cr(NTs)Cl2}2O] (13) [32] and can be envis-
aged as a dinuclear or oligonuclear d1Cr-(µ-O)-Cr d1 com-
plex, probably formed by a combination of the CrVI species
[Cr(NTs)(O)Cl2] with a coordinatively unsaturated CrIV
species [Cr(NTs)Cl2]. The latter may have formed via an
oxen abstraction from the CrVI oxo-imido intermediate by
either SO2 or TsNSO.
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754
747

K. Weber, K. Korn, A. Schorm, J. Kipke, M. Lemke, A. Khvorost, K. Harms, J. Sundermeyer
It is interesting to note, that [CrO2Cl2] is not reduced by
SO2 in CCl4 /25 °C: The methatesis of [CrO2Cl2] with
slightly more electron rich C6F5-NSO leads selectively to
SO2 and [Cr(NC6F5)2Cl2] which has been structurally
characterized [30]. If the methatesis with TsNSO is per-
formed in SO2Cl2 as solvent, the d1 species [Cr(NTs)Cl3]
(13) is formed. We believe, that the d2 intermediate
[Cr(NTs)Cl2] is quenched by chlorine radicals under these
reaction conditions. Both d1 chromium sulfonylimido com-
plexes 12 and 13 are precursors for active Phillips-type cata-
lysts in the MAO induced ethylene polymerisation [32, 33].
Fig. 4 Molecular structure of [Ta(NPtBu3)3Cl2] (15a) (CCDC
117899). Selected bond distances/pm and angles/°:
Ta-N1 190.6(3), Ta-N2 189.3(4), Ta-N3 190.1(4), Ta-N1-P1 170.4(2), Ta-N2-
P2 171.1(2), Ta-N3-P3 168.7(2).
Reactions of N-silyl and N-lithio iminophosphoranes
with oxo, chloro and imido complexes
Phosphoraniminato complexes are known for nearly all
transition metals [19], they are the largest group of N-
heteroimido compounds. As already discussed most com-
pounds containing the nitrido bridge [MϭNϭPR3] have
phenyl substituents at the phosphorus atom. The good
availability of the reactive ligand synthon Me3SiϪNϭPPh3,
its tendency to eliminate Me3SiCl with reactive metal
halides and to form highly crystalline complexes are rea-
sons, that chemistry of iminophosphorane ligands with
of 16a,b and 17a,b is accomplished by condensation of the
precursor complexes [V(NAr)Cl3] with Me3Si-NPR3 (R ϭ
tBu, NMe2).
t
other P-substituents such as R ϭ Bu, Et, Me, and NMe2
The crystal structure of 16a (Fig. 5) reveals a tetrahedral
coordination at vanadium. The bonding distance of the
monoanionic V-N-P ligand V-N(2) 169.5(3) pm is only
slightly longer than that of the dianionic V-N-C ligand V-
N(1) 167.1(3) pm. Having two sterically demanding nitro-
gen donors and two chloro ligands as good leaving groups
these imido vanadium complexes turned out to be active
catalysts in the ethylaluminium sesquichloride induced eth-
ene-propene co-polymerisation leading to the synthetic elas-
tomer EPM or terpolymer EPDM [37].
Tetrahedral molecules that contain two nitrido bridges
M-N-S and M-N-P have been obtained from tungstyl and
molybdyl dichloride via the sulfonyl sufinylamide metha-
tesis followed by salt methatesis with N-lithio iminophos-
phoranes [38].
The molecular structure of 19a is presented in Figure 6.
The average Mo-N bonding distances for [NSO2Tol]2Ϫ and
[NPtBu3]Ϫ ligands are the same within 3σ. In accord with
this observation, both heteroimido ligands should reveal a
similar donor capability.
A rather unexplored strategy for the synthesis of phos-
phoraniminato complexes are reactions of binary metal ox-
ides with N-silyl iminophosphoranes [39]. We used this ap-
proach to synthesize the first chromyl [CrO2]2ϩ complexes
containing the NPR3 ligand and to develop a new most
simple synthesis for chromium(VI) nitrido complexes [40].
have become popular only recently. In 1999 two reports by
Stephan et al. describing the catalytic performance of steri-
cally demanding complexes [CpTi(NPtBu3)Cl2] and
[Ti(NPtBu3)2Cl2] in olefin polymerisation reactions [14]
pointed out the need for steric protection of the catalytically
active species by bulky P-substituents. Both catalysts are
isolobal with metallocene dichlorides of group 4 [Cp2MCl2]
(M ϭ Ti, Zr) [12, 14]. In the same year we summarized the
results of our investigation of phosphoraniminato com-
plexes of group 5 metals containing the sterically de-
t
manding P-substituents R ϭ NMe2 and Bu [34].
14a,b and 15a,b are synthesized by reaction of LiNPtBu3
or LiNP(NMe2)3 with the parent metal pentachlorides. The
crystal structure analysis of 15a fills the missing link be-
tween the known structures [Nb(NPPh3)2Cl3] [35] and
[Ta(NPPh3)4]ϩ[TaCl6]Ϫ [36]. The strong π donors with the
high cone angle are in the equatorial positions of a trigonal
bipyramid (Fig. 4). Even when an excess of LiNPtBu3 is
reacted with 15a for sterical reasons a tetra-substitution is
not observed.
Catalysts of the type [Cp2TiCl2], [CpTi(NPR3)Cl2], and
[Ti(NPR3)2Cl2] are isolobal with novel complexes of the
type [V(NR)(NPR3)Cl2]. Therefore we set out to synthesize
representatives of this class of imido vanadium complexes
with one sterically demanding NPR3 ligand. The synthesis
748
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754

Synthesis of N-Heteroatom Substituted Imido Complexes Containing a Nitrido Bridge [MϭNϪE]
Fig. 5 Molecular structure of [V(N-2,4,6-Cl3C6H2)(NPtBu3)Cl3]
(16a) (CCDC 198833). Selected bond distances/pm and angles/°:
V-N1 167.1(3), V-N2 169.5(3), N2-P1 162.6(3), V-N1-C1 160.6(3), V-N2-P1
170.7(2).
Fig.
7 Molecular structure of [Cr(O)2(NPPh3)(OSiMe3)] 20b
(CCDC 198832). Selected bond distances/pm and angles/°:
Cr-N1 173.2(4), Cr-O1 159.4(3), Cr-O2 157.7(3), Cr-O3 177.5(2), N1-P1
160.6(3), Cr-N1-P1 133.6(2).
is observed: With elimination of R3PO chromium(VI) ni-
trido complexes 21a,b are formed in perfect yield. The re-
sults of crystal structure determinations of 20b and 21a are
shown in Figures 7 and 8.
The most striking result of these crystal structure deter-
minations is the small angle at both Cr-N-P bond axes:
133.6(3)° (20b) and 127.7(4)° (21a). As both molecules do
not show a pronounced sterically congested situation, we
assume that the bending is mainly due to electronic reasons.
The hard Lewis acids [CrO2]2ϩ and [CrN]3ϩ seem prefer π
bonding to the harder oxo and nitrido ligands rather than
to the softer NPR3 ligands. 21a has a crystallographically
imposed mirror plane defined by Cr, N1 and N2.
Fig.
6
Molecular structure of[Mo(NSO2Tol)2(NPtBu3)2] 19a
(CCDC 136826). Selected bond distances/pm and angles/°:
Mo-N1 181.6(2), Mo-N2 182.9(2), Mo-N3 182.0(2), Mo-N4 180.4(2), Mo-
N1-P1 167.75(15), Mo-N2-P2 159.68(14), Mo-N3-S1 154.83(15), Mo-N4-
S2 156.96(16).
Nitrido chromium complexes are rare [41] and there is a
luck of straight forward synthetic methods for their syn-
thesis. We discovered that polymeric CrO3 is cut down to
covalent hexane soluble chromyl silylesters 20a؊c by reac-
tion with one equivalent of the ligand precursors Me3Si-
NPR3.
Oxidative addition of nitrenes generated from element
azides to d2 metal complexes
t
For sterically demanding substituents R ϭ Bu, NMe2
and Ph the reaction stops at this point, however for R ϭ
Me or nBu a subsequent methatesis of the intermediate
chromyl complex with a second equivalent of Me3SiNPR3
The oxidative addition of nitrenes generated from element
azides RnE-N3 to low valent metal complexes is one of the
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754
749

K. Weber, K. Korn, A. Schorm, J. Kipke, M. Lemke, A. Khvorost, K. Harms, J. Sundermeyer
Fig.
8
Molecular structure of [Cr(N)(NPMe3)(OSiMe3)2]
21a(CCDC 198831). Selected bond distances/pm and angles/°:
Cr-N1 173.9(6), Cr-N2 155.1(6), Cr-O1/O1a 179.3(4), N2-P1 161.7(6), Cr-
N1-P1 127.7(4).
Fig. 9 Molecular structure of [W(NMes)2(NBMes2)PMe3)2] 22
(CCDC 198829). Selected bond distances/pm and angles/°:
W-N1 184.6(6), W-N2 183.5(6), W-N3 182.8(6), N1-B 139.0(10), W-N1-B
177.7(5), W-N2-C20 171.7(5), W-N3-C10 176.4(5).
most powerful strategies to synthesize nitrido bridged com-
pounds containing the [M-N-ERn] link. Using this strategy
Wilkinson et al. prepared the first borylimido complex
[W(NBMes2)2Cl2(PMe3)2] [42] whereas Doherty et al. were
able to prepare five-coordinate bis-silylimido complex
[W(NSiMe3)2Cl2(PPh2Me)] [43]. We used the azide route to
synthesize five-coordinate tungsten trisimido complexes
containing silyl- and borylimido ligands and to get access
to a new class of N-heteroimido mesityl vanadium com-
plexes. As highly reducing d2 metal complexes for the oxi-
dative addition of hetero-nitrenes we chose [W(NMes)2-
(PMe3)3] [44] and [VMes3(THF)] [45].
Trisimido complexes tungsten complexes are very rare,
they have been only characterized in a tetrahedral coordi-
nation so far [46]. Addition of the boryl azide Mes2BN3 to
[W(NMes)2(PMe3)3] leads to the borylimido complex 22. In
a similar way, metal induced reduction of triorganosilyl
azides leads to the silylimido complexes 23a,b.
Both compounds 22 and 23 have been structurally
characterized, the results are shown in Figures 9 and 10
[47]. Both complexes contain the three most powerful π-
donor ligands in the equatorial plane of a trigonal bipyra-
mid while the phosphine ligands occupy the axial positions.
The electron withdrawing character of the N-hetero-sub-
stituents ϪSiR3 and ϪBR2 is responsible for the weaker
donor character and longer metal nitrogen bonds of these
heteroimido ligands compared with organoimido counter-
parts. The boryl group in particular is a strong competitor
t
Fig. 10 Molecular structure of [W(NMes)2(NSiMe2 Bu)(PMe3)2]
23b (CCDC 198828). Selected bond distances/pm and angles/°:
W-N1 180.6(4), W-N2 182.7(3), W-N3 181.7(4), W-N1-Si 169.3(2), W-N2-
C20 176.9(3), W-N3-C10 171.7(4).
for the electron density at the nitrogen ligand: This is re-
flected by a W-N distance 4 pm longer than in silylimido
counterpart 23b and in a short B-N bond with a bond order
of two, more than 4 pm shorter than in Wilkinson’ borylim-
ido reference complex [42].
Trimesityl vanadium [45] has been the other d2 complex
chosen for oxidative nitren addition reactions. The follow-
ing scheme presents the scope of this synthetic strategy: V-
N-B, V-N-S, V-N-Si links and the first structurally charac-
terized example 25 of a V-N-Ge nitrido bridge have been
obtained in good yields by reduction of the corresponding
element azides with the vanadium(III) complex [47].
As expected, the organovanadium(V) imides are diamag-
netic, orange or yellow and sensitive to light, as they are
prone to homolytic V-C bond cleavage. The crystal struc-
ture of the germylimido complex 25 is shown in Figure 11.
Vanadium is tetrahedrally coordinated by a linear V-N-
GePh3 group and three mesityl groups, which are twisted
into a propeller-like conformation. A similar reactivity pat-
750
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754

Synthesis of N-Heteroatom Substituted Imido Complexes Containing a Nitrido Bridge [MϭNϪE]
are sometimes hampered by low yields due to impurities
difficult to remove. Transamination reactions of metal am-
ides with NH-functional iminophosphoranes have not been
investigated in detail so far. We found that [Ti(NMe2)4] is
in close to quantitative yields transformed into mono- and
bis-substituted complexes 28 and 29.
In a similar way, [CpTi(NMe2)3] is transformed to 31.
The reaction with BCl3 selectively replaces dimethylamino
groups of these complexes by chloro ligands [49]. Our trans-
amination strategy allows the stepwise introduction of other
anionic N donors by further replacement of the amine
HNMe2: When 28 is treated with an equimolar amount of
HN(C6F5)2 another dimethylamido group is selectively re-
placed. The driving force of such transamination reactions
is the demand of the electropositive metal to preferentially
bind a nitrogen donor of higher group electronegativity.
The molecular structure of 30 (Fig. 12) reveals three anionic
nitrogen donors of different donor strength bonded to the
same metal. With respect to the Ti-N bond distances the
anion [NPtBu3]Ϫ is the strongest donor, followed by
[NMe2]Ϫ and [N(C6F5)2]Ϫ as the weakest donor.
Triphenylphosphoraniminato complexes of the f-block
elements have been investigated by Dehnicke et al. [50], they
reveal catalytic activity in ring-opening of lactones [15, 16,
17, 18]. The catalysts are commonly synthesized by reaction
of the metal halides with Na- or KNPPh3 [50, 51]. This
strategy cannot be used for the synthesis of up to date un-
known f-metal complexes with the sterically more de-
manding ligand [NPtBu3]Ϫ as it is known that its sodium
or potassium salts are unstable with respect to the elimin-
ation of isobutene [52]. For this reason we extended our
transamination studies towards the homoleptic bistrime-
Fig. 11 Molecular structure of [V(NGePh3)Mes3] 25 (CCDC
198830). Selected bond distances/pm and angles/°:
V-N 161.6(3), N-Ge 186.8(3), V-N-Ge 177.8(2).
tern but limited to alkyl and silyl azides has been observed
by Gambarotta et al. using [V(NPh2)3(THF)] as starting
material [48].
Transamination reactions of iminophosphoranes with
metal amides
The synthesis of Stephan-type olefin polymerization cata-
lysts [CpTi(NPtBu3)Cl2] and [Ti(NPtBu3)2Cl2] is ac-
complished by reaction of Me3Si-NPtBu3 with the corre-
sponding chloro complexes [14]. We were interested to de-
velop a new synthesis for this catalyst class, as in our hands
it turned out, that large scale preparations of the catalysts
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754
751

K. Weber, K. Korn, A. Schorm, J. Kipke, M. Lemke, A. Khvorost, K. Harms, J. Sundermeyer
Fig. 12 Molecular structure of [Ti(NPtBu3)(NMe2)2{N(C6F5)2}] 30
(CCDC 198834). Selected bond distances/pm and angles/°:
Ti-N1 188.70(5), Ti-N2 190.61(4), Ti-N3 205.60(4), Ti-N4 178.86(4), N4-P
159.63(4), Ti-N4-P 173.06(10).
Fig. 13 Molecular structure of [Sm(NPtBu3)3(NHPtBu3)] 32
(CCDC 198835). Selected bond distances/pm and angles/°:
Sm-N1 227.0(8), Sm-N2 223.2(8), Sm-N3 227.6(10), Sm-N1-P1 169.4(6), Sm-
N2-P2 170.8(7), Sm-N3-P3 162.6(7).
ting the structure and reactivity of intermediates and pro-
gression products of this transamination cascade [49].
Outlook
The present report offers a short review on selected syn-
thetic strategies suitable to link a transition metal M via a
formal nitride N3Ϫ ion to a main group element E (B, Si,
Ge, P, S). Variation of the substituent E within the linkage
[M-N-E] allows to control the basicity, donor strength and
steric demand of these anionic nitrogen donors. The chem-
isty of complexes with heteroimido spectator ligands is cur-
rently investigated, as some of the complexes with 3d metals
Ti, V, and Cr as well as those of rare earth metals Sm and
Y turned out to be catalysts in polymerisations of olefins
and oxacycles.
Acknowledgments. The authors gratefully acknowledge generous fi-
nancial support by the Deutsche Forschungsgemeinschaft, Schwer-
punktprogramm “Nitridobrücken“.
thylsilyl amides of the rare earth elements. A typical reactiv-
ity pattern is shown below for samarium, a similar is ob-
served for yttrium.
The reaction of [Sm{N(SiMe3)2}3] with HNPtBu3 in a 1:2
molar ratio leads to 31 whereas the reaction in a 1:4 molar
ratio gives an adduct 32 of HNPtBu3 to the Lewis acid
[Sm(NPtBu3)3]. Up to date we were not able to isolate the
free Lewis acid [Sm(NPtBu3)3] without the neutral ligand
HNPtBu3. The molecular structure of 32 is shown in Figure
13. The complex crystallizes in the space group I a b m with
Z ϭ 8. The molecule reveals a crystallographically imposed
mirror plane defined by Sm, N1 and N3 generating N2 on
symmetry equivalent positions.
The position of the hydrogen atom could not be located.
We assume that it resides preferentially at N(3) having the
longest of all Sm-N bonds and the smallest angle Sm(1)-
N(3)-P(3) However we cannot exclude that the hydrogen
atom is statistically distributed over N(1) and N(3) or over
all four nitrogen atoms, as within 3 σ Sm-N(1) has the same
bonding distance as Sm-N(3). Currently we are investiga-
References
[1] Reviews: (a) W. A. Nugent, B. L. Haymore, Coord. Chem. Rev.
1980, 31, 123. (b) K. Dehnicke, J. Strähle, Angew. Chem. 1981,
93, 451; Angew. Chem., Int. Ed. Engl., 1981, 20, 413. (c) D. E.
Wigley, Progr. Inorg. Chem. 1994, 42, 239.
[2] (a) V. C. Gibson, J. Chem. Soc., Dalton Trans. 1994, 1607.
(b) The capability of the cyclopentadienyl ligand for π back-
bonding and its softer character is not taken into consider-
ation.
[3] D. S. Williams, M. H. Schofield, J. Anhaus, R. R. Schrock, J.
Am. Chem. Soc. 1990, 112, 6728.
[4] Selected references: (a) D. N. Williams, J. P. Mitchell, A. D.
Poole, U. Siemeling, W. Clegg, D. C. R. Hockless, P. A. O’Neil,
V. C. Gibson, J. Chem. Soc., Dalton Trans. 1992, 739; (b) D.
S. Glueck, J. C. Green, R. I. Michelman, I. N. Wright, Or-
ganometallics 1992, 11, 4221. (c) A. D. Poole, V. C. Gibson,
W. Clegg, J. Chem. Soc., Chem. Commun. 1992, 237. (d) P. W.
752
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754

Synthesis of N-Heteroatom Substituted Imido Complexes Containing a Nitrido Bridge [MϭNϪE]
Dyer, V. C. Gibson, J. A. K. Howard, B. Whittle, C. Wilson,
J. Chem. Soc., Chem. Commun. 1992, 1666. (e) J. K. Cockcroft,
V. C. Gibson, J. A. K. Howard, A. D. Poole, U. Siemeling, C.
Wilson, J. Chem. Soc., Chem. Commun. 1992, 1668.
[17] S. Agarwal, C. Mast, K. Dehnicke, A. Greiner, Macromol. Ra-
pid Commun. 2000, 21, 195.
[18] T. Gröb, G. Seybert, W. Massa, F. Weller, R. Panislawmi, A.
Greiner, K. Dehnicke, Angew. Chem. 2000, 112, 4542; Angew.
Chem., Int. Ed. Engl. 2000, 39, 4373.
[5] (a) J. Sundermeyer, D. Runge, Angew. Chem. 1994, 106, 1328;
Angew. Chem., Int. Ed. Engl. 1994, 33, 1255. (b) S. Schmidt,
J. Sundermeyer, F. Möller, J. Organomet. Chem. 1994, 475,
157. (c) U. Radius, J. Sundermeyer, H. Pritzkow, Chem. Ber.
1994, 127, 1827. (d) U. Radius, A. Schorm, D. Kairies, S.
Schmidt, F. Möller, H. Pritzkow, J. Sundermeyer, J. Or-
ganomet. Chem. 2002, 655, 96.
[6] (a) M. C. W. Chan, K. C. Chew, C. I. Dalby, V. C. Gibson, A.
Kohlmann, I. R. Little, W. Reed, Chem. Commun. 1998, 16,
1673. (b) P. T. Witte, A. Meetsma, B. Hessen, P. H. M. Bud-
zelaar, J. Am. Chem. Soc. 1997, 119, 10561. (c) C. Lorber, B.
Donnadieu, R. Choukroun, Dalton 2000, 24, 4497. (d) V. J.
Murphy, H. Turner, Organometallics 1997, 16, 2495. (e) K. No-
mura, A. Sagara, Y. Imanishi, Chem. Lett. 2001, 36. (f) V. J.
Murphy, H. Turner, WO 98/34964 Exxon Chemicals, 1998. (g)
R. Krishnamurti, S. Nagy, J. Tyrell, B. Etherton, WO 00/
07725, Equistar Chemicals, 2000. (h) J. Sundermeyer, J. Kipke,
X. Li, M. Arndt-Rosenau, M. Hoch, patent pending, Bayer
AG.
[7] M. P. Coles, V. C. Gibson, Polym. Bull. 1994, 33, 529. (b) M.
P. Coles, C. I. Dalby, V. C. Gibson, W. Clegg, M. R. J. Else-
good, J. Chem. Soc., Chem. Commun. 1995, 1709. (c) M. P.
Coles, C. I. Dalby, V. C. Gibson, I. R. Little, E. L. Marshall,
M. H. Ribeiro da Costa, S. Mastroianni, J. Organomet. Chem.
1999, 591, 78. (d) V. C. Gibson, Eur. Pat. Appl. EP 641804 A2,
BP Chemicals Ltd., UK, 1995. (e) U. Siemeling, L. Kölling, A.
Stammler, H.-G. Stammler, E. Kaminski, G. Fink, J. Chem.
Soc., Chem. Commun. 2000, 1177. (f) M. Schopf, J. Sun-
dermeyer, J. Kipke, K. A. Rufanov, W. Heitz, U. Peucker, DE
19935592 A1, Elenac GmbH, 2001. (g) V. R. Jensen, K. J.
[19] (a) Reviews covering transition metal derivatives: K. Dehnicke,
J. Strähle, Polyhedron 1989, 8, 707. K. Dehnicke, M. Krieger,
W. Massa, Coord. Chem. Rev. 1999, 182, 19. (b) Review cover-
ing main group derivatives: K. Dehnicke, F. Weller, Coord.
Chem. Rev. 1997, 158, 103. (c) Review covering cyclic deriva-
tives: M. Witt, H. W. Roesky, Chem. Rev. 1994, 94, 1163.
[20] (a) J. R. Dilworth, H. J. Liefde Meijer, J. H. Teuben, J. Or-
ganomet. Chem. 1978, 159, 47. (b) I. A. Latham, G. J. Leigh,
G. Huttner, I. Jibril, J. Chem. Soc., Dalton Trans. 1986, 377.
(c) R. Choukroun, D. Gervais, J. R. Dilworth, Transition Met.
Chem. 1979, 4, 249.
[21] D. M. Antonelli, A. Leins, J. M. Stryker, Organometallics
1997, 16, 2500.
[22] A. Schorm, J. Sundermeyer, Eur. J. Inorg. Chem. 2001, 2947.
[23] C. M. Jones, M. E. Lerchen, C. J. Church, B. M. Schomber,
N. M. Doherty, Inorg. Chem. 1990, 29, 1679.
[24] S. Schmidt, J. Sundermeyer, J. Organomet. Chem. 1994, 472,
127.
[25] J. Sundermeyer, J. Putterlik, M. Foth, J. S. Field, N. Ramesar,
Chem. Ber. 1994, 127, 1201.
[26] H. W. Roesky, J. Sundermeyer, J. Schimkowiak, P. G. Jones,
M. Noltemeyer, T. Schroeder, G. M. Sheldrick, Z. Naturforsch.
1985, 40b, 736.
[27] W.-H. Leung, M.-C. Wu, J. L. C. Chim, W.-T. Wong, Poly-
hedron 1998, 17, 457.
[28] K. Korn, A. Schorm, J. Sundermeyer, Z. Anorg. Allg. Chem.
1999, 625, 2125.
[29] K. A. Rufanov, D. N. Zarubin, N. A. Ustynyuk, D. N. Goure-
vitch, J. Sundermeyer, A. V. Churakov, J. A. K. Howard, Poly-
hedron 2001, 20, 379.
[30] K. A. Rufanov, J. Kipke, D. N. Gurevitch, D. N. Zarubin, N.
A. Ustynyuk, J. Sundermeyer, Inorg. Chem. in print.
[31] K. Frick, S. Verma, J. Sundermeyer, M. Hanack, Eur. J. Inorg.
Chem. 2000, 1025.
[32] J. Sundermeyer, J. Kipke, M. Schopf, U. Peuker, W. Heitz, WO
01/09148 A 1, Elenac GmbH, 2001.
[33] J. Kipke, Dissertation, Univ. Marburg 2001.
[34] K. Weber, K. Korn, M. Schulz, J. Sundermeyer, Z. Anorg. Allg.
Chem. 1999, 625, 1315.
[35] F. Weller, D. Nußhär, K. Dehnicke, Z. Anorg. Allg. Chem.
1992, 615, 7.
[36] D. Nußhär, F. Weller, A. Neuhaus, G. Frenking, K. Dehnicke,
Z. Anorg. Allg. Chem. 1992, 615, 86.
[37] J. Sundermeyer, J. Kipke, M. Lemke, M. Arndt-Rosenau, M.
Hoch, patent pending, DE-101 40 202.3, Bayer AG.
[38] K. Korn, A. Schorm, K. Weber, W. Massa, J. Sundermeyer,
Inorg. Chem. submitted.
´
Borve, Organometallics 2001, 20, 616.
[8] G. Li, H.-T. Chang, K. B. Sharpless, Angew. Chem. 1996, 108,
449; Angew. Chem., Int. Ed. Engl. 1996, 35, 451.
[9] (a) C. J. Schaverin, J. C. Dewan, R. R. Schrock, J. Am. Chem.
Soc. 1986, 108, 2771. (b) R. R. Schrock, R. T. DePue, J. Feld-
man, K. B. Yap, D. C. Yang, W. M. Davis, L. Park, M. Di-
Mare, M. Schofield, J. Anhaus, E. Walborsky, E. Evitt, C.
Krüger, P. Betz, Organometallics 1990, 9, 2262. (c) H. H. Fox,
R. R. Schrock, R. O’Dell, Organometallics 1994, 13, 635. (d)
R. R. Schrock, S. Luo, J. C. Lee Jr., N. C. Zanetti, W. M.
Davis, J. Am. Chem. Soc. 1996, 118, 3883.
[10] M. Nobis, B. Drießen-Hölscher, Angew. Chem. 2001, 113,
4105; Angew. Chem., Int. Ed. Engl. 2001, 40, 3983.
[11] E. A. Kretzschmar, J. Kipke, J. Sundermeyer, J. Chem. Soc.,
Chem. Commun. 1999, 2381.
[12] A. Diefenbach, F. M. Bickelhaupt, Z. Anorg. Allg. Chem.
1999, 625, 892.
[13] C. Mast, M. Krieger, K. Dehnicke, A. Greiner, Macromol. Ra-
pid Commun. 1999, 20, 232.
[39] (a) H. W. Roesky, D. Hesse, M. Rietzel, M. Noltemeyer, Z.
Naturforsch. 1990, 45b, 72. (b) S. Schlecht, G. Geiseler, K.
Harms, M. Bickelhaupt, M. Diedenhofen, K. Dehnicke, Z.
Anorg. Allg. Chem. 1999, 625, 887. (c) R. Hasselbring, H. W.
Roesky, M. Noltemeyer, Angew. Chem. 1992, 104, 613; Angew.
Chem., Int. Ed. Engl. 1996, 31, 601.
´
[14] (a) D. W. Stephan, F. Guerin, R. E. v. Spence, L. Koch, X.
Gao, S. J. Brown, J. W. Swabey, Q. Wang, W. Xu, P. Zoricak,
D. G. Harrison, Organometallics 1999, 17, 2046. (b) D. W. Ste-
´
phan, J. C. Stewart, F. Guerin, R. E. v. Spence, W. Xu, D. G.
Harrison, Organometallics 1999, 17, 1116.
[15] S. Agarwal, C. Mast, S. Anfang, M. Karl, K. Dehnicke, A.
Greiner, Polym. Prepr. (Polym. Division) 1998, 39(1), 414.
[16] P. Ravy, T. Gröb, K. Dehnicke, A. Greiner, Macromolecules
2001, 34, 8649.
[40] M. Lemke, Dissertation, Univ. Marburg 2002.
[41] (a) A. L. Odom, C. C. Cummins, J. Am. Chem. Soc. 1995, 117,
6613. (b) A. L. Odom, C. C. Cummins, Polyhedron 1998, 17,
675. (c) C. C. Cummins, Angew. Chem. 1998, 110, 983; Angew.
Chem., Int. Ed. Engl. 1998, 37, 945. (d) H. T. Chiu, Y.-P. Chen,
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754
753

K. Weber, K. Korn, A. Schorm, J. Kipke, M. Lemke, A. Khvorost, K. Harms, J. Sundermeyer
S.-H. Chuang, J. S. Jen, G.-H. Lee, S.-M. Peng, J. Chem. Soc.,
[47] K. Korn, Dissertation, Univ. Marburg 1999.
Chem. Commun. 1996, 139.
[48] J.-I. Song, S. Gambarotta, Chem. Eur. J. 1996, 2, 1258.
[49] A. Khvorost, Dissertation, Univ. Marburg 2003.
[50] (a) S. Anfang, K. Harms, F. Weller, O. Borgmeier, H. Lüken,
H. Schilder, K. Dehnicke, Z. Anorg. Allg. Chem. 1998, 624,
159. (b) S. Anfang, T. Gröb, K. Harms, G. Seybert, W. Massa,
A. Greiner, K. Dehnicke, Z. Anorg. Allg. Chem. 1999, 625,
1853. (c) S. Anfang, G. Seybert, K. Harms, G. Geiseler, W.
Massa, K. Dehnicke, Z. Anorg. Allg. Chem. 1998, 624, 1187.
[51] T. Gröb, Dissertation, Univ. Marburg 2000.
[42] A. A. Danopoulos, C. Redshaw, A. Vaniche, G. Wilkinson, B.
Hussain-Bates, M. B. Hursthouse, Polyhedron 1993, 12, 1061.
[43] J. D. Lichtenhan, S. C. Critchlow, N. M. Doherty, Inorg.
Chem. 1990, 29, 439.
[44] U. Radius, J. Sundermeyer, H. Pritzkow, Chem. Ber. 1994,
127, 1827.
[45] W. Seidel, G. Kreisel, Z. Anorg. Allg. Chem. 1977, 435, 146.
[46] Y. W. Chao, P. M. Rodgers, D. E. Wigley, S. J. Alexander, A.
L. Rheingold, J. Am. Chem. Soc. 1991, 113, 6326.
[52] B. Ross, K.-P. Reetz, Chem. Ber. 1979, 112, 1756.
754
Z. Anorg. Allg. Chem. 2003, 629, 744Ϫ754