Ethylene slurry polymerization using nickel diimine catalysts covalently-attached onto MgCl2-based supports

Article abstract of DOI:10.1016/j.polymer.2010.03.051

MgCl₂/alcohol adducts were recrystallized with alkyl aluminums and used as catalysts supports for nickel diimine complexes functionalized with amine groups. These supported catalysts were used to polymerize ethylene in a slurry reactor. The MgC

Full text of DOI:10.1016/j.polymer.2010.03.051

Polymer 51 (2010) 2271e2276
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Ethylene slurry polymerization using nickel diimine catalysts
covalently-attached onto MgCl2-based supports
*
Yiyoung Choi, João B.P. Soares
Department of Chemical Engineering, University of Waterloo, Ontario, Canada N2L 3G1
a r t i c l e i n f o
a b s t r a c t
Article history:
MgCl2/alcohol adducts were recrystallized with alkyl aluminums and used as catalysts supports for
nickel diimine complexes functionalized with amine groups. These supported catalysts were used to
polymerize ethylene in a slurry reactor. The MgCl2-based supported nickel diimine catalysts had higher
activities than the equivalent SiO2-supported nickel diimine catalysts, even without the use of activators.
These catalysts made polyethylene with melting temperatures and molecular weights higher than those
made with the equivalent homogeneous catalysts. Interestingly, the catalyst activity and polymer
molecular weight could be controlled by changing the support composition. In addition, covalent
chemical bonds between the functionalized nickel diimine complex and the MgCl2-based supports
avoided catalyst leaching during polymerization, leading to the production of polymer particles with
good morphology. The mechanical strength of the resulting polymer particles, however, was lower than
those made with SiO2-supported nickel diimine catalysts.
Received 12 February 2010
Received in revised form
25 March 2010
Accepted 27 March 2010
Available online 2 April 2010
Keywords:
Ethylene polymerization
Nickel diimine catalysts
Supported catalysts
Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved.
1. Introduction
catalysts. They made polyethylene with a spherical particle
morphology and small polydispersity index (PDI) [5,6].
Olefin polymerization processes in slurry and gas phase reactors
require the use of solid-supported catalysts [1,2]. The catalyst
should not leach from the support during polymerization, and the
catalyst structure, activity, and comonomer reactivity must be
maintained after supporting. Presently, the most common type of
support for heterogeneous ZieglereNatta catalysts is MgCl2. These
industrial catalysts are extremely active and produce polymer
particles with high density and excellent morphology.
Several researchers showed that MgCl2 could also be used as an
effective support for early and late transition metal single-site
catalysts. Xu et al. supported Fe(II) and Ni(II) catalysts onto spherical
MgCl2 particles by direct impregnation. The activities of the sup-
ported catalysts and the properties of the resultant polymers were
comparable to those obtained with SiO2-supported catalysts [3].
Nakayama et al. showed that MgCl2-based compounds could act as
good activators as well as supports for FI catalysts (bis(phenox-
yimine) catalysts). When compared to the corresponding homoge-
neous systemactivated with methylaluminoxane (MAO), FI catalysts
had higher activity, and made polymer with higher molecular
weight averages and controlled particle morphology [4]. Chadwick
et al. synthesized MgCl2-based compounds by the recrystallization
method, and then used them as supports for various single-site
Although the catalysts activities reported using physisorbed
SiO2-supported nickel diiminecatalystswereverylowandproduced
polymer with poor morphology [7,8], SiO2-supported nickel diimine
catalysts where the catalyst sites had been covalently bonded to the
support were very active, did not leach appreciably from the surface,
and did not cause reactor fouling [8e10].
In this work, MgCl2/alcohol adducts were crystallized using
several alkyl aluminums and used as supports for nickel diimine
complexes functionalized with amine groups. We propose that the
catalysts were covalently bonded to the support surface through
their amine functionalities, leading to high ethylene polymeriza-
tion activity and the production of polymer powders with good
morphologies. The polymer properties were controlled by varying
the polymerization conditions, the structure of the nickel diimine
complexes, and the chemical components of the support. The
microstructures of the resultant polymers were analyzed by several
techniques and the morphologies of the polymer particles were
characterized by scanning electron microscope (SEM).
2. Experimental
2.1. Materials
All operations were performed under nitrogen (99.999%, from
Praxair) using standard Schlenk techniques or inside a glove-box
* Corresponding author. Fax: þ1 519 746 4979.
E-mail address: jsoares@uwaterloo.ca (J.B.P. Soares).
0032-3861/$ e see front matter Crown Copyright Ó 2010 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2010.03.051

2272
Y. Choi, J.B.P. Soares / Polymer 51 (2010) 2271e2276
(Vacuum Atmosphere Company, Nexus). Polymer grade ethylene
(99.9%, from Praxair) and nitrogen were purified by passing
through columns packed with R3-11 copper catalyst, activated
alumina and 3A/4A mixed molecular sieves. Materials for nickel
diimine complex synthesis, anhydrous MgCl2, methanol (99.8%),
isopropyl alcohol (99.5%) and 2-ethyl hexanol (99.6%) were
purchased from Aldrich and used without further purification.
Trimethyl aluminum (TMA, 2 M in hexane), and triisobutyl
aluminum (TIBA, 1 M in hexane) were purchased from Aldrich.
Ethyl aluminum sesquichloride (EASC, 97%) was kindly donated by
Akzo Nobel. Solvents for catalyst synthesis and polymerization
were purified by passing through columns packed with molecular
sieves (Zeolum Type F-9, Tosoh Co.). All purified solvents were
stored in Schlenk flasks with 3A/4A mixed molecular sieves.
were added to a 100 mL Schlenk flask. A volume of 0.03 mL sulfuric
acid as reaction catalyst was added dropwise to the flask. The
mixture was stirred and refluxed for 3 h. The insoluble residue was
removed by filtration while still hot and dried overnight under
vacuum. A volume of 200 mL cold hexane was added to the flask
containing the dried product and stirred for 2 h. The solid fraction
was filteredand washed with 100 mL cold hexane, and vacuum dried
overnight. A dried yellow powder (2.90 g, 77% yield) was obtained.
An amount of 1.6 mmol nickel(II) dibromide-ethyleneglycol
dimethylether complex (NiBr2(DME)) and 1.8 mmol bis(4-amino-
2,3,5,6-tetramethylimino)butane were added to a 50 mL Schlenk
flask, then dissolved with 20 mL dichloromethane. After the
mixture was stirred for 3 days at room temperature, dichloro-
methane was removed by vacuum. The product was washed three
times with 20 mL diethyl ether each time, and then dried under
vacuum. A dried brown powder (0.7 g, 67% yield) was isolated and
named N2.
2.2. Synthesis of nickel diimine complexes
(Elemental Analysis result (N2, C24H34N4Br2Ni) (599.5): Calcu-
lated: C 48.1, H 6.1, N 9.3; Found: C 47.8, H 6.2, N 9.6).
The syntheses of nickel diimine complexes followed procedures
published in the literature and will be briefly described below
[8,10].
Fig. 1 shows the chemical structures of the N1 and N2 catalysts.
2.2.1. Bis(4-amino-2,3,5,6-tetramethylimino)acenaphtene nickel(II)
dibromide (N1)
2.3. Synthesis of MgCl2-based supported nickel diimine catalysts
An amount of 8.2 mmol acenaphtenequinone, 30.0 mmol
2,3,5,6,-tetramethyl-phenylene-1,4-diamine, and 50 mL toluene as
solvent were added to a 100 mL Schlenk flask. A volume of 0.03 mL
sulfuric acid as reaction catalyst was added dropwise to the flask.
The mixture was stirred and refluxed for 3 h. After the insoluble
residue was filtered while still hot, the product was dried under
vacuum. The dried powder was dissolved in 200 mL ethyl acetate,
400 mL hexane was added, and the mixture was stirred for 30 min.
This solution was kept in the freezer at ꢀ15 ꢁC overnight. The
precipitated product was filtered and washed with 50 mL cold
hexane, and vacuum-dried overnight. A dried red powder (2.75 g,
71% yield) was obtained.
An amount of 1.6 mmol nickel(II) dibromide-ethyleneglycol
dimethylether complex (NiBr2(DME)) and 1.8 mmol bis(4-amino-
2,3,5,6-tetramethylimino)acenaphtene were added to a 50 mL
Schlenk flask, then dissolved with 20 mL dichloromethane. After
the mixture was stirred for 3 days at room temperature, dichloro-
methane was removed by vacuum. The product was washed three
times with 20 mL diethyl ether each time, and then dried under
vacuum. A dried black powder (0.8 g, 73% yield) was isolated and
named N1.
An amount of 10 mmol of anhydrous MgCl2, 30 mmol 2-ethyl
hexanol, and 25 mL hexane were added to a dried 225 mL glass
reactor equipped with a mechanical stirrer (Lab-Crest Glass Pres-
sure Reaction Vessels, Andrews Glass Co.). After the mixture was
maintained at 100 ꢁC overnight and cooled to room temperature,
a clear MgCl2/alcohol adduct solution was obtained. An amount of
50 mmol TMA was added dropwised into the solution under 100
rpm. When 10% of TMA was added, the MgCl2/alcohol adduct begun
to crystallize. After complete addition of the TMA for 1 h, the
mixture was allowed to react for 4 more hours. The stirring was
stopped, the supernatant liquid was removed with a cannula, and
the solid product was washed two times with 20 mL dichloro-
methane. An amount of 0.08 mmol N1 was dissolved in 20 mL
dichloromethane, added the precipitated solid, and allowed to
react for 15 h. After the catalyst particles were precipitated, the
black color of N1 solution disappeared, and a transparent solvent
phase was separated. This indicates that N1 was almost completely
supported onto the MgCl2-based support. The supported catalyst
was filtered, washed with 20 mL dichloromethane, and then dried
under vacuum. This supported catalyst was named MgCl2-based
supported N1 catalyst (N1Mg). MgCl2-based supported N2 catalyst
(N2Mg) was synthesized according to same procedure using N2.
Table 1 lists the chemical components of all supported catalysts
made with this procedure.
(Elemental Analysis result (N1, C32H34N4Br2Ni) (695.6): Calcu-
lated: C 55.3, H 5.3, N 8.1; Found: C 55.0, H 5.2, N 8.0).
2.2.2. Bis(4-amino-2,3,5,6-tetramethylimino)butane nickel(II)
dibromide (N2)
An amount of 10 mmol 2,3 butanedione, 30.0 mmol 2,3,5,6,-
tetramethyl-phenylene-1,4-diamine, and 50 mL benzene as solvent
The N1 catalyst was also supported on SiO2 to serve as a refer-
ence for comparison with the SiO2-supported catalysts. One gram
of silica (Grace Davison, XPO 2410) and 20 mL toluene were
introduced in a same glass reactor. The reactor was heated to 60 ꢁC
H2N
N
N
NH2
H2N
N
N
NH2
Ni
Ni
Br
Br
Br
Br
N1
N2
Fig. 1. Structures of nickel diimine complexes: bis(4-amino-2,3,5,6-tetramethylimino)acenaphtene nickel(II) dibromide (N1) and bis(4-amino-2,3,5,6-tetramethylimino)butane
nickel(II) dibromide (N2).

Y. Choi, J.B.P. Soares / Polymer 51 (2010) 2271e2276
2273
Table 1
Description of prepared MgCl2-based supported nickel diimine catalysts.
Entry
Used Chemical Components
Characterization Results
Supporting Efficiencya
Ni complex
(mmol)
MgCl2
(mmol)
Alcohol
(mmol)
Alkyl
aluminum
(mmol)
Ni
(mmol/g
catalyst)
Mg
(mmol/g
catalyst)
Cl
O
Al
Mg
(wt.-%)
Cl
(wt.-%)
O
Al
(wt.-%)
(mmol/g
catalyst)
(mmol/g
catalyst)
(mmol/g
catalyst)
(wt.-%)
1
2
N1 0.08
N1 0.08
10
10
2-Ethyl
hexanol 30
2-Ethyl
TMA 50
0.053
6.21
8.61
17.59
0.85
93.7
83.1
65.0
56.4
88.5
88.3
2.6
3.6
TIBA 50
0.049
5.09
6.91
16.23
1.11
hexanol 30
Methanol 30
Methanol 30
Isopropyl
alcohol 30
2-Ethyl
3
4
5
N1 0.08
N1 0.08
N1 0.08
10
10
10
TMA 50
TIBA 50
TMA 50
0.053
0.051
0.049
6.05
5.75
5.76
8.44
7.51
9.60
17.88
18.58
19.81
1.04
0.99
0.73
90.9
90.7
94.2
64.0
58.9
78.4
90.0
97.2
92.6
3.2
3.2
2.4
6
N2 0.08
10
TMA 50
0.055
6.14
7.74
16.89
0.82
91.4
56.2
81.9
2.4
hexanol 30
a
Calculated by assuming 100% Ni deposition for the supported chemical components.
at a stirring rate of 100 rpm, then 20 mmol TMA was slowly added
and reacted for 15 h. After stopping the stirrer, the supernatant
liquid was removed with a cannula, the product was washed two
times with 20 mL toluene. An amount of 0.08 mmol N1 was added
to the product with 20 mL dichloromethane, and allowed to react
for 15 h under 100 rpm at room temperature. The produced sup-
ported catalyst (N1Si) was filtered and washed with 20 mL
dichloromethane, and dried under vacuum. The analyzed Ni
content was 0.65 wt.-%.
characterize the sample. The polymer molecular weight distribution
was determined by gel permeation chromatography (GPC, Polymer
Char). Samples were dissolved at 145 ꢁC with 1,2,4-trichlorobenzene
and passed through three linear Polymer Laboratories columns
which were calibrated with polystyrene standards and operated
with a flow rate of 1 mL/min. Short chain branching distributions
(SCBD) were determined by crystallization analysis fractionation
(CRYSTAF, Polymer Char). Samples were dissolved at 160 ꢁC for 30
min followed by 60 min of equilibration at 95 ꢁC. The cooling rate
during crystallizationwas 0.1 ꢁC/min, from 95 ꢁC to 30 ꢁC. A SEM was
used to investigate the morphology of polymer particles.
2.4. Characterization of supported nickel diimine catalysts
Ni, Al and Mg contents on the supported catalysts were
measured by inductively coupled plasma-optical emission spec-
troscopy (ICP, Teledyne Leeman Labs, Prodigy High Dispersion ICP).
A mass of 20 mg of supported catalyst was dissolved in 10 mL 3 N
HNO3 solution. The solution was diluted with distilled water prior
to analysis. A LEO 1530 scanning electron microscope (SEM)
combined energy dispersive X-ray (EDX) analysis was used to
investigate the morphology of the catalysts and organic elements of
the supported catalysts. For the SEM/EDX analysis, the supported
catalysts particles were embedded onto an epoxy resin and cut into
slices of 120 nm thicknesses with a diamond knife.
3. Results and discussion
3.1. Structure of MgCl2-based supported nickel diimine catalysts
Fig. 2 describes the preparation steps for the synthesis of N1Mg.
The reaction of TMA with MgCl2/methanol adduct in hexane
diluent leads to the precipitation of the MgCl2-based support. The
N1 complex is likely supported through one or two of its amine
functional groups, creating covalent bonds with the MgCl2-based
support.
Fig. 3 shows a SEM image of a spherical catalyst particle con-
taining several wide cracks. These cracks may be formed when the
epoxy-embedded particle was microtomed with the diamond
knife, and indicates that its mechanical strength is not very high.
Table 1 lists the chemical components and characterization
results for MgCl2-based supported catalysts investigated in this
2.5. Polymerization
Polymerizations were carried out in a 300 mL semi-batch
autoclave, equipped with a mass flow controller and a temperature
control unit consisting of a cooling coil and an electric heater. The
polymerization temperature was maintained within ꢂ0.2 ꢁC of the
set point. Prior to each reaction, the reactor was purged several
times with nitrogen, then heated to 120 ꢁC under vacuum and
purged again to the set point temperature under nitrogen flow. A
volume of 150 mL hexane and 1 mmol EASC or 0.1 mmol TIBA was
transferred to the reactor. After introducing the supported catalyst
in hexane slurry, polymerization took place under a continuous
ethylene flow to meet the desired pressure. At the end of the
polymerization time, the reactor was rapidly vented and the
resultant polymer was precipitated and washed with acidified
ethanol, then filtered and dried under vacuum.
Me
Me
Al
Me
O
OMe
Cl
Cl
TMA
MgCl2/methanol
adduct solution
Mg
Mg
Mg
O
Me
Recrystallized MgCl2-based support
HN
N
N
NH2
Me
Ni
Br Br
Al
OMe
2.6. Polymer characterization
Me
O
Cl
Cl
N1
Mg
Mg
Mg
The catalyst activity was calculated as the mass of polymer
product divided by mmol catalyst per hour. The melting tempera-
ture (Tm) of the resultant polymer was characterized using differ-
ential scanning calorimetry (DSC, TA Instruments Q2000). The
second scan was done at a heating rate of 10 ꢁC/min and used to
O
Me
MgCl2-based supported nickel diimine catalyst
Fig. 2. Preparation of MgCl2-based supported nickel diimine catalyst (N1Mg).

2274
Y. Choi, J.B.P. Soares / Polymer 51 (2010) 2271e2276
14
homogeneous catalyst (Run 1, Table 2)
MgCl2-based supported catalyst (Run 5, Table 2)
12
10
8
6
4
2
0
-2
35 40 45 50 55 60 65 70 75 80 85 90 95
Temperature(oC)
Fig. 4. Crystaf (SCBD) profiles of polyethylene made with N1 and N1Mg.
functional groups and impurities present on the catalyst surface, or
to steric hindrance caused by the support [11]. In the present
investigation, however, the activities of the supported catalysts
were comparable to those of the homogeneous catalysts. It seems
that the N1 and N2 complexes are covalently bonded to the support
through their amine groups, forming highly active and strongly
immobilized metal centers [8e10,12]. The homogeneous N1 cata-
lyst was more active, and produced polymer with lower Tm, than
the homogeneous N2 catalyst; these trends were the same when
N1 and N2 were supported on SiO2 or MgCl2-based supports.
Interestingly, the N1Mg had higher activity than N1Si (Runs 3 and
5, Table 2), and its activity was nearly 90% of the homogeneous
catalysts (Runs 1 and 5, Table 2). The activity of N2Mg was even
slightlyhigherthanthatof thehomogeneousN2complex(Runs 2and
7, Table 2). BothN1Mgand N2Mgalsohad considerablehighactivities
in the absence of the EASC activator. The Lewis acidic MgCl2-based
supports may be able to activate nickel diimine complexes, as
reported by Marks et al. [13]. The melting temperature (Tm) and
weight average molecular weight (Mw) of the polymer made with the
supported catalysts without activator were also a little higher than
those made in the presence of the EASC, indicating that EASC favored
Fig. 3. SEM micrograph of a N1Mg particle (Entry 1 in the Table 1, scale bar 5 mm).
study. Compared with the content of other elements, the fraction of
Al on the support is relatively low. It seems that the alkyl aluminum
acts mostly as a dealcoholation reagent, helping crystallize the
MgCl2/alcohol adducts, and that only a small fraction of the alkyl
aluminum molecules become chemically-bonded onto the MgCl2-
based support surface. It is also possible that excess alkyl aluminum
molecules are removed during the catalyst washing steps.
3.2. Ethylene polymerization
Ethylene polymerizations using homogeneous and SiO2-sup-
ported nickel diimine catalysts were conducted to serve as a refer-
ence for the polymerizations using MgCl2-based supported nickel
diimine catalysts. Table 2 summarizes these polymerization results.
In general, a significant reduction on catalyst activity is observed
when homogeneous catalysts are supported on several different
carriers. This behavior has been attributed to catalyst poisoning by
b-hydride elimination and chain walking.
Comparative CRYSTAF results are shown in Fig. 4. Poly-
ethylenes made with N1Mg and N1 had different microstruc-
tures. The polymer made with N1 (Run 1, Table 2) had a broader
Table 2
Comparison of ethylene polymerization results with homogeneous and supported
nickel diimine catalysts.
Table 3
Run
Polymerization
Conditions
Results
Ethylene polymerization results using MgCl2-based supported nickel diimine
catalysts in the absence of activator.
Catalyst
Activator
Activitya
Mw
PDIb
Tm (ꢁC)
Run Polymerization Conditions
Results
(g/mol ꢃ 10ꢀ3
)
Catalyst Pressure Temperature Activitya Mw
PDIb Tm
)
(ꢁC)
1
2
3
4
5
6
7
8
N1
EASC
41 200
6900
29 600
1460
36 450
12 550
7250
60
67
2.8
3.3
5.8
4.9
5.5
5.5
5.8
5.9
115.2
122.8
126.7
127.1
125.5
126.6
123.5
124.4
(psig)
(ꢁC)
(g/mol ꢃ 10ꢀ3
202
140
84
110
334
172
211
N2
EASC
N1Si
N1Si
N1Mg
N1Mg
N2Mg
N2Mg
EASC
Not used
EASC
Not used
EASC
Not used
143
167
127
140
154
172
1
2
3
4
5
6
7
N1Mg
N1Mg
N1Mg
N1Mg
N2Mg
N2Mg
N2Mg
150
150
150
50
150
150
50
30
60
80
60
30
60
30
4590
12 550
7690
4400
3620
4.4 129.6
5.5 126.6
5.3 124.1
5.3 123.8
5.2 129.0
5.9 124.4
5.8 126.4
4960
4960
1470
Polymerization conditions: homogeneous catalyst 0.5
mmol, supported catalyst
10e20 mg, EASC 1 mmol, TIBA 0.1 mmol in Runs 4, 6, 8; solvent 150 mL hexane,
ethylene pressure 150 psig, temperature 60 ꢁC, time 15 min; MgCl2-based support:
MgCl2/2-ethyl hexanol/TMA.
Polymerization conditions: catalyst 30e40 mg, TIBA 0.1 mmol, solvent 150 mL
hexane, time 15 min.
MgCl2-based support: MgCl2/2-ethyl hexanol/TMA.
a
a
Activity in kg PE/(mol Ni ꢃ h).
Activity in kg PE/(mol Ni ꢃ h).
b
b
Polydispersity index (Mw/Mn).
Polydispersity index (Mw/Mn).

Y. Choi, J.B.P. Soares / Polymer 51 (2010) 2271e2276
2275
Table 4
Effect of the support components using N1 complex on ethylene polymerization.
Run
Conditions
Results
MgCl2-based supports
Alcohol
Temperature (ꢁC)
Activitya
Mw (g/mol ꢃ 10ꢀ3
)
PDIb
Tm (ꢁC)
Alkyl Aluminum
1
2
3
4
5
6
7
8
9
2-Ethyl hexanol
2-Ethyl hexanol
Methanol
TMA
TIBA
TMA
TIBA
TMA
TMA
TIBA
TMA
TIBA
TMA
30
30
30
30
30
60
60
60
60
60
7590
7290
12 890
8970
204
273
199
213
198
140
154
144
152
145
5.4
5.8
5.6
5.1
5.3
5.5
6.3
6.4
5.7
5.6
129.6
130.8
131.1
129.7
129.9
126.5
128.0
128.2
126.9
127.0
Methanol
Isopropyl alcohol
2-ethyl hexanol
2-Ethyl hexanol
Methanol
Methanol
Isopropyl alcohol
8300
12 550
11 080
18 200
14 290
13 150
10
Polymerization conditions: catalyst 30e40 mg, TIBA 0.1 mmol, solvent 150 mL hexane, time 15 min.
a
Activity in kg PE/(mol Ni ꢃ h).
b
Polydispersity index (Mw/Mn).
SCBD in the temperature range 40e75 ꢁC, while polymer made
with N1Mg (Run 5, Table 2) had a narrower SCBD (80e90 ꢁC)
with a high crystallization temperature peak (>85 ꢁC). It seems
that strong steric effects between support and N1 depress chain
walking [14] rates and, therefore, produce polymer chains having
lower short chain branch content and higher Tm. This is also
a clear indication that the catalyst is firmly anchored onto the
support surface.
Ethylene polymerization results under different polymerization
conditions in the absence of activator are shown in Table 3. As the
polymerization temperature increases, a maximum activity is
reached at 60 ꢁC, and then decreases for N1Mg. On the other hand,
Fig. 5. SEM micrographs of polymer particles: a) polyethylene particles using N1Mg (Run 5 in Table 2, scale bar 200
mm); b) polyethylene particles made with N1Si (Run 3 in Table 2,
scale bar 200 m); c) A closer look at the external surface of a) (scale bar 2 m); d) A closer look at the external surface of b) (scale bar 2 m
m
m
m).

2276
Y. Choi, J.B.P. Soares / Polymer 51 (2010) 2271e2276
the catalyst activity is not very sensitive to temperature variations
for N2Mg. According to the chain walking mechanism, homoge-
neous nickel diimine catalysts produce polymers with higher Tm
and higher Mw as the ethylene pressure increases or as the poly-
merization temperature decreases. The MgCl2-based supported
nickel diimine catalysts, N1Mg and N2Mg, obey the same trends
observed for homogeneous nickel diimine catalysts.
To compare the effect of support structure on catalyst activity,
various alcohols and alkyl aluminums were used to synthesize
MgCl2-based supported nickel diimine catalysts; these polymeri-
zation results are shown in Table 4. MgCl2-based supports, in
combination with methanol/TMA, produced catalysts with the
highest activity. It seems that a favorable electronic environment,
based on the acidic nature of the support, was generated when
methanol/TMA was used. The supported catalyst, in combination
with 2-ethyl hexanol/TIBA, made polymer with the highest Mw. It is
likely that TIBA, with bulky alkyl groups, has a lower chain transfer
rate than TMA, and therefore produces polyethylene with higher
Mw [15]. Regardless of the support type, the catalysts activities at 60
ꢁC were higher than at 30 ꢁC, but Tm and Mw were lower.
4. Conclusions
Ethylene slurry polymerizations were performed using various
MgCl2-based supported nickel-diimine catalysts. Nickel diimine
complexes having amine functional groups were covalently
attached onto MgCl2-based supports, producing highly active and
strongly immobilized nickel diimine complexes. The Lewis acidic
MgCl2-based supports were able to act as activators for the nickel
diimine complexes, leading to active polymerization catalysts even
in the absence of external alkyl aluminum activators. The chemical
structure of the support had an influence on catalyst activity and
microstructure of the resultant polymers. The polymers made with
MgCl2-based supports resulted in free-flowing particles that did
not cause reactor fouling, but were weaker and more porous than
the ones made with the equivalent SiO2-supported catalyst.
References
[1] Mckenna TF, Soares JBP. Chem Eng Science 2001;56:3931e49.
[2] Hlatky GG. Chem Rev 2000;100:1347e76.
[3] Xu R, Liu D, Wang S, Mao B. Macromol Chem Phys 2006;207:779e86.
[4] Nakayama Y, Saito J, Banko H, Fujita T. Chem Eur J 2006;12:7546e56.
[5] Severn JR, Chadwick JC. Macromol Rapid Comm 2004;25:1024e8.
[6] Severn JR, Chadwick JC. Macromolecules 2004;37:6258e9.
[7] Alobaidi F, Ye Z, Zhu S. Macromol Chem Phys 2003;204:1653e9.
[8] Choi Y, Soares JBP. Macromol Chem Phys 2009;210:1979e88.
[9] Preishuber-Pflugl P, Brookhart M. Macromolecules 2002;35:6074e6.
[10] Schrekker HS, Kotov V, Preishuber-Pflugl P, White P, Brookhart M. Macro-
molecules 2006;39:6341e54.
[11] Wilke G. Angew Chem Int Ed 1988;27:185e206.
[12] Schneider H, Puchta GT, Kaul FAR, Sieber GR, Lefebvre F, Saggio G, et al. J Mol
Catal A: Chem 2001;170:127e41.
[13] Finch WC, Gillespie RD, Hedden D, Marks TJ. J Am Chem Soc 1990;112:6221e32.
[14] Gate DP, Svejda SA, Onate E, Killian CM, Johnson LK, White PS, et al. Macro-
molecules 2000;33:2320e34.
[15] Jamjah R, Zohuri GH, Javaheri M, Nekoomanesh M, Ahmadjo S, Farhadi A.
Macromol Symp 2008;274:148e53.
SEM micrographs of the polymer particles made with both
MgCl2-based supported catalyst and SiO2-supported catalyst are
compared in Fig. 5. There was no reactor fouling during the poly-
merizations, and free-flowing polymer particles were always
obtained. Fig. 5a shows the polymer particles made using N1Mg
(Run 5, Table 2) had rough and apparently frail surfaces, resulting in
particles with irregular shapes and debris. In Fig. 5c, the expanded
view of Fig. 5a, the primary particles that compose the larger
secondary particles appear as relatively flat stacks with high
porosity, which lowers the bulk density of the secondary particles.
On the other hand, Fig. 5b depicts polymer particles made with
N1Si (Run 3, Table 2), where denser and roughly spherical primary
and secondary particles (Fig. 5d) are apparent.