Ethylene polymerization catalysts based on nickel(II) 1,4-diazadiene complexes: The influence of the 1,4-diazadiene backbone substituents on structure and reactivity

Article abstract of DOI:10.1016/S0022-328X(98)00784-0

Thermally sensitive dialkyl nickel complexes {DAD(H,H)}Ni(CH₂SiMe₃)₂ and {DAD(Me,Me)}Ni(CH₂SiMe₃)₂ [DAD(X,X)=2,6-iPr₂C₆H₄-N=C(X)-C(X)=N-C ₆H

Full text of DOI:10.1016/S0022-328X(98)00784-0

Journal of Organometallic Chemistry 569 (1998) 159–167
Ethylene polymerization catalysts based on nickel(II) 1,4-diazadiene
complexes: the influence of the 1,4-diazadiene backbone substituents
on structure and reactivity1
Thomas Schleis a, Thomas P. Spaniol a, Jun Okuda a,*, Johannes Heinemann b, Rolf Mu¨lhaupt b
a Institut fu¨r Anorganische Chemie und Analytische Chemie, Johannes Gutenberg-Uni6ersita¨t Mainz, J.-J.-Becher-Weg 24,
D-55099 Mainz, Germany
b Freiburger Materialforschungszentrum und Institut fu¨r Makromolekulare Chemie, Albert-Ludwig-Uni6ersita¨t Freiburg, Stefan-Meier-Strasse 21,
D-79104 Freiburg, Germany
Received 1 April 1998
Abstract
Thermally sensitive dialkyl nickel complexes {DAD(H,H)}Ni(CH2SiMe3)2 and {DAD(Me,Me)}Ni(CH2SiMe3)2 [DAD(X,X)=
2,6-iPr2C6H4–NꢀC(X)–C(X)ꢀN–C6H4iPr2-2,6] were synthesized and were characterized by X-ray diffraction studies on single
crystals. The substituents X on the backbone of the h-diimine ligand significantly influence the conformation of the 2,6-diiso-
propylphenyl substituents. This effect is thought to be of crucial importance for the polymerization of ethylene when
{DAD(X,X)}NiBr2/MAO is used as catalyst. The influence of the catalyst structure, pressure, and temperature on the
polymerization activity, molar mass, glass transition temperature, melting temperature and branching of the polymers has been
studied. The dialkyl complex {DAD(H,H)}Ni(CH2SiMe3)2 underwent rapid reductive carbonylation giving the dicarbonyl
complex {DAD(H,H)}Ni(CO)2 along with both bis(trimethylsilyl)ethane and h,h%-bis(trimethylsilyl)acetone. The dicarbonyl
{DAD(H,H)}Ni(CO)2 was characterized by X-ray crystallography. © 1998 Elsevier Science S.A. All rights reserved.
Keywords: X-ray crystallography; Polymerization catalyst; 1,4-Diazadiene ligand; Nickel alkyl complexes; Reductive carbonylation
1. Introduction
Brookhart et al. ([4]c) introduced highly efficient nickel
and palladium catalysts for the synthesis of branched
polyethylenes [4]. They are based on substituted 1,4-di-
azadiene or h-diimine ligands containing sterically
bulky 2,6-diisopropylphenyl substituents at the imine
nitrogen atoms: 2,6-iPr2C6H4–NꢀC(X)–C(X)ꢀN–
C6H4iPr2-2,6 [DAD(X,X): X=H, Me; X2=ace-
naphtylene]. This type of ligand has long been known,
e.g. as ancillary ligands for group 10 metal catalysts for
the oligomerization of alkynes [5–7]. It was noted [4]
that the presence of the 2,6-diisopropylphenyl sub-
stituents is of critical importance in generating a suit-
able coordination sphere at the cationic nickel(II)
center. The branching was attributed to the migration
of the nickel alkyl along the polymer chain through
repetitive i-H elimination and reinsertion of the vinyl-
In the mid 1980s homogeneous nickel catalysts [1] for
the oligo- and polymerization of ethylene to produce
short-chain branched polyethylene became available [2].
The basis for an unprecedented type of polymerization
was the discovery of the so-called 2,ꢀ-polymerization
by Fink et al. [3]. Using nickel(0) catalysts, h-olefins
such as 1-pentene do not form poly(h-olefin) as would
be expected with Ziegler-type catalysts, but instead
gives polyethylene with methyl branches corresponding
to
poly(ethylene-alt-propylene).
More
recently,
* Corresponding author. Fax: +49 6131 395605.
1 Dedicated to Professor Akira Nakamura with best personal
wishes.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00784-0

160
T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
In order to explore the nature of the nickel-carbon
bond in the dialkyl 3, carbonylation was studied. When
a hexane solution of the dialkyl 3 at 25°C is exposed to
slight excess of CO at 1 bar, followed by partial evapo-
ration of solvent and crystallization at ꢁ30°C, the dicar-
bonyl {DAD(H,H)}Ni(CO)2 (5) can be isolated as red
crystals. The CO ligands give rise to a 13C-NMR reso-
nance at 196.9 ppm and two wCO bands in the IR
spectrum at 2014 and 1954 cm−1. The carbonylation of
3 to give 5 is accompanied by the formation of the
organic byproducts Me3SiCH2CH2SiMe3 (6) and
(Me3SiCH2)2CO (7). It can be regarded as a reductive
elimination, preceded by CO insertion into the nickel–
alkyl bond. This type of reaction has been studied
extensively by Yamamoto et al. ([9]a) using dialkyl
complexes of the type L2NiR2, where L are two-elec-
tron donor, and R are alkyl ligands [9]. These authors
observed the formation of the alkane R-R, ketone
R-CO-R, diketone R-CO–CO-R, and aldehyde R-
CHO as products of reductive elimination depending
on the complex used and the reaction conditions. When
3 is reacted with slight excess of CO at r.t. in benzene-
d6, the reaction mixture clearly contains both the
alkane 6 and the ketone 7. In dichloromethane, the
reaction of 3 at −60°C with a slight excess of CO gave
terminated polymer into the metal-hydride bond. We
report here the synthesis and X-ray diffraction studies
of two nickel dialkyls containing 1,4-diazadiene ligands
derived from glyoxal and diacetyl. Trimethylsilymethyl
was used for the alkyl groups on the nickel center, since
it gives thermally more stable nickel derivatives and it
has been proven to be a useful model for a growing
polymer chain [8]. We also present detailed data on the
ethylene polymerization behavior of nickel catalysts
based on the DAD(X,X) ligands.
2. Results and discussion
Easily accessible, paramagnetic dibromo complexes
{DAD(H,H)}NiBr2 (1) and {DAD(Me,Me)}NiBr2 (2)
[6] can be activated by methylalumoxane (MAO) to
give the cationic species [{DAD(X,X)}NiMe]+
[MAO]− active for the polymerization of olefins [4].
The generation of [{DAD(H,H)}NiMe]+[B{C6H3-
(CF3)2-3,5}4]− was reported by Brookhart et al. ([4]c)
but its isolation was hampered by extreme temperature-
sensitivity. We found that even the dimethyl precursor
{DAD(X,X)}NiMe2, previously reported in the litera-
ture [6] was difficult to handle. When 1 and 2 are
treated with the Grignard reagent Mg(CH2SiMe3)Cl in
ether at −78°C, the new dialkyl complexes
{DAD(H,H)}Ni(CH2SiMe3)2 (3) and {DAD(Me,Me)}-
Ni(CH2SiMe3)2 (4) are isolated in analytically pure
form in 18 and 28% yield, respectively (Scheme 1). The
turquoise extremely air- and moisture-sensitive crystals
are stable at 25°C. However, both compounds rapidly
decompose in solution above 0°C and substantial care
must be taken for their isolation to be successful.
1
According to H- and 13C-NMR spectra, compounds
3 and 4 have apparent C2v-symmetry in solution. The
two separated doublets for the four isopropyl groups of
the aromatic ring indicate both hindered rotation about
the nitrogen ipso carbon and the isopropyl ortho carbon
bond. The trimethylsilymethyl groups in 3 give rise to
two signals at l 0.08 and 2.32 in the ratio 2:9. The
corresponding signals of 4 appear at 0.05 and 1.17. In 3
the methylene protons are recorded at 25°C as a broad
signal, indicating fluxional behavior. If the C2-symmet-
ric structure found in the crystal lattice reflects the
preferred conformation, the methylene protons are
diastereotopic. It is noteworthy that the methyl groups
of the 1,4-diazadiene backbone in 4 have undergone a
high-field shift of almost 2 ppm upon complexation,
indicating their close proximity to the aromatic imine
substituents. Reaction of the dialkyl complexes with the
strong Brønsted acid H(Et2O)[B{C6H3(CF3)2-3,5}4] led
to decomposition, resulting in intractable product
mixtures.
Scheme 1.

T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
161
tween the mean plane defined by the aromatic ring and
the mean plane defined by the atoms N, N(1)A, C(17),
C(17)A, and Ni is 76.6° for 3 and 83.4° for 4. This
shows that in the latter complex, the aromatic ring is
less tilted towards the diazadiene backbone.
For the polymerization of olefins it seems to be
important that the aromatic residues at the 1,4-diazadi-
ene bridge have bulky substituents in the ortho-posi-
tions. Brookhart and coworkers [4] showed that
compounds with methyl substituents give lower molecu-
lar weight and unbranched polyolefins. Isopropyl sub-
stituents increase the molecular weight and allow
isomerization and ‘chain walking’ to take place. It is
therefore important to understand the influence of the
isopropyl groups. Its steric bulk blocks the chain termi-
nation step and allows isomerization to take place. The
available space at the nickel center is the key to the
control of the polymerization behavior. The free space
at the metal center is partly controlled by non-bonded
interaction to the methyl groups of the isopropyl units.
The molecular structures show that the shortest dis-
˚
tances are 4.550(5) and 5.120(3) A for 3 compared to
˚
4.424(4) and 5.029(4) A for 4. In addition, the distances
between the same methyl groups and C(1) are 5.066(4)
˚
and 4.846 (4) A in 3 compared to 4.071(4) and 4.624(5)
Fig. 1. ORTEP diagrams for the molecular structure of
{DAD(H,H)}Ni(CH2SiMe3)2 (3). Thermal ellipsoids are drawn at the
50% probability level. Hydrogen atoms have been omitted for the
sake of clarity.
only the ketone 7, with the dicarbonyl complex 5
decomposing at −20°C.
2.1. Crystal structures
Despite their pronounced sensitivity, single crystals
of the dialkyl complexes 3 and 4 could be obtained and
analyzed by X-ray diffraction. Figs. 1 and 2 show the
ORTEP-plots of the compounds 3 and 4. Selected bond
lengths and angles are listed in Table 1.
The overall molecular structure of both compounds
can be described as square planar, typical for a d8
nickel(II) center with strong ligands. Both compounds
reside on a crystallographic C2-axis with the two bulky
trimethylsilyl groups arranged on opposite sides of the
planar Ni{DAD(X,X)} fragment. Upon closer inspec-
tion, however, the conformation of the ligand periphery
is different for the two compounds. The ORTEP plot of
3 shows how the isopropyl groups of the 2,6-diiso-
propylphenyl substituent are tilted away from the
nickel center. This effect can be explained by steric
repulsion of the bulky alkyl substituents. In 4 on the
other hand, the isopropyl groups are turned towards
the nickel dialkyl unit. This reduction of the space at
the nickel center is evidently caused by the two methyl
groups of the 1,4-diazadiene backbone. The angle be-
Fig. 2. ORTEP diagrams for the molecular structure of
{DAD(Me,Me)}Ni(CH2SiMe3)2 (4). Thermal ellipsoids are drawn at
the 50% probability level. Hydrogen atoms have been omitted for the
sake of clarity.

162
T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
Table 1
˚
Selected bond lengths (A) and angles (°) of {DAD(H,H)}Ni-
(CH2SiMe3)2 (3) and {DAD(Me,Me)}Ni(CH2SiMe3)2 (4)
3 (X=H)
4 (X=Me)
NiꢁC(1)
NiꢁN
NꢁC(17)
NꢁC(5)
C(10)ꢁC(14)
C(11)ꢁC(12)
C(11)ꢁC(13)
C(14)ꢁC(16)
C(14)ꢁC(15)
C(17)ꢁC(17)A
1.950(2)
1.964(1)
1.293(2)
1.449(2)
1.512(3)
1.517(4)
1.528(4)
1.529(5)
1.538(4)
1.432(4)
1.955(2)
1.974(2)
1.294(3)
1.445(3)
1.513(4)
1.499(4)
1.516(5)
1.502(7)
1.531(5)
1.471(5)
C(1)ꢁNiꢁC(1)A
C(1)ꢁNiꢁN(1)A
C(1)AꢁNiꢁN(1)A
N(1)AꢁNiꢁN
C(17)ꢁNꢁNi
SiꢁC(1)ꢁNi
88.0(1)
174.84(7)
95.71(1)
80.78(9)
114.1(1)
116.1(1)
115.5(1)
87.9(1)
174.45(9)
96.34(8)
79.6(1)
115.9(2)
119.4(1)
114.0(1)
Fig. 3. ORTEP diagrams for the molecular structure of
{DAD(H,H)}Ni(CO)2 (5). Thermal ellipsoids are drawn at the 50%
probability level. Hydrogen atoms have been omitted for the sake of
clarity.
NꢁC(17)ꢁC(17)A
as generally expected for zero-valent dicarbonyl nickel
complexes. For instance, it is similar to the structure of
(dicarbonyl)(4,6-dimethyl-2,2%-bipyridyl)nickel [11]. The
nickel–imine nitrogen distances are slightly longer than
those in 3, which can be explained by the presence of
the strong y-acceptor ligand CO at the nickel center.
The angle N–Ni–N(1)A of 80.3(1)° can be regarded as
a consequence of the chelating 1,4-diazadiene ligand,
distorting the idealized tetrahedral configuration of the
nickel.
Thermal ellipsoids are drawn at 50% probability level.
˚
A in 4. This shows that the isopropyl methyl groups are
closer to the central metal atom in 4 than in 3. We
anticipate that the catalytically active species derived
from the closely analogous precursors 1 and 2 should
also provide more space for polymerization reactions in
the former.
For both structures the angle C(1)–Ni–C(1)A of 88°
is the same, within the standard deviation, but the angle
at the methylene carbon Si–C(1)–Ni in 4 is larger by
3°. Thus, the trimethylsilyl groups of 4 are forced away
further from the nickel center than those in 3. Interest-
ingly the carbon–carbon bond lengths C(17)–C(17)A
2.2. Polymerizations
Polymerizations of h-olefins using MAO-activated
catalysts {DAD(H,H)}NiBr2 (1) and {DAD(H,H)}-
NiBr2 (2) have been reported by Brookhart et al. [4] to
give polymers with a randomly distributed multiply
branched structure, with predominantly methyl
˚
differ quite substantially, with 1.432(4) A for 3 and
˚
1.471 (5) A for 4. The larger bond length in 4 seems to
reflect the electronic and steric effects of the two addi-
tional methyl groups at the 1,4-diazadiene bridge. This
difference can also be seen in the angles at the imine
nitrogen atom and nickel atom C(17)–N–Ni and N–
Ni–N(1)A.
There are no published crystal structure analyses of
dialkyl nickel complexes with sterically hindered 1,4-di-
azadienes, but the molecular structure of
bis(trimethylsilylmethyl) complex, py2Ni(CH2SiMe3)2,
has been described [10]. The nickel atom finds itself in
a nearly perfect square planar, unconstrained coordina-
tion sphere with values of the angles at the metal
N–Ni–C(1) of 92.5° and N–Ni–N(1) of 87.5°.
The crystal structure analysis of the dicarbonyl com-
pound 5 could also be established. Fig. 3 depicts the
ORTEP diagram of the dicarbonyl 5. Selected bond
lengths and angles are listed in Table 2. The molecular
structure of 5 shows a distorted tetrahedral geometry,
Table 2
˚
Selected bond lengths (A) and angles (°) of {DAD(H,H)}Ni(CO)2 (5)
NiꢁC(15)
NiꢁC(14)
NiꢁN
NꢁC(1)
NꢁC(2)
O(1)ꢁC(14)
O(2)ꢁC(15)
1.756(5)
1.761(4)
1.971(2)
1.291(3)
1.444(3)
1.135(5)
1.134(5)
a
C(15)ꢁNiꢁC(14)
C(15)ꢁNiꢁN
C(14)ꢁNiꢁN
NꢁNiꢁN(1)A
C(1)ꢁNꢁC(2)
C(1)ꢁNꢁNi
105.6(2)
117.6(1)
117.4(1)
80.3(1)
119.5(2)
113.8(2)
126.5(2)
115.8(1)
C(2)ꢁNꢁNi
NꢁC(1)ꢁC(1)A

T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
163
Table 3
Ethylene polymerization data using MAO-activated 1 and 2
Run
Catalyst precursora
Pb
(bar)
fEc
TPd
(°C)
Ae
Mnf
Tmg
(°C)
Nh
(mol l−1
)
(kg (molNi ·h·mol l−1 −1
)
)
(g mol−1
)
1
2
3
4
5
6
7
8
1
1
1
1
1
1
1
1
2
2
2
1.5
4.6
6.1
5.2
8.0
10.6
3.3
6.2
1.5
4.6
6.1
0.22
0.66
0.88
0.44
0.66
0.88
0.66
0.66
0.22
0.66
0.88
20
20
20
60
60
60
0
40
20
20
20
2700
2090
850
800
860
240
740
2420
1500
3080
2180
56 300
44 000
57 700
23 000
30 300
28 300
83 900
122
128
129
104
115
117
129
120
44
8
6
4
23
12
15
1
37 400
7
i
9
10
11
54
31
11
I
I
86
105
a [Ni]=20 mmol l−1, [Al]=20 mmol l−1
b Pressure in the polymerization vessel.
c Ethylene concentration in toluene [14].
d Polymerization temperature.
.
e Activity, calculated from the mass of the polymer, catalyst concentration, monomer concentration and polymerization time.
f Molar mass determined by GPC against polyethylene standard.
g Determined by DSC (heating rate 20 K min−1).
h Number of branching carbons per 1000 carbon atoms, determined by 1H-NMR and 13C-NMR spectroscopy.
i Samples only partially soluble due to high molecular weight (\107 g mol−1).
ethylene concentration the catalytic activity drops. By
increasing the reaction temperature from 20 to 60°C the
catalytic activity decreases. Nevertheless, at a reaction
temperature of 60°C, a higher initial activity is ob-
served, but fast deactivation of the catalyst causes loss
of overall polymerization activity. At a reaction tem-
perature of 20°C deactivation of the catalyst is slower
and the overall polymerization activity is higher.
As seen in Fig. 5 there is a dependence of the number
of branches/1000C (N) and polymer melting tempera-
ture (Tm) on reaction temperature. By increasing the
reaction temperature more branches in the obtained
polymer are observed. Because of the direct dependence
branches. In agreement with those results, the catalytic
activity was determined as a function of temperature,
monomer concentration, and catalyst structure in some
detail for the above catalysts. The polymerization re-
sults are summarized in Table 3 and can be interpreted
by the reaction steps depicted in Scheme 2.
Fig. 4 shows the dependence of the catalytic activity
on the concentration of ethylene in the polymerization
solution. The catalytic activity varies depending on
ethylene pressure and reaction temperature. At higher
Fig. 4. Polymerization activity of 1/MAO versus concentration of
Scheme 2. Polymerization and branching mechanism.
ethylene at different reaction temperatures.

164
T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
Fig. 5. Polymer melting temperature, Tm, and number of branches/1000C, N, of polyethylene obtained with 1/MAO versus the reaction
temperature (at a polyethylene concentration of 0.66 mol l−1).
stronger dependence of N and Tm on ethylene pressure
than the analog generated from 1.
between the number of branches/1000C (N) and the
polymer melting temperature, it is obvious that when
the number of branches in the polymer rises the poly-
mer melting temperature drops.
The termination reaction appears to be more feasible
for 1 than for 2. The molecular weights of the
polyethylenes generated with 2 are higher than those
formed using 1. This may suggest that the sterically
more hindered nickel center of the catalytic species
derived from 2 are longer-lived and may suppress ter-
mination by associative olefin exchange of the
monomer. Provided that the insertion reaction of
ethylene is the rate determining step [4,15,16], the sig-
nificantly higher steric hindrance of the isopropyl
groups of the catalyst formed from 2 compared to that
of the catalyst formed from 1 may result in a more
reactive, destabilized resting state. This is in line with
the conformational differences of the {DAD(X,X)}Ni
fragment found for the model compounds 3 and 4.
In agreement with previous observations [4,12], the
degree of branching N, determined by NMR spec-
troscopy as the number of methyl groups per 1000
carbon atoms, decreases with increasing ethylene pres-
sure. The difference in ligand structure also influences
the number of branches, N. This number is higher for
all polymers produced by catalyst 2 compared to the
polymers generated with 1 under the same polymeriza-
tion conditions (see entry 1 and 9, 2 and 10). Branches
are formed when i-hydride elimination of the growing
polymer chain, followed by isomerization, occurs
(Scheme 2). Isomerization is possible when the polymer
chain at the nickel center undergoes i-hydride elimina-
tion forming an intermediate olefin hydride complex
and the olefin reinserts in a 2,1-manner into the nickel–
hydride bond before insertion of the next monomer
Fig. 6 shows the dependence of the polymer melting
temperature (Tm) on the ethylene pressure. Decreasing
the ethylene pressure leads to more branches in the
polymer and so to a lower polymer melting tempera-
ture. This is the case for the catalytic systems derived
from both 1 and 2. The dependence of N and Tm on the
ethylene pressure at different reaction temperatures is
shown. At elevated reaction temperature more branches
are formed and as a result the polymer melting temper-
ature drops. The catalyst generated from 2 shows a
Fig. 6. Polymer melting temperature Tm versus the ethylene concen-
tration obtained with 1/MAO and 2/MAO at different reaction
temperatures.

T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
165
takes place. Somewhat counterintuitive, more isomer-
ization and branching occurs if there is less space at the
catalytic site. One would assume that the 2,1-insertion
of the olefin should be sterically more hindered. Thus
with the sterically more hindered, but more active
catalyst precursor 2 higher branching frequencies are
observed. In calculations [15,16], based on QM or
combined QM/MM methods, where the nickel center
with the attached ligands is calculated by quantum
mechanics and the steric demanding groups are refined
by molecular mechanics, a lower DG‡ value was found
for isomerization than for insertion. This leads to the
conclusion that a more stable resting state of the cata-
lyst allows more isomerization of the polymer chain to
take place resulting in a higher degree of branching of
the generated polymer.
1H-NMR (200 MHz): l 0.08 (s, 18 H, SiCH3), 0.88 (d,
3
3JHH=6.8 Hz, 6 H, CH(CH3)2), 1.52 (d, JHH=6.8 Hz,
6 H, CH(CH3)2), 2.32 (s, 4 H, CH2SiMe3), 3.56 (sept,
3JHH=6 8 Hz, 4 H, CH(CH3)2) 7.35 (m, 6 H, C6H3),
9.07 (s, 2H, NꢀCH). 13C{1H}-NMR (100.6 MHz): l
−3.2 (CH2SiMe3), 3.4 (SiCH3), 23.0, 25.3 (CH(CH3)2),
28.7 (CHMe2), 124.1 (meta-C), 127.0 (para-C), 140.1
(ortho-C), 147.1 (ipso-C), 162.9 (NꢀCH); 29Si{1H}-
NMR (79.5 MHz):
l
4.42. Anal. Calcd. for
C34H58N2NiSi2: C 66.98, H 9.59, N 4.59. Found: C
66.93, H 9.73, N 7.91.
3.2. {DAD(Me,Me)}Ni(CH2SiMe3)2 (4)
This compound was prepared following a similar
procedure as described for
3 using {DAD(Me,-
Me)}NiBr2 (2) (640 mg, 1.0 mmol) and trimethylsilyl-
methyl magnesiumchloride (2 ml of 1 M solution in
diethyl ether) to give 175 mg (0.28 mmol) of turquoise
prisms, yield: 28%. 1H-NMR (400 MHz): l 0.05 (s, 6 H,
NꢀCCH3), 0.23 (s,18 H, SiCH3), 0.93 (d, 3JHH=6.8 Hz,
6 H, CH(CH3)2), 1.17 (s, 4H, CH2 Si(CH3)3), 1.53 (d,
3. Experimental
All operations were performed under an inert atmo-
sphere of argon using standard Schlenk-line and glove-
box techniques. Diethyl ether and hexane were dried
over sodium/triglyme benzophenone ketyl and distilled
under argon. DAD(H,H), DAD(Me,Me), {DAD-
(H,H)}NiBr2 (1), {DAD(Me,Me)}NiBr2 (2) were pre-
pared according to literature procedures [6].
Trimethylsilylmethyl magnesiumchloride was used as
1.0 M solution in diethyl ether (Aldrich). 1H- and
13C{H}-NMR spectra were recorded on a Bruker
ARX200, ARX300, ARX400, and DRX400 spectrome-
ter in benzene-d6 at r.t. and in C2D2Cl4 at 100°C for the
polymers. Elemental analyses were determined on a
Heraeus CHN-Rapid instrument. IR-spectra were
recorded in concentrated dichloromethane solutions us-
ing a closed sample cell equipped with KBr windows on
a Mattson Galaxy 2030 FT-IR spectrometer.
3
3JHH=6.8 Hz, 6 H, CH(CH3)2), 3.58 (sept, JHH=6.8
Hz, 2 H, CH(CH3)2), 7.3 (m br, 6 H, C6H3); 13C{1H}-
NMR (100.6 MHz): l −5.6 (NꢀCCH3), 4.1 (SiCH3),
22.8 (CH2SiMe3), 24.2, 24.6 (CH(CH3)2), 28.6
(CHMe2), 124.1 (para-C), 127.0 (para-C), 140.1 (ortho-
C), 147.1 (ipso-C), 162.9 (NꢀCMe). Anal. Calcd. for
C36H62N2NiSi2: C 67.80, H 9.80, N 4.05, Found: C
66.26, H 9.19, N 4.39.
3.3. {DAD(H,H)}Ni(CO)2 (5)
A solution of {DAD(H,H)}Ni(CH2SiMe3)2 (3) (71
mg, 0.1 mmol) in 50 ml of hexane was reacted with
carbon monoxide in a 100 ml Schlenk flask by exchang-
ing the argon atmosphere over the solution by carbon
monoxide (1 bar) under stirring. The color of the
solution changed immediately from blue green to red.
After exchanging the atmosphere over the solution
back to argon, the solution was concentrated to 5 ml
and allowed to crystallize for 2 days at −30°C. Purple-
red crystals suitable for crystal structure analysis were
3.1. {DAD(H,H)}Ni(CH2SiMe3)2 (3)
To a stirred suspension of {DAD(H,H)}NiBr2 (555
mg, 0.93 mmol) in 20 ml of diethyl ether at −78°C was
dropwise added a solution of trimethylsilylmethyl mag-
nesiumchloride (1.86 ml of the 1 M solution in diethyl
ether, 1.86 mmol). The color of the suspension changed
to dark blue. After 1 h the suspension was allowed to
warm to 0°C and stirring was continued for another
hour. During this time, the color of the reaction mix-
ture changed to green. All volatiles were removed in
vacuo and the blue-green brittle foam was extracted at
0°C with 100 ml of cold hexane. The blue-green hexane
extracts were filtered through celite to remove the mag-
nesium halides, concentrated to 10 ml in vacuo, and
cooled to −30°C to induce crystallization. After 2 days
turquoise prisms (100 mg, 0.17 mmol) were obtained in
18% yield, suitable for a crystal structure analysis.
1
obtained in 43% yield. H-NMR (200 MHz): l 0.08 (s,
3
18 H, SiCH3), 0.88 (d, JHH=6.8 Hz, 6 H, CH(CH3)2),
3
1.52 (d, JHH=6.8 Hz, 6 H, CH(CH3)2), 2.32 (s, 4 H,
CH2SiMe3), 3.56 (sept, 3JHH=6 8 Hz, 4 H, CH(CH3)2,)
7 35 (m, 6 H, C6H3), 9.07 (s, 2H, NꢀCH). 13C{1H}-
NMR (100.6 MHz): l 23.7, 25.5 (CH(CH3)2). 27.9
(CHMe2), 123.7 (meta-C), 126.5 (para-C), 139.1 (ortho-
C), 152.3 (ipso-C), 163.4 (NꢀCH), 196.9 (CO). IR
(CH2Cl2): wCO (cm−1): 2014, 1954.
The mother liquor was analyzed by NMR spec-
1
troscopy and contains (Me3SiCH2)2CO (7): H-NMR
(200 MHz): l 0.21 (s, 9 H, SiCH3), 2.22 (s, 2 H,
CH2SiMe3). 13C{1H}-NMR (100.6 MHz): l −1.2

166
T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
Table 4
Crystallographic data for 3, 4, and 5
Compound
3
4
5
Chemical formula
Formula weight
Crystal size (mm)
Crystal system
C34H58N2NiSi2
609.71
0.45×0.5×0.5
Monoclinic
C2/c (No.15)
C36H62N2NiSi2
637.77
0.3×0.4×0.7
Monoclinic
C2/c (No.15)
C28H36N2NiO2
491.30
0.4×0.4×0.5
Orthorhombic
Pnma (No. 62)
Space group
Unit cell dimensions
˚
a (A)
17.899(4)
10.343(2)
20.197(3)
98.57(1)
3697(1)
8/2
1.095
0.612
1328
10.503(3)
20.481 (2)
18.482(5)
102.92(3)
3875(2)
8/2
1.093
0.586
12.123(2)
21.454(3)
10.614(4)
˚
b (A)
˚
c (A)
i (°)
3
˚
V (A )
2761 (1)
8/2
1.182
0.727
Z
zcalcd (g cm−3
)
v (mm−1
F(000)
)
1392
1048
Temperature (K)
2qmax [°]
296(2)
56.0
296(2)
58.0
296(2)
60.0
Index ranges
05h523, 05kB13,
−265l526
−105h514, −195k527,
−255l524
05h517, 05k530,
05l514
Absorption correction empirical
(C-scans)
Transmission (min/max)
Reflections
99.18/99.95
97.94/99.95
98.13/99.93
4111/4111
Collected/independent
4585/4449
7149/5145
(Rint=0.0180)
3613
293
(Rint=0.0132)
3597
194
Independent with I\2|(I)
No. of parameters
2450
180
R1/wR2/|; (all data)
R1/wR2/| [I\2|(I)]
0.0491/0.0974/1.070
0.0316/0.0808/0.995
0.268/−0.218
0.0875/0.1351/1.041
0.0468/0.1135/1.053
0.456/−0.283
0.1079/0.1209/1.142
0.0455/0.0895/1.111
0.435/−0.278
Resid. Electron density (e A−3
)
(SiCH3), 39.8 (CH2SiMe3), 147.9 (CꢀO). IR (CH2Cl2): w
(cm−1): 2965, 1669 (CꢀO), 1256. Experiments on
NMR-tube-scale in benzene-d6 showed the presence of
the second byproduct Me3SiCH2CH2SiMe3 (6), charac-
was precipitated in 1.5 1 of methanol acidified with 10
ml of 10 wt.% hydrochloric acid, filtered, and dried at
60°C under vacuum.
1
terized by its resonances. H-NMR (400 MHz): l 0.08
3.5. Crystal strucure determination of {DAD(H,H)}-
Ni(CH2SiMe3)2 (3), {DAD(Me,Me)}Ni(CH2SiMe3)2 (4)
and {DAD(H,H)}Ni(CO)2 (5)
(s, 18H, SiCH3), 0.44 (s, 4H, CH2) [17].
3.4. Polymerization procedure
{DAD(H,H)}Ni(CH2SiMe3)2 (3) and {DAD(Me,-
Me)}Ni(CH2SiMe3)2 (4) were obtained as turquoise
prisms by cooling concentrated hexane solutions of the
compounds to −30°C. {DAD(H,H)}Ni(CO)2 (5) was
obtained as red prisms by cooling a concentrated hex-
ane solution of the compound to −30°C. X-ray dif-
fraction data were collected with an Enraf-Nonius
CAD4 diffractometer at room temperature by using
Mo–Kh radiation and ꢀ-scans. Data reduction was
carried out using the program system MolEN ([13]a).
The structures were solved by direct methods
(SHELXS-86) ([13]b) and refined by fullmatrix least
squares (SHELXL-93) ([13]c) against F2. Non-hydro-
gen atoms were refined anisotropically. For 3, all hy-
drogen atoms were located in difference Fourier maps
and refined in their position. Isotropic temperature
The ethylene polymerization reactions using
{DAD(H,H)}NiBr2 (1) and {DAD(Me,Me)}NiBr2 (2)
precatalysts were performed in a mechanically stirred 1
l glass reactor. Typically, 400 ml of toluene was
pumped into the reactor. After thermal equilibration of
the reactor system ethylene was continuously added by
a mass-flow meter (F-111 C, Bronkhorst, NI-7261 AK
Ruurlo, Netherlands) until the reaction mixture was
saturated with ethylene. The polymerization was started
by adding 8 mmol of 1 or 2 in 8 ml of a 1:1 mixture of
toluene and MAO solution, so that [Ni]=20 mmol l−1
and [Al]=20 mmol l−1. The vapor pressure of the
solvent was considered when calculating the ethylene
concentration. Typically, polymerization was quenched
by adding 10 ml isopropanol after 1 h. The polymer

T. Schleis et al. / Journal of Organometallic Chemistry 569 (1998) 159–167
167
Preprints 37 (2) (1996) 254. (d) Brookhart/DuPont Pat. Appl.
WO 96/23010. (e) C.M. Killian, L.K. Johnson, M. Brookhart,
Organometallics 16 (1997) 2005. (e) S. Mecking, L.K. Johnson,
L. Wang, M. Brookhart, J. Am. Chem. Soc. 120 (1998) 888.
[5] For a review on 1,4-diazadiene complexes, see (a) G. van Koten,
K. Vrieze, Adv. Organomet. Chem. 21 (1982) 151, and references
therein. (b) J.H. Groen, C.J. Elsevier, K. Vrieze, W.J.J. Smeets,
A.L. Spek, Organometallics 15 (1996) 3445. (c) H.F. Haarman,
F.R. Bregman, J.M. Ernsting, N. Veldman, L.A. Spek, K.
Vrieze, Organometallics 16 (1997) 54.
[6] (a) H. tom Dieck, M. Svoboda, Chem. Ber. 109 (1976) 1657. (b)
M. Svoboda, Ph. D. Thesis 1978, University of Frankfurt, Main,
Germany. (c) M. Svoboda, H. tom Dieck, Organomet. Chem.
191 (1980) 321.
[7] (a) H. tom Dieck, M. Svoboda, T.Z. Greiser, Naturforsch. 36B
(1981) 823. (b) M. Svoboda, H. tom Dieck, C. Kruger, Y.H.
Tsay, Z. Naturforsch. B 36B (1981) 814. (c) H. tom Dieck, M.
Svoboda, J. Kopf, Z. Naturforsch. 33B (1978) 1381.
factors were also refined. For 5, the hydrogen atoms
not belonging to methyl groups were refined in their
position; the other hydrogen atoms (as well as all
hydrogen atoms in 4 were included in calculated posi-
tions with fixed thermal parameters. Crystallographic
data are summarized in Table 4. Further details of the
crystal structure determinations are available on request
from
the
Fachinformationzentrum
Karlsruhe,
Gesellschaft fu¨r wissenschaftlich-technische Informa-
tion mbH, D-76344 Eggenstein-Leopoldshafen, on
quoting the depository number CSD-408809 (3),
408808 (4) and 408810 (5), the names of the authors,
and the journal citation.
[8] J. Okuda, F.J. Schattenmann, S. Wocadlo, W. Massa,
Organometallics 14 (1995) 789.
Acknowledgements
[9] (a) T. Yamamoto, T. Kohara, A. Yamamoto, Chem. Lett.
(1976) 1217. (b) T. Kohara, S. Komiya, T. Yamamoto, A.
Yamamoto, Chem. Lett. (1979) 1513. (c) S. Komiya, Y. Abe, A.
Yamamoto, T. Yamamoto, Organometallics 2 (1983) 1466. (d)
K. Tatsumi, A Nakamura, S. Komiya, A. Yamamoto, T. Ya-
mamoto, J. Am. Chem. Soc. 106 (1984) 8181.
[10] E. Carmona, F. Gonzalez, M.L. Poveda, J.L. Atwood, R.D.
Rogers, J. Chem. Soc. Dalton Trans. (1981) 777.
[11] (a) G. Sonnek, E. Dinjus, Z. Anorg. Allg. Chem. 441 (1978) 58.
(b) J. Sieler, R. Benedix, N.N. Than, E. Dinjus, D. Walther, Z.
Anorg. Allg. Chem. 522 (1985) 131.
This research was supported by the German Federal
Ministry for Education, Science, Research, and Tech-
nology (BMBF) (Project no. 03N10280) and the BASF
AG as well as the Fonds der Chemischen Industrie. We
thank Professor M. Brookhart for some discussion.
WITCO GmbH, Berkamen, kindly supplied us with
MAO solutions.
[12] (a) R.F. de Souza, R.S. Mauler, L.C. Simon, F.F. Nunes, D.V.S.
Vescia, A. Cavagnolli, Macromol. Rapid. Commun. 18 (1997)
795. (b) D. Pappalardo, M. Mazzeo, C. Pellecchia, Macromol.
Rapid. Commun. 18 (1997) 1017.
[13] (a) C.K. Fair, MolEN, An Interactive Structure Solution Proce-
dure; Enraf-Nonius, Delft, The Netherlands, 1990. (b) G.M.
Sheldrick, SHELXS-86, Program for the Solution of Crystal
Structures; University of Go¨ttingen, Go¨ttingen, Germany, 1986.
(c) G.M. Sheldrick, SHELXL-93, Program for the Refinement of
Crystal Structures, University of Go¨ttingen, Go¨ttingen, Ger-
many, 1993.
[14] W. Krauss, W. Gestrich, Chem. Tech. 6 (1977) 513.
[15] (a) L. Deng, P.M. Margl, T. Ziegler, J. Am. Chem. Soc. 119
(1997) 1094. (b) L. Deng, T.K. Woo, L. Cavallo, P.M. Margl, T.
Ziegler, J. Am. Chem. Soc. 119 (1997) 6177.
[16] (a) D.G. Musaev, R.D.J. Froese, M. Svensson, K. Morokuma, J.
Am. Chem. Soc. 119 (1997) 364. (b) D.G. Musaev, M. Svensson,
K. Morokuma, S. Stromberg, K. Zetterberg, P.E.M. Siegbahn,
Organometallics 16 (1997) 1933. (c) R.D.J. Froese, D.G.
Musaev, K. Morokuma, J. Am. Chem. Soc. 120 (1998) 1581.
[17] (a) I. Fleming, A. Pearce, J. Chem. Soc. Perkin (1981) 251. (b)
D. Djahanbini, B. Cazes, J. Gore, F. Gobert, Tetrahedron 41
(1985) 867.
References
[1] (a) G. Wilke, Angew. Chem Int. Ed. Engl. 27 (1988) 185. (b)
P.W. Jolly, G. Wilke, The Organic Chemistry of Nickel, vol. 1,
Academic Press, New York, 1975. (c) Wilke, G.; Herrmann, G.
Angew. Chem. 1966, 78, 591.
[2] (a) W. Keim, R. Appel, A. Storeck, C. Kruger, R. Goddard,
Angew. Chem. 93 (1981) 91. (b) M. Peuckert, W. Keim,
Organometallics
2 (1983) 594. (c) For a recent review on
oligomerization of ethylene see, D. Vogt in Applied Homoge-
neous Catalysis with Organometallic Compounds (Eds.: B.
Cornils, W.A. Herrmann), vol.1, chapter 2.3.1.3, VCH Verlags-
gesellschaft, Weinheim, 1996. (d) U. Klabunde, S.D. Ittel, J.
Mol. Cat. 41 (1987)123. (e) K.A. Ostaja Starzewski, J. Witte,
Angew. Chem. 97 (1985) 610. (f) K.A. Ostaje Starzewski, J.
Witte, Angew. Chem. 99 (1987) 76.
[3] G. Fink, V.M. Mohring, Angew. Chem Int. Ed. Engl. 24 (1985)
1001.
[4] (a) L.K. Johnson, C.M. Killian, M. Brookhart, J. Am. Chem.
Soc. 117 (1995) 6414. (b) L.K. Johnson, S. Mecking, M.
Brookhart, J. Am. Chem. Soc. 118 (1996) 267. (c) M. Brookhart,
L.K. Johnson, C.M. Killian, S. Mecking, D.J. Tempel, Polymer
.