Note: Descriptions are shown in the official language in which they were submitted.
CA 02228923 1998-02-06
"Process For Preparing
Supported Olefin Polymerization Catalyst"
FIELD OF THE INVENTION
This invention relates to a process to prepare a supported
catalyst which is highly active for ethylene polymerization. The catalyst
is particularly useful in slurry or gas phase polymerization processes.
BACKGROUND OF THE INVENTION
:L 0
The synthesis of supported catalyst components using an
organometallic complex having a cyclopentadienyl ligand and a
phosphinimine ligand is disclosed in a co-pending and commonly
assigned patent application entitled "Supported Phosphinimine-CP
Catalysts" ("Stephan et aP').
The Stephan et a! reference teaches the use of two different
2 o types of activators, namely methyl alumoxane ("MAO") or
triphenylcarbenium tetrakis (pentafluorophenyl) borate
("[Ph3C][B(C6F5)4]") and further teaches that the alumoxane (especially
MAO) is highly preferred because of the excellent catalyst activity
which MAO provides.
However, as will be appreciated by those skilled in the art, the
use of MAO has been associated with reactor continuity problems
(particularly reactor fouling) when used in supported form.
Accordingly, an active catalyst which utilizes a so-called "ionic
activator" would represent a useful addition to the commercial art.
Hlatky and Turner made a very elegant invention relating to the
use of "ionic activators" as co-catalysts for bis-cyclopentadienyl type
m:\mawsonj\psc\spec\9145can.doc
CA 02228923 1998-02-06
metallocenes (as disclosed in United States Patents ("USP") 5,153,157
and 5,198,401 ). Hlatky et al subsequently discovered that this type of
catalyst is useful in supported form, as disclosed in PCT patent
application WO 91I09882. The '9882 application teaches at examples
10-15 that the metal oxide support material may be pre-treated with an
aluminum alkyl prior to the deposition of the catalyst/co-catalyst.
io Similar treatment of the support with aluminum alkyl when using an
ionic activator is also disclosed in the following literature: USP
5,474,9962 (Takahashi et an; European Patent Application ("EPO")
628574 (Inatomi et a~; PCT application 97I31038 (Lynch et an; and
Polymer Preprints 1996, 37(1 ), p. 249 (Hlatky and Upton). In an
analogous disclosure, PCT application 94I07928 teaches the use of
MAO pretreatment of a silica support for a monocyclopentadienyl
catalyst which is activated with an ionic activator.
SUMMARY OF THE INVENTION
The invention provides a process for preparing a supported
olefin polymerization catalyst consisting of:
Step (1 ) reacting a particulate metal oxide support having surface
hydroxyl groups with a reactive organometallic agent so
3 o as to eliminate substantially all of said surface hydroxyl
groups;
Step (2) depositing onto the reaction product from said step (1 ) a
combination of:
(2.1 ) a catalyst which is an unbridged organometallic
complex comprising:
m:lmawsonj\psc\spec\9145can.doc 3
CA 02228923 1998-02-06
(i) a group 4 metal selected from Ti, Hf, and
Zr;
(ii) a cyclopentadienyl-type ligand;
(iii) a phosphinimine ligand; and
(iv) two univalent ligands, and
(2.2) an ionic activator.
p DETAILED DESCRIPTION
The organometallic complex of this invention includes a
cyclopentadienyl ligand. As used in this specification the term
"cyclopentadienyl" refers to a 5-member carbon ring having delocalized
bonding within the ring and typically being bound to the group 4 metal
(M) through covalent r~5 -bonds.
An unsubstituted cyclopentadienyl ligand has a hydrogen
bonded to each carbon in the ring. ("Cyclopentadienyl-type" ligands
also include hydrogenated and substituted cyclopentadienyls, as
discussed in detail later in the specification.)
In more specific terms, the group 4 metal complexes of the
present invention (also referred to herein as "group 4 metal complex"
or "group 4 OMC") comprise a complex of the formula:
so Cp
I
I(R')3-P=Nl - M - (~1)2
wherein M is selected from the group consisting of Ti, Zr, and Hf; Cp is
a cyclopentadienyl-type ligand which is unsubstituted or substituted by
up to five substituents independently selected from the group
consisting of a C1-1o hydrocarbyl radical or two hydrocarbyl radicals
m:\mawsonj\psclspec19145can.doc
CA 02228923 1998-02-06
taken together may form a ring which hydrocarbyl substituents or
cyclopentadienyl radical are unsubstituted or further substituted by a
halogen atom, a C1_8 alkyl radical, C1_8 alkoxy radical, a C6_10 aryl or
aryloxy radical; an amido radical which is unsubstituted or substituted
by up to two C1_$ alkyl radicals; a phosphido radical which is
unsubstituted or substituted by up to two C1_s alkyl radicals; silyl
o radicals of the formula -Si-(R2)3 wherein each R2 is independently
selected from the group consisting of hydrogen, a C1_s alkyl or alkoxy
radical, C6_1o aryl or aryloxy radicals; germanyl radicals of the formula
Ge-(R2)3 wherein R2 is as defined above; each L1 is independently
selected from the group consisting of a hydrogen atom, of a halogen
atom, a C1_10 hydrocarbyl radical, a C1-1o alkoxy radical, a C5_1o aryl
oxide radical, each of which said hydrocarbyl, alkoxy, and aryl oxide
radicals may be unsubstituted by or further substituted by a halogen
atom, a C1_s alkyl radical, C1_8 alkoxy radical, a C6_10 aryl or aryloxy
radical, an amido radical which is unsubstituted or substituted by up to
two C1_s alkyl radicals; a phosphido radical which is unsubstituted or
substituted by up to two C1_8 alkyl radicals, provided that L' may not be
a Cp radical as defined above.
so For reasons of cost, the Cp ligand in the group 4 metal complex
is preferably unsubstituted. However, if Cp is substituted, then
preferred substituents include a fluorine atom, a chlorine atom, C1_s
hydrocarbyl radical, or two hydrocarbyl radicals taken together may
form a bridging ring, an amido radical which is unsubstituted or
substituted by up to two C1_4 alkyl radicals, a phosphido radical which is
m:\mawsonj\psc\spec\9145can.doc rJ
CA 02228923 1998-02-06
unsubstituted or substituted by up to two C1_4 alkyl radicals, a silyl
radical of the formula -Si-(R2)3 wherein each R2 is independently
selected from the group consisting of a hydrogen atom and a C1_4 alkyl
radical; a germanyl radical of the formula -Ge-(R2)3 wherein each R2 is
independently selected from the group consisting of a hydrogen atom
and a C1_4 alkyl radical.
o Referring to the above formula, the [(R1)3-P=N] fragment is the
phosphinimine ligand. The ligand is characterized by (a) having a
nitrogen phosphorous double bond; (b) having only one substituent on
the N atom (i.e. the P atom is the only substituent on the N atom); and
(c) the presence of three substituents on the P atom. Each R1 is
preferably selected from the group consisting of a hydrogen atom, a
halide, preferably fluorine or chlorine atom, a C1_4 alkyl radical, a C1-4
;2 0
alkoxy radical, a silyl radical of the formula -Si-(R2)3 wherein each R2 is
independently selected from the group consisting of a hydrogen atom
and a C1_4 alkyl radical; and a germanyl radical of the formula -Ge-(R2)3
or an amido radical of the formula -N-(R2)2 wherein each R2 is
independently selected from the group consisting of a hydrogen atom
and a C1_4 alkyl radical. It is particularly preferred that each R' be a
so tertiary butyl radical.
The organometallic complex is "unbridged" (which is intended to
convey a plain meaning, namely that the phosphinimine ligand is not
bonded or bridged to the Cp ligand).
Each L' is a univalent ligand. The primary performance criterion
for each L' is that it doesn't interfere with the activity of the catalyst
m:\mawsonj\psc\spec\9145can.doc
CA 02228923 1998-02-06
system. As a general guideline, any of the non-interfering univalent
ligands which may be emplayed in analogous metallocene compounds
(e.g. halides, especially chlorine, alkyls, alkoxy groups, amido groups)
phosphido groups, etc.) may be used in this invention.
In the group 4 metal complex preferably each L1 is
independently selected from the group consisting of a hydrogen atom,
:LO a halogen, preferably fluorine or chlorine atom, a C1_6 alkyl radical, a
C1_6 alkoxy radical, and a C6_1o aryl oxide radical. For reasons of cost
and convenience it is preferred that each L1 is a halogen (especially
chlorine).
The supported catalyst components of this invention are
particularly suitable for use in a slurry polymerization process or a gas
phase polymerization process.
A typical slurry polymerization process uses total reactor
pressures of up to about 50 bars and reactor temperatures of up to
about 200°C. The process employs a liquid medium (e.g. an aromatic
such as toluene or an alkane such as hexane, propane or isobutane) in
which the polymerization takes place. This results in a suspension of
solid polymer particles in the medium. Loop reactors are widely used
so in slurry processes. Detailed descriptions of slurry polymerization
processes are widely reported in the open and patent literature.
The gas phase process is preferably undertaken in a stirred bed
reactor or a fluidized bed reactor. Fluidized bed reactors are most
preferred and are widely described in the literature. A concise
description of the process follows.
m:\mawsonjlpsc\spec\9145can.doc 7
CA 02228923 1998-02-06
In general, a fluidized bed gas phase polymerization reactor
employs a "bed" of polymer and catalyst which is fluidized by a flow of
monomer which is at least partially gaseous. Heat is generated by the
enthalpy of polymerization of the monomer flowing through the bed.
Unreacted monomer exits the fluidized bed and is contacted with a
cooling system to remove this heat. The cooled monomer is then
:1o recirculated through the polymerization zone, together with "make-up"
monomer to replace that which was polymerized on the previous pass.
As will be appreciated by those skilled in the art, the "fluidized" nature
of the polymerization bed helps to evenly distribute/mix the heat of
reaction and thereby minimize the formation of localized temperature
gradients (or "hot spots"). Nonetheless, it is essential that the heat of
reaction be properly removed so as to avoid softening or melting of the
polymer (and the resultant -and highly undesirable - "reactor chunks").
The obvious way to maintain good mixing and cooling is to have a very
high monomer flow through the bed. However, extremely high
monomer flow causes undesirable polymer entrainment.
An alternative (and preferable) approach to high monomer flow
is the use of an inert condensable fluid which will boil in the fluidized
3 o bed (when exposed to the enthalpy of polymerization), then exit the
fluidized bed as a gas, then come into contact with a cooling element
which condenses the inert fluid. The condensed, cooled fluid is then
returned to the polymerization zone and the boiling/condensing cycle is
repeated.
m:\mawsonj\psc\spec\9145can.doc
CA 02228923 1998-02-06
The above-described use of a condensable fluid additive in a
gas phase polymerization is often referred to by those skilled in the art
as "condensed mode operation" and is described in additional detail in
USP 4,543,399 and USP 5,352,749. As noted in the '399 reference, it
is permissible to use alkanes such as butane, pentanes or hexanes as
the condensable fluid and the amount of such condensed fluid should
io not exceed about 20 weight per cent of the gas phase.
Other reaction conditions for the polymerization of ethylene
which are reported in the '399 reference are:
Preferred Polymerization Temperatures: about 75°C to about
115°C (with the lower temperatures being preferred for lower
melting copolymers - especially those having densities of less
than 0.915 g/cc - and the higher temperatures being preferred
for higher density copolymers and homopolymers); and
Pressure: up to about 1000 psi (with a preferred range of from
about 100 to 350 psi for olefin polymerization).
The '399 reference teaches that the fluidized bed process is well
adapted for the preparation of polyethylene but further notes that other
monomers may also be employed. The present invention is similar
3 o with respect to choice of monomers.
Preferred monomers include ethylene and C3_12 alpha olefins
which are unsubstituted or substituted by up to two C1_6 alkyl radicals,
C$_12 vinyl aromatic monomers which are unsubstituted or substituted
by up to two substituents selected from the group consisting of C1-4
alkyl radicals, C4_12 straight chained or cyclic diolefins which are
m:\mawsonjlpsclspec19145can.doc
CA 02228923 1998-02-06
unsubstituted or substituted by a C1_4 alkyl radical. Illustrative non-
limiting examples of such alpha-olefins are one or more of propylene,
1-butene, 1-pentene, 1-hexene, 1-octene, and 1-decene, styrene,
alpha methyl styrene, p- t-butyl styrene, and the constrained-ring cyclic
olefins such as cyclobutene, cyclopentene, dicyclopentadiene
norbornene, alkyl-substituted norbornenes, alkenyl-substituted
to norbornenes and the like (e.g. 5-methylene-2-norbornene and 5-
ethylidene-2-norbornene, bicyclo-(2,2,1)-hepta-2,5-diene).
The polyethylene polymers which may be prepared in
accordance with the present invention typically comprise not less than
60, preferably not less than 70 weight % of ethylene and the balance
one or more C4_1o alpha olefins, preferably selected from the group
consisting of 1-butene, 1-hexene and 1-octene. The polyethylene
prepared in accordance with the present invention may be linear low
density polyethylene having a density from about 0.910 to 0.935 g/cc
or high density polyethylene having a density above 0.935 g/cc. The
present invention might also be useful to prepare polyethylene having
a density below 0.910 g/cc - the so-called very low and ultra low
density polyethylenes.
3 o The present invention may also be used to prepare co- and ter-
polymers of ethylene, propylene and optionally one or more diene
monomers. Generally, such polymers will contain about 50 to about
75 weight % ethylene, preferably about 50 to 60 weight % ethylene and
correspondingly from 50 to 25 weight % of propylene. A portion of the
monomers, typically the propylene monomer, may be replaced by a
m:\mawsonj\psclspec\9145can.doc 1 ~
CA 02228923 1998-02-06
conjugated diolefin. The diolefin may be present in amounts up to
weight % of the polymer although typically is present in amounts
from about 3 to 5 weight %. The resulting polymer may have a
composition comprising from 40 to 75 weight % of ethylene, from 50 to
weight % of propylene and up to 10 weight % of a diene monomer
to provide 100 weight % of the polymer. Preferred but not limiting
:~o examples of the dienes are dicyclopentadiene, 1,4-hexadiene, 5-
methylene-2-norbornene, 5-ethylidene-2-norbornene and 5-vinyl-2-
norbornene. Particularly preferred dienes are 5-ethylidene-2-
norbornene and 1,4-hexadiene.
The present invention unequivocally requires the use of a metal
oxide support. An exemplary list of support materials include metal
oxides such as silicas, alumina, silica-alumina, alumina-phosphate,
titanic and zirconia.
These metal oxide support materials initially contain surface
hydroxyl groups. Whilst not wishing to be bound by any particular
theory, it has been postulated that reactions between the surface
hydroxyl and the catalyst and/or ionic activator may "diminish or
extinguish catalyst activity" (Ref. Hlatky and Upton, Polymer Preprints
1996) 37(1 ), 249). Thus, the process of the present invention requires
a step in which these surface hydroxyls are treated with a "reactive
organometallic agent" so as to substantially eliminate the surface
hydroxyls. As used herein, the term "reactive organometallic agent" is
meant to describe any organometallic which will react with the surface
m:\mawsonj\psc\spec\9145can.doc 1 1
CA 02228923 1998-02-06
hydroxyls without producing a subsequent adverse affect upon the
activity of the catalyst. Most metal alkyls should satisfy these criteria.
An exemplary list includes aluminum alkyls (particularly the
inexpensive and commercially available aluminum alkyls such as
triethylaluminum, triisobutyl aluminum and tri n-hexyl aluminum) and
magnesium alkyls.
o The preferred support material is silica. It will be recognized by
those skilled in the art that silica may be characterized by such
parameters as particle size, pore volume and initial silanol
concentration. The pore size and silanol concentration may be altered
by heat treatment or calcining prior to treatment with the reactive
organometallic agent.
The preferred particle size, preferred pore volume and preferred
residual silanol concentration may be influenced by reactor conditions.
Typical silicas are dry powders having a particle size of from 1 to 200
microns (with an average particle size of from 30 to 100 being
especially suitable); pore size of from 50 to 500 Angstroms; and pore
volumes of from 0.5 to 5.0 cubic centimeters per gram. As a general
guideline, the use of commercially available silicas, such as those sold
3o by W.R. Grace under the trademarks Davison 948 or Davison 955, are
suitable.
The invention also requires an ionic activator. The ionic
activator is an activator capable of ionizing the group 4 metal complex
and may be selected from the group consisting of:
m:\mawsonj\psc\spec\9145can.doc 12
CA 02228923 1998-02-06
(i) compounds of the formula [R5]+ [B(R')4J- wherein B is a boron
atom, R5 is a cyclic C5_~ aromatic cation or a triphenyl methyl
cation and each R' is independently selected from the group
consisting of phenyl radicals which are unsubstituted or
substituted with from 3 to 5 substituents selected from the group
consisting of a fluorine atom, a C1_4 alkyl or alkoxy radical which
o is unsubstituted or substituted by a fluorine atom; and a silyl
radical of the formula -Si-(R9)3; wherein each R9 is
independently selected from the group consisting of a hydrogen
atom and a C1_4 alkyl radical; and
(ii) compounds of the formula [(R8)tZH]+[B(R')4]- wherein B is a
boron atom, H is a hydrogen atom, Z is a nitrogen atom or
phosphorus atom, t is 2 or 3 and R8 is selected from the group
consisting of C1_$ alkyl radicals, a phenyl radical which is
unsubstituted or substituted by up to three C1_4 alkyl radicals, or
one R$ taken together with the nitrogen atom may form an
anilinium radical and R' is as defined above; and
(iii) compounds of the formula B(R')3 wherein R' is as defined
above.
3 o In the above compounds preferably R' is a pentafluorophenyl
radical, and R5 is a triphenylmethyl cation, Z is a nitrogen atom and R8
is a C1_4 alkyl radical or R$ taken together with the nitrogen atom forms
an anilium radical which is substituted by two C1_4 alkyl radicals.
While not wanting to be bound by theory, it is generally believed
that the activator capable of ionizing the group 4 metal complex
m:~nawsonjlpsclspec\9145can.doc 13
CA 02228923 1998-02-06
abstract one or more L' ligands so as to ionize the group 4 metal
center into a cation (but not to covalently bond with the group 4 metal)
and to provide sufficient distance between the ionized group 4 metal
and the ionizing activator to permit a polymerizable olefin to enter the
resulting active site. In short the activator capable of ionizing the group
4 metal complex maintains the group 4 metal in a +1 valence state,
:LO while being sufficiently liable to permit its displacement by an olefin
monomer during polymerization. In the catalytically active form, these
activators are often referred to by those skilled in the art as
substantially non-coordinating anions ("SNCA").
Examples of compounds capable of ionizing the group 4 metal
complex include the following compounds:
triethylammonium tetra(phenyl)boron,
tripropylammonium tetra(phenyl)boron,
tri(n-butyl)ammonium tetra(phenyl)boron,
trimethylammonium tetra(p-tolyl)boron,
trimethylammonium tetra(o-tolyl)boron,
tributylammonium tetra(pentafluorophenyl)boron)
tripropylammonium tetra (o,p-dimethylphenyl)boron,
so tributylammonium tetra(m,m-dimethylphenyl)boron,
tributylammonium tetra(p-trifluoromethylphenyl)boron,
tributylammonium tetra(pentafluorophenyl)boron,
tri(n-butyl)ammonium tetra (o-tolyl)boron
N,N-dimethylanilinium tetra(phenyl)boron,
N,N-diethylanilinium tetra(phenyl)boron,
m:\mawsonj\psc\spec\9145can.doc 14
CA 02228923 1998-02-06
N,N-diethylanilinium tetra(phenyl)n-butylboron)
N,N-2,4,6-pentamethylanilinium tetra(phenyl)boron
di-(isopropyl)ammonium tetra(pentafluorophenyl)boron,
dicyclohexylammonium tetra (phenyl)boron
triphenylphosphonium tetra)phenyl)boron,
tri(methylphenyl)phosphonium tetra(phenyl)boron,
o tri(dimethylphenyl)phosphonium tetra(phenyl)boron,
tropillium tetrakispentafluorophenyl borate,
triphenylmethylium tetrakispentafluorophenyl borate,
benzene (diazonium) tetrakispentafluorophenyl borate,
tropillium phenyltris-pentafluorophenyl borate,
triphenylmethylium phenyl-trispentafluorophenyl borate,
benzene (diazonium) phenyltrispentafluorophenyl borate,
tropillium tetrakis (2,3,5,6-tetrafluorophenyl) borate,
triphenylmethylium tetrakis (2,3,5,6-tetrafluorophenyl) borate,
benzene (diazonium) tetrakis (3,4,5-trifluorophenyl) borate,
tropillium tetrakis (3,4,5-trifluorophenyl) borate,
benzene (diazonium) tetrakis (3,4,5-trifluorophenyl) borate,
tropillium tetrakis (1,2,2-trifluoroethenyl) borate,
3 o triphenylmethylium tetrakis (1,2,2-trifluoroethenyl) borate,
benzene (diazonium) tetrakis (1,2,2-trifluoroethenyl) borate,
tropillium tetrakis (2,3,4,5-tetrafluorophenyl) borate,
triphenylmethylium tetrakis (2,3,4,5-tetrafluorophenyl) borate,
and
benzene (diazonium) tetrakis (2,3,4,5-tetrafluorophenyl) borate.
m:\mawsonj\psc\spec\9145can.doc 15
CA 02228923 1998-02-06
Readily commercially available activators which are capable of
ionizing the group 4 metal complexes include:
N,N- dimethylaniliumtetrakispentafluorophenyl borate
("[Me2NHPh][B(C6F5)4]");
triphenylmethylium tetrakispentafluorophenyl borate
("[Ph3C][B(C6F5)4]"); and
to trispentafluorophenyl boron.
Catalysts prepared by the process of this invention are highly
active in the polymerization of ethylene as illustrated in the
accompanying examples. High catalyst activity is desirable because it
reduces the level of catalyst residue contained in the final product and
because it reduces the concentration of transition metal in the
polymerization reactor. However, the low concentration of transition
metal in the reactor also means that the polymerization process is
highly sensitive to trace amounts of impurities. Accordingly, it is
preferred to use poison scavengers in the polymerization process
when using the catalysts of this invention. The use of an
organometallic scavenger (especially an aluminum alkyl) is especially
preferred. Moreover, when the catalyst is used in the preferred
3o dichloride form, the organometallic scavenger may also serve as an
alkylating agent.
As previously noted, the metal oxide support must be initially
treated with the reactive organometallic agent so as to eliminate
substantially all of the surface hydroxyls on the support. This initial
pretreatment may be conveniently completed by adding a solution of
m:\mauvsonj\psc\spec19145can.doc 1 6
CA 02228923 1998-02-06
the reactive organometallic agent to the metal oxide support followed
by stirring for a sufficient amount of time to allow the organometallic
agent to react with the hydroxyls. It will be apparent to those skilled in
the art that this is a fairly trivial procedure. As a general guideline, a
stirring time of 30 minutes to 10 hours will be sufficient.
The treated support may then be recovered from the slurry by
:~o conventional techniques (such as filtration or evaporation of solvent)
followed by an optional wash of the treated support to remove any free
or excess amount of the reactive organometallic agent.
The catalyst and ionic activator are then co-deposited on the
.z o
treated support. Again, this is a trivial procedure for a skilled chemist.
A preferred method is to first prepare a solution of the catalyst and
activator in a hydrocarbon solvent and to then add this solution to the
treated support. This results in a slurry which is preferably stirred for
from 30 minutes to 8 hours, followed by recovery of the supported
catalyst by filtration and/or solvent evaporation. The mole ratio of the
ionic activator to the catalyst component is preferably from 0.5/1 to 2I1;
most preferably 1I1 (with the basis being the moles of group 4
transition metal in the catalyst to moles of substantially non-
so coordinating anion provided by the ionic activator).
The catalysts produced by the process of this invention are
highly active for ethylene polymerization. This is desirable because it
effectively reduces the amount of support material contained in the
polyethylene product. It will be appreciated by those skilled in the art
that it is desirable for supported catalysts to produce at least 3 x 103
m:\mawsonjlpsc\spec\9145can.doc 17
CA 02228923 1998-02-06
grams of polyethylene per gram of support material (otherwise, plastic
film which is subsequently produced form the polyethylene may have a
gritty and/or sandy appearance and texture). The productivity of a
supported catalyst (expressed on a support basis) may be influenced
within a certain range by increasing or decreasing the amount of the
transition metal catalyst on the support. For example, even if a
to transition metal catalyst has low activity, it may be possible to produce
a commercially useful supported catalyst by increasing the level of
transition metal on the support. However, there are limits to this
approach due to problems which are associated with obtaining a
satisfactory dispersion of the transition metal on the support. In
particular, it is preferred to use a transition metal concentration of less
than 5 millimoles per gram of support, especially less than 2, and most
preferably less than 1.
Further details are illustrated in the following non-limiting
examples.
EXAMPLES
Catalyst Preparation and Polymerization Testing
Using a Semi-Batch, Gas Phase Reactor
The catalyst preparation methods described below employ
typical techniques for the synthesis and handling of air-sensitive
materials. Standard Schlenk and drybox techniques were used in the
preparation of ligands, metal complexes, support substrates and
supported catalyst systems. Solvents were purchased as anhydrous
materials and further treated to remove oxygen and polar impurities by
m:lmawsonj\psc\spec19145can.doc i
CA 02228923 1998-02-06
contact with a combination of activated alumina, molecular sieves and
copper oxide on silica/alumina.
All the polymerization experiments described below were
conducted using a semi-batch, gas phase polymerization reactor of
total internal volume of 2.2 liters. Reaction gas mixtures, including
separately ethylene or ethylene/butene mixtures were measured to the
o reactor on a continuous basis using a calibrated thermal mass flow
meter, following passage through purification media as described
above. A pre-determined mass of the catalyst sample was added to
the reactor under the flow of the inlet gas. The catalyst was treated in-
situ (in the polymerization reactor) at the reaction temperature in the
presence of the monomers, using a metal alkyl complex which has
been previously added to the reactor to remove adventitious impurities.
Purified and rigorously anhydrous sodium chloride was used as a
catalyst dispersing agent.
The internal reactor temperature is monitored by a
thermocouple in the polymerization medium and can be controlled at
the required set point to +/- 1.0°C. The duration of the polymerization
experiment was one hour. Following the completion of the
3o polymerization experiment, the polymer was separated from the
sodium chloride and the yield determined.
Catalyst Preparation
Part 1.1
A commercially available silica support material (sold under the
tradename "Davison 955" by W.R. Grace) was mixed with a 35 weight
m:\mawsonj\psc\spec\9145can.doc 1 g
CA 02228923 1998-02-06
solution of triisobutyl aluminum ("TIBAL") in hexane. The
TIBAUsilica weight ratio was about 2/1 which provided a large molar
excess of the TIBAL to the hydroxyl groups on the silica. The mixture
was stirred overnight, followed by recovery of the TIBAL-treated
support by filtration and final washing.
Part 1.2
:LO In an inventive experiment, cyclopentadienyl titanium [tri (tertiary
butyl) phesphinimine] dichloride ("catalyst") was mixed with
[Me2NHPh][B(C6F5)4] ("ionic activator") in toluene (with the
catalyst/ionic activator mole ratio being 1I1 ).
Subsequently, the mixture was added to a toluene slurry of the
:20
TIBAL-treated silica support from Part 1.1 (0.1 millimole of titanium per
gram of silica). The resulting mixture was heated for 30 minutes at
80°C with stirring followed by removal of the solvent under vacuum.
Part 1.3 (Comparative)
Metallocene catalysts in which the cyclopentadienyl ligands are
substituted with alkyl groups, such as n-butyl, are well known to be
highly active (as disclosed in USP 5,324,800, "Welborn"). Thus, for
the comparative experiment, the procedures described in Part 1.2
3 o above were repeated except that bis [(n-butyl)-cyclopentadienyl]
zirconium dichloride was used as the catalyst.
Polymerization
Part 2.1 (Inventive)
The above described 2.2 liter polymerization reactor was initially
charged with a 160 g bed of sodium chloride (table salt, as a seed bed)
m:\mawsonj\psc\spec\9145can.doc 2p
CA 02228923 1998-02-06
and 0.5 ml of a 25 weight % solution of tri n-hexyl aluminum in hexane
and 20 mg of the supported catalyst from Part 1.2 above.
Polymerization was undertaken for 1 hour at 90°C and an ethylene
pressure of 200 pounds per square inch gauge. 120 grams of
polyethylene was produced, corresponding to a productivity of about
6 x 103 g of polyethylene per gram of catalyst per hour. This is
o substantially in excess of the 3 x 103 g of polyethylene per gram of
catalyst which is desirable for high quality film resins. In addition, the
very high activity corresponds to a residual titanium concentration in
the polyethylene of less than 1 part per million by weight.
Part 2.2
In a comparative polymerization experiment using 50 mg of the
catalyst from Part 1.3 and 1.0 ml of a 25 weight % solution of tri n-
hexyl aluminum in hexane, a catalyst productivity of 1.5 x 103 g
polyethylene per gram of catalyst per hour was observed (using the
same ethylene pressure and temperature as used in Part 2.1 ). This
polyethylene would not be suitable for producing high quality film due
to the high concentration of catalyst support material in the resin.
m:\mawsonj\psc\spec\9145can.doc 21
