Transition metal complexes containing fnnctionalized organoimido and phosphaneiminato ligands

Article abstract of DOI:10.1002/zaac.200390140

Two different types of modified imido and phosphaneiminato ligands are investigated, namely chelate ligands and redox-functionalised ligands. The first examples of di(organoimido)chromium as well as di(phosphaneiminato)titanium and niobium chelates are described. Furthermore, the first complexes containing redox-functionalised organoimido ligands are presented, together with the first structurally characterised redox-functionalised phosphaneiminato complex. Compounds of the type [(RN)₂M-(CH₂Ph)₂] (M = Cr, Mo) are used as catalysts for the (co-)polymerisation of the polar olefins methyl methacrylate, acrylonitrile and vinyl acetate. A range of X-ray crystal structure determinations provide clear evidence for the quantum-chemical result that, similar to organoimido complexes, the potential energy well for the angle at the nitrogen atom is very shallow for phosphaneiminato complexes.

Full text of DOI:10.1002/zaac.200390140

Transition Metal Complexes Containing Functionalized Organoimido and
Phosphaneiminato Ligands
U. Siemeling*,a, L. Köllinga, O. Kuhnertb, B. Neumannb, A. Stammlerb, H.-G. Stammlerb, G. Finkc,
E. Kaminskic, A. Kieferc, and R. R. Schrockd
a Kassel, Fachbereich Physik der Universität
b Bielefeld, Fakultät für Chemie der Universität
c
Mülheim an der Ruhr, Max-Planck-Institut für Kohlenforschung
d Cambridge, MA/USA, Department of Chemistry, Massachusetts Institute of Technology
Received October 21st, 2002.
Abstract. Two different types of modified imido and phosphaneim-
inato ligands are investigated, namely chelate ligands and redox-
functionalised ligands. The first examples of di(organoimido)chro-
mium as well as di(phosphaneiminato)titanium and niobium che-
lates are described. Furthermore, the first complexes containing
redox-functionalised organoimido ligands are presented, together
with the first structurally characterised redox-functionalised phos-
phaneiminato complex. Compounds of the type [(RN)2M-
(CH2Ph)2] (M ϭ Cr, Mo) are used as catalysts for the (co-)poly-
merisation of the polar olefins methyl methacrylate, acrylonitrile
and vinyl acetate. A range of X-ray crystal structure determinations
provide clear evidence for the quantum-chemical result that, similar
to organoimido complexes, the potential energy well for the angle
at the nitrogen atom is very shallow for phosphaneiminato com-
plexes.
Keywords: Titanium; Niobium; Chromium; Molybdenum; Iron;
Imido complexes; Phosphaneiminato complexes; Chelate com-
plexes; Ferrocenes
Übergangsmetall-Komplexe mit funktionalisierten Organoimido- und
Phosphaniminato-Liganden
Inhaltsübersicht. Zwei unterschiedliche Typen modifizierter Imido-
und Phosphaniminato-Liganden werden untersucht, nämlich Che-
latliganden und redoxfunktionalisierte Liganden. Die ersten Bei-
spiele für Di(organoimido)chrom- sowie Di(phosphaniminato)-
titan- und -niob-Chelate werden beschrieben. Weiterhin werden die
ersten Komplexe mit redoxfunktionaliserten Organoimido-Ligan-
den sowie der erste strukturell charakterisierte redoxfunktionali-
sierte Phosphaniminato-Komplex vorgestellt. Verbindungen des
Typs [(RN)2M(CH2Ph)2] (M ϭ Cr, Mo) werden als Katalysatoren
für die (Co-)Polymerisation der polaren Olefine Methylmethacry-
lat, Acrylnitril und Vinylacetat benutzt. Eine Reihe von Röntgen-
strukturanalysen liefert klare experimentelle Belege für den quan-
tenchemischen Befund, daß, ähnlich wie bei Imido-Komplexen, die
Potentialenergiemulde für den Winkel am Stickstoffatom in Phos-
phaniminato-Komplexen sehr flach ist.
1 Introduction
opment has spurred our investigations concerning the de-
sign of functionalised organoimido and phosphaneiminato
ligands, which, in comparison to the related functionalised
cyclopentadienyl ligands [6], is only at its beginning. We
have studied two classes of functionalised ligands in this
context, namely (i) chelate ligands (i. e. ligands func-
tionalised with themselves) and (ii) redox-functionalised li-
gands.
Organoimido [1] and phosphaneiminato ligands [2] may be
viewed as anionic 6-electron donors of the 1σ2π type, simi-
lar to cyclopentadienyl ligands. This electronic analogy has
repercussions on the chemistry of complexes containing
such ligands and will provide a leitmotif of this paper.
The chemistry of organoimido [3] and phosphaneiminato
complexes [4] has clearly attained a state of maturity now,
and the relevance of such compounds for technologically
important areas of applied chemistry, such as, for example,
olefin polymerisation [5], is beginning to emerge. This devel-
2 Results
2.1 Chelate Complexes
Chelate ligands with two identical metal binding regimes
connected by a suitable linker represent the simplest type
of functionalised ligands. The chelate effect [7] generally
leads to greater complex stability. Furthermore, chelates
with a rigid ligand architecture may have special stereo-
chemical properties absent in their unchelated analogues. Fi-
* Prof. Dr. U. Siemeling
Universität Kassel
Fachbereich 18 (Physik)
D-34109 Kassel
Fax: ϩ49-(0)561-8044777
E-mail: siemeling@uni-kassel.de
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792  2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 0044Ϫ2313/03/629/781Ϫ792 $ 20.00ϩ.50/0
781

U. Siemeling et al.
nally, the reactivity of chelates may be influenced by the
“bite angle” of the chelate ligand. In the chemistry of
di(cyclopentadienyl) chelates, these different aspects, and
their interplay, have already been put to good use, an excel-
lent example being the stereospecific polymerisation of α-
olefins by ansa-zirconocene derivatives [8].
2.1.1 Di(organoimido) chelates
There is a well-established isolobal relationship between the
metal-ligand fragments [(C5R5)2M] (M ϭ Group 4 metal)
and [(RN)2MЈ] (MЈ ϭ Group 6 metal) [9], and this has been
used to rationalise the finding that, similar to titanocene
and zirconocene dichloride, complexes of the type
[(RN)2CrCl2] and [(RN)2MoCl2] polymerise ethylene in the
presence of Et2AlCl, the chromium system being more ac-
tive than the molybdenum one [10]. [(tBuN)2Cr(CH2Ph)2]
has been reported to be an active ethylene polymerisation
catalyst in the presence of a suitable activator. The cationic
η2-benzyl species [(tBuN)2Cr(CH2Ph)]ϩ has been put for-
ward as the catalytically active species [10a, 11]. In this con-
text [(tBuN)2Cr(CH2Ph)]ϩ corresponds to cations of the
type [(C5R5)2Zr(CH2Ph)]ϩ, which have been demonstrated
to exhibit excellent polymerisation properties [12]. A recent
quantum-chemical study, however, has advocated a species
containing a reduced chromium atom as the active cata-
lyst [13].
Our efforts to produce isolobal analogues of ansa-titano-
cene and ansa-zirconocene derivatives have resulted in the
synthesis of the first di(organoimido)chromium chelates as
well as closely related Mo compounds (Scheme 1).
The
(OSiMe3)2] (2), [(NCMe2CH2CH2CMe2N)Cr(CH2Ph)2] (4),
[(NCMe2CH2CH2CMe2N)MoCl2(DME)] (5) and
structures
of
[(NCMe2CH2CH2CMe2N)Cr-
[(NCMe2CH2CH2CMe2N)Mo(CH2Ph)2] (6) have been de-
termined by X-ray crystallography and are shown in Fig-
ures 1 Ϫ 4. The coordination of the metal atom of 5 is best
described as pseudo-octahedral and closely resembles that
of a TADDAMIN-derived molybdenum chelate [(TADDA-
MINat)MoCl2(DME)], which also contains a seven-mem-
bered molybdacycle [14]. The coordination of the metal
atom of the other complexes is pseudo-tetrahedral. A com-
mon feature of all these compounds is the remarkably small
NϪMϪN bite angle (ca. 100°) as well as the strong devi-
ation of the MϪNϪC angle (< 150°) from linearity. These
values are in contrast to those of open-chain analogues
such as [(tBuN)2Cr(CH2Ph)2] [NϪCrϪN 116.09(8)°,
CrϪNϪC 166.07(13) and 160.60(13)°] [11], [(tBuN)2-
Scheme 1 Synthesis of di(organoimido) chelates of chromium and
molybdenum. Reagents and conditions: (i) nBuLi (2 equiv.),
Me3SiCl (2 equiv.), diethyl ether; (ii) CrO2Cl2 (0.5 equiv.), hexane;
(iii) BCl3 (2 equiv.), dichloromethane; (iv) PhMgCl (2 equiv.), Et2O;
(v) Na2MoO4 (1 equiv.), NEt3 (4 equiv.), Me3SiCl (8 equiv.), DME;
(vi) PhMgCl (2 equiv.), Et2O.
0.5 mmol/L in toluene or methylcyclohexane with 100
equivalents of iBu3Al in conjunction with 1 equivalent of
[CPh3][B(C6F5)4] at 30 °C under 4 bar of ethylene. This
catalyst formulation produces linear polyethylene. After a
polymerisation time of 1 h its activity has been determined
as 19.5 g/(mmol·h·bar). Under essentially identical con-
ditions the chelate [(NCMe2CH2CH2CMe2N)Cr(CH2Ph)2]
(4) leads to a more active catalyst [47.3 g/(mmol·h·bar)].
This result may be due to both steric and electronic effects,
similar to the so-called constrained-geometry olefin polym-
erisation catalysts [6c].
Mo(CH2Ph)2]
160.00(12)
[NϪMoϪN
113.64(7)°,
[15]
MoϪNϪC
and [(2,6-
and 162.83(12)°]
iPr2C6H3N)(tBuN)MoCl2(DME)] [NϪMoϪN 105.91(9)°,
MoϪNϪCaryl 174.3(2), MoϪNϪCalkyl 157.9(2)°] [16]. Each
of the dibenzyl complexes, whether chelate or not, contains
one η1-coordinated and one η2-coordinated CH2Ph ligand.
We have been able to confirm the results concerning the
ethylene polymerisation behaviour of [(tBuN)2Cr(CH2Ph)2]
[10a, 11], using this compound in a concentration of
Owing to their comparatively low Lewis acidity, catalysts
based on Group 6 metals may be more tolerant towards
782
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792

Transition Metal Complexes Containing Functionalized Organoimido and Phosphaneiminato Ligands
Fig. 3 Molecular structure of [(NCMe2CH2CH2CMe2N)MoCl2-
(DME)] (5) in the crystal. Selected bond lengths/pm and angles/°:
Mo(1)ϪN(1) 172.8(4), Mo(1)ϪN(2) 173.6(3), Mo(1)ϪO(1) 239.3(3),
Mo(1)ϪO(2) 240.5(3), Mo(1)ϪCl(1) 242.84(12), Mo(1)ϪCl(2) 241.36(12),
N(1)ϪC(1) 145.1(5), N(2)ϪC(4) 145.5(5); N(1)ϪMo(1)ϪN(2) 99.46(17),
Fig.
1 Molecular structure of [(NCMe2CH2CH2CMe2N)Cr-
(OSiMe3)2] (2) in the crystal. Selected bond lengths/pm and
angles/°:
Cr(1)ϪN(1) 164.5(2), Cr(1)ϪO(1) 181.1(2), Si(1)ϪO(1) 164.6(2), N(1)ϪC(1)
145.6(4); O(1)ϪCr(1)ϪO(1A) 114.45(14), Si(1)ϪO(1)ϪCr(1) 136.93(13),
N(1)ϪCr(1)ϪN(1A) 102.83(17), Cr(1)ϪN(1)ϪC(1) 147.8(2).
Cl(1)ϪMo(1)ϪCl(2)
158.38(4),
O(1)ϪMo(1)ϪO(2)
68.67(10),
Mo(1)ϪN(1)ϪC(1) 149.8(3), Mo(1)ϪN(2)ϪC(4) 147.9(3).
Fig.
4 Molecular structure of [(NCMe2CH2CH2CMe2N)Mo-
Fig.
2
Molecular structure of [(NCMe2CH2CH2CMe2N)-
(CH2Ph)2] (6) in the crystal. Selected bond lengths/pm and angles/°:
Cr(CH2Ph)2] (4) in the crystal. Selected bond lengths/pm and
angles/°:
Cr(1)ϪN(1) 163.1(3), Cr(1)ϪN(2) 163.3(3), Cr(1)ϪC(1) 242.0(4),
Cr(1)ϪC(7) 207.0(4), Cr(1)ϪC(14) 208.5(4), N(1)ϪC(17) 146.6(5),
N(2)ϪC(20) 145.9(5); N(1)ϪCr(1)ϪN(2) 104.19(16), Cr(1)ϪN(1)ϪC(17)
146.8(3), Cr(1)ϪN(2)ϪC(20) 147.3(3), C(1)ϪCr(1)ϪC(7) 37.03(14),
Cr(1)ϪC(7)ϪC(1) 84.6(2), C(7)ϪCr(1)ϪC(14) 130.89(16), Cr(1)ϪC(14)Ϫ
C(8) 113.9(3).
Mo(1)ϪN(1) 173.56(8), Mo(1)ϪN(2) 173.65(8), Mo(1)ϪC(9) 217.19(10),
Mo(1)ϪC(10) 252.31(9), Mo(1)ϪC(16) 218.77(11), N(1)ϪC(1) 145.40(13),
N(2)ϪC(6) 145.59(13); N(1)ϪMo(1)ϪN(2) 102.42(4), Mo(1)ϪN(1)ϪC(1)
146.11(7), Mo(1)ϪN(2)ϪC(6) 146.38(7), C(9)ϪMo(1)ϪC(10) 35.33(4),
Mo(1)ϪC(9)ϪC(10)
Mo(1)ϪC(16)ϪC17) 112.13(7).
85.56(6),
C(9)ϪMo(1)ϪC(16)
127.68(4),
with Ziegler-Natta type Group 4 ansa-metallocene catalysts
is fraught with many problems [18].
polar, functionalised olefins than Group 4 metal systems.
For example, complexes of the type [(RN)2CrCl2] have been
shown to tolerate ester groups in the cyclopropanation of
olefins [17]. We have therefore investigated chromium and
molybdenum chelates in the polymerisation of polar olefins,
namely methyl methacrylate (MMA), acrylonitrile (AN),
and vinyl acetate (VA). Their stereospecific polymerisation
Both the chelate [(NCMe2CH2CH2CMe2N)Cr(CH2Ph)2]
(4) and the open-chain analogue [(tBuN)2Cr(CH2Ph)2] do
not polymerise methyl methacrylate under the conditions
used for ethylene polymerisation. However, polymerisation
does occur in the absence of any activator. After 24 h ca.
23 % conversion is reached with the chelate 4, whereas only
ca. 14 % is observed with [(tBuN)2Cr(CH2Ph)2], since this
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792
783

U. Siemeling et al.
Fig. 5 Conversion/time diagram for the polymerisation of MMA
with [(NCMe2CH2CH2CMe2N)Cr(CH2Ph)2] (4). Conditions:
[Cr] ϭ 2.0·10Ϫ4 M, TP ϭ r.t.
Fig. 7 Temperature dependence of monomer conversion in the poly-
merisation of MMA with [(NCMe2CH2CH2CMe2N)Cr(CH2Ph)2]
(4). Conditions: [Cr] ϭ 2.0·10Ϫ4 M, [MMA] ϭ 3.12 M, tP ϭ 6 h.
Fig. 6 Conversion/time diagram for the polymerisation of MMA
with [(tBuN)2Cr(CH2Ph)2]. Conditions: [Cr] ϭ 2.0·10Ϫ4 M, TP ϭ r.t.
Fig. 8 Temperature dependence of monomer conversion in the poly-
merisation of MMA with [(tBuN)2Cr(CH2Ph)2]. Conditions: [Cr] ϭ
2.0·10Ϫ4 M, [MMA] ϭ 3.12 M, tP ϭ 6 h.
ceases to be productive after ca. 6 h (Figs. 5 and 6). So, the
conversion/time diagrams prove that the chelate 4 is more
stable than the open-chain analogue and therefore remains
active for prolonged periods of time. The molybdenum che-
late [(NCMe2CH2CH2CMe2N)Mo(CH2Ph)2] (6) also poly-
merises methyl methacrylate, albeit with much lower ef-
ficiency than its chromium analogue (3 % conversion after
6 h).
The enhanced stability of the chromium chelate 4 is also
reflected in the temperature dependence. The maximum
conversion is reached at ca. 50 °C with 4 (Fig. 7), whereas
the less robust open-chain analogue [(tBuN)2Cr(CH2Ph)2]
has a slightly lower optimum at ca. 30 °C (Fig. 8). In both
cases, a remarkably narrow maximum is followed by a steep
decrease caused by thermal deactivation.
The polymer microstructure has been investigated by 13
C
Fig. 9 Temperature dependence of the pentad sequence distribution
in the polymerisation of MMA with [(NCMe2CH2CH2CMe2N)-
Cr(CH2Ph)2] (4). Conditions: [Cr] ϭ 2.0·10Ϫ4 M, [MMA] ϭ
3.12 M, tP ϭ 6 h.
NMR spectroscopy. In all cases studied the PMMA con-
tains syndiotactic pentad sequences, with rather similar re-
sults being obtained for the three catalysts [(tBuN)2-
Cr(CH2Ph)2], [(NCMe2CH2CH2CMe2N)Cr(CH2Ph)2] (4)
and [(NCMe2CH2CH2CMe2N)Mo(CH2Ph)2] (6). Figs. 9
and 10 show the temperature dependences of the pentad
sequences distribution of the polymers obtained by the
polymerisation of MMA with the two chromium catalyst.
In both cases the percentage of the syndiotactic rrrr pentad
decreases from more than 60 % at Ϫ45 °C to ca. 40 % at
75 °C, that is with increasing polymerisation temperature
the polymer becomes more atactic.
The molecular weights Mn obtained with the chromium
chelate 4 range from 148000 to 168000 g/mol with a PDI of
784
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792

Transition Metal Complexes Containing Functionalized Organoimido and Phosphaneiminato Ligands
Fig. 10 Temperature dependence of pentad sequence distribution
in the polymerisation of MMA with [(tBuN)2Cr(CH2Ph)2]. Con-
ditions: [Cr] ϭ 2.0·10Ϫ4 M, [MMA] ϭ 3.12 M, tP ϭ 6 h.
Fig. 11 Catalytic efficiency of [(NCMe2CH2CH2CMe2N)-
Cr(CH2Ph)2] (4) and [(tBuN)2Cr(CH2Ph)2] in the polymerisation of
AN. Conditions: [Cr] ϭ 2.0·10Ϫ4 M, [AN] ϭ 5 M, TP ϭ r.t.
1.5. With the open-chain species [(tBuN)2Cr(CH2Ph)2] even
higher molecular weights are obtained (195000 Ϫ 220000 g/
mol, PDI 1.6). The molybdenum chelate 6 leads to lower
molecular weights (101000 Ϫ 162000 g/mol) and consider-
ably higher PDI values (3.0 Ϫ 4.9).
The polymerisation of acrylonitrile is also possible with
these imido complexes, leading to atactic PAN.
[(tBuN)2Cr(CH2Ph)2] affords higher molecular weights
(Mn ϭ 219000 Ϫ 254000 g/mol, PDI 2.2 Ϫ 2.4) than the
chromium (154000 Ϫ 180000 g/mol, PDI 2.0 Ϫ 2.2) and
molybdenum chelates (57000 Ϫ 84000 g/mol, PDI 2.9).
Again, the productivity of the Mo complex is much lower
than that of the chromium catalysts. 2 Ϫ 5 % conversion
was reached with [(NCMe2CH2CH2CMe2N)Mo(CH2Ph)2]
(6), whereas 12 Ϫ 20 % was obtained with its chromium
analogue 4, and 13 Ϫ 26 % with [(tBuN)2Cr(CH2Ph)2].
The catalytic efficiency has been determined for both
chromium compounds as a function of time (Fig. 11). It is
defined as the number of polymer chains generated per
catalyst molecule. The number of polymer chains can be
calculated by dividing the yield of the obtained polymer by
the number average molecular weight of the polymer deter-
mined by GPC. Obviously, the amount of catalytically ac-
tive species increases with time, the efficiency rising gradu-
ally from small values to ca. 0.8 after several hours. Even
after 6 hours only 80 % of the catalyst is active in poly-
merising AN. The simplest explanation is a composed reac-
tion in which the formation of the active species is involved.
The three catalysts investigated are also able to poly-
merise vinyl acetate without any added cocatalyst, leading
to rather similar results. Conversions of 5Ϫ6 % were ob-
tained over the course of 24 h.
Fig. 12 Copolymer diagram of the copolymerisation of AN and
MMA with [(tBuN)2Cr(CH2Ph)2]. Conditions: [Cr] ϭ 2.0·10Ϫ4 M,
TP ϭ r.t., tP ϭ 5 h.
of both monomers into the polymer chain, indicated as a
dashed line in Figure 12.
The molecular weight distributions of the resulting poly-
mers have been determined by GPC (Fig. 13). When the
feed mole fraction of AN was over 50 %, bimodal behav-
iour was observed. Solubility tests have shown that the
minor peak is most likely due to the homopolymer PAN,
which, in contrast to PMMA and the copolymer, is in-
soluble in chloroform. The main peak corresponds to the
copolymer and to over 95 % of the measured sample.
Finally, DSC studies prove that copolymers, rather than
blends of homopolymers, have been obtained. The low glass
transition temperatures Tg in comparison with those of the
respective homopolymers, PAN and PMMA, indicate that
the products obtained must be copolymers. Tg is dependent
on the feed mole fraction and passes through a minimum
for both [(tBuN)2Cr(CH2Ph)2] and [(NCMe2CH2CH2-
CMe2N)Cr(CH2Ph)2] (4) (Fig. 14).
The chromium compounds are also effective in the co-
polymerisation of MMA and AN. The copolymer compo-
sition was calculated from the percentage of nitrogen in the
copolymers by elemental analyses. A copolymerisation dia-
gram shows the dependence of the copolymer composition
Fx on the feed mole fraction fx (Fig. 12). There is always a
slight preference for incorporation of MMA, corresponding
to only a small deviation from a true random incorporation
The mechanism of the polymerisation of the polar olefins
with these chromium complexes is not yet understood in
detail. A light-induced or spontaneous polymerisation of
the monomers can be excluded, since they do not poly-
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792
785

U. Siemeling et al.
NMR experiments using equimolar amounts of mono-
mer and catalyst so far have failed to shed light on the
mechanism. Further mechanistic work is currently in pro-
gress and will be reported in due course.
2.1.2 Di(phosphaneiminato) chelates
Di(phosphaneiminato)titanium(IV) complexes of the type
[(tBu3PN)2TiX2] (X ϭ Cl, Me) represent the first non-
cyclopentadienyl, single-site catalysts competitive with deri-
vatives of metallocenes under commercially relevant poly-
merisation conditions [5b]. Similar to di(organoimido) com-
plexes of the Group 6 metals, they are isolobal analogues
of bent metallocene derivatives of the Group 4 metals.
The number of ansa-di(phosphaneiminato) complexes
known to date is very small, and no such compound has
been described in the chemistry of the Group 4 metals. A
convenient method for the synthesis of phosphaneiminato
complexes is the metathesis of silylated iminophosphoranes
with metal halides [4]. For example, the reaction of two
equivalents of Ph3P(NSiMe3) [19] and tBu3P(NSiMe3) [5b],
respectively, with TiCl4 affords [(Ph3PN)2TiCl2] and
[(tBu3PN)2TiCl2] in 85 % and 91 % yield, respectively. In
contrast, the reaction of (Me3SiN)Ph2PCH2CH2PPh2-
(NSiMe3) with TiCl4 does not lead to a di(phosphaneimin-
ato) chelate. Instead, reaction of a CH2 group with the
strongly Lewis acidic metal halide takes place, and a bis(tri-
methylsilylphosphoranimino) chelate containing one car-
bon-titanium bond is obtained [20]. A similar reaction takes
place between [VCl3(THF)3] and (Me3SiN)Ph2PCH2PPh2-
(NSiMe3) [21]. Dinuclear complexes containing a bridging
di(phosphaneiminato) ligand are formed in the reaction of
(Me3SiN)Ph2P(CH2)nPPh2(NSiMe3) with the compara-
tively weak Lewis acids Cp*TiF3 (n ϭ 2) [22] and Cp*TiCl3
(n ϭ 1 [23], 2 [24]).
Fig. 13 Dependence of the average molecular weight of the copoly-
mer on the initial monomer ratio. Conditions: [Cr] ϭ 2.0·10Ϫ4 M,
TP ϭ r.t., tP ϭ 5 h.
Fig. 14 Glass transition temperature of the polymers obtained from
the copolymerisation of AN and MMA as a function of the feed
mole fraction. Conditions: [Cr] ϭ 2.0·10Ϫ4 M, TP ϭ r.t., tP ϭ 5 h.
merise in the absence of the chromium species under the
conditions employed.
A redox-chemical initiation of a radical polymerisation
seems to be possible with chromium(VI) compounds. How-
ever, neither chromates nor dichromates led to the polym-
erisation of the polar monomers. The fact that polymeri-
sation is also achieved with the molybdenum system pro-
vides further evidence against a redox-initiated radical
mechanism, since molybdenum(VI) is a much weaker oxi-
dant than chromium(VI). Furthermore, the polymerisation
results were unaffected by stoichiometric amounts of the
radical scavenger 2,6-di-tert-butyl-4-methylphenol (BHT).
An anionic polymerisation, which, in contrast to a cat-
ionic mechanism, is possible with the electron-poor olefins
MMA and AN, is equally unlikely. The high molecular
weights obtained, in conjunction with rather low monomer
conversion, are not characteristic for anionic polymeri-
sations. Furthermore, isotactic PMMA is obtained by
anionic polymerisation in apolar solvents, whereas in our
case a syndiotactic polymer is produced.
(Me3SiN)tBu2P(CH2)3PtBu2(NSiMe3) (7) reacts with
TiCl4 in boiling toluene under dilution conditions, affording
the di(phosphaneiminato) chelate [(NtBu2P(CH2)3PtBu2N)-
TiCl2] (8) (Scheme 2). The hydrocarbyl derivatives
[(NtBu2P(CH2)3PtBu2N)TiMe2] (9) and [(NtBu2P(CH2)3-
PtBu2N)Ti(CH2Ph)2] (10) were obtained from 8 by reaction
with methyllithium and benzylmagnesium chloride, respec-
tively (Scheme 2).
A single-crystal X-ray structure analysis was performed
for 9 and 10 (Figs. 15 and 16).
The TiϪN [9, 183.5(2) and 181.0(2) pm; 10, 180.7(2) and
182.3(2) pm] and PϪN distances [9, 158.9(3) and 156.9(2);
10, 158.7(3) and 158.4(3) pm] are very similar to the re-
spective values found for [(tBu3PN)2TiMe2] [5b] and com-
patible with bond orders close to two. The NϪTiϪN bite
angle is 114.81(12)° for 9 and 114.17(12)° for 10, which is
only slightly smaller than the value of 117.24(11)° found
for [(tBu3PN)2TiMe2]. This finding suggests that the eight-
membered titanacycle present in 8Ϫ10 is essentially un-
strained. Two quite different TiϪNϪP angles are found for
both 9 [140.30(19)° vs. 159.82(19)°] and 10 [140.95(17)° vs.
162.35(17)°], while angles of 171.2(2)° and 175.3(2)° have
An insertion polymerisation with a cationic metal centre,
similar to that operative in the polymerisation of ethylene,
must be excluded, since the activated complexes do not
polymerise the polar olefins used.
786
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792

Transition Metal Complexes Containing Functionalized Organoimido and Phosphaneiminato Ligands
Scheme 2 Synthesis of di(phosphaneiminato)titanium chelates. Re-
agents and conditions: (i) Me3SiN3, toluene; (ii) TiCl4, toluene; (iii)
9: MeLi (> 2 equiv.), diethyl ether; 10: PhCH2MgCl (> 2 equiv.),
diethyl ether.
Fig. 16 Molecular structure of [(NtBu2P(CH2)3PtBu2N)Ti-
(CH2Ph)2] (10) in the crystal. Selected bond lengths/pm and
angles/°:
Ti(1)ϪC(4) 215.7(5), Ti(1)ϪC(11) 216.5(4), P(1)ϪC(1) 180.5(3), P(1)ϪC(18)
187.0(3), P(1)ϪC(22) 187.4(3), P(2)ϪC(3) 184.0(3), P(2)ϪC(26) 186.0(3),
P(2)ϪC(30) 186.3(3), C(1)ϪC(2) 153.2(4), C(2)ϪC(3) 154.7(4);
C(4)ϪTi(1)ϪC(11) 105.47(19), P(1)ϪC(1)ϪC(2) 116.6(2), C(1)ϪC(2)ϪC(3)
111.8(3), C(2)ϪC(3)ϪP(2) 113.7(2).
mium compound the titanium complex does not polymerise
polar olefins.
2.2 Redox-functionalised Complexes
The development of compounds with redox-tunable proper-
ties is a challenging field of research and is attracting in-
creased interest [26]. So far the main focus has been on host
molecules which bear redox-active substituents and can be
utilised for the electrochemical detection of ions in solution.
A quite different field, which is still in its infancy, is the
development of catalysts carrying a redox switch, since such
catalysts offer a scope for new aspects and applications in
catalysis [27]. The basic principle of redox-switchable tran-
sition metal catalysts is straightforward: The oxidation of a
redox-active substituent induces a higher electrophilicity
and hence a change in reactivity of the catalytically active
transition metal centre. Catalyst activity is a crucial factor
in olefin polymerisation processes, which is relevant to such
varied aspects as polymer microstructure and process en-
gineering.
Fig. 15 Molecular structure of [(NtBu2P(CH2)3PtBu2N)TiMe2] (9)
in the crystal. Selected bond lengths/pm and angles/°:
Ti(1)ϪC(4) 212.4(3), Ti(1)ϪC(5) 212.8(3), P(1)ϪC(1) 181.3(3), P(1)ϪC(6)
186.9(3), P(1)ϪC(10) 186.6(3), P(2)ϪC(3) 184.7(3), P(2)ϪC(14) 186.7(3),
P(2)ϪC(18) 185.9(3), C(1)ϪC(2) 155.3(5), C(2)ϪC(3) 152.3(4);
C(4)ϪTi(1)ϪC(5) 107.00(15), P(1)ϪC(1)ϪC(2) 115.8(2), C(1)ϪC(2)ϪC(3)
110.5(3), C(2)ϪC(3)ϪP(2) 114.1(2).
been reported for [(tBu3PN)2TiMe2] [25]. This lends further
credence to the quantum-chemical result that the potential
energy well for the angle at the nitrogen atom is very shal-
low for such compounds [2]. A similar structural feature
has previously been observed for a related niobium chelate,
exhibiting NbϪNϪP angles of 145.3(5)° and 168.6(5)°
(vide infra).
The markedly different TiϪNϪP angles (ca. 140° vs. ca.
160°) found in the structurally characterized alkyl deriva-
tives 9 and 10 should not be interpreted in terms of linear
vs. bent phosphaneiminato ligands, but instead reflect the
“softness” of the angle at the nitrogen atom.
2.2.1 Imido complexes
We have chosen imido alkylidene complexes of the type
[(DippN)Mo(CHCMe2Ph)(OR)2] (Dipp ϭ 2,6-diisopro-
pylphenyl) as study objects. Such compounds are olefin
metathesis initiators and polymerise strained cycloolefins
[28]. Their catalytic activity and chemical selectivity are
known to be governed by the electrophilicity of the metal
centre [29]. Owing to the well-behaved redox chemistry of
There is an isolobal relationship between [(NtBu2P-
(CH2)3PtBu2N)Ti(CH2Ph)2] (10) and [(NCMe2CH2CH2-
CMe2N)Cr(CH2Ph)2] (4). However, in contrast to the chro-
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792
787

U. Siemeling et al.
Fig. 17 Molecular structure of [(FcdippN)2Mo(CH2Ph)2] (13) in
the crystal. Selected bond lengths/pm and angles/°:
Mo(1)ϪN(1) 176.4(4), Mo(1)ϪN(2) 175.9(4), Mo(1)ϪC(45) 214.8(5),
Mo(1)ϪC(55) 212.9(5), N(1)ϪC(14) 140.5(6), N(2)ϪC(36) 140.4(6),
C(10)ϪC(11) 148.5(7), C(32)ϪC(33) 147.8(7), C(45)ϪC(46) 155.9(7),
C(55)ϪC(56) 155.5(7); N(1)ϪMo(1)ϪN(2) 111.39(18), N(1)ϪMo(1)ϪC(45)
103.74(19), N(1)ϪMo(1)ϪC(55) 111.1(2), N(2)ϪMo(1)ϪC(45) 108.23(19),
N(2)ϪMo(1)ϪC(55) 104.5(2), C(45)ϪMo(1)ϪC(55) 118.0(2), Mo(1)Ϫ
N(1)ϪC(14) 161.4(4), Mo(1)ϪN(2)ϪC(36) 157.0(3), Mo(1)ϪC(45)ϪC(46)
119.6(4), Mo(1)ϪC(55)ϪC(56) 122.7(4).
Table 1 Formal electrode potentials (vs. ferrocene/ferrocenium) for
the oxidation of the ferrocenyl-funtionalised compounds in di-
chloromethane solution. Conditions: 298 K; nBu4NPF6 (0.1 M)
supporting electrolyte; Pt disk working electrode, Ag wire pseudo-
reference electrode, Pt wire counter electrode; scan rate 0.1V/s; sol-
ute concentration: 10Ϫ3 M; confidence limit 0.01 V; ∆E values ref-
erenced to FcdippNH2 (11)
Compound
E0/V
∆E/V
∆Ep/V
Scheme 3 Synthesis of the redox-functionalised “Schrock type” ole-
fin metathesis initiator 15. Reagents and conditions: (i)
[NH4]2Mo2O7 (0.25 equiv.), NEt3 (4 equiv.), Me3SiCl (5 equiv),
DME; (ii) PhMe2CCH2MgCl (2 equiv.), Et2O; triflic acid (3 equiv.),
DME; (iv) KOtBu (2 equiv.), Et2O.
FcdippNH2 (11)
12
13
Ϫ0.10
0.03
Ϫ0.02
0.07
0.12
0.09
0.13
0.08
siderable degree of π-delocalisation over both rings. Such a
delocalisation is beneficial for the interaction between the
ferrocenyl units and the molybdenum centre, which was in-
vestigated by electrochemical methods (Table 1).
ferrocene [30], the ferrocenyl group (Fc) was used as redox-
active unit.
The ferrocenyl-substituted aniline FcdippNH2 (11)
(Fcdipp ϭ 4-ferrocenyl-2,6-diisopropylphenyl) was used as
The oxidation of the ferrocenyl substituents gives rise to
a single two-electron redox wave in the cyclic voltammo-
gram. The half-wave potential of 13 is shifted significantly
by ϩ0.08 V with respect to the potential of FcdippNH2
(11). The ferrocene-centred redox reaction of 13 is cathod-
ically shifted by 0.05 V with respect to the corresponding
potential of 12. These results prove that the potentials are
indeed sensitive to the fragment bound to the nitrogen atom
of the FcdippN group. Consequently, in 13 and related
species the molybdenum centre will be sensitive to the ferro-
cenyl redox state. The electron-donating effect of a ferro-
cenyl group has been described to be larger than that of
OMe, whereas the electron-withdrawing effect of an oxi-
dised ferrocenyl group resembles that of COOMe [26d].
a
precursor for the imido alkylidene complex
[(FcdippN)Mo(CHCMe2Ph)(OtBu)2] (15) by following a
well-established strategy (Scheme 3) [29a, 31]. 15 is the first
example of a complex belonging to the family of “Schrock
type” olefin metathesis initiators which contains a ferrocene
unit bound to a spectator ligand.
A single-crystal X-ray structure analysis was performed
for 13. A view of the molecule is shown in Figure 17.
The molybdenum atom is coordinated by two imido and
two alkyl ligands in a slightly distorted pseudo-tetrahedral
environment. Each C6 ring plane is almost coplanar to the
plane of the cyclopentadienyl ring attached to it (dihedral
angles are 9.1° and 21.3°, respectively), indicating a con-
788
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792

Transition Metal Complexes Containing Functionalized Organoimido and Phosphaneiminato Ligands
Fig. 18 Molecular structure of [NbCl3(NPPh2Fc)2] (17) in the crys-
tal. Selected bond lengths/pm and angles/°:
Nb(1)ϪCl(1) 245.7(3), Nb(1)ϪCl(2) 244.3(3), Nb(1)ϪCl(3) 244.3(3),
Fig. 19 Molecular structure of [NbCl3{[NPPh2(C5H4)]2Fe}] (18) in
the crystal. Selected bond lengths/pm and angles/°:
Nb(1)ϪCl(1) 243.1(2), Nb(1)ϪCl(2) 244.4(2), Nb(1)ϪCl(3) 244.5(2),
P(1)ϪC(1) 178.9(8), P(1)ϪC(11) 182.0(8), P(1)ϪC(17) 181.1(8), P(2)ϪC(6)
177.2(8), P(2)ϪC(23) 180.0(7), P(2)ϪC(29) 180.9(7); N(1)ϪNb(1)ϪN(2)
109.8(3), N(1)ϪNb(1)ϪCl(2) 135.5(2), N(2)ϪNb(1)ϪCl(2) 114.7(2),
N(1)ϪP(1)ϪC(1) 112.7(4), N(1)ϪP(1)ϪC(11) 110.5(4), N(1)ϪP(1)ϪC(17)
P(1)ϪC(5) 176.8(9), P(1)ϪC(16) 179.9(9), P(1)ϪC(22) 180.8(9), P(2)ϪC(27)
178.8(9), P(2)ϪC(38) 181.2(9), P(2)ϪC(44) 180.5(9); N(1)ϪNb(1)ϪN(2)
109.0(4), N(1)ϪNb(1)ϪCl(3) 122.8 (2), N(2)ϪNb(1)ϪCl(3) 128.1(3),
N(1)ϪP(1)ϪC(5) 110.7(4), N(1)ϪP(1)ϪC(16) 111.6(4), N(1)ϪP(1)ϪC(22)
112.4(4), N(2)ϪP(2)ϪC(27) 109.6(4), N(2)ϪP(2)ϪC(38) 110.3(4),
N(2)ϪP(2)ϪC(44) 109.2(4).
109.8(4),
N(2)ϪP(2)ϪC(29) 109.1(3).
N(2)ϪP(2)ϪC(6)
112.9(3),
N(2)ϪP(2)ϪC(23)
110.6(3),
The reaction of 13 with three equivalents of triflic acid
proceeds cleanly and swiftly affording [(FcdippN)-
Mo(CHCMe2Ph)(OTf)2(DME)] (14) in good yield. Reac-
tion of 14 with a slight excess of KOtBu yielded crude
[(FcdippN)Mo(CHCMe2Ph)(OtBu)2] (15) as a brown solid.
Only one alkylidene species was formed, as proven by the
occurrence of a single signal for the alkylidene Hα proton
(δ ϭ 11.34 ppm). Unfortunately, despite many efforts, it has
not proved possible to obtain the compound sufficiently
pure for further studies.
also compare well with the respective values observed for
[NbCl3(NPPh3)2] (NbϪN 182.7(4)/183.0(6) pm, PϪN
160.2(5)/159.5(6) pm) and [NbCl3(NPiPr3)2] (NbϪN
183.5(7)/185.3(9) pm, PϪN 156.4(9)/159.0(7) pm). As for 9
and 10, the molecular structure should not be interpreted
in terms of linear vs. bent phosphaneiminato ligands, but
instead reflects the “softness” of the angle at the nitrogen
atom.
2.2.2 Phosphaneiminato complexes
The chelate analogue [NbCl3{[NPPh2(C5H4)]2Fe}] (18)
was obtained by reaction of [Fe{(C5H4)PPh2NSiMe3]}2]
[34] with niobium pentachloride in acetonitrile at 40 °C as
an orange solid in 49 % yield. A single-crystal X-ray struc-
ture analysis was performed for this complex. A view of the
molecule is shown in Fig. 19. The ferrocene unit is essen-
tially undistorted with eclipsed cyclopentadienyl rings. The
Nb atom is coordinated by three chloro and two phos-
phaneiminato ligands in a distorted trigonal-bipyramidal
arrangement. Two Cl atoms occupy the axial positions for-
ming an angle of 165.63(8)° with the Nb atom, which lies
in the plane of the equatorial atoms N(1), N(2) and Cl(2).
Both the niobium-nitrogen [Nb(1)ϪN(1) 184.9(6),
Nb(1)ϪN(2) 182.9(6) pm] and the phosphorus-nitrogen dis-
tances [P(1)ϪN(1) 158.8(6), P(2)ϪN(2) 159.3(6) pm] are in
accord with double bonds. In contrast to the unchelated
analogue 17, the NbϪNϪP bond angles of 18 are very simi-
lar [159.1(4)° at N(1) and 153.3(4)° at N(2)].
Reaction of [Me3SiNPPh2Fc] (16) with niobium pentachlo-
ride in acetonitrile at 70 °C afforded [NbCl3(NPPh2Fc)2]
(17) as a yellow solid in 76 % yield. A single-crystal X-ray
structure analysis was performed for this compound. A
view of the molecule is shown in Fig. 18. The structure is
similar to that of [NbCl3(NPiPr3)2] [32] and
[NbCl3(NPPh3)2] [33]. The Nb atom is coordinated by three
chloro and two phospaneiminato ligands in a distorted trig-
onal-bipyramidal arrangement. Two Cl atoms occupy the
axial positions forming an angle of 166.17(9)° with the Nb
atom, which lies in the plane of the equatorial atoms N(1),
N(2) and Cl(3). Both the niobium-nitrogen [Nb(1)ϪN(1)
184.5(7), Nb(2)ϪN(2) 183.5(8) pm] and the phosphorus-ni-
trogen distances [P(1)ϪN(1) 159.7(8), P(2)ϪN(2) 159.5(8)
pm] are in accord with double bonds.
A striking feature of the structure is the fact that the
NbϪNϪP bond angles are quite different, namely 145.3(5)°
at N(1) and 168.6(5)° at N(2). A similar feature has been
observed for the related titanium complexes 9 and 10 (vide
supra). The values reported for [NbCl3(NPPh3)2] and
[NbCl3(NPiPr3)2] are 156.4(3)/161.9(3)° and 168.5(5)/
168.2(4)°. The NbϪN and the PϪN bond lengths of 17 are
identical for both ligands within experimental error and
The redox-functionalised di(phosphaneiminato)niobium
complexes polymerise ethylene in the presence of MAO
(Nb/Al 1/750, 55 °C, 5 bar ethylene, toluene solvent) with
high initial activity. However, the catalytically active species
is very short-lived and polymerisation ceases within ca.
1 min.
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792
789

U. Siemeling et al.
69.3 mmol) in diethyl ether (100 mL) cooled to 0 °C. After 3 h at
room temperature the mixture was cooled to 0 °C. Chlorotrime-
thylsilane (15.1 g, 139 mmol) was added. After 14 h at room tem-
perature the mixture was filtered and the solvent removed in vacuo,
leaving the crude product as a light yellow liquid which was puri-
fied by distillation. Yield: 15.2 g (76 %), bp (4 mbar) 90 Ϫ 93 °C.
3 Summary and Outlook
Complexes containing organoimido and phosphaneiminato
ligands are prominent examples of compounds containing
a nitrido bridge between a main group element (C and P,
respectively) and a transition metal. The chemistry of such
species has flourished over the past decades and is now be-
ing extended to the use of functionalised ligands. Such a
development is by no means unusual and has been encoun-
tered before with, for example, cyclopentadienyl and phos-
phane ligands [6a].
We have investigated two different types of modified im-
ido and phosphaneiminato ligands, namely chelate ligands
and redox-functionalised ligands (including also an ex-
ample of a redox-functionalised chelate ligand), and have
obtained the following main results.
We have synthesised the first di(organoimido)chromium
chelates and have further shown that complexes of the type
[(RN)2M(CH2Ph)2] (M ϭ Cr, Mo) are active catalysts for
the polymerisation of the polar olefins methyl methacrylate,
acrylonitrile and vinyl acetate, with copolymerisations be-
ing also possible.
1H NMR (CDCl3): δ ϭ 0.04 (s, 18 H, SiMe3), 1.07 (s, 12 H, CMe2), 1.29 (s,
4 H, CH2). 13C{1H} NMR (CDCl3): δ ϭ 2.9 (SiMe3), 31.4 (CMe2), 41.3
(CH2), 51.5 (CMe2). 29Si{1H} NMR (CDCl3): δ ϭ 2.4. EI-MS (70 eV): m/z ϭ
289 (Mϩ, 100 %).
[(NCMe2CH2CH2CMe2N)Cr(OSiMe3)2] (2): A solution of chromyl
chloride (4.08 g, 26.3 mmol) in hexane (10 mL) was added dropwise
to a stirred solution of 1 (15.2 g, 52.7 mmol) in hexane (100 mL)
cooled to Ϫ20 °C. The mixture was allowed to warm to room tem-
perature and was subsequently heated to reflux for 4 h. After cool-
ing to room temperature the mixture was filtered and the volume
of the filtrate reduced to ca. 40 mL in vacuo. Crystallisation at
Ϫ60 °C afforded the product as a dark red solid. Yield: 2.90 g
(30 %). Elemental analysis: C14H34N2CrO2Si2 (370.6): C 44.50
(calcd. 45.37), H 9.00 (9.25), N 7.85 (7.56) %.
1H NMR (C6D6): δ ϭ 0.29 (s, 18 H, SiMe3), 0.98 (s, 12 H, CMe2), 2.18 (s,
4 H, CH2). 13C{1H} NMR (C6D6): δ ϭ 2.7 (SiMe3), 27.7 (CMe2), 41.2
(CH2), 79.4 (CMe2). 29Si{1H} NMR (C6D6): δ ϭ 15.2. EI-MS (70 eV): m/z ϭ
371 (Mϩ, 100 %).
We have synthesised the first di(phosphaneiminato)titan-
ium and niobium chelates. The first complexes containing
redox-functionalised organoimido ligands have also been
prepared as well as the first structurally characterised ex-
amples of complexes containing redox-functionalised phos-
phaneiminato ligands.
We have provided clear experimental evidence for the
quantum-chemical result that the potential energy well for
the angle at the nitrogen atom is very shallow for phos-
phaneiminato complexes. Great care should therefore be
taken in the interpretation of structural results in terms of
linear vs. bent ligands.
[(NCMe2CH2CH2CMe2N)CrCl2] (3): Boron trichloride (15.6 mL
of a 1.00 M solution in hexane, 15.6 mmol) was added dropwise to
a stirred solution of 2 (2.90 g, 7.82 mmol) in dichloromethane
(30 mL) cooled to 0 °C. After 2 h at room temperature volatile
components were removed in vacuo. The residue was extracted with
hot hexane (5 ϫ 20 mL). The combined extracts were reduced to
dryness in vacuo. Sublimation (60 °C, 10Ϫ3 mbar) of the crude
product afforded a reddish black solid. Yield: 1.20 g (58 %). Ele-
mental analysis: C8H16N2Cl2Cr (263.1): C 35.83 (calcd. 36.52), H
5.89 (6.13), N 10.96 (10.65) %.
1H NMR (C6D6): δ ϭ 0.77 (s, 12 H, Me), 1.97 (s, 4 H, CH2). 13C{1H} NMR
(C6D6): δ ϭ 25.7 (Me), 40.7 (CH2), 84.0 (CMe2). EI-MS (70 eV): m/z ϭ 263
(Mϩ, 100 %).
It is clear that our investigations have merely scratched
the surface and that the chemistry of complexes containing
organoimido, phosphaneiminato or related ligands will
continue to be a bonanza.
[(NCMe2CH2CH2CMe2N)Cr(CH2Ph)2] (4): Benzylmagnesium
chloride (7.14 mL of a 1.00 M solution in diethyl ether, 7.14 mmol)
was added dropwise to a stirred solution of 3 (0.94 g, 3.57 mmol)
in diethyl ether (100 mL) cooled to Ϫ45 °C. The mixture was al-
lowed to warm to room temperature. After 15 h volatile compo-
nents were removed in vacuo. The residue was extracted with pen-
tane (4 ϫ 30 mL). The volume of the combined extracts was re-
duced to ca. 40 mL. Crystallisation at Ϫ60 °C afforded the product
as a red solid. Yield: 0.95 g (71 %). Owing to the high sensitivity
of the product towards air and moisture, no reproducible microana-
lytical data could be obtained.
4 Experimental
All manipulations were performed in an inert atmosphere (purified
argon or dinitrogen) by using standard Schlenk and cannula tech-
niques or a conventional glovebox. Solvents and reagents used were
commercially available and were appropriately dried and purified
by using standard procedures. NMR spectra were recorded with a
Bruker Avance DRX 500 or a Varian Unity INOVA 500 spec-
1H NMR (C6D6): δ ϭ 0.99 (s, 12 H, Me), 1.83 (s, 4 H, CH2CH2), 2.53 (s,
4 H, CH2Ph), 7.00 (m, 10 H, Ph). {1H} NMR (C6D6): δ ϭ 27.4 (Me), 40.6
(CH2CH2), 43.7 (CH2Ph), 74.2 (CMe2), 124.9 (Ph), 129.0 (Ph), 130.8 (Ph),
141.0 (Ph). EI-MS (70 eV): m/z ϭ 375 (Mϩ, 100 %).
1
trometer operating at 500.13 MHz for H. Mass spectra were ob-
[(NCMe2CH2CH2CMe2N)MoCl2(DME)] (5): Sodium molybdate
(6.00 g, 29.1 mmol) was suspended in DME (150 mL). Triethyl-
amine (11.8 g, 117 mmol), chlorotrimethylsilane (25.5 g, 235 mmol)
and a solution of 2,5-diamino-2,5-dimethylhexane [40] (4.20 g,
29.1 mmol) in DME (40 mL) were added. The mixture was heated
to reflux for 15 h. After cooling to room temperature the mixture
was filtered and the volume of the filtrate reduced to ca. 50 mL in
vacuo. Crystallisation at Ϫ30 °C afforded the product as a pale
green solid. Yield: 6.04 g (52 %). Elemental analysis:
C12H26N2Cl2MoO2 (397.2): C 36.29 (calcd. 36.14), H 6.60 (6.44),
N 7.14 (7.05) %.
tained with a VG Autospec instrument. Elemental analyses were
performed by the Microanalytical Laboratory of the Universität
Bielefeld, the Microanalytical Laboratory Beller (Göttingen) and
the Microanalytical Laboratory H. Kolbe (Mülheim an der Ruhr).
The syntheses and properties of the following compounds have al-
ready been published: 7Ϫ10 [35], 11 [36], 12Ϫ15 [37], 16Ϫ17 [38],
18 [39].
Me3Si(H)NCMe2CH2CH2CMe2N(H)SiMe3 (1): nBuLi (87 mL of
a 1.60 M solution in hexane, 139 mmol) was added dropwise to
a stirred solution of 2,5-diamino-2,5-dimethylhexane [40] (10.0 g,
790
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792

Transition Metal Complexes Containing Functionalized Organoimido and Phosphaneiminato Ligands
Table 2 X-ray crystallographic data
Compound
2
C
5
6
Chem. formula
Formula weight
Crystal size
14H34CrN2O2Si2
C12H26Cl2MoN2O2
C22H30MoN2
370.61
397.19
418.42
0.4ϫ0.4ϫ0.1 mm3
0.6ϫ0.3ϫ0.2 mm3
0.4ϫ0.4ϫ0.3 mm3
Space group
monoclinic, C2/c
21.031(12)
7.631(3)
13.578(5)
orthorhombic, Pbca
13.443(4)
15.764(4)
16.912(6)
monoclinic, P21/n
8.4600(1)
16.7250(2)
14.7170(3)
˚
a/A
˚
b/A
˚
c/A
β/°
100.20(6)
2144.7(17)
90
99.5490(16)
2053.51(5)
3
˚
V/A
3583.9(19)
Z
4
8
4
ρ(calcd.)/g/cm3
1.148
0.650
1.472
1.030
1.353
0.645
µ/mmϪ1
F(000)
Index ranges
θ range/°
Reflections collected
Independent reflections
Data/restraints/parameters
Goodness-of-fit on F2
Final R1a) [I>2σ(I)]/wR2b)
800
1632
872
Ϫ27ՅhՅ27, Ϫ9ՅkՅ0, Ϫ17ՅlՅ17
2.8Ϫ27.5
4928
2469 [R(int.) ϭ 0.0701]
2469/0/101
1.003
0.0479/0.1050
0.310/Ϫ0.354
0ՅhՅ18, 0ՅkՅ22, 0ՅlՅ23
2.3Ϫ30.0
5224
5224
5224/0/178
1.018
Ϫ13ՅhՅ13, Ϫ26ՅkՅ26, Ϫ23ՅlՅ23
2.7Ϫ35.0
62984
9031 [R(int.) ϭ 0.0400]
9031/0/346
1.056
0.0246/0.0499
0.485/Ϫ0.600
0.0505/0.1279
0.715/Ϫ1.091
3
˚
Largest diff.peak/hole/e/A
a) R1 ϭ ΣʈFoΗϪΗFcʈ/ΣΗFoΗ; b) wR2 ϭ {Σ[w(Fo2ϪFc2)2]/Σ[w(Fo2)2]}0.5
.
1H NMR (CDCl3): δ ϭ 1.25 (s, 12 H, CMe2), 2.30 (s, 4 H, CH2CMe2), 3.66
(s, 6 H, OMe), 3.75 (s, 4 H, OCH2).
cooled and re-heated, typically in a temperature range from 25 °C
to 180 °C with a heating rate of 10 K·minϪ1
.
[(NCMe2CH2CH2CMe2N)Mo(CH2Ph)2] (6): Benzylmagnesium
chloride (3.52 mL of a 1.00 M solution in diethyl ether, 3.52 mmol)
was added dropwise to a stirred solution of 5 (0.70 g, 1.76 mmol)
in diethyl ether (70 mL) cooled to Ϫ60 °C. The mixture was al-
lowed to warm to room temperature. After 15 h volatile compo-
nents were removed in vacuo. The residue was extracted with hex-
ane (3 ϫ 20 mL). The volume of the combined extracts was reduced
to ca. 30 mL. Crystallisation at Ϫ30 °C afforded the product as a
light orange solid. Yield: 0.54 g (73 %). Elemental analysis:
C22H30N2Mo (418.4): C 63.13 (calcd. 63.15), H 7.34 (7.23), N
6.65 (6.69) %.
Gel permeation chromatographic (GPC) analyses of PMMA, PAN
and the AN-MMA copolymer were performed by the Bayer AG
Dormagen and by PSS Standards Service GmbH / Mainz, whose
support is thankfully acknowledged.
X-ray Crystal Structure Determinations: The crystal structures of 4
[41], 9 [35], 10 [35], 13 [37], 17 [38] and 18 [39] have already been
reported. The crystal structures of 2 and 5 were obtained with a
Siemens P21 four-circle diffractometer at 173 K, that of 6 was ob-
tained with a Nonius Kappa CCD diffractometer at 100 K. Graph-
˚
ite-monochromated Mo-Kα radiation (λ ϭ 0.71073 A) was used in
each case. The structures were solved by direct methods. The pro-
gram package used was SHELXL 97 [42]. Full-matrix least-squares
refinement on F2 was carried out anisotropically for the non-hydro-
gen atoms. Hydrogen atoms were refined isotropically for 6 and
were included at calculated positions using a riding model for 2
and 5. Further X-ray structure analyses data are given in Table 2.
1H NMR (C6D6): δ ϭ 1.08 (s, 12 H, Me), 1.83 (s, 4 H, CH2CH2), 2.53 (s,
4 H, CH2Ph), 7.00 (m, 10 H, Ph). {1H} NMR (C6D6): δ ϭ 28.0 (Me), 41.6
(CH2CH2), 43.7 (CH2Ph), 71.8 (CMe2), 124.7 (Ph), 128.3 (Ph), 129.6 (Ph),
140.7 (Ph).
Polymerisation Studies: All polymerisations of polar monomers
were carried out under argon in a 100 ml flask equipped with a
magnetic stirrer and with a cooling jacket for the temperature
experiments. Typically, the solution had a total volume of 60 ml
and polymerisations were run with 0.012 mmol of catalyst and the
appropriate amount of monomer in toluene solution at ambient
temperature. The reaction was quenched with methanol at a
specific time and the polymer was precipitated in methanol, filtered
off and dried in vacuo at 60 °C until a constant weight was
achieved. Copolymerisation experiments with acrylonitrile and
methyl methacrylate were performed under essentially identical
conditions.
Full details (excluding structure factors) have been deposited with
the Cambridge Crystallographic Data Centre (2 CCDC 195080, 5
CCDC 195081, 6 CCDC 905082). Copies of the data can be ob-
tained free of charge on application to The Director, CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (Fax: int.code
ϩ(1223)336-033; e-mail: deposit@ccdc.cam.ac.uk).
This work was generously supported by the DFG (Si 429/6) and in
part also by the BMBF (03C0276C/3). O. K. thanks the DAAD for
a travel grant.
The 13C NMR spectra were measured in 10 mm NMR tubes with
proton broad-band decoupling at 120 °C on a Bruker AMX 300
spectrometer at 75 MHz. The samples of the polymer (200 Ϫ
300 mg) were dissolved in 3 mL of a mixture of 1,2,4-trichlorben-
zene and 1,1,2,2-tetrachloroethane-d2 (volume ratio 2.5 : 1). The
chemical shifts were calibrated relative to 1,1,2,2-tetrachloro-
ethane-d2 (dc ϭ 74.2 ppm).
References
[1] See, for example: V. C. Gibson, J. Chem. Soc., Dalton Trans.
1994, 1607.
[2] A. Diefenbach, F. M. Bickelhaupt, Z. Anorg. Allg. Chem.
1999, 625, 892.
[3] Review: D. E. Wigley, Prog. Inorg. Chem. 1994, 42, 239.
[4] Reviews: a) K. Dehnicke, M. Krieger, W. Massa, Coord. Chem.
Rev. 1999, 182, 19; b) K. Dehnicke, J. Strähle, Polyhedron 1989,
8, 707.
Glass transition temperatures were measured on a Mettler-Toledo
DSC 820 calorimeter. The sample was first heated, gradually
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792
791

U. Siemeling et al.
[5] a) Review: G. J. P. Britovsek, V. C. Gibson, D. F. Wass, Angew.
Chem. 1999, 111, 448; Angew. Chem. Int. Ed. 1999, 38, 428; b)
[20] M. J. Sarsfield, M. Thornton-Pett, M. Bochmann, J. Chem.
Soc., Dalton Trans. 1999, 3329.
´
D. W. Stephan, F. Guerin, R. E. v. H. Spence, L. Koch, X.
[21] G. Aharonian, K. Feghali, S. Gamborotta, G. P. A. Yap, Or-
ganometallics 2001, 20, 2616.
[22] M. Sotoodeh, I. Leichtweis, H. W. Roesky, M. Noltemeyer,
H.-G. Schmidt, Chem. Ber. 1993, 126, 913.
[23] R. P. Kamalesh Babu, R. McDonald, R. G. Cavell, J. Chem.
Soc., Dalton Trans. 2001, 2210.
Gao, S. J. Brown, J. W. Swabey, Q. Wang, W. Xu, P. Zoricak,
D. G. Harrison, Organometallics 1999, 18, 2046.
[6] Reviews: a) U. Siemeling, Chem. Rev. 2000, 100, 1495; b) H.
Butenschön, Chem. Rev. 2000, 100, 1527; c) J. Okuda, T. Eb-
erle, in Metallocenes, Vol. 1, A. Togni, R. L. Halterman (eds),
Wiley-VCH, Weinheim, 1998, Chap. 7; d) P. Jutzi, T. Redeker,
Eur. J. Inorg. Chem. 1998, 663; e) B. Wang, D. Deng, C. Qian,
New J. Chem. 1995, 19, 515; f) P. Jutzi, U. Siemeling, J. Or-
ganomet. Chem. 1995, 500, 175; g) P. Jutzi, J. Dahlhaus, J.
Coord. Chem. Rev. 1994, 137, 179; h) J. Okuda, Comments
Inorg. Chem. 1994, 16, 185.
[24] M. J. Sarsfield, M. Said, M. Thornton-Pett, L. A. Gerrard,
M. Bochmann, J. Chem. Soc., Dalton Trans. 2001, 822.
[25] Ref. [5b] erroneously gives a value of 167.8(3)° for this com-
pound. The correct angles may be found in the deposited CIF.
[26] Reviews: a) P. D. Beer, Chem. Soc. Rev. 1989, 18, 409; b) R.
A. Bissell, A. P. de Silva, H. Q. N. Gunaratne, P. L. M. Lynch,
G. E. M. Maguire, K. R. A. S. Sandanayake, Chem. Soc. Rev.
1992, 21, 187; c) L. Fabrizzi, A. Poggi, Chem. Soc. Rev. 1995,
24, 197; d) A. M. Allgeier, C. A. Mirkin, Angew. Chem. 1998,
110, 936; Angew. Chem. Int. Ed. 1998, 37, 894; e) V. Amendola,
L. Fabrizzi, M. Lichelli, C. Mangano, P. Pallavicini, L. Parodi,
A. Poggi, Coord. Chem. Rev. 1999, 190Ϫ192, 649.
[27] I. M. Lorkovic, R. R. Duff, M. S. Wrighton, J. Am. Chem.
Soc. 1995, 117, 3617.
[28] Reviews: a) R. R. Schrock, Pure Appl. Chem. 1994, 66, 1447;
b) R. R. Schrock, Tetrahedron 1999, 55, 8141; c) M. R. Buch-
meiser, Chem. Rev. 2000, 100, 1565.
[29] a) R. R. Schrock, J. S. Murdzek, G. C. Bazan, J. Robbins, M.
DiMare, M. OЈRegan, J. Am. Chem. Soc. 1990, 112, 3875; b)
J. H. Oskam, R. R. Schrock, J. Am. Chem. Soc. 1992, 114,
7588; c) J. H. Oskam, R. R. Schrock, J. Am. Chem. Soc. 1993,
115, 11831; d) W. J. Feast, V. C. Gibson, E. L. Marshall, J.
Chem. Soc., Chem. Commun. 1992, 1157.
[7] L. H. Gade, Koordinationschemie, Wiley-VCH, Weinheim,
1998, Chap. 20.
[8] Reviews: a) Organometallic Catalysts and Olefin Polymeriz-
ation, R. Blom, A. Follestad, E. Rytter, M. Tilset, M. Ystenes
(eds), Springer, Berlin, 2001; b) W. Kaminsky, J. Chem. Soc.,
Dalton Trans. 1998, 1413; c) C. Janiak, in Metallocenes, Vol.
2, A. Togni, R. Halterman (eds), Wiley-VCH, Weinheim, 1998,
Chap. 9; d) M. Bochmann, J. Chem. Soc., Dalton Trans. 1996,
255; e) H. H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger,
R. Waymouth, Angew. Chem. 1995, 107, 1255; Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1143; f) Ziegler Catalysts, G. Fink, R.
Mülhaupt, H. H. Brintzinger (eds), Springer, Berlin, 1995.
[9] See, for example: a) J. Sundermeyer, D. Runge, Angew. Chem.
1994, 106, 1328; Angew. Chem. Int. Ed. Engl. 1994, 33, 1255;
b) D. S. Williams, M. H. Schofield, R. R. Schrock, Organomet-
allics 1993, 12, 4560; c) D. S. Williams, J. P. Mitchell, A. D.
Poole, U. Siemeling, W. Clegg, D. C. R: Hockless, P. A.
O’Neill, V. C. Gibson, J. Chem. Soc., Dalton Trans. 1992, 739;
d) D. S. Williams, M. H. Schofield, J. T. Anhaus, R. R.
Schrock, J. Am. Chem. Soc. 1990, 112, 6728.
[30] P. Zanello, in Ferrocenes, A. Togni, T. Hayashi (eds), VCH,
Weinheim 1995, Chap. 7.
[31] J. H. Oskam, H. H. Fox, K. B. Yap, D. H. McConville, R.
O’Dell, B. J. Lichtenstein, R. R. Schrock, J. Organomet. Chem.
1993, 459, 185.
[10] a) M. P. Coles, C. I. Dalby, V. C. Gibson, I. R. Little, E. L.
Marshall, M. H. Ribeiro da Costa, S. Mastoianni, J. Or-
ganomet. Chem. 1999, 591, 78; b) M. P. Coles, V. C. Gibson,
Polymer Bulletin 1994, 33, 529.
[32] G. Philipp, K. Harms, K. Dehnicke, C. Maichle-Mössmer, U.
Abram, Z. Anorg. Allg. Chem. 1996, 622, 1927.
[11] M. P. Coles, C. I. Dalby, V. C. Gibson, W. Clegg, M. R. J.
Elsegood, J. Chem. Soc., Chem. Commun. 1995, 1709.
[12] a) M. Bochmann, S. J. Lancaster, M. B. Hursthouse, K. M.
A. Malik, Organometallics 1994, 13, 2235; b) M. Bochmann,
S. J. Lancaster, Organometallics 1993, 12, 633; c) M. Boch-
mann, S. J. Lancaster, Makromol. Chem., Rapid Commun.
1993, 14, 807.
[13] V. R. Jensen, K. J. Borve, J. Chem. Soc., Chem. Commun.
2002, 542.
[14] E. A. Kretzschmar, J. Kipke, J. Sundermeyer, J. Chem. Soc.,
Chem. Commun. 1999, 2381.
[33] F. Weller, D. Nußhär, K. Dehnicke, Z. Anorg. Allg. Chem.
1992, 615, 7.
[34] H. W. Roesky, K. Swerat, F. Edelmann, Z. Naturforsch. 1988,
B43, 231.
[35] U. Siemeling, L. Kölling, A. Stammler, H.-G. Stammler, J.
Chem. Soc., Dalton. Trans. 2002, 3277.
[36] U. Siemeling, B. Neumann, H.-G. Stammler, O. Kuhnert,
Polyhedron 1999, 18, 1815.
[37] U. Siemeling, R. R. Schrock, A. Stammler, H.-G. Stammler,
O. Kuhnert, Z. Anorg. Allg. Chem. 2001, 627, 925.
[38] U. Siemeling, A. Stammler, H.-G. Stammler, O. Kuhnert, Z.
Anorg. Allg. Chem. 1999, 625, 845.
[39] U. Siemeling, B. Neumann, H.-G. Stammler, O. Kuhnert, Z.
Anorg. Allg. Chem. 2000, 626, 825.
[40] W. Duismann, C. Rüchard, Chem. Ber. 1978, 111, 596.
[41] U. Siemeling, L. Kölling, A. Stammler, H.-G. Stammler, E.
Kaminski, G. Fink, J. Chem. Soc., Chem. Commun. 2000,
1177.
[15] W. Clegg, M. R. J. Elsegood, P. W. Dyer, V. C. Gibson, E. L.
Marshall, Acta Crystallogr. 1999, C55, 916.
[16] R. C. B. Copley, P. W. Dyer, V. C: Gibson, J. A. K. Howard, E.
L. Marshall, W. Wang, B. Whittle, Polyhedron 1996, 15, 3001.
[17] D. Jan, F. Simal, A. Demonceau, A. F. Noels, K. A. Rufanov,
N. A. Ustynyuk, D. N. Gourevitch, Tetrahedron Lett. 1999,
40, 5695.
[18] H. Frauenrath, H. Keul, H. Höcker, in: ref. [8a], pp. 97Ϫ108.
[19] T. Rübenstahl, F. Weller, S. Wocaldo, W. Massa, K. Dehnicke,
Z. Anorg. Allg. Chem. 1995, 621, 953.
[42] G. M. Sheldrick, SHELX 97, University of Göttingen, Ger-
many, 1997.
792
Z. Anorg. Allg. Chem. 2003, 629, 781Ϫ792