Almost all polar substances exert a strong negative inuence on the polymerization.

COS and hydrogen sulde, particularly, are considered to be strong catalyst poisons, of
which traces of more than 0.2 vol ppm aect a catalysts activity. Neither the solvent nor
the gaseous monomer should contain water, carbon dioxide, alcohols, or other polar
substances in excess of 5 ppm. Purication may be carried out by means of molecular
sieves.
The termination of the polymerization reaction by the addition of carbon monoxide
is used to determine the active centers (sites) of the catalyst. Hydrogen is known to slightly
reduce the catalysts activity. Yet it is commonly used as an important regulator to lower
the molecular weights of the polyethene or polypropene produced.
II. POLYETHENE
The polymerization of ethene can be released by radical initiators at high pressures as
well as by organometallic coordination catalysts. The polymerization can be carried out
either in solution or in bulk. For pressures above 100 MPa, ethene itself acts as a solvent.
Both low- and high-molecular-weight polymers up to 10
6
g/mol can be synthesized by
either organometallic coordination or high pressure radical polymerization. The structure
of the polyethene diers with the two methods. Radical initiators give more-or-less
branched polymer chains, whereas organometallic coordination catalysts synthesize linear
molecules.
A. Radical Polymerization
Since the polymerization of ethene develops excess heat, radical polymerization on a
laboratory scale is best carried out in a discontinuous, stirred batch reactor. On a technical
scale, however, column reactors are widely used. The necessary pressure is generally
kept around 180 to 350 MPa and the temperature ranges from 180 to 350

C [2429].
Solvent polymerization can be performed at substantial lower pressures and at tem-
peratures below 100

C. The high-pressure polymerization of ethene proceeds via a radical

chain mechanism. In this case chain propagation is regulated by disproportionation or
recombination.
1
2
3
Copyright 2005 by Marcel Dekker. All Rights Reserved.
The rate constants for chain propagation and chain termination at 130

and
180 MPa can be specied as follows [30]:
M
p
5:93 103 L mol
1
s
1
M
t
2 108 L mol
1
s
1
Intermolecular and intramolecular chain transfer take place simultaneously. This
determines the structure of the polyethene. Intermolecular chain transfer results in long
exible side chains but is not as frequent as intramolecular chain transfer, from which
short side chains mainly of the butyl type arise [31,32].
Intermolecular chain transfer:
4
5
Intramolecular chain transfer:
6
7
Radically created polyethene typically contains a total number of 10 to 50 branches
per 1000 C atoms. Of these, 10% are ethyl, 50% are butyl, and 40% are longer side
chains. With the simplied formulars (6) and (7), not all branches observed could be
explained [33,34]. A high-pressure stainless steal autoclave (0.1 to 0.51 MPa) equipped
with an inlet and outlet valve, temperature conductor, stirrer, and bursting disk is used for
the synthesis. Best performance is obtained with an electrically heated autoclave [3541].
To prevent self-degeneration, the temperature should not exceed 350

C. Ethene
and intitiator are introduced by a piston or membrane compressor. An in-built sapphire
window makes it possible to observe the phase relation. After the polymerization is
nished, the reaction mixture is released in two steps. Temperature increases are due to
a negative JouleThompson eect. At 26 MPa, ethene separates from the 250

C hot
polymer melt. After further decompression down to normal pressure, the residual ethene
is removed [4246]. Reaction pressure and temperature are of great importance for
the molecular weight average, molecular weight distribution, and structure of the
polymer. Generally, one can say that with increasing reaction pressure the weight average
increases, the distribution becomes narrower, and short- and long-chain branching both
decrease [47].
Copyright 2005 by Marcel Dekker. All Rights Reserved.
Oxygen or peroxides are used as the initiators. Initiation is very similar to that
in many other free-radical polymerizations at dierent temperatures according to their
half-live times (Table 1). The pressure dependence is low. Ethene polymerization can also
be started by ion radiation [4851]. The desired molecular weight is best adjusted by the
use of chain transfer reagents. In this case hydrocarbons, alcohols, aldehydes, ketones, and
esters are suitable [52,53].
Table 2 shows polymerization conditions for the high-pressure process and density,
molecular weight, and weight distribution of the polyethene (LDPE). Bunn [54] was
the rst to study the structure of polyethene by x-ray. At a time when there was still
considerable debate about the character of macromolecules, the demonstration that
wholly synthetic and crystalline polyethene has a simple close-packed structure in which
the bond angles and bond lengths are identical to those found in small molecules such
Table 1 Peroxides as initiators for the high-pressure polymerization of ethene.
Peroxide Molecular weight Half-time period of 1 min
by a polymerization
temperature (

C)
(H
3
C)
3
-COOC(CH
3
)
3
146.2 190
174.2 110
146 115
216.3 130
286.4 120
230.3 160
246.4 100
194.2 120
194.2 170
234.3 90
Copyright 2005 by Marcel Dekker. All Rights Reserved.
as C
36
H
74
[5557], strengthened the strictly logical view that macromolecules are a
multiplication of smaller elements joined by covalent bonds. LDPE crystallizes in single
lamellae with a thickness of 5.0 to 5.5 nm and a distance between lamellae of 7.0 nm which
is lled by an amorphous phase. The crystallinity ranges from 58 to 62%.
Recently, transition metals and organometallics have gained great interest as
catalysts for the polymerization of olens [58,59] under high pressure. High pressure
changes the properties of polyethene in a wide range and increases the productivity
of the catalysts. Catalyst activity at temperatures higher than 150

C is controlled
primarily by polymerization and deactivation. This fact can be expressed by the practical
notion of catalyst life time, which is quite similar to that used with free-radical initiators.
The deactivation reaction at an aluminum alkyl concentration below 5 10
5
mol/l
seems to be rst order reaction [60]. Thus for various catalyst-activator systems, the
approximate polymerization times needed in a continuous reactor to ensure the best use
of catalyst between 150 to 300

C are between several seconds and a few minutes.

Several studies have been conducted to obtain ZieglerNatta catalysts with good thermal
stability. The major problem to be solved is the reduction of the transition metal
(e.g., TiCl
3
) by the cocatalyst, which may be aluminum dialkyl halide, alkylsiloxyalanes
[60], or aluminoxane [59].
Luft and colleagues [61,62] investigated high-pressure polymerization in the presence
of heterogeneous catalysts consisting of titanium supported on magnesium dichloride
or with homogeneous metallocene catalysts. With homogeneous catalysts, a pressure of
150 MPa (80 to 210

C) results in a productivity of 700 to 1800 kg PE/cat, molecular

weights up to 110 000 g/mol, and a polydispersity of 5 to 10, with heterogeneous catalysts,
whereas the productivity is 3000 to 7000 kg PE/cat, molecular weight up to 70 000 g/mol,
and the polydispersity 2.
B. Coordination Catalysts
Ethene polymerization by the use of catalysts based on transition metals gives a polymer
exhibiting a greater density and crystallinity than the polymer obtained via radical
polymerization. Coordination catalysts for the polymerization of ethene can be of very
dierent nature. They all contain a transition metal that is soluble or insoluble in
hydrocarbons, supported by silica, alumina, or magnesium chloride [5,63]. In most cases
cocatalysts are used as activators. These are organometallic or hydride compounds of
group I to III elements; for example, AlEt
3
, AlEt
2
Cl, Al(i-Bu)
3
, GaEt
3
, ZnEt
2
, n-BuLi,
amyl Na [64]. Three groups are used for catalysis:
1. Catalysts based on titanium or zirconium halogenides or hydrides in connection
with aluminum organic compound (Ziegler catalysts)
Table 2 Polymerization conditions and product properties of high-pressure polyethene (LDPE).
Pressure
(MPa)
Temp.
(

C)
Regulator
(propane) (wt%)
Density
(g/cm
3
)
Molecular
weight MFI
Distribution
165 235 1.6 0.919 1.3 20
205 290 1.0 0.915 17.0 10
300 250 3.9 0.925 2.0 10
Source: Ref. 29.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
2. Catalysts based on chromium compounds supported by silica or alumina
without a coactivator (Phillips catalysts)
3. Homogeneous catalysts based on metallocenes in connection with aluminoxane
or other single site catalysts such as nickel ylid, nickel diimine, palladium, iron or
cobalt complexes.
Currently, mainly Ziegler and Phillips catalysts as well as some metallocene catalysts
[63] are generally used technically.
Three dierent processes are possible: the slurry process, the gas phase process, and
the solvent process [6568]:
1. Slurry process. For the slurry process hydrocarbons such as isobutane,
hexane, n-alkane are used in which the polyethene is insoluble. The polymer-
ization temperature ranges from 70 to 90

C, with ethene pressure varying

between 0.7 and 3 MPa. The polymerization time is 1 to 3 h and the yield is
95 to 98%. The polyethene produced is obtained in the form of fine particles in
the diluent and can be separated by filtration. The molecular weight can be
controlled by hydrogen; the molecular weight distribution is regulated by
variation of the catalyst design or by polymerization in several steps under
varying conditions [6973]. The best preparation takes place in stirred vessels or
loop reactors.
In some processes the polymerization is carried out in a series of cascade
reactors to allow the variation of hydrogen concentration through the operating
steps in order to control the distribution of the molecular weights. The slurry
contains about 40% by weight polymer. In some processes the diluent is
recovered after centrifugation and recycled without purification.
2. Gas phase polymerization. Compared to the slurry process, polymerization
in the gas phase has the advantage that no diluent is used which simplifies
the process [7476]. A fluidized bed that can be stirred is used with supported
catalysts. The polymerization is carried out at 2 to 2.5 MPa and 85 to 100

C.
The ethene monomer circulates, thus removing the heat of polymerization
and fluidizing the bed. To keep the temperature at values below 100

C, gas
conversion is maintained at 2 to 3 per pass. The polymer is withdrawn periodi-
cally from the reactor.
3. Solvent polymerization. For the synthesis of low-molecular-weight poly-
ethene, the solvent process can be used [77,78]. Cyclohexane or another
appropriate solvent is heated to 140 to 150

C. After addition of the catalyst,

very rapid polymerization starts. The vessel must be cooled indirectly by water.
Temperature control is also achieved via the ethene pressure, which can be
varied between 0.7 and 7 MPa.
In contrast to high-pressure polyethene with long-chain branches, the polyethene
produced with coordination catalysts has a more or less linear structure (Figure 1) [79].
A good characterization of high-molecular-weight-polyethenes gives the melt rheological
behaviour [80] (shear viscosity, shear compliance). The density of the homopolyethenes
is higher but it can be lowered by copolymerization. Polymers produced with unmodi-
ed Ziegler catalysts showed extremely high molecular weight and broad distribution
[81]. In fact, there is no reason for any termination step, except for consecutive
reaction. Equations (8) to (11) show simplied chain propagation and chain termination
steps [11].
Copyright 2005 by Marcel Dekker. All Rights Reserved.
Chain propagation:
8
Chain termination:
(a) By b elimination with H transfer to monomer
9
(b) By hydrogenation
10
Figure 1 Comparison of various polyethenes.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
(c) By b elimination forming hydride
11
Termination via hydrogenation gives saturated polymer and metal hydride. The
termination of a growing molecule by an a-elimination step forms a polymer with an
olenic end group and a metal hydride. In addition, an exchange reaction with ethene
forming a polymer with an olenic end group and an ethyl metal is observed.
1. Titanium Chloride-Based Catalysts
The rst catalyst used by Ziegler et al. [5,82] for the polymerization of ethene was a
mixture of TiCl
4
and Al(C
2
H
5
)
3
, each of which is soluble in hydrocarbons. In combination
they form an olive-colored insoluble complex that is very unstable. Its behavior is very
sensitive to a number of experimental parameters, such as Al/Ti ratio, temperature and
time of mixing of all components, and absolute and relative concentrations of reactants
[83]. After complexation, TiCl
4
is reduced by a very specic reduction process. This
reduction involves alkylation of TiCl
4
with aluminum alkyl molecules followed by a
dealkylation reduction to a trivalent state:
Complexation: TiCl
4
AlEt
3
TiCl
4
AlEt
3
12
Alkylation: TiCl
4
:AlEt
3
EtTiCl
3
AlEt
2
Cl 13
Reduction: 2EtTiCl
3
2TiCl
3
Et
2
14
Under drastic conditions, TiCl
3
can be reduced to TiCl
2
in a similar way. The actual
TiCl
3
product is a compound alloyed with small amounts of AlCl
3
and probably some
chemisorbed AlEt
2
Cl. The mechanistic process is very complex and not well understood.
Instead of Al(C
2
H
5
)
3
, also Al(C
2
H
5
)
2
Cl, Al
2
(C
2
H
5
)
3
Cl
3
, or Al(i-Bu)
3
could be used.
These systems, called rst-generation catalysts, are used for the classic process of olen
polymerization. In practice, however, the low activity made it necessary to deactivate
the catalyst after polymerization, remove the diluent, and then remove the residues of
catalyst with HCl and alcohols. This treatment is followed by washing the polyethene
with water and drying it with steam. Purication of the diluent recovered and feedback
of the monomer after a purication step involved further complications. The costs of
these steps reduced the advantage of the low-pressure polymerization process. Therefore,
it was one of the main tasks of polyolen research to develop new catalysts (second
generation catalysts) that are more active, and can therefore remain in the polymer
without any disadvantage to the properties (Table 3) [84]. The process is just as sensitive
to perturbation, it is cheaper, and energy consumption as well as environmental loading
are lower. It is also possible to return to the polymerization vessel diluent containing
a high amount of the aluminum alkyl. The second generation is based on TiCl
3
compounds or supported catalysts MgCl
2
/TiCl
4
/Al(C
2
H
5
)
3
or CrO
3
(SiO
2
) (Phillips).
Copyright 2005 by Marcel Dekker. All Rights Reserved.
2. Unsupported Titanium Catalysts
There is a very large number of dierent combinations of aluminum alkyls and titanium
salts to make high mileage catalysts for ethene polymerization, such as a-TiCl
3
AlEt
3
,
AlEt
2
Cl, Al(i-Bu)
3
, and Ti(III)alkanolate-chloride Al(i-
hexyl
)
3
[85]. TiCl
3
exists in four
crystalline modications, the a, b, g, and d forms [86]. The composition of these TiCl
3
s
can be as simple as one Ti for as many as three Cl, or they can have a more complex
structure whereby a second metal is cocrystallized as an alloy in the TiCl
3
. The particular
method of reduction determines both composition and crystalline modication. a-TiCl
3
can be synthesized by reduction of TiCl
4
with H
2
at elevated temperatures (500 to 800

C)
or with aluminum powder at lower temperatures (about 250

C); in this case the a-TiCl

3
contains Al cations [87]. More active are g- and d-TiCl
3
modications. They are formed
by heating the a-TiCl
3
to 100 or 200

C. The preferred a-TiCl

3
contains Al and is
synthesized by reducing TiCl
4
with about 1/3 part AlEt
3
or 1 part AlEt
2
Cl. A modem
TiCl
3
catalyst has a density of 2.065 g/cm
3
, a bulk density of 0.82, a specic surface area
(BET) of 29 m
2
/g, and a particle size of 10 to 100 mm. The polymerization activity is in the
vicinity of 500 Lmol
1
s
1
[88].
3. Supported Catalysts
MgCl
2
/TiCl
4
catalysts. Good progress in increasing the polymerization activity was made
with the discovery of the MgCl
2
/TiCl
4
-based catalysts [89]. Instead of MgCl
2
, Mg(OH)Cl,
MgRCl, or MgR
2
[9094] can be used. The polymerization activity goes up to
10 000 Lmol
1
s
1
. At this high activity the catalyst can remain in the polyethene. For
example, the specic volume (BET) of the catalystis 60 m
2
/g [95]. The high activity is
accomplished by increasing the ethene pressure. The dependence is not linear as it was
for rst-generation catalysts, and the morphology is also dierent. The polyethene has a
cobweb-like structure, whereas rst generation catalysts produced a worm-like structure
[90,91]. The cobweb structure is caused by the fact that polymerization begins at the
surface of the catalyst particle. The particle is held together by the polymer. While
polymerization is in progress, the particle grows rapidly and parts of it break. Cobweb
structures are formed by this fast stretching process of the polyethene.
Table 3 Comparison of various catalyst processes for ethene polymerization.
First generation Second generation
Catalyst preparation Catalyst preparation
Polymerization Polymerization
Limited inuence to molecular weight and
weight distribution
Great variation of molecular weight and
weight distribution
Catalyst deactivation with alcohol
Filtration Filtration
Washing with water (HCl), wastewater treatment,
purication, and drying of diluent
Feedback of diluent
Drying of PE Drying of PE
Finishing Finishing
Thermal degradation of molecular weight, blending
Stabilization Stabilization
Source: Ref. 84.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
It is known that in the case of these supported catalysts the higher activity is linked
to a higher concentration of active titanium. In contrast to rst-generation catalysts in
which only 0.1 to 1% of all titanium atoms form active sites, in supported catalysts 20 to
80% of them are involved in the formation of active sites [97,98].
Solvay workers [99] have investigated extensively the supported Mg(OH)Cl/TiCl
4
/
AlEt
3
catalyst and related systems including MgSO
4
, MgOSiO
2
, and MgO. It is not clear
whether all of the Ti centers in the supported catalysts are isolated. The high activity
suggests the incorporation of small TiCl
3
crystallites into the Mg(OH)Cl. Fink and
Kinkelin [100] prepared a high-activity catalyst by combination of MgH
2
and TiCl
4
.
The MgH
2
has a much greater surface area (90 m
2
/g). It reacts with the TiCl
4
under
the evolution of hydrogene. By 30

C and 2 bar ethene pressure, 110 kg of PE per gram of

Ti could be obtained.
4. Phillips Catalyst
The widely investigated Phillips catalyst, which is alkyl free, can be prepared by impreg-
nating a silica-alumina (87:13 composition [101103] or a silica support with an aqueous
solution of CrO
3
). High surface supports with about 400 to 600 g/m
2
are used [104]. After
the water is removed, the powdery catalyst is uidized and activated by a stream of dry air
at temperatures of 400 to 800

C to remove the bound water. The impregnated catalysts

contain 1 to 5 wt% chromium oxides. When this catalyst is heated in the presence of
carbon monoxide, a more active catalyst is obtained [105]. The Phillips catalyst specically
catalyzes the polymerization of ethene to high-density polyethene. To obtain polyethene of
lower crystallinity, copolymers with known amounts of an a-olen, usually several percent
of 1-butene can be synthesized. The polymerization can be carried out by a solution,
slurry, or gas-phase (vapor phase) process.
The chromium oxide-silica is inactive for polymerizing ethylene at low temperatures
but becomes active as the temperature is increased from 196

C (the melting point for

CrO
3
) to 400

C. Interactions of chromium oxide with SiO

2
and Al
2
O
3
take place.
Hogan [103] calculated that for a silica support of 600 m
2
/g and about 5% Cr(VI),
the average distance between adjacent Cr atoms is 10 A

. This corresponds to the accepted

population of silanol groups on silica after calcination. The structures (15) and (16) are
proposed:
15
16
Copyright 2005 by Marcel Dekker. All Rights Reserved.
It has been calculated that between 0.1 and 0.4 wt% of the total chromium
forms active centers [105]. A dicult question relates to the valences of chromium in the
active sites. Valences of II, III, IV, V, and VI have been established [106]. Because of the
small number of total chromium atoms that are active centers, it has not been possible to
unequivocally assign the active valence [107,108]. Krauss and Hums [109] concluded that
the reduction of hexavalent chromium centers linked to support produced coordinately
unsaturated Cr(II) surface compounds. A speciality of the Phillips catalyst is that there is
no inuence of hydrogen to control the molecular weight of the polyethylene. Only by
higher activation temperatures can the molecular weight be lowered.
5. Homogeneous (Single Site) Catalysts
Among the great number of Ziegler catalysts, homogeneous systems have been
preferentially studied in order to understand the elementary steps of the polymerization
which is simpler in soluble systems than in heterogeneous systems. The situation
has changed since in recent years homogeneous catalyst based on metallocene and
aluminoxane [12,110], nickel and palladium diimin complexes [111], and iron and cobalt
compounds were discovered which are also very interesting for industrial and laboratory
synthesis. Some special polymers can only be synthesized with these catalysts.
In comparison to Ziegler systems, metallocene catalysts represent a great develop-
ment: they are soluble in hydrocarbons, show only one type of active site and their
chemical structure can be easily changed. These properties allow one to predict accurately
the properties of the resulting polyolens by knowing the structure of the catalyst used
during their manufacture and to control the resulting molecular weight and distribution,
comonomer content and tacticity by careful selection of the appropriate reactor condi-
tions. In addition, their catalytic activity is 10100 times higher than that of the classical
ZieglerNatta systems.
Metallocenes, in combination with the conventional aluminum alkyl cocatalysts used
in Ziegler systems, are indeed capable of polymerising ethene, but only at a very low
activity. Only with the discovery and application of methylaluminoxane (MAO) it was
possible to enhance the activity, surprisingly, by a factor of 10 000 [113]. Therefore, MAO
plays a crucial part in the catalysis with metallocenes.
Kinetic studies and the application of various methods have helped to dene the
nature of the active centers, to explain the aging eects of Ziegler catalysts, to establish
the mechanism of interaction with olens, and to obtain quantitative evidence of some
elementary steps [9,112115]. It is necessary to dierentiate between the soluble catalyst
system itself and the polymerization system. Unfortunately, the well-dened bis(cyclo-
pentadienyl)titanium system is soluble, but it becomes heterogeneous when polyethylene
is formed [116].
The polymerization of olens, promoted by homogeneous Ziegler catalysts based
on biscyclopentadienyltitanium(IV) or analogous compounds and aluminum alkyls,
is accompanied by a series of other reactions that greatly complicate the kinetic inter-
pretation of the polymerization process:
17
Copyright 2005 by Marcel Dekker. All Rights Reserved.
18
19
20
Concomitant with continued olen insertion into the metalcarbon bond of the
transition metal aluminum complex, alkyl exchange and hydrogen-transfer reactions
are observed. Whereas the normal reduction mechanism for transition metal organic
complexes is initiated by release of olens with formation of a hydride followed by hydride
transfer to an alkyl group, a reverse reaction takes place in the case of some titanium
and zirconium acompounds. A dimetalloalkane is formed by the release of ethane. In
second step, ethene is evolved from the dimetalloalkane:
TiIVCH
2
CH
2
TiIV !CH
2
CH
2
2TiIII 21
leaving two reduced metal atoms. Some of the aging processes occurring with homo-
geneous and heterogeneous Ziegler catalysts can be explained with the aid of these side
reactions.
Table 4 summarizes important homogeneous Ziegler catalysts. The best known sys-
tems are based on bis(cyclopentadienyl)titanium(IV), bis(cyclopentadienyl)zirconium(IV),
terabenzyltitanium, vanadium chloride, allyl metal, or chromium acetylacetonate with
trialkylaluminum, alkylaluminum halides, or aluminoxanes. Breslow [126] discovered that
bis(cyclopentadienyl)titanium(IV) compounds, which are easily soluble in aromatic
hydrocarbons, could be used instead of titanium tetrachloride as the transition metal
compound together with aluminum alkyls for ethene polymerization. Subsequent research
on this and other systems with various alkyl groups has been conducted by Natta [127],
Belov et al. [128,129], Patat and Sinn [130], Shilov [131], Henrici-Olive and Olive [132],
Reichert and Schoetter [133], and Fink et al. [134,135]. With respect to the kinetics of
polymerization and side reactions, this soluble system is probably the one that is best
understood. It is found that the polymerization takes place primarily if the titanium exists
as titanium(IV) [136,137]. According to Henrici-Olive and Olive [138], the speed
of polymerization decreases with increasing intensity of ESR signals of the developing
titanium(III) compound.
The increase in length of the polymer chain occurs by insertion of the monomer in to
a metalcarbon bond of the active complex. Dyachkovskii et al. [139] and Eisch et al. [140]
were the rst to believe, based on kinetic measurements and synthesis, that the insertion
takes place on a titanium cation. An ion of the type (C
5
H
5
)
2
Ti

-R, derived from

Copyright 2005 by Marcel Dekker. All Rights Reserved.
complexing and dissociation,
C
5
H
5

2
TiRCl AlRCl
2
C
5
H
5

2
TiRCl AlRCl
2
22
C
5
H
5

2
TiRCl AlRCl
2
C
5
H
5

2
TiRCl
3

AlRCl
3

23
could be the active species of polymerization. Sinn and Patat [137] drew attention to the
electron-decient character of those main-group alkyls that aord complexes with the
titanium compound. Fink and co-workers [141] showed by
13
C-NMR spectroscopy with
13
C-enriched ethene at low temperatures (where no alkyl exchange was observed) that
in higher halogenated systems, insertion of the ethene takes place only into a titanium
carbon bond.
At low polymerization temperatures with benzene as a solvent, Hocker and Saeki
[142] could prepare polyethene with a molecular weight distribution M
W
/M
n
1.07 using
the bis(cyclopentadienyl)titanium dichloride/diethylaluminum chloride system. The mole-
cular weight could be varied in a wide range by changing the polymerization temperature.
Using ally
4
Zr(allylZrBr
3
) at a polymerization temperature of 160

C (80

C) yields poly-
ethene with a density of 0.966 g/cm, M
n
of 10,500, (700), 3.0 CH
3
groups per 1000

Cand 0.4
vinyl groups. The benzene- and allyl-containing transition metals are working without any
cocatalyst and therefore are alkyl free. If transition metal organometallic compounds such
as Cr(allyl)
3
, Zr(allyl)
4
, Zr(benzyl)
4
, Ti(benzyl)
4
, and Cr(cyclopentadienyl)
2
are supported
on Al
2
O
3
Or SiO
2
, the activity increases by a factor of more than 100 [124,143].
Apparently, soluble catalysts are obtained by reaction of Ti(OR)
4
with AlR
3
[144].
High-molecular-weight polyethene is obtained in variable amounts, with Al/Ti ratios
ranging between 10 and 50. Similar results are attained by replacing titanium alkoxide
by Ti(NR
2
)
4
[145]. Soluble catalytic systems are also obtained by reaction of Ti(acac)
3
[146] and Cr(acac)
3
[147] with AlEt
3
as well as by reaction of Cr(acac)
3
and VO(acac)
2
with
AlEt
2
Cl in the presence of triethyl phosphite [121]. With vanadium catalysts the activity
reaches its maximum at Al/V ratio 50. Under these conditions up to 67% vanadium is in
the bivalent oxidation state. Bivalent and trivalent compounds will be active.
Table 4 Homogeneous catalysts for ethene polymerization.
System Transition metal
(M) compound
Polymerization
temperature (

C)
Normalized
activity
Catalyst
yield
Refs
Cp
2
TiCl
2
/AlMe
2
Cl
a
1:2.51:6 30 40200 117
Cp
2
TiCl
2
/AlMe
2
Cl/H
2
O 1:6:3 30 2000 117
Cp
2
TiCl
2
/AlEt
2
Cl 1:2 15 745 118
Cp
2
TiMe
2
/MAO 1:10
5
.5 10
2
20 35 000 >15 000 110
Cp
2
TiMe
2
/MAO 1:100 20 200 >5 000 119
Cp
2
ZrCl
2
/MAO 1:1000 70 400 000 >10 000 120
VO(acac)
2
/Et
2
AlCl/activator 1:50 20 180 121
Cp
2
VCl
2
/Me
2
AlCl 1:5 50 13 122
Zr(allyl)
4
80 2.0
Hf(allyl)
4
160 0.6
Cr(ally)
3
80 0.3 123
Cr(acac)
3
/EtAlCl 1:300 20 150 121
Ti(benzyl)
4
20(80) 8 10
3
(0.2) 124,125
Ti(benzyl)
3
Cl 20 0.4 124,125
Ti(benzyl)
4
Copyright 2005 by Marcel Dekker. All Rights Reserved.
6. Aluminoxane as Cocatalysts
The use of metallocenes and alumoxane as cocatalyst results in extremely high poly-
merization activities (see Tables 4 and 5). This system can easily be used on a laboratory
scale. The methylalumoxane (MAO) is prepared by careful treatment of trimethylalumi-
num with water [148]:
24
MAO is a compound in which aluminum and oxygen atoms are arranged alternately
and free valences are saturated by methyl substituents. It is gained by careful partial
hydrolysis of trimethylaluminum and, according to investigations by Sinn [149] and
Barron [150], it consists mainly of units of the basic structure [Al
4
O
3
Me
6
], which contains
four aluminum, three oxygen atoms and six methyl groups. As the aluminum atoms in
this structure are co-ordinatively unsaturated, the basic units (mostly four) join together
forming clusters and cages. These have molecular weights from 1200 to 1600 and are
soluble in hydrocarbons.
If metallocenes, especially zirconocenes but also titanocenes, hafnocenes and other
transition metal compounds (Figure 2) are treated with MAO, then catalysts are acquired
that allow the polymerization of up to 100 tons of ethene per g of zirconium [151153].
At such high activities the catalyst can remain in the product. The insertion time (for the
insertion of one molecule of ethene into the growing chain) amounts to some 10
5
s only
(Table 6). A comparison with enzymes is not far-fetched.
As shown by Tait under these conditions every zirconium atom forms an active
complex and produces about 20 000 polymer chains per hour. At temperatures above
50

C, the zirconium catalyst is more active than the hafnium or titanium system; the latter
is decomposed by such temperatures. Transition metal compounds containing some
halogene show a higher activity than systems that are totally free of halogen. Of the
cocatalysts, methylalumoxane is much more eective than the ethylaluminoxane or
isobutylalumoxane.
It is generally assumed that the function of MAO is rstly to undergo a fast ligand
exchange reaction with the metallocene dichloride, thus rendering the metallocene methyl
Table 5 Ethene polymerization
a
with metallocene/methylaluminoxane catalysts.
Metallocene
b
Structure Activity
[kg PE/(mol Zr.h.c
mon
]
Molecular weight
(g/mol)
Cp
2
ZrCl
2
6 60 900 62 000
[Me
2
C(Ind)(Cp)]ZrCl
2
8 3330 18 000
[En(IndH
4
)
2
]ZrCl
2
9 22 200 1 000 000
[Em(Ind)
2
]ZrCl
2
11 12 000 350 000
[En(Ind)
2
]HfCl
2
12 2900 480 000
[Me
2
Si(Ind)
2
]ZrCl
2
13 36 900 260 000
[Me
2
Si(2,4,7-Me
3
Ind)
2
]ZrCl
2
15 111 900 250 000
[Me
2
C(Flu)(Cp)]ZrCl
2
18 2000 500 000
a
Ethene pressure 2.5 bar. temp. 30

C. [metallocene] 6.25 10
6
M. Metaliocene/MAO250. Solvent
toluene;
b
Cp cyclopentadienyl; Indindenyl; EnC
2
H
4
; Fluuorenyl.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
and dimethyl compounds (Figure 3). In the further step, either Cl

or CH

3
is abstracted
from the metallocene compound by al Al-center in MAO, thus forming a metallocene
cation and a MAO anion [156,157]. The alkylated metallocene cation represents the active
center (Figure 4). Meanwhile, other weakly coordinating cocatalysts, such as
tetra(peruorophenyl)borate anions [(C
6
F
5
)
4
B]

, have been successfully applied to the

activation of metallocenes [158161].
Polyethenes synthesized by metallocene-alumoxane have a molecular weight dis-
tribution of M
w
/M
n
2, 0.9 to 1.2 methyl groups per 1000 C atoms, 0.11 to 0.18 vinyl
groups, and 0.02 trans vinyl group per 100 C atoms. The molecular weight can easily be
lowered by increasing the temperature, increasing the metallocene concentration, or
Figure 2 Some classes of metallocene catalysts used for olefin polymerization.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
Table 6 Polymerization activity of bis(cyclopentadienyl)zirconium dichloride/
methylalumoxane catalyst applied to ethene in 330 ml of toluene.
Activity (95

C), 8 bar 39.8 10

6
g PE/g Zr h
[Zirconocene] 6.2 10
8
mol/l
[Alumoxane] (M1200) 7.1 10
4
mol/l
Molecular weight of the polyethene obtained 78 000
Degree of polymerization 2800
Macromolecules per Zr atom per hour 46 000
Rate of growth of one macromolecule 0.087 s
Turnover time 3.1 10
5
s
Figure 3 Reactions of zirconocenes with MAO.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
decreasing the ethene concentration. The molecular weight distribution can be decreased
up to 1.1 (living polymerization) by bis(phenoxy-imine)titanium complexes [161].
Molecular weights of 170 000 were obtained. The molecular weight is also lowered by
the addition of small amounts) (0.1 to 2 mol%) of hydrogen (e.g., without H
2
,
M
w
170 000; adding 0.5 mol% H
2
, M
w
42 000) [155].
7. Late Transition Metal Catalyst
Brookhart et al. [57,58] described square planar nickel and palladium-diimine systems
which are capable of polymerizing ethene to high molecular weight polymers with activ-
ities comparable to the metallocene catalyst systems when activated with methyl-
aluminoxane.
25 26
Important for the polymerization activity is the substituent 1 which has to be a bulky
aryl group. The task of this substituent is to ll up the coordination spheres below and
above the square plane of the complex and thus enable the growing polymer chain to stay
coordinated to the metal center. This is one of the main dierences to the well-known
SHOP catalysts invented by Keim et al. [164] and Ostoja-Starzewski and Witte [165] which
produces mainly ethene oligomers.
Figure 4 Mechanism of the polymerization of olefins by zirconocenes. Step 1: The cocatalyst
(MAO: methylalumoxane) converst the catalyst after complexation into the active species that has a
free coordination position for the monomer and stabilizes the latter. Step 2: The monomer (alkene) is
allocated to the complex. Step 3: Insertion of the alkene into the zirconium alkyl bond and provision
of a new free coordination position. Step 4: Repetition of Step 3 in a very short period of time (about
2000 propene molecules per catalyst molecule per second), thus rendering a polymer chain.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
27
The use bis(ylid)nickel catalysts by reaction of nickel oxygen complexes and
phosphines [166].
For the one-component catalyst, it is possible to use solvents of various polarities.
Even in THF or acetone there is good activity. The best solvents are methylene chloride
or hexane. If the hydrogen next to the oxygen in the ylid is replaced by larger groups,
the activity increases and reaches at 10-bar ethene pressure and 100

C about 50 000 mol of

reacted ethene per mole of nickel [167].
A very interesting feature of this new catalyst generation is that chain isomerization
processes can take place during the polymerization cycles. This results in more or less
branched polymers with varying product properties depending on polymerization
conditions and catalyst type. The number of isomerization cycles which are carried out
directly one after another determines the nature of the branching formed. Branches
ranging from methyl to hexyl and longer can be formed.
The extent of branching can be tailored precisely by tuning the polymerization
conditions and products, from highly crystalline HDPE to completely amorphous poly-
mers with glass transition temperatures of about 50

C. These products are dierent to

all known conventionally produced copolymers due to their content and distribution
pattern of short chain branching [168].
Another new catalyst generation based on iron and cobalt. The direct iron analogs
of the nickel-diimine catalysts derived from structures (25) and (26) did not seem to be
very active in olen polymerization at all. The electronic and steric structure analysis
shows why: the nickel d
8
-system favors a square planar coordination sphere but the
iron d
6
-system favors a tetrahedral one. It is very likely that these tetrahedral coordination
sites are not available for olen insertion, and hence no polymerization can take place.
The next logical step was the employment of another electron donating atom in
the ligand structure in order to obtain a trigonal-bipyramidal coordination sphere.
Gibson and Brookhart both succeeded with a catalyst system based on an iron
bisiminopyridyl complex. The structures (28)(30) illustrate the three types of catalysts
[169,170].
28
Square planar
Copyright 2005 by Marcel Dekker. All Rights Reserved.
29
Tetrahedral
30
Trigonal-bipyramidal
The ethene polymerization activity of these new family of catalysts is comparable
with the one obtained with the most productive metallocenes under similar conditions
if activated with methylaluminoxane. Again, the nature of the aryl substituents R1 plays a
major role in controlling the molecular weight of the polymers.
In contrast to nickel-diimine catalysts no chain isomerization takes place and thus
only linear HDPE is formed.
In 1998, Grubbs [171,172] reported on a new type of neutral nickelII-complexes with
salicylaldimin ligands (structure (31)). With these catalysts low branched polyethylenes
were obtained with a narrow molecular weight distribution. The copolymerization of
ethene and norbornene is possible.
31
C. Copolymers of Ethene
The properties of polyethene could be varied in a wide range by copolymerization of
ethene with other comonomers. Most commercial products contain at least small amounts
of other monomers. In general, adding comonomers to the polymerization reduces the
polyethenes crystallinity, thereby reducing the melting point, the freezing point, and in
many cases the tensile strength and modulus. At the same time, optical properties are
Copyright 2005 by Marcel Dekker. All Rights Reserved.
improved and polarity is increased. The architecture of the copolymer can be controlled
experimentally by the following factors: operating conditions, chemical composition and
physical state of used catalyst, physical state of the copolymer being formed, and structure
of the comonomers.
The practically most important copolymer is made from ethene and propene.
Titanium- and vanadium-based catalysts have been used to synthesize copolymers that
have a prevailingly random, block, or alternating structure. Only with Ziegler or single
site catalyst, longer-chain a-olens can be used as comonomer (e.g., propene, 1-butene,
1-hexene, 1-octene). In contrast to this, by radical high-pressure polymerization it is also
possible to incorporate functional monomers (e.g., carbon monoxide, vinyl acetate). The
polymerization could be carried out in solution, slurry, or gas phase. It is generally accepted
[173] that the best way to compare monomer reactivities in a particular polymerization
reaction is by comparison of their reactivity ratios in copolymerization reactions.
The simplest kinetic scheme of binary copolymerization in the case of olen insertion
reaction is
CatM
1
polymer M
1
!
k
11
CatM
1
M
1
polymer 32
CatM
1
polymer M
2
!
k
12
CatM
2
M
1
polymer 33
CatM
2
polymer M
1
!
k
21
CatM
1
M
2
polymer 34
CatM
2
polymer M
2
!
k
22
CatM
2
M
2
polymer 35
r
1

k
11
k
12
r
2

k
22
k
21
36
where k
11
and k
22
are the homopolymerization propagation rates for monomers M
1
and M
2
and k
12
and k
21
are cross-polymerization rate constants. The denition of reactivity ratios is
dM
1

dM
2

M
1
r
1
M
1
M
2

M
2
M
1
r
2
M
2

37
The product r
1
r
2
usually ranges from zero to 1. When r
1
r
2
1, the copoly-
merization is random. As r
1
r
2
approaches zero, there is an increasing tendency toward
alternation.
1. Radical Copolymerization
At elevated temperatures, ethene can be copolymerized with a number of unsaturated
compounds by radical polymerization [174180] (Table 7). The commercially most
important comonomers are vinyl acetate [181], acrylic acid, and methacrylic acid as well as
their esters. Next to these carbon monoxide is employed as a comonomer, as it promotes
the polymers degradability in the presence of light [182].
As a consequence of the diversied nature of the comonomers, a large number of
variants of copolymer composition can be realized, thus achieving a broad variation of
properties. The copolymerization can be carried out in the liquid monomer, in a solvent,
or in aqueous emulsion. When high molecular mass is desired, solvents with low chain
transfer constants (e.g., tert-butanol, benzene, 1,4-dioxane) are preferred. Solution
Copyright 2005 by Marcel Dekker. All Rights Reserved.
polymerization permits the use of low polymerization temperatures and pressures.
Poly(ethylene-co-vinyl acetate, for instance, is produced at 100

C and 14 to 40 MPa [183].

For the polymerization of ethene with vinyl acetate and vinyl chloride, emulsion
polymerization in water is particularly suitable. The polymerizates have gained some
importance as adhesives, binding materials for pigments, and coating materials [184,185].
2. Linear Low-Density Polyethene (LLDPE)
In contrast to LDPE produced with the high-pressure process, the tensile strength in
LLDPE is much higher. Therefore, there has been a considerable boost in the production
of LLDPE [186]. All Ziegler catalysts listed earlier are suitable for the copolymerization of
ethene with other monomers. Monomers that decrease the melting point and crystallinity
of a polymer at low concentrations are of great interest. Portions of 2 to 5 mol% are
used. Longer-chained monomers such as 1-hexene are more eective at the same weight
concentration than smaller units such as propene. It results in a branched polyethene with
methyl branching (R) if propene is used, ethyl if butene is used, and so on.
38
Important for the copolymerization are the dierent ractivities of the olens. The
principal order of monomer reactivities is well known [187]; ethene >propene >1-butene >
linear a-olens > branched a-olens. Normally propene reacts 5 to 100 times slower than
ethene, and 1-butene 3 to 10 times slower than propene. Table 8 shows the reactivity ratios
for the copolymerization of ethene with other olens. The data imply that the reactivity of
the polymerization center is not constant for a given transition metal compound but
depends on the structure of the innermost monomer unit of the growing polymer chain
and on the cocatalyst.
On a laboratory scale, single site catalysts based on metallocene/MAO are highly
useful for the copolymerization of ethene with other olens. Propene, 1-butene, 1-pentene,
1-hexene, and 1-octene have been studied in their use as comonomers, forming linear low-
density polyethene (LLDPE) [188,189]. These copolymers have a great industrial potential
and show a higher growth rate than the homopolymer. Due to thee short branching from
Table 7 Copolymerization of ethene (M
1
) with various comonomers (M
2
).
Comonomer r
1
r
2
Pressure (MPa) Temp. (

C)
Propene 3.2 0.62 102170 120220
1-Butene 3.2 0.64 102170 130220
Isobutylene 2.1 0.49 102170 130220
Styrene 0.7 1 150250 100280
Vinyl acetate 1 1 110190 200240
Vinyl chloride 0.16 1.85 30 70
Acrylic acid 0.09 196204 140226
Acrylic acid methylester 0.12 13 82 150
Acrylnitrile 0.018 4 265 150
Methacrylic acid 0.1 204 160200
Methacrylic acid methylester 0.2 17 82 150
Copyright 2005 by Marcel Dekker. All Rights Reserved.
the incorporated a-olen, the copolymers show lower melting points, lower crystallinities,
and lower densities, making lms formed from these materials more exible and better
processible. Applications of the copolymers can be found in packaging, in shrink lms
with a low steam permeation, in elastic lms, which incorporate a high comonomer
concentration, in cable coatings in the medical eld because of the low part of extractables,
and in foams, elastic bers, adhesives, etc. The main part of the comonomers is randomly
distributed over the polymer chain. The amount of extractables is much lower than in
polymers synthesized with Ziegler catalysts.
The copolymerization parameter r
1
, which says how much faster an ethene unit is
incorporated into the growing polymer chain than an a-olen, if the last inserted monomer
was an ethene unit, lies between 1 and 60 depending on the kind of comonomer
and catalyst. The product r
1
r
2
is important for the distribution of the comonomer and
is close to one when using C
2
-symmetric catalysts [190] (Table 9).
Under the same conditions, syndiospecic (C
s
-symmetric) metallocenes are more
eective in inserting a-olens into an ethene copolymer than isospecic working
(C
2
-symmetric) metallocenes or unbridged metallocenes. In this particular case,
hafnocenes are more ecient than zirconocenes, too.
An interesting eect is observed for the polymerization with ethylene(bisindenyl)-
zirconium dichloride and some other metallocenes. Although the activity of the homo-
polymerization of ethene is very high, it increases when copolymerizing with propene [191].
The copolymerization of ethene with other olens is eected by the variation of
the Al/Zr ratio, temperature and catalyst concentration. These variations change the
molecular weight and the ethene content. Higher temperatures increase the ethene content
and lower the molecular weight.
Table 8 Reactivity ratios of ethene with various comonomers and heterogeneous TiCl
3
catalyst
by 70

C.
Comonomer Cocatalyst r
1
r
2
Ref.
Propene Al(C
6
H
13
)
3
15.7 0.11 174
Propene AlEt
3
9.0 0.10 174
1-Butene AlEt
3
60 0.025 178
4-Methyl-1-pentene AlEt
2
Cl 195 0.0025 177
Styrene AlEt
3
81 0.012 179
Table 9 Results of ethene reactivity ratio determinations with soluble catalysts
a
.
Metallocene Temp. (

C) a-Olen r
1
r
2
r
1
r
2
Cp
2
ZrMe
2
20 Propene 31 0.005 0.25
[En(Ind)
2
]ZrCl
2
50 Propene 6.61 0.06 0.40
[En(Ind)
2
]ZrCl
2
25 Propene 1.3 0.20 0.26
Cp
2
ZrCl
2
40 Butene 55 0.017 0.93
Cp
2
ZrCl
2
60 Butene 65 0.013 0.85
Cp
2
ZrCl
2
80 Butene 85 0.010 0.85
[En(Ind)
2
]ZrCl
2
30 Butene 8.5 0.07 0.59
[En(Ind)
2
]ZrCl
2
50 Butene 23.6 0.03 0.71
Cp
2
ZrMe
2
60 Hexene 69 0.02 1.38
[Me
2
Si(Ind)
2
]ZrCl
2
60 Hexene 25 0.016 0.40
Copyright 2005 by Marcel Dekker. All Rights Reserved.
Studies of ethene copolymerization with 1-butene using the Cp
2
ZrCl
2
/MAO catalyst
indicated a decrease in the rate of polymerization with increasing comonomer concen-
tration.
3. Ethene-Propene Copolymers
The copolymers of ethene and propene, with a molar ratio of 1:0.5 up to 1:2, are of great
industrial interest. These EP-polymers show elastic properties and, together with 25 wt%
of dienes as third monomers, they are used as elastomers (EPDM). Since there are
no double bonds in the backbone of the polymer, it is less sensitive to oxidation reaction.
Ethylidenenorbornene, 1,4-hexadiene and dicyclopentadiene are used as dienes. In most
technical processes for the production of EP and EPDM rubber, soluble or highly disposed
vanadium components have been used in the past (Table 10) [192195]. Similar elastomers
which are less coloured can be obtained with metallocene/MAO catalyst at a much
higher activity [196]. The regiospecicity of the metallocene catalysts towards propene
leads exclusively to the formation of head-to-tail enchainments. Ethylidenenorbornene
polymerizes via vinyl polymerization of the cyclic double bond and the tendency of
branching is low. The molecular weight distribution of about 2 is narrow [197].
At low temperatures the polymerization time to form one polymer chain is long
enough to consume one monomer and then to add another one. So, it becomes possible
to synthesize block copolymers if the polymerization, catalyzed especially by hafnocenes,
starts with propene and, after the propene is nearly consumed, continues with ethene.
High branching, which is caused by the incorporation of long chain olens into the
growing polymer chain, is obtained with silyl bridged amidocyclopentadienyltitanium
compounds (structure (39)) [198200].
39
Table 10 Results of ethene reactivity ratio determinations with soluble catalysts
a
.
Catalyst Cocatalyst Temp. (

C) r
1
(M
l
) r
2
(M
2
) r
1
r
2
Ref.
VCl
4
AlEt
2
Cl 21 3.0 0.073 0.23 192
VCl
4
Al-i-Bu
2
Cl 20.0 0.023 0.46 193
VOCl
3
Al-i-Bu
2
Cl 30 16.8 0.052 0.87 192
V(acac)
3
Al-i-Bu
2
Cl 20 16.0 0.04 0.64 193
VOCl
2
(OEt) Al-i-Bu
2
Cl 30 16.8 0.055 0.93 194
VOCl
2
Al-i-Bu
2
Cl 30 18.9 0.069 1.06 194
VO(OBu)
3
Al-i-Bu
2
Cl 30 22.0 0.046 1.01 194
VO(OEt)
3
Al-i-Bu
2
Cl 30 15.0 0.070 1.04 194
VO(OEt)
3
AlEt
2
Cl 30 26.0 0.039 1.02 195
a
Monomer 1 ethene, monomer 2 propene.
Copyright 2005 by Marcel Dekker. All Rights Reserved.
These catalysts, in combination with MAO or borates, incorporate oligomers with
vinyl endgroups which are formed during polymerization by b-hydrogen transfer resulting
in long chain abranched polyolens. In contrast, structurally linear polymers are obtained
when catalysed by other metallocenes. Copolymers of ethylene with 1-octene are very
exible materials as long as the comonomer content is less than 10%.
With higher 1-octene content they show that elastic properties polyolen elastomers
(POE) are formed [201]. EPDM is a commercially important synthetic rubber. The dienes
as terpolymers are curable with sulfur. This rubber shows a higher growth rate than the
other synthetic rubbers [202]. The outstanding property of ethene-propene rubber is its
weather resistance since it has no double bonds in the backbone of the polymer chain and
thus is less sensitive to oxygen and ozone. Other excellent properties of this rubber are its
resistance to acids and alkalis, its electrical properties, and its low-temperature
performance [203].
EPDM rubber is used in the automotive industry for gaskets, wipers, bumpers,
and belts. In the tire industry, EPM and EPDM play a role as a blending component,
especially for sidewalls. Furthermore, EPDM is used for cable insulation and in
the housing industry, for roong as well as for many other purposes, replacing special
rubbers [204].
For technical uses, the molecular weight (M
w
) is in the range 100 000 to 200 000.
EPDM rubber, synthesized with vanadium catalyst, show a molecular weight distribution
between 3 and 10, indicating that two and more active centers are present.
The properties of the copolymers depend to a great extent on several structural
features of the copolymer chains as the relative content of comonomer units, the way the
comonomer units are distributed in the chain, the molecular weight and molecular weight
distribution, and the relative content of normal head-to-tail addition or head-to-head/
tail-to-tail addition.
4. Ethene-Cycloolefin Copolymers
Metallocene/methylaluminoxane (MAO) catalysts can be used to polymerize and copoly-
merize strained cyclic olens such as cyclobutene, cyclopentene, norbornene, DMON and
other sterically hindered olens [205210]. While polymerization of cyclic olens by
ZieglerNatta catalysts is accompanied by ring opening [10], homogeneous metallocene
[211], nickel [212,213], or palladium [214,215], catalysts achieve exclusive double bond
opening polymerization.
40 41 42
Copolymerization of these cyclic olens with ethylene or a-olens cycloolen
copolymers (COC) can be produced, representing a new class of thermoplastic amorphous
materials [217220]. Early attempts to produce such copolymers were made using
heterogeneous TiCl
4
/VAlEt
2
Cl or vanadium catalysts, but rst signicant progress was
Copyright 2005 by Marcel Dekker. All Rights Reserved.
made by utilizing metallocene catalysts for this purpose. They are about ten times more
active than vanadium systems and by careful choice of the metallocene, the comonomer
distribution may be varied over a wide range by selection of the appropriate cycloolen
and its degree of incorporation into the polymer chain. Statistical copolymers become
amorphous at comonomer incorporations beyond 1015 mol% cycloolen.
COCs are characterized by excellent transparency and very high, long-life service
temperatures. They are soluble, chemically resistant and can be melt-processed. Due to
their high carbon/hydrogen ratio, these polymers feature a high refractive index, e.g. 1.53
for ethene-norbornene copolymer at 50 mol% norbornene incorporation. Their stability
against hydrolysis and chemical degradation, in combination with their stiness lets them
become desirable materials for optical applications, e.g. for compact disks, lenses, optical
bers and lms. The rst commercial COC plant run by Ticona GmbH with a capacity
of 30 000 tons a year commerced production in September 2000 and is located in
Oberhausen, Germany.
The rst metallocene-based COC material was synthesized from ethene and cyclo-
pentene [218]. While homopolymerization of cyclopentene results in 1,3-enchainment of
the monomer units [219], isolated cyclopentene units are incorporated into the ethene-
cyclopentene copolymer chain by 1,2-insertion. Ethylene is able to compensate the steric
hindrance at the a-carbon of the growing chain after and before the insertion of
cyclopentene [220].
Ethene-norbornene copolymers are most interesting for technical applications
as they can be made from easily available monomers and provide glass transition
temperatures up to 200

C. Table l1 presents the activities and comonomer ratios for the

several applied catalysts of C
2
- and C
s
-symmetry. C
s
-symmetric zirconocenes are more
active in the copolymerization than for the homopolymerization of ethene. Under the
chosen conditions, [En(Ind)
2
]ZrCl
2
develops the highest activity while the highest
comonomer incorporation is achieved by [Ph
2
C(Ind)(Cp)]ZrCl
2
.
Due to dierent incorporation ratios of the cyclic olen into the copolymer, the glass
transition temperature can vary over a wide range which is basically independent of the
applied catalyst. A copolymer containing 50 mol% of norbornene yields a material with
a glass transition point of 145

C. Considering COCs of dierent comonomers with equal

comonomer ratios, increased T
g
values can be observed for the bulkier comonomer, for
instance 72

C for ethene-norbornene and 105

C for ethene-DMON at comonomer mole

ratio X
Co
0.30 each.
The copolymerization parameters r
1
and r
2
were calculated from the rates of incor-
poration, determined by
13
C NMR spectroscopy, dependent on the reaction temperature.
Table 12 shows the temperature dependence of the copolymerization parameters
r
l
and r
2
and of the inuence of the catalyst systems. Metallocene catalysts show low r
1
values, which increases with the temperature and allows the easy incorporation of bulky
cycloolens into the growing polymer chain. Surprisingly, the copolymerization parameter
r
1
1.83.1 for cyclopentene and norbornene is surprisingly low. The r
1
value of 2 means
that ethylene is inserted only twice as fast as norbornene.
The product r
1
r
2
shows whether statistical insertion (r
1
r
2
) or alternating one
(r
1
r
2
0) has occurred. The dierent catalysts produce copolymers with structures that
are between statistical and alternating.
Due to dierent incorporation values of the cyclic olen in the copolymer, the glass
transition temperature can vary over a wide range that is independent of most of the used
catalysts (Figure 5). Acopolymer with 50 mol% of norbornene yields a material with a glass
transition point of 145

C. A T
g
of 205

C can be reached by higher incorporation rates.

Copyright 2005 by Marcel Dekker. All Rights Reserved.
The T
g
values are raised with a bulkier cycloolen, regarding the same incorporation
rate of 30% (norbornene: T
g
72

C; DMON: T
g
105

C). The highest glass transition

temperature with 229

C was reached by a copolymer of ethene and 5-phenylnorbornene.

Copolymerization of ethene and norbornene with [Me
2
C(Flu)(tert-BuCp)]ZrCl
2
leads to a strong alternating structure [221]. This copolymer is crystalline and shows a
melting point of 295

C, good heat resistance and resistance against unpolar solvents.

5. Ethene-Copolymerization by Styrene or Polar Monomers
The copolymerization of ethene and styrene is possible by single site catalysts such as
metallocenes and amido (see structure (33)) [222,223]. Amounts of more than 50 mol%
of styrene could be incorporated into the copolymer. In dependence of the styrene
content the copolymers show elastic to sti properties. The polymerization happens by
both 1,2- and 2,1-insertion of the styrene unit; the regioselectivity is low.
While it is dicult to copolymerize ethene and polar monomers by Ziegler- or single-
site catalysts because of the great reactivity of the active sites to polar groups, it is
commercialized to use free radical polymerization by high ethene pressure.
Vinyl acetate and acrylate esters used as comonomers containing sucient stabilizer
to prevent the homopolymerization. The eect of the copolymerization with polar
Table 11 Copolymerization of norbornene (N) and ethene (E) by different metallocene/MAO
catalysts at 30

C. Conditions: MAO/Zr 200, c(Zr) 5 10

6
mol/l; p(E) 2.00 bar,
c(N) 0.05 mol/l.
Catalyst t [min] Activity
[kg/mol h]
Incorp. of norbornene
[weight %]
Cp
2
ZrCl
2
30 1200 21.4
[En(Ind)
2
]ZrCl
2
10 9120 26.1
[Me
2
Si(Ind)
2
]ZrCl
2
15 2320 28.4
[En(IndH
4
)
2
]ZrCl2 40 480 28.1
[Me
2
C(Flu)(Cp)]ZrCl
2
10 7200 28.9
[Ph
2
C(Flu)(Cp)]ZrCl
2
10 6000 27.3
[Ph
2
C(Ind)(Cp)]ZrCl
2
15 2950 33.3
Table 12 Copolymerization parameters r
1
and r
2
of ethene/cycloolefin copolymerization with
different metallocene/MAO catalysts.
Cycloolen Catalyst Temp. in

C r
1
r
2
r
1
r
2
Cyclopentene [En(IndH
4
)
2
]ZrCl
2
0 1.9 <1 $1
Cyclopentene [En(IndH
4
)
2
]ZrCl
2
25 2.2 <1 $1
Norbornene [Me
2
Si(Ind)
2
]ZrCl
2
30 2.6 <2 $1
Norbornene [Me
2
C(FIu)(Cp)]ZrCl
2
30 3.4 0.06 0.2
Norbornene [Ph
2
C(Flu)(Cp)]ZrCl
2
0 2.0 0.05 0.1
Norbornene [Ph
2
C(Flu)(Cp)]ZrCl
2
30 3.0 0.05 0.15
Norbornene [Me
2
C(Flu)(t-BuCp)]ZrCl
2
30 3.1 0 0
DMON [Ph
2
C(Flu)(Cp)]ZrCl
2
50 7.0 0.02 0.14
DMON [Ph
2
C(Ind)(Cp)]ZrCl
2
50 6.4 0.10 0.64
DMON [Ph
2
C(Flu)(Cp)]HfCl
2
50 7.1 0.04 0.28
Copyright 2005 by Marcel Dekker. All Rights Reserved.
monomers is to reduce the crystallinity and to receive materials for blending. Acrylate
esters such as methyl acrylate, ethyl acrylate, butyl acrylate, methyl methacrylate form
exible copolymers. They provide enhanced adhesion, particularly in coextruded lms or
laminates.
Late transition metal complexes are more ecient in the copolymerization of ethene
and polar monomers. Nickel or palladium complexes (see strtuctures (30)(32) are
functional-group tolerant allowing the copolymerization of ethene and methyl methacry-
late or CO [224227].
III. POLYPROPENE
A. Homopolymerization
In contrast to the polymerization of ethene, only coordination catalysts are successful in
polymerizing propene to a crystalline polymer. The cationic polymerization of propene
with concentrated sulfuric acid leads to oily or waxy amorphous polymers of low
molecular weight [228]. Next to strong acids, catalysts such as complex Lewis acids
may serve as initiators in the cationic polymerization of propene. The polymerization
is conducted at temperatures between 100 and 80

C. Chlorinated hydrocarbons are

commonly used as solvents.
Under cationic conditions, migration of the CC double bond is observed. Like all
other a-olens, propene cannot be polymerized via an anionic route. The same applies to
free-radical polymerization. In polymerization with ZieglerNatta catalysts, propene or
longer-chained a-olens are inserted into the growing chain in a head-to-tail fashion
with high selectivity. Every CH
2
-group (head) is followed by a CH(R)-group (tail) with
Figure 5 Glass transition temperatures of norbornene/ethene copolymers catalyzed with different
zirconocenes.
Copyright 2005 by Marcel Dekker. All Rights Reserved.