Vol.

11 PROPYLENE POLYMERS 1

PROPYLENE POLYMERS
Introduction

The stereospecific polymerization of propylene by Giulio Natta in 1954 is one of the most
commercially significant scientific breakthroughs in polymer chemistry. Natta’s discovery
that a Ziegler catalyst could be used to produce highly isotac- tic polypropylene led to the
first commercial processes for the production of this polymer by Montecatini in Italy and
Hercules in the United States in 1957. The attractive properties and relatively low cost of
polypropylene produced using this technology led to its rapid commercial acceptance.
Consequently, Karl Ziegler and Giulio Natta were awarded the Nobel Prize in Chemistry in
1963. The commercial potential of olefin polymerization was recognized by many of the
leading compa- nies, leading to a tremendous amount of activity and the invention of a
number of competing technologies in the early 1950s. These technologies were not eco-
nomically competitive with those based on Ziegler–Natta catalysts. However, the competing
claims of a number of companies led to a massive patent interference in the United States
that continued for almost 30 years, eventually resulting in the award of a U.S. patent for the
composition of matter of crystalline polypropy- lene to Hogan and Banks of Phillips
Petroleum in 1983. Continued interest in olefin polymerization led to the invention of the
supported high yield, high stere- oregularity catalyst systems by Montedison and Mitsui
Petrochemical. This led to the development of a number of low cost polymerization
processes, spurring a dramatic increase in production capacity. Today, these catalyst systems
are used to produce polypropylene in every major region of the world. The use of homo-
geneous organometallic (metallocene) olefin polymerization catalysts has led to the
development of a number of unique propylene polymers. Currently, these polymers are
produced in relatively limited amounts for a number of specialty applications.
Polypropylene is one of the most widely used polymers in the world because of the
widespread availability and low cost of monomer, low manufacturing cost, and attractive
polymer properties. These properties can be modified to be suitable for a wide variety of
applications. Polypropylene can be processed by almost all commercial fabrication
techniques. Approximately 30,000,000 ton was consumed worldwide in 2001.

Encyclopedia of Polymer Science and Technology. Copyright John Wiley & Sons, Inc. All rights reserved.
 2 PROPYLENE Vol.

Table 1. Physical Properties of Propylene

Property Value Reference
Molecular weight 42.078 1
Boiling point at 101.3 kPaa, ◦C −47.7 3
−185.3 3
Melting point, ◦C
Critical temperature, ◦C 92 1
Critical pressure, MPab 4.65 1
Critical density, g/mL 0.233 2
Critical compressibility
Refractive index, nD 0.275 3
Dipole moment, 10 − 30 C·mc 1.3 3
1.3567 3
Explosion limit, % by volume in air
Lower 2.4 1
Upper 11.1 1
Autoignition temperature, ◦C 224 1
Solubility in water (at 20◦C, 101.3 kPaa), mL gas/100 mL 44.6 3
a
To convert kPa to mm Hg, multiply by 7.5.
b
To convert MPa to atm, multiply by 9.87.
c
To convert C·m to D, divide by 3.336 x 10 − 30.
Properties. Propylene is an olefin hydrocarbon that is a gas under am- bient
conditions but is normally stored as a liquid under pressure. The physical properties of
propylene are given in Table 1. Thermodynamic properties are widely reported in the
Monomer
literature. Vapor–liquid equilibria of mixtures of propylene with other hydrocarbons and
hydrogen are accurately represented by correlations for hydrocarbon mixtures, such as the
Chao–Seader correlation.
The reactivity of propylene is a result of the olefinic double bond in H2C CHCH2,
which gives rise to addition reactions. Consequently, propylene is used in the synthesis of
many industrially important compounds, including propylene oxide, acrylonitrile, cumene,
and isopropyl alcohol. The consumption of propylene in the production of polymers,
however, is greater than the total for all other chemicals. Propylene is also used in alkylating
feedstocks for gasoline.
Manufacture. The major commercial sources of propylene are processes for the
cracking of hydrocarbons. Initially, these processes were designed for the manufacture of
other products, and propylene was considered an undesirable by- product. Now, propylene is
often an equally desirable co-product. Ethylene is pro- duced by the steam cracking of
hydrocarbons at high temperatures and very short residence times. The two most common
feedstocks are naphtha and ethane. This process produces a variety of by-products; the ratio
of ethylene to by-products is dependent on the type hydrocarbon feedstock and the reaction
conditions. Propy- lene production is higher when naphtha is used as a feedstock rather than
ethane. Catalytic cracking of hydrocarbons is often used to increase the production of gaso-
line in oil refining. Increasing consumption of gasoline in the United States has led refiners
to increase the severity of the cracking process, resulting in an in- crease of the production
of propylene as a by-product. Consequently, the ratio of
Vol. 11 PROPYLENE POLYMERS 3
Table 2. Recommended Maximum Limits of Impurities
in Propylene Used in Polymerizationa
Impurity Maximum limit
Acetylene, ppm vol 5
Methylacetylene, ppm vol 3
Propadiene, ppm vol 5
Butadiene, ppm vol 50
Green Oils (C6-C12), ppm vol 20
Oxygen, ppm vol 2
CO, ppm vol 0.03
CO2, ppm vol 5
COS, ppm vol 0.02
Total sulphur, ppm wt 1
Methanol, ppm vol 5
Isopropanol, ppm vol 15
Water, ppm wt 2
Arsine, ppm vol 0.03
Phosphine, ppm vol 0.03
Ammonia, ppm wt 5
a
Source: Basell.

propylene isolated from refineries to propylene from steam cracking is greater in the United
States than in Western Europe and other parts of the world. The propylene isolated from the
cracking processes is purified by distillation to remove propane and other impurities.
Propylene is commercially available as chemical grade (approximately 95% propylene) and
polymerization grade (>99.5% propy- lene) where propane is the major impurity. However,
the suitability of propylene as polymerization monomer is dependent on the levels of trace
impurities rather than on the propane content. Commonly occurring impurities are
acetylenes, di- enes, CO, COS, water, and alcohols. These impurities affect the activity and
stere- ospecificity of propylene polymerization catalysts. Typical limits for the level of
impurities acceptable in polymerization grade propylene are given in Table 2.
Propylene is also produced by the metathesis of butene and ethylene. Usually, this
process is installed as an addition in a refinery or steam cracking facility to increase the
production of propylene. The first commercial metathesis process for the production of
propylene was developed by Phillips, and is now licensed by ABB Lummus.
The relative abundance of propane and other light hydrocarbons in certain locations
has increased interest in the production of propylene by the catalytic dehydrogenation of
propane. Currently only a few propane dehydrogenation plants are operating, producing a
small fraction of the world supply of propy- lene; however, it is anticipated that propane
dehydrogenation will be the major source of propylene in the Middle East. The two major
processes available are the Catofin process originally developed by Houdry and now
licensed by ABB Lum- mus and the Oleflex process licensed by UOP. Natural gas can be
used as the feedstock for the production of propylene by adding a Lurgi MTP process
facility to a conventional methanol plant.
 4 PROPYLENE Vol.
Transportation, Storage, and Handling. Propylene is stored and trans-
ported as a liquid under pressure The preferred method of transporting propy- lene is
through pipelines. Commercial quantities of propylene are also shipped by tanker and rail.
The most extensive pipeline system for the transportation of propylene is in the Gulf Coast
region of the United States. This pipeline system is also connected to a series of
underground brine caverns used for propylene storage. Pressure vessels are also used for the
storage of propylene.
Propylene is a flammable gas under normal ambient conditions and is not hazardous at
low concentrations, but is an asphyxiant, which is a concern in closed environments. Direct
contact with liquid propylene can cause skin burns from freezing. Fire or explosion is the
greatest potential hazard associated with the storage and handling of propylene. Care must
be exercised to avoid creating an explosive atmosphere when disconnecting lines and
unloading vessels containing propylene. Explosions of vapor clouds formed from large
leaks of liquid or gaseous propylene are the greatest potential hazard. The fire and explosion
hazards asso- ciated with propylene are similar to those of propane and liquified petroleum
gas (LPG).

Molecular Structure

The unique feature of propylene polymerization, versus ethylene polymerization, is the

symmetry of the monomer insertion into the growing polymer chain. It is the presence of the
methyl group in the propylene monomer that is responsible for this difference. This gives
the monomer insertion an orientation (the monomer has a “head” and a “tail”) and a
stereochemical configuration with respect to the other units in the chain backbone.
Regularity with respect to monomer orientation is termed the regiospecificity of the
polymerization. Regularity of the methyl group placement relative to the other methyl
groups along the chain backbone is termed the stereospecificity of the polymerization. The
three limiting classifications of stereospecificity in polypropylene (PP) are illustrated in
Figure 1. In isotactic polypropylene (iPP), all of the methyl groups have the same
configuration with respect to the polymer backbone. In syndiotactic polypropylene (sPP),
the methyl groups have an alternating configuration. Atactic polypropylene (aPP) has a ran-
dom configuration. An additional configuration, hemi-isotactic polypropylene, is discussed
elsewhere (4). iPP is overwhelmingly the most commercially significant form of PP. In
practice, the degree of stereoregularity (and tacticity microstruc- ture) can vary considerably
within these general classifications. For the specific case of isotactic polymerization, Figure
2 shows three idealized limiting sample microstructures. Figure 2a shows a case where an
isolated stereo error is imme- diately corrected along the growing chain. Figure 2b shows a
case where a stereo error is propagated along the growing chain, and Figure 2c shows an
iPP/aPP multiblock structure (5–7). Figure 2a is most common. Further discussions of
stereo defects are available in the literature (4–22). In general terms, stereoregu- larity (or
stereospecificity) refers to the content of defects that disrupt the regular placement of methyl
groups. The intrachain defect content can also vary widely between chains (interchain
distribution) (23) for different catalyst systems. The
Vol. 11 PROPYLENE POLYMERS 5

Fig. 1. Polypropylene stereochemical configurations: (a) isotactic; (b) syndiotactic;

(c) atactic.

control of stereoregularity by catalyst and polymerization process is a critical de- terminant

of PP properties.
Kinetic arguments suggest that iPP polymerization shows a strong pref- erence for
primary “head-to-tail” insertion of monomer into the growing chain (24–26). Regioirregular
insertion of monomer renders the active site kinetically dormant (24–30), and its
incorporation in the growing polymer chain occurs infre- quently. With Ziegler–Natta
polymerization catalysts, regio misinsertion errors are generally not present in the isotactic
fraction (30). Metallocene polymeriza- tion catalysts can show appreciable levels of regio
misinsertion errors in chains of high stereospecificity (4,27,31–41) (see METALLOCENES; SINGLE-
SITE CATALYSTS).
 6 PROPYLENE Vol.

Fig. 2. Sample tacticity microstructures in isotactic polymerization: (a) isolated stereo error is
corrected in growing chain; (b) stereo error is propagated in growing chain; (c) an iPP/aPP multi-
block structure.

The stereo- and regioregularity of PP is best characterized by solution 13C NMR

spectra (42). These spectra can also provide information on the poly- merization mechanism
(4,7,9,11,12,14–22,30), and have been modeled in terms of the interchain tacticity
distribution (14,43–45). Infrared spectroscopy (IR) has also been used to characterize the
stereospecificity of iPP. Commonly the ratio of the 998 and 973 cm − 1 band absorbance is
used (46,47). The result is sensitive to thermal history (46). Solvent fractionation techniques
are commonly used in industrial practice. These methods are based on the fact that chains
with vary- ing stereospecificity have different crystallinity and solubility in hydrocarbon sol-
vents. In nominally isotactic PP, the atactic fractions of the interchain tactic- ity distribution
are soluble, whereas the isotactic fractions are not. The isotactic index suggested by Natta
represents the percentage of polymer insoluble in boiling n-heptane (48). Other procedures
require preliminary total dissolution of the poly- mer in high boiling hydrocarbons such as
xylene, and subsequent cooling in order to separate the precipitated crystalline portion (an
example is ASTM D5492-94). The insoluble fraction is a measure of the stereospecificity.
Often the term stere- oblock is used to refer to fractions of the interchain distribution of
intermediate solubility. Related solvent fractionation techniques include preparative and an-
alytical temperature rising elution fractionation (TREF) (49). The use of pulsed proton
solid-state NMR has been demonstrated as an alternative to solubility mea- surements of
stereoregularity (50). Other secondary measurements which probe melting and/or
crystallinity behavior can also be applied.
In addition to tacticity, important additional parameters of the PP chain which
influence properties include the molecular weight, polydispersity, and composition and
comonomer distribution in blends and copolymers. PP, like all synthetic and most natural
polymers, consists of a distribution of macro- molecules of different lengths. Therefore, the
polymers exhibit a molecular weight
Vol. 11 PROPYLENE POLYMERS 7
distribution. The number-average molecular weight (Mn) is determined by gel- permeation
chromatography (GPC), more generally referred to as size-exclusion chromatography (51).
Because of the strength of the GPC technique, the older techniques of osmometry (52,53),
cryoscopy (54), and ebulliometry (54) are less commonly used today. For PP, the GPC
method typically utilizes high temper- atures (135–145◦C) in solvents such as o-
dichlorobenzene and trichlorobenzene. The weight-average molecular weight (Mw) can be
obtained by light scattering
(55) and GPC. Higher moments of the molecular weight distribution can also be obtained by
GPC. The viscosity-average molecular weight Mν can be obtained by measurement of the
solution intrinsic viscosity [η]0 (52). Mν and [η]0 are related by the Mark–Houwink
equation:

[η]0 = K·(Mν )a

Generally, values of Mν lie between Mn and Mw values. Values of the constant K and
exponent a are determined by calibration with homogeneous fractions of known molecular
weight for specific solvent/temperature conditions. In practice, for polypropylene,
homogeneous fractions are not available and values for some moment of the molecular
weight distribution (Mn or Mw ) are used in equation 1 on an empirical basis to determine K
and a. This allows for the correlation of viscosity to molecular weight, though corrections
need to be applied for samples with polydispersity much different from the correlation
standards (56). Represen- tative values using Mw from relatively narrow polydispersity
fractions in decalin at 135◦C are K 2.38 10 − 4, a 0.725 (57) and K 1.0–1.10 10 − 4, a 0.80
(58,59), and in tetralin at 135◦C are K 9.42 10 − 5, a 0.784 (60). A common in- dustrial
measure of molecular weight is the melt-flow = rate (MFR)
× (ASTM= D1238), which is related
to the melt viscosity. The MFR technique measures the amount of molten polymer extruded
by a standardized apparatus
= ×in 10 min. = Higher MFR values are= obtained for
×lower molecular
=
weight polymer. The breadth of the molec- ular weight distribution (polydispersity) has a
large effect on polymer properties. The polydispersity is commonly characterized by the
ratio Mw/Mn as measured by GPC or by semiempirical correlation of the dynamic melt
oscillitory shear be- havior of the polymer (61). This ratio has a value of 1 for
macromolecules of equal molecular weight (monodisperse) and increases with increasing
polydispersity. Rheological measurements, including the steady-state compliance, can be
very sensitive to higher moment components (62) important in flow processes.
The molecular structure of PP can be further modified by the introduction of
comonomers during the polymerization process, most commonly ethylene or butene. This
comonomer incorporation greatly expands the property range of PP. Table 3 illustrates the
broad range of copolymer microstructures possible for the specific case of propylene–
ethylene copolymerization. Similar to the case of stere- oregular defects in PP
homopolymer, comonomer defects introduced into “ran- dom” PP copolymers (PP-RACO)
can show an interchain composition distribution depending on the catalyst and process. In
some cases it can be advantageous, in terms of quality and the process conditions, to
achieve a homogeneous distri- bution of the comonomer molecules. Metallocene catalysts
show a greater ten- dency for homogeneous comonomer incorporation (4). In this case, each
chain, re- gardless of length, contains the same percentage of comonomer. Impact-modified
 8 PROPYLENE Vol.
Table 3. Propylene–Ethylene Copolymerization Microstructure
Microstructurea Products
PPPPPPPPPPPPPPPP (PP) Homopolymer polypropylene
PPPEPPPPPPPPEPPP (PP-RACO) Polypropylene randomly modified with
ethylene
PPPPPPPP EPEPEP (PP-EPR) Impact-resistant polypropylene PPPPP
+
EPEP EEEE (PP-EPR PE) Impact-resistant polypropylene (with
+ + + polyethylene
EPEPPEEPEPEPEPE (EPR) Ethylene–propylene rubber (elastomer)
EEEPEEEEEEEPEEE (PE-RACO) Polyethylene randomly modified with
propylene
EEEEEEEEEEEEEEE (PE) Homopolymer polyethyle
a
P= propylene, E= ethylene.

“copolymers” (PP-EPR) are better described as polymer blends or alloys. The composition
of the PP-EPR formulations varies over a wide range in industrial practice. However, a
typical formulation would contain 60–90% homopolymer (or PP-RACO) and 10–40%
ethylene–propylene copolymer rubber (EPR) with ethy- lene concentration of 30–60%. The
EPR component has a lower glass-transition temperature (Tg), lower crystallinity, and
typically exists as a separate phase. PP-EPR formulations were first produced by mechanical
blending of two compo- nents. Today they are commonly synthesized directly in a
multistage process to obtain better economics, and often better distribution of the
elastomeric phase in the polypropylene matrix and thus better quality. In some cases
polyethylene is present as a third phase. The characterization of PP copolymers is often
compli- cated by the multiplicity of structural species present. An in-depth characteriza- tion
requires knowledge of the distributions of molecular weight and composition. Different
solvent fractionation techniques are combined, and each fraction is an- alyzed. Composition
is most frequently determined by solution 13C NMR, by IR, or inferred from secondary
measurements which probe melting behavior, crys- tallinity, or glass-transition temperature.

Morphology

Crystallography and Polymorphism. The stereochemistry of PP plays a

critical role governing the packing of chains in the crystalline regions of the morphology
(qv). Figure 3 shows the wide-angle X-ray scattering (WAXS) pat- terns of iPP, sPP, and
aPP (23). The regular molecular structure of iPP and sPP readily enables crystallization of
the chains, leading to well-defined crystalline re- flections differing in unit cell symmetry.
aPP lacks a regular molecular structure, and does not crystallize. This leads to a very broad
and diffuse scattering from X-rays. The iPP chain adopts a helical conformation in the
crystalline unit cell, as shown in Figure 4 (63). The helix repeats itself after three
monomeric units, with an identity period of 0.65 nm. Four helical arrangements are possible
by right- or left-handed rotation about the central axis with unique (non-identical) “up” and
“down” inclinations independent of the handedness (23,64,65). The dominant
Vol. 11 PROPYLENE POLYMERS 9

Fig. 3. Wide-angle X-ray scattering (WAXS) patterns of iPP, sPP, and aPP. The shaded region
illustrates the separation of crystalline and amorphous scattering contributions. Adapted from Ref. 23.

crystallographic form for iPP is the α-form. The elementary unit cell of α-form iPP is
monoclinic, containing 4 chains and 12 monomeric units with specific packing of helical
arrangements. Crystallographic densities are generally in the range of 0.936–0.946 gm/cm3.
Representative cell constants are given in Table 4, and have been reviewed elsewhere (23).
The helical conformation of sPP differs from that of iPP, and has an identity period of 0.74
nm. The elementary unit cell of the most stable crystallographic form of sPP is
orthorhombic, containing 4 chains and 16 monomeric units with specific packing of helical
arrangements (23,64,66–69). The crystallographic density is 0.930 gm/cm3. Representative
cell constants are given in Table 4 (69).

Table 4. Unit Cell Constants of Polypropylene

Cell constants Isotactica Syndiotacticb
Unit cell monoclinic orthorhombic
a, nm 0.664 1.45
b, nm 2.084 1.12
c, nm 0.651 0.74
β 99.0◦ 90◦
a
Averages from literature compilation of P21/c space group
symmetry refinements (23).
b
From Ref. 69.
 10 PROPYLENE Vol.
Both iPP and sPP exhibit polymorphism, which is the tendency to crystallize into
different crystallographic forms depending on crystallization conditions. In iPP, the
dominant form is the α-form. Other forms include the β-, γ -, and mesomorphic (smectic)
forms. All of these crystalline forms maintain the same helical conformation with 0.65 nm
repeat distance, but differ in unit cell symmetry, interchain packing, and structural
disorder. The mesomorphic form is formed by rapid quenching conditions, and has
important property implications, particularly in film and fiber applications.
Crystallographically, the mesomorphic form has disordered interchain packing relative to
the other polymorphic unit cell symmetries (65,70,71). The mesomorphic form converts
rapidly to the α-form on heating (23). Formation of the β-form results from the addition of
specific nucleators and additives, specific crystallization conditions, crystallization under
controlled temperature gradients, and in some cases crys- tallization under shear (20). The
β-form has lower unit cell density, higher rate of crystallization, and lower apparent melting
point relative to the α-form which

Fig. 4. Chain conformation of isotactic polypropylene (iPP).

Vol. 11 PROPYLENE POLYMERS 11
is due, in-part, to its unique crystallography (64,72,73). The γ -form is rarely observed in
pure form in commercially significant homopolymer from Ziegler– Natta catalysts under
typical processing conditions. A variety of conditions for γ -form formation are outlined
below (23). The γ -form occurs in low molecular weight fractions, homopolymer from some
homogeneous (metallocene) catalysts (74–76), random copolymers, random copolymer and
metallocene homopolymer at high crystallization temperature (76), samples with high
stereoblock con- tent, and homopolymer crystallized at high pressure (77). Crystal structure
refinements suggest that γ -form iPP has nonparallel chains (64,65,78–81), a conclusion
not previously cited in polymeric crystal structure. Because of its importance, further
discussion of iPP morphology is restricted to the α-form of iPP.
sPP also exhibits polymorphism (23,64,66–69,82,83). Polymorphism is re- lated to
both the intrachain conformation and interchain packing. Crystallization at lower
temperature can lead to defect structures relative to the unit cell symme- try of the dominant
form (66–68,84–86). Other unit cell variations are formed from sPP with low
syndiospecificity (87,88), as-polymerized sPP (89–92), sPP copoly- mers (92), oriented sPP
(23,87,93), sPP crystallized at low temperature (94,95), and nonhelical structures associated
with cold drawing (87,96) and exposure of cold drawn samples to solvent vapor (97).
Crystallinity. The degree of crystallininity varies between 0 for a com- pletely
amorphous material (such as aPP) and 1 for a completely crystalline mate- rial. As with
most semicrystalline polymers (qv), the degree of crystallinity plays a critical role in
determining properties of PP. Commonly measured properties such as modulus, yield stress,
oxygen and moisture barrier resistance, and hard- ness, to name a few, all increase with
increasing crystallinity. In iPP (and sPP) tacticity is a critical factor influencing the
crystallinity (23,47,98). This is due to the role of stereo defects in disrupting the length of
the crystallizable isotactic sequences. Details of the interchain tacticity distribution affect
not only the crys- tallinity at room temperature, but also the partial melting behavior (and
hence crystallinity) at elevated temperature, influencing hot drawing characteristics.
Solvent fractionation techniques, which separate the interchain tacticity distri- bution of the
whole iPP homopolymer into fractions of varying tacticity, show that the lower tacticity
fractions have lower crystallinity (47). With metallocene cat- alysts (4,18,30), PP chain
microstructure can now be varied continuously with decreasing isospecific sequencing from
iPP (stiff thermoplastic) to aPP (very low modulus lacking dimensional integrity) to sPP
(stiff thermoplastic) correlating to changes of either isotactic or syndiotactic crystallinity.
The crystallinity of sPP is less than that of iPP with currently available catalysts. In addition
to tacticity, crystallinity generally increases with decreasing molecular weight (in- creased
chain mobility), and is promoted by slower cooling rates from the melt. It should be
emphasized that the crystalline fraction of perfectly isotactic PP is still much less than
unity because of the long-chain nature. Copolymeriza- tion (PP-RACO in Table 3) is also
used to modify the polymer crystallinity in a controlled manner. In this case, the comonomer
is the source of irregularity in the polypropylene chain. The introduction of comonomer
decreases crystallinity, reduces stiffness and melting temperature, and increases impact
resistance (99).
 12 PROPYLENE Vol.
The degree of crystallinity is determined by a number of analytical tech- niques based
on different criteria (see CRYSTALLINITY DETERMINATION). The most widely used are fractional
solubilization (previously discussed), density deter- mination, X-ray diffraction, and thermal
analysis. The density of iPP in the α- form varies between the limit of 100% amorphous (ρa
0.850 to 0.855 g/cm3) and 100% crystalline (ρc 0.936 to 0.946 g/cm3) (20). In this way, the
measured mass density ρ gives a measure of the crystallinity. Values of ρ are most often
measured by the density gradient technique (ASTM D1505-85). = Similarly, dilatometric
meth- ods are used to measure
= variations of density as a function of temperature (100). X-
ray diffraction (101) is also widely used to determine PP crystallinity. Figure 3 shows the
contributions of the crystalline scattering and amorphous scattering in the unoriented pattern
of the α-form of iPP. The shaded region represents the area attributable to the noncrystalline
fraction. From the relationship of the peak area to the total area, crystallinity can be
evaluated by a number of numerical means (101). Orientation complicates the analysis
(101). Differential scanning calorime- try (DSC) provides useful information on PP
structure by the determination of related parameters, such as transition temperature, heat of
fusion, crystalliza- tion temperature, and others. The crystalline fraction is given by the
ratio of the measured heat of fusion to the value for 100% crystalline material (see THERMODY-
NAMIC PROPERTIES OF POLYMERS). The shape of the melting curve gives information on the
melting distribution.
Lamellar and Spherulitic Morphology. The crystal habit of iPP quies-
cently crystallized from the melt, like other semicrystalline homopolymers, is that of folded
chain lamellae (Fig. 5). The lamellar thickness ( lc ) increases with in- creasing
crystallization temperature, and is generally in the range of 5–20 nm ( for iPP. iPP in the α-
form exhibits a tendency, unique among semicrystalline poly- mers, to form a “cross-
hatched” pattern (23,64). Radial lathlike lamellae coexist with “cross-hatched” tangential
lamellae oriented nearly orthogonal to the radial direction. This homoepitaxy (102–107) is
related to the relatively modest mis- match between the c- and a-axis unit cell parameters
(Table 4), and the detailed molecular aspects have been reviewed (64). Spherulites are larger
scale aggre- gates made up of the smaller scale lamellar building blocks. Depending on
the

Fig. 5. Schematic of chain-folded lamellar structure in semicrystalline polymers.

Vol. 11 PROPYLENE POLYMERS 13
crystallization conditions, the dimensions of spherulites can vary from a few mi- crometers
to 100 µm, or larger. In the α-form of iPP, there is a strong link be- tween lamellar and
spherulitic morphologies. The α-form iPP spherulites have been classified according to
different optical characteristics in cross-polarized light (106,108,109). The optical
characteristic is highly sensitive to crystalliza- tion temperature and resin type, and has been
linked to the lamellar mor- phology through the balance of cross-hatched radial and
tangential lamellae (23,103,104,106,109,110). Optical and mechanical properties depend on
the di- mension and number of spherulites, which can be modified with nucleating agents
that act as crystallization centers. Optical properties (such as clarity) improve with
decreasing spherulite size. The borders among spherulites can represent weak zones (111),
with large spherulites adversely affecting properties such as impact. Unlike iPP,
pronounced cross-hatched lamellar morphologies in bulk crys- tallized material appears
limited in sPP (112,113). As with iPP, well-developed small-angle X-ray (SAXS) long
spacings, characteristic of well-developed lamellar morphologies, are observed for sPP of
high syndiospecificity (113–116). In gen- eral terms, spherulite formation in as-polymerized
metallocene based sPP tends to be more restricted than in iPP (117–119), and the spherulite-
scale morpholo- gies are generally smaller than in iPP in bulk-crystallized specimens.
Spherulitic growth rate measurements have been summarized (120), with rates lower than
those of Ziegler–Natta iPP (113). Unique morphologies (relative to iPP) are ob- served in
sPP with special preparation procedures, including a transition from single-crystal like
entities to spherulitic structures on cooling (68,117–119).
Macromorphology and Processing Relationships. Other morpholo-
gies in iPP include the macroscopic phase morphologies in multiphase structures such as
rubber-modified iPP (an example being PP-EPR in Table 3), and addi- tional morphologies
associated with the specific method of polymer processing. Unique morphologies and
structure/processing relationships are associated with fiber, film, and injection molding
processes to name a few. In parts formed by the in- jection molding process, pronounced
morphological gradients exist in a processed part which can be crudely partitioned into
highly oriented skin layers (near the mold surface) and less oriented core layers (near the
part center). These layers can approach macroscopic dimensions ( mm) depending on the
part geometry, resin, and molding conditions. The properties of moldings can often be
correlated to the orientation and crystallinity of the part (23,121). A literature review of
more detailed aspects of the morphology of iPP injection moldings is available (23). In
mixtures of iPP homopolymer and EPR (PP-EPR ∼ in Table 3), the two components are
generally immiscible in the melt. In the solid state, the rubber has low glass-transition
temperature (Tg) and imparts impact resistance to the blend. The effectiveness of the rubber
phase as an impact modifier depends crit- ically on the dispersion (particle size). The
dispersion is a complex function of
(1) matrix and rubber phase composition, (2) viscosity ratio of matrix and dis- persed
phases, (3) compounding history, and (4) melt deformation history during fabrication of the
final processed part (23). In impact-modified copolymers for in- jection molding
applications, the dimension of the dispersed rubber phase in the core (reduced melt
deformation) is typically on the order of 0.5–2 µm in formula- tions with optimal impact
performance. Near the surface of moldings (high melt deformation), highly anisotropic
dispersed phase morphologies can be observed
 14 PROPYLENE Vol.
depending on blend composition. Generally rubber phase dispersion worsens (in- creased
phase size) with increasing melt temperature and melt time prior to mold- ing due to
coalescence.
High speed melt spinning of fibers represents another of the most com- mon
processing methods for iPP. The final morphology results from the com- plex
interrelationships of polymer structure (polydispersity, molecular weight, tacticity) and
processing conditions (melt temperature, melt throughput, spin- nerette design, spin speed,
cooling rate). For high spinning speeds, the orienta- tion of melt-spun fibers greatly exceeds
that of injection-molded parts. The orien- tation of crystalline regions is greater than the non
crystalline regions (122). As the spinning speed increases, orientation increases. The
orientation can be fur- ther increased with post-spinning drawing operations (122). As with
unoriented material, the as-spun fiber shows alternating crystalline and amorphous regions
on lamellar size scales (10–30 nm), but generally smaller than hot-drawn articles. This
periodicity is highly oriented along the fiber axis, and characterized by a high fraction of tie
molecules (the same chain passing through several crystalline regions).
Drawn film, and more specifically biaxially oriented polypropylene (BOPP) film, is
another of the most common fabrication methods (see F ILMS, MANUFAC- TURE; FILMS,
ORIENTATION). During BOPP film formation, a cast sheet of iPP is drawn below the melting
point in two directions. This drawing process creates a film of useful dimension while
imparting desired properties such as barrier and stiffness. These properties are related to the
imposed orientation during drawing. During unidirectional drawing of semicrystalline
polymers at elevated temper- ature, the initial spherulitic morphology of the cast sheet is
transformed to an oriented fibrillar morphology (122–125). The yielding process, during
which the lamellar crystallites are disrupted, is strongly correlated to the crystallinity at the
draw temperature. As with melt-spun fibers, the orientation of the crystalline regions is
greater than that of the noncrystalline regions (122). Orientation is generally much higher
than in injection moldings. For BOPP, the molecular orien- tation is dominated by the
transverse orientation stage for sequentially oriented film (126), and tends to be planar
(126,127) for both sequential and simultane- ous oriented film. Generally, the lamellar
crystals in the initial as-cast sheet are highly disrupted in the final BOPP film.
Polymerization Particle Morphology. Morphological control of as-
polymerized polymer particles influences the polymerization process. Aspects of the particle
morphology include the shape, size, size distribution, and porosity. Control of particle
morphology is also a critical concern to technologies which utilize the as-polymerized
particle in subsequent polymerization processes (128). Because of the replication
phenomenon (129–137), the morphological character- istics are strictly dependent on the
catalyst. The catalyst reproduces its shape in the polymer on an obviously greater scale
according to its activity. The activity is the quantity of polymer produced by a unit of
catalyst.
The replication phenomenon of iPP polymerization from magnesium chloride
supported Ziegler–Natta catalysts is illustrated in Figure 6. The polymer particle has the
same shape as the catalyst particle, although its diameter is approximately 20–100 times
larger. The growth of the polymer particle during polymerization has been extensively
modeled (129–137). In the multigrain model (131–135,137), the
Vol. 11 PROPYLENE POLYMERS 15

Fig. 6. Ziegler–Natta catalyst particle (a) and corresponding polymer particle (b).

catalyst undergoes fragmentation into microparticles evenly distributed within the

macroscopic particle due to forces exerted by the growing polymer layers. In Ziegler–Natta
catalysts, fragmentation occurs at very low polymer yields (137–139), which provides a
large active surface from the beginning of polymer- ization. The growth of the
macroparticle results from the cumulative uniform ex- pansion of microparticle layers.
Although the actual morphology of Ziegler–Natta polymerized particles shows a more
complicated hierarchy than the “two-level” approach of the multigrain model, a key feature
is the replication of the porous spherical catalyst morphology into a porous spherical
polymer particle during polymerization. In practice, a careful morphological analysis of the
polymeriza- tion particle requires, beyond microscopic investigations, analyses of particle
size distribution (ASTM D1921-96), poured bulk density (ASTM D1895-96, Method A),
and pourability (ASTM D1895-96, Method A).

Thermodynamic Properties

Melting. The melting point of α-form iPP is strongly influenced by the

stereoregularity (19,23,47,99,140–144) and regioregularity (4,33,34,145,146). Melting point
increases with improved regularity. T 0 is the theoretical equilib- rium melting point of a
perfect and infinitely large crystal. This value exceeds experimentally
m observed melting
points because of kinetic effects leading to small crystal size (23) (Fig. 5). The value of T 0
is sensitive to stereoregularity. However, the value for 100% isotactic material is not
expected to differ significantly from that of highly isotactic commercial materials (23). For
highly regular materials, literature summariesm of T 0 suggest that values of 185–188◦C seem
reasonable (23), though there remains disagreement in the literature (147). Melting points
in the 160–168◦C range are typical for commercial homopolymer samples under nor- mal
analysis conditions. Introduction of comonomer (ethylene, butene, and higher α-olefins)
reduces the melting point, and can
m vanish at intermediate compositions
 16 PROPYLENE Vol.
as the material approaches an amorphous rubber (Table 3). Literature summaries of the heat
of fusion of 100% crystalline material, OH0, generally lie in the range of 148–209 J/g (23).
The most often applied values are clustered around 165 J/g (77,140,148,149) and 209 J/g
(150,151).
As with iPP, the melting point of sPP is highly sensitive to stereoregularity. While
comparison of T 0 for 100% syndiotactic sPP is ambigous, with current catalysts and in
current commercial sPP materials,
m the observed melting point is generally less than that of
Ziegler–Natta iPP under practical crystallization conditions when compared at comparable
stereospecificity (23). This difference is often of the order of 10–15◦C. Introduction of
comonomer also reduces the melting point of sPP. Syndiospecific copolymerization with
butene has the unique feature of being crystalline at all compositions because of the
similarity of crystal structures of sPP and syndiotactic polybutene (152). Literature
summaries of OH0 lie in the range of 105–190 J/gm (23). Values in the lower end of this
range agree with extrapolated heats of fusion versus density (23) and X-ray crystallinity
(116).
Glass-Transition Temperature. The value of the glass transition (qv)
temperature (Tg) is dependent on the crystallinity of the polymer, the molecular weight, and
the measurement techniques used. DSC measurements of Tg for high molecular weight aPP
and sPP are generally similar and usually close to 0◦C. The DSC glass-transition
temperature in highly stereospecific iPP is often difficult to distinguish because of the high
crystallinity. Transition temperatures are gener- ally in the range of 13 to 0 ◦C. Other
techniques, such as dynamic mechanical analysis, are often more sensitive to the Tg of iPP.
Copolymerization with ethy- lene reduces the glass-transition temperature. Figure 7 shows
DSC measured Tg values − of propylene/ethylene copolymers prepared with an aspecific
catalyst. In this case, all samples are noncrystalline and not affected by crystallinity. A simi-
lar trend is generally observed for isospecific and syndiospecific catalysts as well.
Copolymerization with butene is less effective at lowering the Tg .
Heat Capacity and Thermal Expansion. Extensive tabulations of the
extrapolated heat capacities of crystalline and amorphous iPP as a function of temperature
are available (147). The specific volume (153) and heat capacity (154) of iPP in the melt as
a function of temperature and pressure, and typical thermal expansion coefficients in the
solid state as a function of temperature (151) are also reported.

Chemical Properties

iPP is soluble in high boiling aliphatic and aromatic hydrocarbons at high tem- perature. sPP
shows solubility at lower temperature and in lower boiling hydro- carbons. aPP shows the
highest solubility of the three forms. Extensive chemical resistance data is available for iPP
(155). The high chemical resistance of iPP re- sults in exceptional stain resistance, and has
led to the use of iPP in automobile batteries (156). iPP has outstanding resistance to water
and other inorganic en- vironments (155). iPP resists most strong mineral acids and bases,
but like other polyolefins is subject to attack by oxidizing agents including 98% sulfuric acid
and 30% hydrochloric acid at high temperature ( 100 ◦C), and fuming nitric acid (ambient
temperature) (155). Inorganic chemicals produce little or no effect over

∼
Vol. 11 PROPYLENE POLYMERS 17

Fig. 7. Glass-transition temperature (Tg) versus ethylene concentration in propylene– ethylene

random copolymer using aspecific catalyst (no crystallinity). Values were deter- mined by differential
scanning calorimetry (DSC).

a period of 6 months at temperatures up to 120 ◦C (155). For organic media, ab- sorption is
greater for higher temperatures and decreased polarity, and is higher in copolymers than in
homopolymer.
PP reacts with oxygen in several ways, causing chain scission and brittleness that is
associated with the loss in molecular weight. This action is promoted by high temperatures,
light, or mechanical stress. A wide variety of stabilization packages are added for
protection, depending on the application (157).
The reactivity of iPP can also be usefully exploited. For example, treatment with
peroxides has led to controlled rheology resins (61,158,159) with reduced molecular weight
and narrow polydispersity relative to as-polymerized product from Ziegler–Natta catalysts.
For a fixed final melt-flow rate (MFR), lower pre- cursor MFR gives narrower
polydispersity. These resins are used in some fiber- spinning and injection-molding
applications. The creation of radical sites along the polymer backbone, most often through
peroxide-based initiation, is also an essential condition for many functionalization/grafting
chemistries. The func- tionalization chemistry of post-polymerized iPP has been extensively
reviewed (160). The focus has been on the incorporation of polar functional groups into the
polymer chain and/or the graft polymerization of monomers which are generally
incompatible with Ziegler–Natta catalysts. The incorporation of polar function- ality has
been sought to improve paintability, printability, and metal adhesion characteristics
(important to coatings); to act as coupling agents in composites such as glass-reinforced iPP;
to improve oxygen barrier resistance; and to act as compatibilizers in polymer alloys.
Recent advancements in metallocene and
 18 PROPYLENE Vol.
related transition metal catalysts are showing promise for the direct copolymer- ization of
polar monomers with ethylene and propylene (161). Other chemical modifications include
plasma and acid treatment technologies.
Radiation chemistry of PP has been extensively reviewed (162). Under suit- able
conditions, radiation treatment can lead to desirable branching reactions
(163) with rheological characteristics favorable for thermoforming, foaming, blow molding,
and extrusion coating. Resistance to radiation sterilization is an impor- tant requirement in
some medical applications, and requires specialized formula- tion and stabilization
packages.

Physical Properties

A vast number of PP grades are sold in the United States, including homopolymers with
varying molecular weight (melt-flow rate) and tacticity, filled grades, grades with varying
stabilization packages, and grades which incorporate comonomers with varying architecture
(Table 3). The basic categories of iPP polymers are ho- mopolymers, random copolymers,
impact or heterophasic copolymers, and filled polymers. A number of grades are approved
for food contact (164). Many poly- mers carry a UL94 flammability class HB rating (164),
and some specialized for- mulations have improved ratings. Additional specialty grades
include nucleated polymers for improved clarity and mechanical properties, radiation-
resistant for- mulations, chemically visbroken “controlled rheology” grades (61,158,159),
high melt strength grades, and a large number of formulations with specialized addi- tive
functions including slip agents, antiblocking agents, antistats, and pigments to name a few
(157). Standardized test methods such as ASTM or ISO are used in commercial
specifications. The properties that distinguish iPP from high density polyethylene are high
heat distortion temperature, stiffness, hardness, and lower density.
Properties of various homopolymer grades are given in Table 5. In this table, polymer
molecular weight and polydispersity is tailored to give the best processing characteristics for
each fabrication process. The polymers shown in Table 5 are produced by direct
polymerization. Higher melt flow grades, and “controlled rheology” grades with narrow
polydispersity, can be important in other applications, including fibers. Narrow
polydispersity resins can also be produced by metallocene catalysts (4). These catalysts can
also produce iPP with a narrow interchain tacticity distribution and low levels of
extractables. High crystallinity (tacticity) grades, and high polydispersity grades, provide
additional stiffness and higher heat distortion temperature in injection molding applications
(23,120,166,167). Properties of selected random copolymer grades are given in Table 6.
These copolymers have ethylene as the comonomer, although butene copolymerization is
also possible. The stiffness of these poly- mers is lower than that of homopolymers. The
impact resistance is improved, particularly at refrigeration temperatures. Clarity is also
enhanced. The low melting point allows the use of some copolymer grades as sealant layers
in PP films. Properties of impact-resistant, or heterophasic, copolymers are given in Table
7. Much of this class of material is used in injection-molding applications,
 Table 5. Properties of Polypropylene Homopolymersa
Flexural Notched
Melt modulus Deflection Izod
flow, Tensile 1% tempeature Impact Rockwell
(g/10 strength,b secant, at 455 at 23◦C, hardness,
min) Mpac Elongation,b Mpac kPa,d ◦C J/me R
(ASTM (ASTM % (ASTM (ASTM (ASTM (ASTM (ASTM
D1238) D638) D638) D790A) D648) D256A) D785A) Products/Applications
0.5 33 13 1200 96 81 86 Extrusion, sheet, profiles
4 34 12 1400 93 39 86 Injection molding, general-purpose
12 34 10 1400 92 35 88 Injection molding, general-purpose
22 36 10 1500 93 34 93 Injection molding, general-purpose, thin wall
3

35 32 12 1200 95 32 89 Controlled rheology injection molding, thin-wall

a
Ref. 165.
b
At yield.
c
To convert Mpa to psi, multiply by 145. dTo
convert kPa to psi, multiply by 0.145. eTo convert
J/m to ft·lbf/in., divide by 53.38.
 Table 6. Properties of Polypropylene Random Copolymersa,b
Flexural Notched
Melt flow, modulus Deflection Izod
Tensile 1% tempeature Impact
(g/10 strength,c secant, at 455 at 23◦C,
min) Mpad Elongation,c Mpad kPa,e ◦C J/mf
(ASTM (ASTM % (ASTM (ASTM (ASTM (ASTM
D1238) D638) D638) D790A) D648) D256A) Products/Applications
2 28 13 940 79 330 High clarity blow molding, extrusion,
thermoforming
6.5 28 13 920 76 56 High clarity cast film
11 30 13 1000 84 66 High clarity injection molding and
injection-stretch blow molding
3

35 28 13 940 83 140 Controlled rheology high clarity injection molding

a
Ref. 165
b
Based on ethylene comonomer.
c
At yield.
d
To convert Mpa to psi, multiply by 145. eTo
convert kPa to psi, multiply by 0.145. f To
convert J/m to ft·lbf/in, divide by 53.38.
 Table 7. Properties of Polypropylene Impact Copolymers a,b
Flexural Notched
Melt flow, modulus Deflection Izod
Tensile 1% tempeature Impact
(g/10 strength,c secant, at 455 at 23◦C,
min) Mpad Elongation,c Mpad kPa,e ◦C J/mf
(ASTM (ASTM % (ASTM (ASTM (ASTM (ASTM
D1238) D638) D638) D790A) D648) D256A) Products/Applications
0.45 27 11 1100 88 No break Extrusion
2 27 9 1200 90 270 Injection molding medium impact
4 27 8 1200 90 110 Injection molding medium impact
35 27 6 1400 100 70 High flow injection molding and thin-wall medium
impact
50 26 6 1200 107 42 High flow injection molding and thin-wall medium
impact
3

4 21 8 1000 81 No break Extrusion and injection molding high impact

8 26 8 1200 87 100 Blush-resistant injection molding medium impact
22 24 7 1000 83 100 Blush-resistant controlled rheology injection
molding medium impact
12 22 8 900 82 340 Blush-resistant injection molding high impact
a
Ref. 165
b
Based on ethylene comonomer.
c
At yield.
d
To convert Mpa to psi, multiply by 145. eTo
convert kPa to psi, multiply by 0.145. f To
convert J/m to ft·lbf/in, divide by 53.38.
 30 PROPYLENE Vol.
◦
providing impact resistance well below 0 C. Copolymers containing high flexi- ble product.
Mineral fillers, such as talc or calcium carbonate, and other rein- forcements such as glass
fiber or mica, increase the stiffness and heat-distortion temperature. Some properties of
mineral-filled iPP are given in Table 8. Filled formulations based on impact copolymers
(PP-EPR) are also common.

Catalysts for Polymerization

TiCl3-Based Catalyst. Isotactic polypropylene was first synthesized by Natta in

1954 by employing a system consisting of TiCl 4 and Al(C2H5)3 activator (169–174). It was
based on Ziegler’s catalyst system (175), which was used for the polymerization of
ethylene. Only 30–40% of the polypropylene produced by Natta’s catalyst had the typical
characteristics of isotactic polypropylene; for ex- ample, it was insoluble in boiling heptane
and had a melting point of about 165◦C. The remaining product was atactic with poor
structural uniformity and a rub- bery consistency. Natta quickly realized that the polymer
isotacticity was directly connected to the uniformity of the catalyst surface. Thus for the
polymerization of propylene, he employed solid crystalline TiCl3 (obtained by the reduction
of TiCl4)
(176) with Al(C2H5)2Cl or Al(C2H5)3 and obtained a higher percentage of isotactic product
(see ZIEGLER-NATTA CATALYSTS).
The various TiCl3 structural forms, α, β, γ , and δ, were identified in subse- quent
studies (177); the δ-form, in combination with Al(C2H5)2Cl, gave the best results. The high
polymer isotacticity (ca 90%) permitted a scaled-up industrial process by Montecatini,
which had supported Natta’s research at the Politechnic in Milan. This first plant (and
subsequent others) contained a large section for the separation of the undesirable atactic
fraction from the isotactic fraction and a sec- tion for the removal of catalyst residues that
affected product quality. Substantial progress was achieved in a short time (178,179).
Hercules discovered the role of hydrogen as a molecular weight regulator (180). Esso
improved performance by using AlCl3 in TiCl3 solid solution instead of pure TiCl3 (181).
Mitsubishi increased isotacticity to 92–94% by adding an electron donor, such as carboxylic
acid ester, to the TiCl3 (182).
The treatment of the TiCl3 produced from the reaction between TiCl 4 and
Al(C2H5)2Cl, first with the electron donor, diisoamyl ether, and then with TiCl4, gave a
highly stereospecific catalyst. This catalyst system was four to five times more active than
δ-TiCl3 (183) and capable of producing a polymer with narrow particle size distribution.
This system can be referred to as the second-generation catalyst (Table 9). However, the
catalyst yield was still insufficient to reduce cat- alyst residues enough to eliminate the
deashing step in the production process.
In the meantime it was established that only a small percentage of the tita- nium on
the catalyst was actually active. The active titanium was located on the lateral faces and
edges, and along the crystal defects (184). This led to the realiza- tion that much of the
catalyst mass acted as the support (185). Decisive progress could be achieved by depositing
the active Ti on a support whose residues, unlike those of TiCl3 would not be detrimental to
polymer properties.
MgCl2-Supported Catalysts. Magnesium chloride, in the active form as a
support (186,187), increases catalyst yield and allows for the simplification of the
 Table 8. Properties of Filled Homopolymera
Flexural Notched
Melt modulus Deflection Izod
flow, Tensile 1% tempeature Impact Rockwell
(g/10 strength,b secant, at 455 at 23◦C, hardness,
min) Mpac Elongation,b Mpac kPa,d ◦C J/me R
(ASTM (ASTM % (ASTM (ASTM (ASTM (ASTM (ASTM
D1238) D638) D638) D790A) D648) D256A) D785A) Products/Applications
4 34 12 1400 93 39 86 Injection molding homopolymer, general-purpose
3 83 — 4500 157 85 — 20% glass filled homopolymer
4 31 4 1900f 109 37 84 20% talc filled homopolymer
4 29 4 2900f 125 27 94 40% talc filled homopolymer
3

4 30 7 1500f 96 37 89 20% calcium carbonate filled homopolymer

a
Ref. 168.
b
At yield.
c
To convert Mpa to psi, multiply by 145. dTo
convert kPa to psi, multiply by 0.145. eTo
convert J/m to ft lbf/in, divide by 53.38. f Tangent
method. ·
 31 PROPYLENE Vol.
Table 9. Different Ziegler–Natta Catalyst Generations—Composition, Performance,
Morphology and Process Requirements
Productivity,
kg of PP/g of Isotactic Morphology Process
Generation Composition catalyst index control requirements
1 δ-TiCl3 0.33AlCl3
AlEt2Cl· 0.8–1.2 90–94 Not possible Deashing
+
and atactic
removal
2 δ-TiCl3 + AlEt2Cl 3–5 (10–15) 94–97 Possible Deashing
3 TiCl4/ester/MgCl2 5–10 (15–30) 90–95 Possible Atactic
AIR3/ester +
removal
4 TiCl4/diester/MgCl2 10–25 (30–60) 95–99 Possible None
AlEt3/silane
5 TiCl+4/diether/MgCl2
25–35 (70–120) 95–99 Possible None
+ AlEt3
a
Polymerization productivity: hexane slurry, 70◦C, 0.7 MPa, 4 h, with hydrogen for molecular weight control.
Values in parentheses are polymerizations performed in liquid propylene at 70◦C for 2 h with hydrogen.
b
Only possible with alkyl aluminum reduced TiCl3 to produce a catalyst within a 200–300 µm size.

polymerization process in hydrocarbon slurry with the elimination of the costly deashing
step. This discovery and further improvements led to the development of the superactive,
third-generation catalysts (188–193) (Table 9). Better yield, higher stereospecificity, and
morphology control resulted in simplified processes in which the monomer is the
polymerization medium.
Other catalyst systems that are usually soluble in the reaction medium are of high
scientific importance but not used industrially. Among these is the system consisting of the
product of the reaction of AlR2Cl and Al2R2Cl4 with vanadium compounds. This system is
utilized in the production of syndiotactic polypropylene at very low temperature (194,195).
Isotactic polypropylene has been obtained, although with very low yields, from Ti and
Zr benzyl compounds (196) and allylic derivatives (197).
Other highly active catalyst systems are based on single-site catalyst, for example (η-
5
C5H5)2M(CH3)2; (M Ti, Zr). These systems will be discussed in more detail in the
Homogeneous Catalyst section. =
Heterogeneous Catalyst Preparation.
TiCl3-Based Catalysts. First-generation catalysts are prepared by the re- duction
of TiCl4 with metallic aluminum in aromatic solvent between 100 and 200◦C (198,199). A
solid solution with the composition of AlCl 3 3TiCl3 (200) is formed. Milling converts the
crystalline TiCl3 from the α- to the δ-form, increas- ing its surface area · and radically
improving its performance (201); Al(C2H5)2Cl is the best activator. Hydrogen and Zn(C2H5)2
are the molecular weight regulators, but diethyl zinc is rarely used. Typical performance for
this type of catalyst is reported in Table 9. Milling produces a catalyst, and thus a polymer,
with broad particle size distribution.
A second-generation catalyst (183) is prepared by reducing TiCl4 with Al(C2H5)2Cl in
a hydrocarbon at 0◦C. The 3TiCl3·AlCl3 product with surface area
 Vol. PROPYLENE 31
around 1 m2/g thus obtained is washed at 35◦C with diisoamyl ether to remove most of the
AlCl3. The product is then treated with TiCl4 at 65◦C and washed with hydrocarbons. The
catalyst has a high surface area (up to 150 m2/g). In polymer- ization with Al(C2H5)2Cl and
hydrogen, it gives the average performance shown in Table 9. The polymer has regular
shaped, compact particles and narrow particle size distribution, with an average diameter
around 300–400 µm. Catalyst activity is maintained by storage under refrigeration.
MgCl2-Supported Catalysts. The scientific and patent literature reports several
methods for the preparation of the MgCl2-supported catalysts of third and superactive third
generations. They can be classified as follows:

(1) Catalysts obtained by milling mixtures of anhydrous MgCl2 with an electron donor
and a titanium compound (186,187)
(2) Catalysts obtained by milling anhydrous MgCl2 with an electron donor and treated
with a titanium compound (TiCl4) above 80◦C, followed by washing with
hydrocarbons (188–192)
(3) Catalysts obtained by treatment of “active” MgCl 2 with an electron donor and a
titanium compound (TiCl4) under conditions similar to those of the previous case
(202)

The second method gives the average results reported in Table 9 with regard to third-
generation catalysts. In practice, anhydrous MgCl2 and an aromatic ester, usually ethyl
benzoate (EB) in molar ratios between 2 and 15, are shaken in a vibrating mill containing
steel spheres for 20–100 h. The MgCl2 is activated; that is, it is converted from the
crystalline ordered form to the disordered δ-form while the crystallite size is reduced.
The product is treated twice with an excess of TiCl 4 between 80 and 130◦C, washed
repeatedly with hydrocarbons, and dried. During the treatment with TiCl4, the base is
partially extracted, and TiCl4 enters the support. The final composition of the solid includes
0.5–3.0 wt% Ti and 5–15% EB; the remainder is MgCl2. The surface area is in excess of
100 m2/g. The active or δ-form of MgCl2 re- sembles the δ-form of TiCl3 (203). It can also
be obtained by direct contact of anhy- drous MgCl2 with Lewis base or by chlorination of
magnesium organic compounds (188–192). The degree of activation can be determined by
X-ray diffraction (203), where the passage from α to δ gives rise to a widening and
lowering of the diffrac- tion peaks. The active form, because of the corners, edges, and
surface defects of the crystallites, binds the titanium compound strongly, and repeated
washing or treatment under vacuum cannot remove it. This strong binding is attributed to
the closeness of the Mg and Ti ionic radii (204) and to the similarity among the
crystallographic forms of their halogens. The Ti atoms located in exposed sites give rise to
polymerization active centers (205–208). Magnesium chloride may easily interact with
electron donors, probably affecting the stereospecificity of the active site. It can be easily
transformed into particles with controlled morphol- ogy because of the low melting point of
its adducts with alcohols or water. These catalysts are used with aluminum trialkyl alone or
in blends with chlorinated alu- minum alkyls and with a second electron donor, which can
be equal to or different from that contained in the solid catalyst.
 31 PROPYLENE Vol.
Heterogeneous Catalyst Evaluation. To evaluate a catalyst system, the
following characteristics must be known: yield or productivity, which is the amount of
polymer produced per unit of total catalyst or single component; the polymer- ization
kinetics, or yield development with time; stereospecificity; sensitivity to the molecular
weight regulator; the molecular weight distribution of the polymer produced; the
microtacticity of the resulting polymer; ability to copolymerize; and the final polymer
morphology.
To forecast the behavior of a catalyst system in an industrial continuous
polymerization, these characteristics should be determined within a wide range of
conditions; eg, temperature, concentration, and ratios of the various components (activator,
external donor, solid catalyst, etc). A laboratory batch-scale test can provide most of this
information. A small, simple reactor suitable for these studies is shown in Figure 8. The
polymerization can be carried out in a hydrocarbon or liquid propylene. In some cases the
polymerization test can be performed in the gas phase, provided the reactor is prepared with
a suitable heat transfer and catalyst dispersing bed (eg, a salt bed).
The stainless steel reactor is provided with a jacket for temperature control in the
range 0.5◦C using a steam and chilled water mixture. The reactor con- tents are agitated by
a magnetic stirrer± (500–900 rpm). The consumption of the propylene is followed by
weighing the vessel from time to time or through gas flow meters.
A test using hexane as diluent is the simplest to perform because it does not require
the safety measures required for tests in liquid propylene. It can be carried out as follows.

Fig. 8. Lab-scale polymerization reactor.

 Vol. PROPYLENE 31
◦
The reactor is kept under nitrogen at 40–50 C to exclude oxygen and mois- ture.
Anhydrous, deaerated hexane is introduced followed by a fixed amount of catalyst. The
reactor is closed and heated to 60–70 ◦C, and hydrogen and propy- lene are fed at the desired
concentrations. These feeds are maintained for a certain time. Data on the polymerization
kinetics are obtained from the mass of propylene fed over time. At the end of the test the
temperature is reduced, the reactor de- gassed, and the slurry (hexane/polypropylene)
discharged from the reactor (either through a dip tube or a bottom valve). The polymer is
separated from the hexane by filtration or by evaporation. The polymer is dried at 60◦C
under nitrogen, weighed, and analyzed.
Typical graphs follow which represent the performance of a superactive, third-
generation catalyst system as a function of processing variables. The ac- tivity is expressed
as kilograms polymer per gram catalyst, and the isotacticity as polymer percentage residue
from a boiling heptane extraction in a modified Soxhlet extractor for 12 h.
Figure 9 shows the trend of activity for polymerizations in hexane with a propylene
concentration of 14% at 70◦C; Figure 10 shows the isotactic index.
Polymerization temperature, as reported in Figure 11, influences perfor- mance; an
increase from 50 to 80◦C increases both activity and isotacticity.
Isotacticity should be between 94 and 98% to meet various application re- quirements.
The ratio between aluminum and electron donor controls stereospeci- ficity. At very high
levels of electron donor the activity will be reduced, as shown in Figure 12 (209).
The aluminum compound alkylates the transition metal and can act as a transfer agent.
It also removes impurities from the reaction medium (eg, water, CO2, alcohols), thereby
avoiding catalyst poisoning. In addition, aluminum alkyl reduces the transition metal to
lower valence, thus affecting its activity (210). In the case of supported catalysts, aluminum
alkyl plays a fundamental role by complexing the electron donors (204).

Fig. 9. Activity vs polymerization time.

 31 PROPYLENE Vol.

Fig. 10. Isotactic index vs polymerization time.

Fig. 11. Activity and isotactic index vs polymerization temperature.

Polymerization Reaction Mechanisms

The rate Rp of the primary propagation step of monomer to polymer in Ziegler– Natta
catalysis is represented by

Rp = kp[C∗][M]

where kp is the propagation constant, [C∗] the concentration of active sites, and
[M] the monomer concentration.
 Vol. PROPYLENE 31

Fig. 12. Activity and isotactic index vs Al:donor ratio.

The global polymerization rate changes with time. A period of increasing rate is
usually followed by a decline and eventually by a stationary state. The rate decay is often
attributed to the change in concentration of active sites in the context of the above noted
expression. The initial increase may be due to the progressive activation of new active
centers. The deactivation, particularly evident in supported catalysts (211,212), is attributed
to variations in both number and chemical nature of the centers (213).
Global polymerization rate and rate of primary propagation are affected by the catalyst
system and polymerization conditions (203,214). Effects are due to the chemical and
physical structure of the catalyst as well as to the nature of the activator. Important
parameters include the ratio between catalyst and activator, and their concentrations,
hydrogen concentration, temperature, stirring rate, and type and amount of Lewis base. The
effects vary with the polymerization medium; ie, diluent or the monomer in liquid or gas
phase (215–218).
The complexity of the catalyst systems and their heterogeneous nature make an
accurate analysis of the parameters associated with the primary propagation step difficult.
This is one reason why literature analyses are rarely in agreement; also, the data are
obtained by methods and under conditions that are rarely com- parable. A typical example is
the determination of the concentration of the active sites and the propagation constant,
which can be carried out by various meth- ods (chemical, radiochemical) that are not
completely reliable. Nonetheless the parameters are valuable in determining catalyst
potential and performance.
The same uncertainty is involved in the determination of the activation en- ergy of the
propagation reaction and the average lifetime of the polymer chain.
A value of 23 kJ/mol (5.5 kcal/mol) for TiCl3-based catalysts is reported for the
former (219). For the latter, values of 360–600 s (220) and 160 s (219) for TiCl3-
based catalysts are reported at 70◦C and 5 s (221) for MgCl2-supported catalysts
at 45◦C.
A short time after the discovery of Ziegler–Natta catalysts, it was suggested (222,223)
that chain propagation occurred by monomer insertion into a Ti–carbon
 31 PROPYLENE Vol.
bond of the catalyst. This bond was considered to be polarized, with a weak neg- ative
charge on the carbon atom; this hypothesis is still widely supported today. It was confirmed
by IR analysis of chain terminal groups (224) as well as by 13C enriched aluminum alkyls
(225).
In an attempt to provide a model to explain the growth of the polymer chain
satisfactorily, several hypotheses have been suggested. In Boor’s book on Ziegler– Natta
catalysis (179), he exhaustively reviewed the literature and addressed four mechanisms for
chain growth. The four mechanisms are based on the description of the center where chain
growth takes place:

(1) transition metal–carbon bond,

(2) activator metal–carbon bond,
(3) bound radical center, and
(4) bound anion center.

Of these four, the majority of the research tends to support the transition metal–
carbon center model, as the mechanistic scheme of choice. Using a homo- geneous catalyst
(Cp2TiEt2) as a model, Breslow and Newburg proposed that poly- merization growth
occurred at the Ti C center. Somewhat later, Cossee extended this concept into a more
elaborate mechanism for supported catalysts, which he substantiated with molecular orbital
calculations (see Ref. 179, Chapt. 13).
According to this widely accepted model (226–230) for the surface site and surface
coordination environment, the active site is a titanium atom with octahe- dral shape and a
vacant position (the other positions being occupied by an alkyl group derived from the
alkylation by the aluminum alkyl and the remaining chlo- ride atoms). In total the active site
is made up of four ligands, which, in the case of TiCl3, are chlorine atoms, the alkyl group,
and the vacant site.

Monomer insertion occurs through a first step of monomer coordination to the

transition metal, with formation of a π -complex, subsequent weakening of
 Vol. PROPYLENE 31
the Ti C bond, and finally insertion of the monomer coordinated between the transition
metal and the C atom. Because the two positions are not equivalent in the crystal lattice of
the catalyst, the vacancy and the growing chain exchange positions. These phases are
repeated at the insertion of each monomer unit. On the basis of molecular orbital theory, a
semiquantitative interpretation of the mech- anism was provided by assuming weakening of
the metal–carbon bond during olefin complexation in the case of metal ions with 0–3 d-
electrons (as in Ti, V, Cr), and with more difficulty when the number of d-electrons is
greater. This mech- anism is called monometallic; it requires only the participation of the
transition metal, attributing to the aluminum compound the role of forming the active cen-
ter by alkylating the titanium atom. The aluminum compound is also involved, as
demonstrated by the different effects that various aluminum alkyls have on the performance
of these catalyst systems.
Some researchers believe that the active metal (aluminum) directly partici- pates in
directing of the incoming monomer, in which case this system would be considered a bi
metallic site. However, Boor makes a convincing argument that be- cause of the size of the
polypropylene helix, there is no room at the active site for the aluminum to participate. The
titanium site occupies roughly 0.16 nm2 (diameter
0.45 nm) and the cross-sectional area of a polypropylene helix is about 0.35 nm2 (diameter
∼ nm). This would eliminate the titanium and aluminum from shar- ing a common chloride
0.7
∼ preclude aluminum from directly partic- ipating in directing the incoming
bridge and thus
monomer (179). It has been suggested that the effect the different aluminum alkyls have on
polymer microstructure and molec- ular weight is due to the way in which the alkyl
aluminum sets up the titanium chloride surface. These differences in polymer microtacticity
and molecular weight could be due to the fact that the activator may be generating different
isotactic sites.
The basic concept of monomer insertion has now been discussed and for the
polymerization of ethylene this simple description of a monometallic, transition metal–
carbon centered active site would be sufficient to describe the polymeriza- tion. However,
with α-olefins the matter of stereo- and regio-chemistry must be addressed and will be done
in the next section.
Stereo- and Regiochemistry of Monomer Insertion. For substituted α-
olefins a number of issues concerning monomer coordination/insertion must be considered.
The way in which the monomer inserts itself into the polymer chain determines the
microstructure of the polymer and subsequently the properties. Does the methyl group on
the propylene end up towards the chain or away? What is the position of the methyl group
on the propylene molecule with respect to co- ordination site of the metal and the polymer
chain, cis versus trans? The way in which the propylene molecule may be inserted in the
polymer chain by var- ious mechanisms can give rise to complex phenomena of structural
and steric isomerism.
Two possible regio-insertion mechanisms for α-olefins exist—primary and secondary.
In a primary insertion the unsubstituted end (methylene group) of the monomer attaches to
titanium center. This is also known as a 1,2-insertion. In a secondary insertion the
substituted end of the monomer attaches the titanium. This is also called a 2,1-insertion.
 31 PROPYLENE Vol.

The experimental evidence indicates that for the production of isotactic polypropylene
the primary insertion mechanism is predominant. This is substan- tiated via end group
analysis by the presence of isopropyl groups as the end groups after chain transfer by
hydrogen. The regioselectivity of Ziegler–Natta catalysts is very high as have been proved
by NMR analyses (231,232). Syndiotactic polypropy- lene can be prepared by the low
temperature polymerization of propylene using a vanadium catalyst. In this case the
secondary insertion mechanism is operative (233,234).

For stereoregular insertion there are two modes to consider—cis insertion and trans
insertion. For both isotactic and syndiotactic production, the cis mecha- nism has been
determined to be in operation. This was established by polymerizing with cis-, and trans-1-
deuteriopropylene or related monomers. The expected stere- ochemistry was demonstrated
when deuteriopropylene was polymerized. The cis monomers produce erythro monomer
units whereas the trans monomer yields the threo units when cis- and trans-1-d-propylene is
polymerized. In some cases the nomenclature appearing in the literature can be confusing
and contradictory, but all indicate cis insertion. To be specific, as defined below,
stereochemical struc- tures from cis and trans addition to the double bond of cis-(1-d1) and
trans-(1-d1)- propylene to isotactic polypropylene are as follows (229):
 Vol. PROPYLENE 31
Two potential mechanisms have been proposed as governing factors regu- lating the
stereochemistry of the isotactic insertion, the first being the asymmet- ric structure of the
active site (enantiomeric site control) and the second being the asymmetric carbon atom of
the last chain-inserted monomer unit (chain end control). It has been proved experimentally
that the regulating factor is the ac- tive center (enantiomeric site control) because the steric
order is transferred also through ethylene units and because an occasional error is not
perpetuated in the chains (231,232). This mechanism is valid for Ziegler–Natta catalysts and
many (if not most) homogeneous metallocene catalysts. In addition it has been shown (235–
240) how, in the polymerization of racemic α-olefins, a mixture of optically active polymer
can be obtained with an asymmetric catalyst.

Examination of the crystalline structure of the catalyst components reveals the

asymmetric structure of the active center. In TiCl3 six chlorine atoms at the vertices of an
octahedron surround each titanium atom. These six chlorine atoms are then chelated, by
pairs, to three other titanium atoms. This creates two enan- tiomorphic structures.

The next section will discuss some of the concepts associated with models for the
catalytic sites.

Ziegler–Natta Active Site Models

An important characteristic of a polymerization catalyst is its morphology or struc- tural

architecture. Titanium chloride exhibits four different crystalline modifica- tions: α, β, γ
and δ forms depending upon the method of preparation. In each
 32 PROPYLENE Vol.
case the titanium atom is octahedrally coordinated, except at various defects and edges. The
violet α-form, and the γ and δ forms consist of regular stacking of Cl Ti Cl layers
containing titanium atoms between two layers of chloride ions. The α-TiCl3 form, which
will be discussed, is hexagonally close packed.
Stereospecific behavior of the catalyst site is related to the chirality of the surface
sites of the solid TiCl3. Models by Corradini can explain a number of observations—the
type of tacticity errors along a predominantly isotactic chain, stereospecificity of the
initiation reaction, and the maintenance of isotacticity af- ter the insertion of ethylene
monomer in the chain (241). Furthermore, since violet TiCl3 and MgCl2 solid-state crystal
structures are similar, Corradini’s model re- lates well to both TiCl3 and MgCl2-supported
catalysts.
The structure of titanium trichloride and other Ziegler–Natta catalysts con- sisting of
transition metal halides (VCl3, CrCl3, etc) is comprised of layers of close- packed chlorine
atoms. The crystalline modifications are based on the different ways in which the layers are
stacked on top of one another. Within each layer the metal atoms reside in an ordered
arrangement. The metal atoms occupy two thirds of the octahedral positions. The adjacent
metal atoms, which are bridged by two chlorine atoms, have opposite chirality. The chirality
of these metal atoms is designated ▲ and O according to the International Union of Pure
and Applied Chemistry (IUPAC) nomenclature (242). In the MgCl 2 structure there are no
va- cancies in the lattice and every octahedral position is occupied by a magnesium atom;
however, it can be seen that the gross structures of TiCl 3 and MgCl2 are similar. This
similarity in structure between TiCl3 and MgCl2 is what makes mag- nesium chloride a
suitable substrate for the deposition of titanium chloride to form active sites. It was
determined early on that only a small portion of the titanium on TiCl3 catalysts was active;
the rest was essentially acting as a support. The bulk of the titanium chloride was the culprit
in causing polymer instability and degra- dation problems. When using MgCl2 as a support
the yield of polymer per weight of catalyst is much greater. Along with the fact that the
magnesium chloride is practically inert, this produced a polymer that was inherently more
stable.
A diagram of the TiCl3 structure of the (110) cut and the (100) cut of MgCl 2 is shown
in Figure 13; the chloride atoms are omitted for clarity.
Magnesium chloride can be treated either chemically or physically (ball milling) to
achieve activation. Activated MgCl2 has a very disordered structure, which consists of very
small lamellae. In the bulk, magnesium atoms are coor- dinated to six chlorine atoms, but at
the lateral edges or cleaved surfaces the coordination is with 4 or 5 chorine atoms. These
lateral cuts correspond to the
(110) and (100) faces respectively for magnesium chloride. Upon treatment of ac- tivated
MgCl2 with TiCl4, the bridged dinuclear Ti2Cl8 species coordinate on the
(100) surface while the single TiCl 4 species usually prefer the (110) faces. Treat- ment of
the catalyst by aluminum alkyls will reduce the Ti 2Cl8 to Ti2Cl6 species and alkylate the
titanium. This will generate both TiCl3 and Ti2Cl6 species on the MgCl2 support. The
placement of these Ti2Cl6 units on the (100) lateral surface of MgCl2 produces sites very
similar to those on the (110) surface of TiCl3 catalysts. These Ti2Cl6 sites are chiral and
stereospecific. Coordination of the TiCl4 species on the (110) faces and reduction by
aluminum alkyls produce TiCl3 sites, which lack chirality and are nonstereospecific
producing sites for propylene polymeriza- tion. In Figure 14, is shown both the TiCl3
catalyst and MgCl2/TiCl4 catalyst after
 Vol. PROPYLENE 32

Fig. 13. Comparison of the structures of α-TiCl3 and MgCl2.

reduction with AlR3. Now with the basics of the active centers and structures described, the
details of the polymerization mechanism can be discussed.
As mentioned previously, a two-stage reaction mechanism for the polymer- ization of
propylene was proposed which consists of a coordination stage of the olefin then followed
by an insertion step of the monomer into the Ti–polymer bond. Although this picture (Fig.
15) reveals the basic mechanism of monomer coordination and then insertion, it does not
provide any insight into the mecha- nism for stereocontrol of the propylene monomer. This
is addressed by chirality considerations.

Fig. 14. Crystal structures of α-TiCl3 and MgCl2 with Ti2Cl6 clusters. Chlorine atoms are shown at the
catalyst sites, and substrate chlorine atoms are omitted for clarity.
 32 PROPYLENE Vol.

Fig. 15. Propylene coordination to titanium and then insertion of propylene into the Ti–polymer bond.

As discussed previously the active sites on TiCl 3-based catalysts or MgCl2- supported
catalysts are chiral. In all, there are three elements of chirality that are considered with
respect to the polymerization of α-olefins and stereospecificity of the polymer:

(1) α or δ chirality of the titanium atoms

(2) The si or re chirality of the coordinated propylene monomer
(3) The chirality of the tertiary carbon atoms of the growing polymer chain

For this discussion the bridged dimers of Ti2Cl6 on the lateral face of a MgCl2 crystal
will be considered. If Figure 16 (below) is studied in detail, it can be seen that at one of
titanium atoms of the Ti2Cl6 cluster there are two positions that are available to the growing
polymer chain and the coordinating monomer. One position can be considered outside or
away from the bulk of the crystal and the other can be considered inside, or towards the
bulk of the crystal lattice. These two positions are not sterically equivalent. Through
computer modeling calculations, Corradini and co-workers demonstrated that the most
favorable energy position for the growing chain was at the inside position (241).
If the most favorable energetic position for the chain is at the inward position, then the
monomer must occupy the outside location. For the ▲ titanium site, the si outward
coordination of propylene is favored (re for the outward O titanium site). In this model the
production of stereospecific polypropylene can be explained.
For syndiotactic propagation it was proved that the steric control of the in- sertion
comes from the last inserted monomer unit, whose hindrance affects the
 Vol. PROPYLENE 32

Fig. 16. Inward and outward positions available for occupation by the monomer and polymer chain.

insertion of the subsequent unit in such a way as to prevent a monomer molecule with the
same configuration of the previous molecule from entering (243,244). Recall that
syndiotactic polymerization from Ziegler–Natta vanadium catalysts differs from this
isotactic polymerization monomer insertion. For the vanadium syndiotactic catalysts
secondary or 2,1-insertion is operative (233,234). More re- cently developed homogeneous
syndiotactic catalysts follow 1,2-insertion.
Summarizing, isotactic and syndiotactic propagation highlights the cata- lyst site as
the entity controlling stereoregularity for iPP and the last monomer unit of growing chain
for syndiotactic propagation with vanadium catalysts. The monomer insertion type is
primary (1,2) for isotactic and secondary (2,1) for this latter syndiotactic catalyst.

Electron Donors

There are two classifications of electron donors, internal and external. Electron donors are
thus named for their ability to act as Lewis bases and donate electrons to Lewis acid sites.
For the TiCl3 type catalyst, the electron donors are tradition- ally referred to as just donors
and are generally amines, esters, ethers, alcohols, etc. Their roles range from modifying the
catalyst site and structure of the TiCl3 substrate to complexing with the alkyl aluminum
either during catalyst prepara- tion or catalyst activation.
For MgCl2-supported catalysts (the third- and fourth-generation types; see Table 9 the
electron donors are classified as internal and external based on their sequence of addition
during catalyst preparation or during activation, respectively. For the preparation of the
MgCl2-supported catalyst an electron donor is added
 32 PROPYLENE Vol.
during the process of adding the TiCl4 (titanation of the support) to the activated MgCl2-
support; this is referred to as the internal donor. At the stage of activa- tion of the MgCl2-
supported catalyst, another donor is added during the addition of the alkyl aluminum; this
donor is referred to as the external donor. For the third-generation MgCl2-supported
catalyst, the internal donors are usually ph- thalates and the external donors are alkyl
alkoxy silanes. For the fifth-generation type catalysts, the internal donors are diethers and
the external donor (alkyl alkoxy silane) may or may not be used. In most cases electron
donors are used to increase the activity and stereospecificity of the catalyst system.
Donors work in a variety of ways and the roles of both internal and external donors as
they relate to the MgCl 2-supported catalysts will be discussed in this section. As mentioned
previously, MgCl2 can be activated through mechanical or chemical means. The process of
activating magnesium chloride through mechani- cal means can be achieved by ball milling.
This process usually involves co-milling the magnesium chloride with a Lewis base (233).
The preparation of the catalyst is followed by treatment of the activated magnesium chloride
with TiCl4 and heat. The subsequent treatment of the MgCl2/internal donor support with
TiCl4 usually removes some to the internal donor. In other cases the activated MgCl2-support
can be treated with TiCl4 and the Lewis base at the same time, and then given a heat
treatment. The internal donor helps structurally stabilize the activated magnesium chloride,
prepare sites for the TiCl4, and possibly block certain sites on the MgCl 2 surface, thus
making them unavailable for titanium chloride.
Whereas third-generation Ziegler–Natta catalysts use an ester (ethyl ben- zoate, EB) as
internal donors and another ester as the external donor (methyl para-toluate, MPT), the
fourth-generation catalysts use a phthalate (usually di- iso-butylphthalate, DIBP) as the
internal donor and an alkyl alkoxy silane (eg phenyl triethoxy silane, PES) as an external
donor (233). In both cases the inter- nal and external donors play similar roles. The internal
donors stabilize the MgCl2 support and set up positions for the TiCl 4 complex. The purpose
of employing an external donor in both the third- and fourth-generation catalyst systems is
that without them, the stereospecificity of the catalyst would be very poor for the poly-
merization of propylene. This is due to the fact that activation of the catalyst with the alkyl
aluminum extracts a portion of the internal donor out of the catalyst. The external donor is
added during the activation of the catalyst for the purpose of occupying the site left vacant
by the extracted internal donor (233).
From the discovery of the high yield MgCl2-supported catalysts it was ap- parent that
the presence of both an internal and external donor was a neces- sary condition for a highly
active, highly stereospecific catalyst. Soon after, it was learned that an interrelationship
existed between the two donors and, further- more, specific donor pairs gave optimum
results. However, with the advent of the fifth-generation catalysts, those employing diethers,
there was no need for the ad- dition of an external silane donor (234). These catalysts,
MgCl2/TiCl4/diether, did not lose the internal donor during activation with the alkyl
aluminum and con- sequently maintained relatively good stereocontrol and excellent
activities (2 to 3 times that of the conventional MgCl2-supported/phthalate catalysts)
(245,246) (see Table 9). Interestingly enough, these diether donors can also be used as ex-
ternal donors in the more traditional MgCl2-supported/phthalate/TiCl4 catalyst systems and
they produce the same type of homopolymer. This supports the the- ory that there is a
phthalate/external donor exchange during activation where
 Vol. PROPYLENE 32
the external donor (diether) occupies the site left vacant by the phthalate, very similar to the
phthalate/alkoxy silane exchange. In general the following model best describes the
interactions between the internal and external donors with the catalyst and activator
(246,247):

(1)

(2)

(3)

(4)

(5)
As previously mentioned, the alkyl activator will extract internal donor from the catalyst;
therefore equilibrium 1 is always present even in the absence of an exter- nal donor.
Depending upon the location of the titanium site on the magnesium chlo- ride support, the
free site (Cat-Q) could be isospecific or aspecific. Equilibrium 2 is also present because of
the propensity of the external donor and AlR 3 to form acid/base adducts. These ED AlR3
adducts may also form exchange products (248). In fact, these exchange products are
important because they reduce the concen- tration of free external donor present in the
·
system, which also acts as a poison. In the presence of excess aluminum, there should be
very little free external donor present in the system; therefore equilibrium 3, (the interaction
of catalyst with free external donor) can be neglected. For the external donor to interact with
the free titanium site there are two possibilities—through the ED AlR 3 complex (equilibrium
4) and through the ED AlR3 complex initially as a carrier for the ED (equilibrium 5). In
equilibrium 5, the ED AlR3 complex releases the ED to the vacant titanium site.
Experimental results indicate equilibrium 5 as the most likely case. The alkyl aluminum
activates the titanium and removes the inter- nal donor but does not participate in the actual
stereochemical regulation of the polymerization of propylene (246). ·
· internal and external. For the third- generation
To recap, there are two types of donors,
catalysts the internal/external pair is an ester/ester pair,
· while for the fourth-generation it is
the phthalate/alkoxy silane pair. The fifth-generation Ziegler–Natta catalysts are comprised
of an internal donor, which is a diether compound and may be used with or without the
additional external donor (see Table 9).
There has been a large effort towards elucidating the roles of the external donor in the
polymerization of propylene. Some of this work has centered on the
 32 PROPYLENE Vol.
structural considerations of the external donor. Since the MgCl2/TiCl4/phthalate– alkoxy
silane/AlR3 catalyst system is the most widely used commercial catalyst system for
polymerization of propylene, the focus in the next section will be on the alkoxy silane as the
external donor.
Structural Considerations of Electron Donors. The most effective
alkoxy silanes used in the polymerization of propylene give the greatest activity along with
the highest selectivity. These types of silanes are of the general for- mula R1R2Si(OR)2 and
R1Si(OR)3, with R methoxy or ethoxy (any larger alkoxy group is ineffective). Early on,
PhSi(OEt)3 was used; however, the more effective = silanes for catalyst performance are those
that contain methoxy groups (R OMe). Interestingly enough, silanes with only one alkoxy
group were poor performers, producing polypropylene with low isotacticity. The best
activity/selectivity balance was found for silanes having two methoxy groups and two = alkyl
groups (R1 and R2) that are relatively large—in other words, sterically bulky. The
importance of steric volume was investigated by Okano and co-workers (249). They
correlated silane molecular volume and electron density (calculated by molecular modeling)
with polymerization performance (activity) and stereocontrol (isotactic index). In their study
they found a straight-line correlation with silane volume and isotacticity. Both Okano’s
group and Ha¨ rko¨nen (250) found a decrease in atactic polymer for- mation with an
increase in electron density.
In addition to affecting polymer stereoregularity and catalyst activity, silane donors
influence other aspects of propylene polymerization. It was also found that external
silane donors influenced molecular weight of the polymer via the use of hydrogen as a
chain-transfer agent. When using hydrogen to decrease the molecular weight of the polymer,
there was also an increase in activity when us- ing an external alkoxy silane with the
MgCl2/TiCl4/phthalate ester catalyst sys- tem (251–254). While hydrogen is used as a
molecular weight control agent, this level of control varies from external donor to external
donor. In some cases, un- der the same hydrogen charge, two different donors will give
polypropylene with two very different molecular weights (255). These differences between
donors in molecular weight control are not unlike the differences observed in stereocontrol.
Some donors are better at stereocontrol than others. The general observation is that donors
with good stereocontrol or high selectivity (usually those containing bulky hydrocarbon
groups) are usually poor at effectively using hydrogen to lower the molecular weight of the
polymer. Donors with poor regio- or stereocontrol exhibit good hydrogen response.
Chadwick and co-workers have explained this phenomenon by the process of chain
transfer after a regioirregular insertion of propylene (256). With a 2,1-insertion it is harder
for the next propylene unit to insert into the Ti CH(CH3)CH2–polymer bond; however,
there is a probability of hydrogen inserting to affect a chain transfer. Therefore those donors
with low regio-control permit more 2,1-insertions, thus making chain transfer with hydro-
gen easier. Donors with good regio- control make far fewer 2,1-insertions, thus denying
an easy path and an opportunity towards chain transfer with hydrogen. Molecular
Weight and Molecular Weight Distribution Effects of Electron Donors.
The production of broad molecular weight distributed polypropylene is a function of the
multiplicity of active centers, which differ in stereo- and regiospecificity, and the
propagation rate. Lewis bases in general, and external alkoxy silane donors in particular
can influence the molecular weight
 Vol. PROPYLENE 32
distribution of polypropylene by associating with these multiple active centers
(245,257,258). The regio- and stereoselectivity of the external donors affects the molecular
weight distribution of the polymer indirectly. Normally the atactic frac- tion of
polypropylene has a lower molecular weight. This is in part due to re- gioirregular insertions
of the monomer, which in turn leads to the easier path- way of hydrogen complexation and
subsequent chain transfer (259). It has been shown that the different fractions of
homopolypropylene separated by tempera- ture rising elution fractionation (TREF) not only
differ in microtacticity but also by molecular weight (245). For the MgCl2/TiCl4/diether
based catalyst, Chadwick and co-workers found that the higher temperature eluting fractions
also had higher molecular weights. Not unexpectedly, each fraction from TREF exhibited a
narrow molecular weight distribution (259).
Microtacticity Considerations. In the early days of Ziegler–Natta poly-
merization of propylene, two concerns were activity of the catalyst and the stere-
ospecificity of the polymer. Stereospecificity or stereoregularity of the polymer was
measured by the amount of insolubles produced versus the amount of solubles. However,
another important feature affecting the performance of polypropylene is the microtacticity of
the polymer chain. Nowadays, with the aid of TREF and 13C NMR spectroscopy, the overall
stereospecificity of polypropylene can be fur- ther defined by observing stereoirregular and
regioirregular insertions. The type and quantity of these disruptions in the chain lead to
differences in the physical properties and processing performances.
Chadwick and co-workers demonstrated that the quantities of the three ma- jor
fractions derived from TREF vary with different donors (256). The fraction of polymer that
elutes from 26 to 95◦C, normally referred to as the stereoblock frac- tion, can vary from as
little as 4% up to 21% by weight of the total polypropylene sample (260). The differences in
homopolymer made with two different donors can be seen by looking at the mmmm pentads,
via 13C NMR spectroscopy, of the TREF fraction eluting from 96 to 125◦C.
In summary, donors, both internal and external (alkoxy silanes and diethers), affect
catalyst performance and polypropylene properties in the following ways:
(1) The internal donors restrict the placement of titanium on the (110) face of MgCl2.
(2) The internal donors shift the equilibrium between the aspecific monomeric species
(TiCl3) to the stereospecific dimeric species (Ti2Cl6) on the (100) face of the MgCl2.
(3) The external donors transform nonstereospecific sites into isospecific sites by
blocking open coordination sites near the titanium active centers.
(4) External donors control the path of the incoming monomer to varying de- grees:
a. Little control produces stereo- and regioirregularities that result in mi- crotacticity
changes; this in turn affects crystallinity and thus the sol- ubles levels.
b. Lack of monomer insertion control also affects regioirregular insertions (2,1-
insertions), which leads to chain transfer by hydrogen and lower molecular
weight.
 32 PROPYLENE Vol.
Metallocenes

Metallocene catalysts for the polymerizations of olefins have been known since early 1957
when Natta and co-workers first reacted triethyl aluminum (AlEt 3) and
bis(cyclopentadienyl) titanium dichloride (η5-C5H5)2TiCl2 to form a complex that
polymerized ethylene. The structure of this complex was described and the polymerization
results reported. With ethylene they reported to have made 7 g of crystalline polyethylene in
about 8 h at 95◦C with 40 atm ethylene pressure in n-heptane. Later in the same year,
Breslow and co-workers repeated Natta’s exper- iments. They found that the blue complex
described by Natta was a somewhat poor catalyst (in agreement with Natta’s findings), but
discovered that small amounts of oxygen in the ethylene boosted polymerization activity.
When compared to the heterogeneous Ziegler–Natta catalyst system, these metallocene
catalysts were poor with respect to polymerization activity. They were used essentially for
mech- anistic studies because of their simplicity and ease of structure elucidation.
In 1975 Kaminsky found that a slight amount of water added to a mix- ture of
biscyclopentadienyl titanium dimethyl and trimethyl aluminum rapidly polymerized
ethylene. Eventually it was determined that the addition of water produced
methylaluminoxane (MAO), which was responsible for the boost in ac- tivity (261).
Although these Cp2MX2/MAO catalyst systems rapidly polymerized ethylene and
copolymerized other monomers with ethylene, they were less effec- tive for the
polymerization of propylene. There were several critical shortcom- ings with these early
metallocene catalysts toward the polymerization of propy- lene: low activity, poor
stereospecificity, and low molecular weight polypropylene production.
Because these metallocene catalysts are discrete, single molecules, their structures can
easily be determined by X-ray crystallography. This allows the catalyst chemist to begin the
process of relating the structure of the metallocene catalyst to polymer properties (molecular
weight, stereoregularity, stereospeci- ficity, etc) and polymerization activity. This ability to
relate catalyst structure to polymer properties allows the process of elucidating
polymerization mechanisms and designing catalysts to tailor make polymers and
copolymers. This is a big advantage over the traditional Ziegler–Natta catalysts, which
contain multiple sites (see METALLOCENES; SINGLE-SITE CATALYSTS).
General Description of Structures. A brief description of metallocene
catalyst structures is necessary before going any further in discussions concerning these
systems. In their most basic form these metallocenes consist of a Group IV metal (M Ti, Zr,
Hf), two π -bonded cyclopentadienyl rings, and two sigma- bonded groups (these rings are
also referred to=as carbocyclic π -ligands while all the groups attached to the metal are
generally referred to as ligands). Because of the nature of the π -bonding between the metal
atom and the carbocyclic π -ligands, there is free rotation of these ring groups about their
metal-to-ring centroid axes. This free rotation makes these metallocenes nonstereorigid and
produces atactic polypropylene.
A bridged metallocene contains a linking unit (usually CH2, CH2CH2 , or SiR2)
between the two carbocyclic ligands. These bridged metallocenes are stere- origid, but will
not necessarily produce stereospecific polypropylene.
 Vol. PROPYLENE 32

Furthermore, these bridged metallocenes can become more complicated structurally

with more substitutions on the carbocyclic π -ligands. In fact, the more complicated
bridged metallocenes are of commercial interest be- cause they provide for the
production of isotactic and syndiotactic forms of polypropylene at high molecular
weights. Because of the carbocyclic π -ligand’s complexity, catalyst stereochemistry
will be addressed in the next few paragraphs.
Bridged metallocenes can be further classified based on their stereochem- istry. Using
an ethylene-bridged bis-indenyl metallocene (CH2CH2[Ind]2MX2) as an example, one can
follow the spatial relationships between the ligands and their effects on monomer insertion
mechanisms and polymer properties. The CH2CH2[Ind]2MX2 metallocene can exist in two
forms, mesomeric and racemic (meso and rac). The meso-CH2CH2[Ind]2MX2 catalyst has a
mirror plane and is nonstereospecific in the polymerization of propylene; it is designated
as Cs-symmetric. The rac-CH2CH2[Ind]2MX2 catalyst has a nonsuperimposable mir- ror
image and, to a certain degree, each enantiomer is a stereospecific catalyst site; it is labeled
as C2-symmetric.

Finally, two other structural types of metallocenes need to be mentioned. One type
consists of carbocyclic π -ligands, which are not identical but each possessing a plane of
symmetry (orthogonal to the ring plane). These types of metallocenes are labeled as Cs-
symmetric. The other type of metallocene contains two noniden- tical carbocyclic ligands,
one symmetrical and one asymmetrical. These types of
 33 PROPYLENE Vol.
metallocenes are labeled as C1-symmetric.

Implications of Metallocene Catalyst Structure on Polypropylene

Structure. The previous section gave a brief description of the various types of
metallocenes. In this section a general relationship between metallocene struc- ture and type
of polypropylene produced will be made. It is important to note that these are
generalizations. While the stereochemistry of the metallocene plays an important role in
mechanism of monomer insertion and ultimately the stereo- and regiospecificity of the
polymer, the substituents and location of the substituents on the carbocyclic π -ligands also
effect the microstructure of the polymer.
In general, metallocenes that are either bridged or nonbridged and posses C2v
symmetry (eg, Cp2TiCl2 or Me2Si[fluorenyl]2ZrCl2) will produce atactic polypropy- lene.
This is because these types of catalysts have low stereocontrol. The only stereocontrol
mechanism operating in these systems is chain end control from the polymer. Consequently
the polymer is predominately atactic at normal polymer- ization temperatures. In fact, these
metallocenes are the best source for producing high molecular weight, high atactic
polypropylene (very low to almost zero crys- talline polymer).
To produce highly isotactic polypropylene the metallocene catalyst should be a
bridged chiral metallocene (preferably containing zirconium) having C2 symme- try with
some alkyl substitution in the ring (262). However, with the right sub- stitution others have
shown that a C1-symmetric metallocene can also produce highly isotactic polypropylene
with high Tm (161◦C) (263).
Early metallocenes were less than desirable polypropylene catalysts because they
produced polymer with low stereocontrol and low molecular weight. Recently there has
been much progress in making high molecular weight, high stereo- and regioregular
polypropylene with relatively high melting points. In general, all these metallocenes are
structurally complicated and the reader should refer to references cited.
Advantages of Metallocenes. Because they are discreet molecules, one of the
important features of metallocenes is that their structures are easily de- duced. This allows
almost direct correlation between the catalyst’s structure and the microstructure of the
polymer produced. This fact allows for a rather rapid evolution of focused catalyst design
to tailor polymer properties to spe- cific needs. For example, one of the early successful
metallocenes used in the isospecific polymerization of propylene was [Et(Ind)2]ZrCl2. By
structurally char- acterizing this compound and making some changes in the bridging group,
a new metallocene was prepared [Me2Si(Ind)2ZrCl2] which produced a higher molecular
 Vol. PROPYLENE 33
Table 10. Evolution of the Carbocyclic π-ligands and the Effect on Polypropylene
Propertiesa
kg PP/mmol MW, Melting Isotacticity,%
Metallocene Zr·hr
g/mol point, ◦C mmmm pentads

[Et(Ind)2]ZrCl2 188 24,000 132 78.5

[MeSi2(Ind)2]ZrCl2 190 36,000 137 81.7
[MeSi2(2-Me-4,6-iPr2(Ind)2]ZrCl2 245 213,000 150 88.6
[MeSi2(2-Me-4-PhInd)2]ZrCl2 755 729,000 157 95.2
a
Bulk polymerization in1L of liquid propylene at 70◦C, Al/Zr = 15,000 using MAO as activator.

weight polypropylene with high isotacticity and, consequently, a higher melting point. The
evolution of the carbocyclic ligand and subsequent improvement in isotactic polypropylene
properties can be seen in Table 10.
From the standpoint of monomer insertion control, judicious choice of groups and
location on the carbocyclic π -ligand produce polypropylene with higher iso- tacticity,
greater molecular weight, and higher melting points.
Polymer Property Advantages. Metallocene catalysts, through varia- tion of
catalyst structure, can produce a broad spectrum of polymer microstruc- tures leading to a
very wide property envelope which is potentially accessible. Because of the defined
molecular structure of the catalyst, once a catalyst is cho- sen for a given application, the
properties can be precisely controlled.
Some of the property advantages and types of polypropylene and copolymers of
propylene that can be realized with metallocenes are as follows:

(1) Highly tunable tacticity microstructure (isotactic, syndiotactic, atactic,

isoblock/stereoblock, regiospecific microstructures)
(2) Narrow molecular weight distribution
(3) Absence of oligomers and extractables
(4) Improved melting point/extractable balance
(5) Narrow interchain composition distribution in copolymers
(6) Polypropylene with vinyl end groups
(7) Polymerization of expanded comonomer types (including dienes and cyclo- olefins)

Activators/Cocatalysts. As in conventional Ziegler–Natta catalysts,

metallocene catalysts must be activated before polymerization of olefins can pro- ceed.
Methyl aluminoxane (MAO) has been the activator of choice since its discov- ery by
Kaminsky in 1975 (261). MAO is prepared by the controlled hydrolysis of trimethyl
aluminum. The product is a relatively difficult species to characterize. It consists of
oligomers, both linear and cyclic. The composition of the numer- ous oligomer structures
varies depending upon the preparation methods, concen- tration of the reactants,
temperature, and time. The two simplified linear and cyclic oligomeric alumoxane structures
(with a general alleyl group) are shown in Figure 17.
 33 PROPYLENE Vol.

Fig. 17. Aluminoxane oligomer structures, linear and cyclic.

Fig. 18. Aluminoxane activating Cp2ZrCl2.

The role of MAO in the activation of a metallocene is essentially the same as in the
traditional Ziegler–Natta catalyst that is to alkylate the halogenated metal center. The MAO
forms a cationic complex with the metallocene and a dispersed anionic charge on the
aluminoxane. An excess of MAO will lead to dialkylation of the metallocene metal center.
One of the main disadvantages of aluminoxane activators is the high aluminum
concentration level needed; typical AlZr ratios are over 1000:1. The basic mechanism for
the alkylation and activation of a Cp2ZrCl2 is shown in Figure 18.
Other cocatalysts can be used to activate single-site catalysts. Some other typical
activators, in addition to other alumoxanes, include tetraphenylb- orate [(C6H5)4B− ],
tetra(perfluorophenyl)borate [(C6F5)4B− ], and carborane (C2B9H12 − ).
Supportation of Single-Site Catalysts. Without the ability to support
single-site catalysts, the commercial use of these systems would be eliminated in the
numerous bulk and gas-phase polymerization facilities used today. In essence, they would
be restricted to a limited number of slurry processes. The strategy of supporting a single-site
catalyst for an industrial process is much the same as that for the heterogeneous Ziegler–
Natta catalysts. The target is a morphologi- cally uniform catalyst particle that is easy to
feed into a slurry, bulk-monomer, or gas-phase process, which produces a polymer that is
roughly the same shape as the catalyst particle but that is 20 to 200 times in volume. In
addition the catalyst particle should be robust and not easily fractured during
polymerization. Frac- turing of the polymer particle during polymerization produces fines
that foul the process.
There are a number of materials suitable for supporting single-site or metal- locene
catalysts. These are inorganic oxides, metal halides, and polymers. Among the inorganic
oxides, silica or silica gels have been the supports most widely used because of their wide
range of particle sizes, porosities, etc. More about
 Vol. PROPYLENE 33
silica supports will be discussed later. Other types of supports, which will not be discussed
in detail, are the other types of inorganic oxides such as alumi- nas (zeolites type materials),
MgO, etc. However, it should be noted that zeo- lites have shown some promise as supports
because of their ordered structures and precisely known pore sizes. Other support materials
are polymers which in- clude polystyrene, derivatized polystyrenes, polysiloxanes, and
various polyolefins (porous types of polyethylene and polypropylene) and copolymers. The
following discussions will be predominately focused on silica gel since it is the most widely
used support for metallocenes.
There are essentially three methods in which to make a supported- metallocene
catalyst:

(1) Supporting the metallocene, then treating with an activator

(2) Supporting the activator, then treating with the metallocene
(3) Preparing the metallocene/activator complex in solution, then treating the support
with the complex

Supporting the Metallocene, then Treating with the Activator. Be-

cause of their method of preparation, silicas have a high concentration of surface hydroxyl
groups and complexed water. Normally both hydroxyl groups and wa- ter will promote
decomposition of a metallocene or metallocene/activator complex. Therefore it is necessary
to remove the water either by thermal or chemical de- hydration. Full dehydration of silica
can occur at 150◦C. At this point the silica gel will have a surface that is fully hydroxylated.
Even these hydroxyl groups can decompose some types of metallocenes. In most cases the
silica gel can be further dehydrated at 400◦C (see Fig. 19).
Much work has been reported in the literature on the preparation of sil- ica at various
temperatures and under various conditions. Needless to say, the calcination conditions play
a large part in preparing the silica support for the metallocene.
As far as the actual mechanism of how the metallocene is supported, various theories
abound. However, the metallocene can be “connected” in two ways to

Fig. 19. The steps to the dehydration of silica gel.

 33 PROPYLENE Vol.

Fig. 20. Methods of attachment of a metallocene to silica.

the support: (1) through bonding to the metal and the oxygen of the Si O to create the Si O
Metal complex, or (2) by connection of the metallocene to the support via a linkage through
the carbocyclic ligand moiety (Fig. 20). Tethering the metallocene to the support through the
ring system is the method of choice, but requires preparing specialized carbocyclic ligands,
which have spacers with groups that are able to react with residual OH groups on the silica.
In this case it is necessary to have some remaining hydroxyl groups on the silica.
Supporting the Activator, then treating with the Metallocene. Sup-
porting the alumoxane on the silica support first and then treating with the met- allocene was
one of the first methods used for supporting metallocenes. Many methods of silica treatment,
prior to contacting with the MAO, have been re- ported. In addition, treatment of the silica
gel with solutions of MAO at various temperatures and pressures, heating with dry MAO,
etc, have been reported. All of these methods have advantages and disadvantages depending
on the metallocene, monomers polymerized, and polymerization conditions used.
Supporting the Metallocene/Activator Complex. The last method to
discuss is the supportation of the Metallocene/MAO complex. In this method the
metallocene and the MAO are reacted in solution to form a complex. The silica is then
treated with the metallocene/MAO solution. The solvent is then evaporated from the silica
support to leave a dry, free-flowing catalyst. In some cases the sup- port is held under low
pressure while treated with the metallocene/MAO solution to assist in the impregnation of
the catalyst complex into the pores of the silica. Other methods have been developed which
improve the impregnation of the met- allocene/MAO complex into the interstitial pores of
the support. Concentrating the catalytic complex into the interior of the support rather than
on the surface improves flowability of the supported catalyst system, decreases fouling, and
im- proves the final polymer morphology. While much is known about the structure of the
single-site catalyst and the properties of the resulting polymers, much of this changes upon
supportation.
Commercialization Aspects of Metallocene/Single-Site Catalysts.
While metallocenes have been known to polymerize olefins since the 1950s, only in the last
10 years have they been introduced commercially. Exxon introduced its first generation
metallocene in 1989 for the limited production of polyethy- lene. This was a homogeneous
catalyst used in solution. Other companies have followed with their own metallocene
catalyst technologies, which are supported and are used in solution, gas phase, and
supercondensed phases and processes.
 Vol. PROPYLENE 33
Table 11. Metallocene/Single–Site Technologies
Company Technology Name Type of Polymer
Exxon/Mobil Univation (Exxpol/Unipol) iPP, impact co-PP (commercial)
Basell Metocene iPP, impact co-PP (in development)
Dow INSPiRE iPP (in development)
Borealis Borecene iPP, impact co-PP (commercial)
JPC/Mitsubishi Proprietary usea iPP, impact co-PP (in development)
TotalFina Proprietary usea iPP (commercial)
Mitsui Proprietary usea sPP (in development)
Chisso Proprietary usea iPP, sPP (in development)
BP Amoco Proprietary usea EPP (in development)
a
Implies that technology is not commercially available, but the polymer may be commercially available.

The nonsupported metallocenes are used only in solution processes. The strategies for
market penetration run from targeting commodities to specialty grades.
In the infancy of metallocene catalysis, many companies spent considerable research
dollars on metallocene preparations and understanding the structure of the metallocene in
relation to the polymer properties. In addition to these research dollars, large expenditures
were also made to establish strong patent positions. In an effort to recoup these expenses for
this budding technology, to be competitive, and to fill voids in their technology portfolios,
many have established cooperative alliances or have consolidated.
At the time of this writing a number of metallocene/single-site catalyst technologies
are available with which to produce polypropylene (see Table 11). Metallocene-based
polypropylenes are commercially available and even catalyst licenses are available (264).

Manufacturing Processes

The first commercial processes for the production of polypropylene were batch
polymerization processes using TiCl3 catalysts activated by Al(C2H5)2Cl in a hy- drocarbon
medium. The hydrocarbon, usually hexane or kerosene, maintained the isotactic
polypropylene in suspension and dissolved the undesirable atactic frac- tion. After
polymerization, the suspension is treated with alcohol to deactivate and solubilize the
catalyst residues, and filtered to separate the residues and atactic fraction from the desirable
polymer, which is then dried. The alcohol and diluent are recovered by multiple distillations,
and the atactic fraction is sold as a by- product. As the demand for polypropylene increased,
these batch polymerization processes were rapidly replaced by continuous ones, such as the
Hercules process shown in Figure 21. In this process, typical of those used throughout the
1960s and 1970s, a suspension of TiCl3 catalyst in Al(C2H5)2Cl and kerosene diluent is
continuously fed to the first of a series of continuous stirred overflow reactors. Monomer is
fed to the first reactors and allowed to react out in the later ones, obvi- ating the requirement
for monomer recycle. Typical polymerization temperatures were in the range of 55–70◦C
and maximum pressures as high as 0.5 MPa (75 psig).
 33 PROPYLENE Vol.

Fig. 21. Hercules slurry process for polypropylene.

Other similar processes, such as Montedison’s, operated at pressures as high as

1.3 MPa (200 psig) with monomer recycle (265). Hydrogen is added to the reac- tors as
required to achieve the desired polymer molecular weight (266). Following polymerization,
the slurry is contacted with isopropyl alcohol, then aqueous caus- tic to decompose and
neutralize catalyst residues. The aqueous phase containing the alcohol and catalyst residues
is separated from hydrocarbon, polymer slurry phase. The suspended isotactic polymer is
separated from the diluent containing the atactic polymer by continuous filtration or
centrifugation, then dried. The al- cohol and kerosene are each purified by a series of
distillations, then recycled. Atactic polymer is dried using a thin-film evaporator and sold as
by-product. The aqueous stream containing catalyst residues is treated prior to disposal of
wastew- ater and inorganic solids. The products available from this technology were lim-
ited to homopolymers with relatively high molecular weights (MFR< 15 dg/min), random
copolymers containing low amounts of ethylene, and impact-resistant copolymers of high
molecular weight and low rubber content. Excessive produc- tion of soluble polymer
causing fouling of heat-transfer surfaces, was the primary cause of this limitation, more so
than the loss of monomer to the production of less valuable by products. This limitation,
and the high energy cost of recycling
 Vol. PROPYLENE 33
diluent and alcohol, led to the development of processes that eliminated the need for diluent.
Polymerization in liquid monomer was pioneered by Rexall Drug and Chem- ical and
Phillips Petroleum. In the Rexall process, liquid propylene is polymerized in a stirred
reactor to form a polymer slurry. This suspension is transferred to a cyclone to separate the
polymer from gaseous monomer under atmospheric pres- sure. The gaseous monomer is
then compressed, condensed and recycled to the polymerizer (267). In the Phillips process,
polymerization occurs in loop reactors, increasing the ratio of available heat-transfer surface
to reactor volume (268). In both of these processes, high catalyst residues necessitate post-
reactor treatment of the polymer.
Gas-phase polymerization of propylene was pioneered by BASF, who devel- oped the
Novolen process, using stirred-bed reactors (269). Unreacted monomer is condensed and
recycled to the polymerizer, providing effective removal of the heat of reaction. As in the
early liquid-phase systems, post-reactor treatment of the polymer is required to remove
catalyst residues (270). The high content of atactic polymer in the final product limits its
usefulness in many markets.
In the 1970s, Solvay introduced an advanced TiCl3 catalyst with high activ- ity and
stereoregularity (271). The level of atactic polymer was sufficiently low so that its removal
from the product was not required. When this catalyst was used in liquid monomer
processes, residues were sufficiently reduced so that simplified systems for post-reactor
treatment were acceptable. Montedison and Mitsui Petro- chemical introduced MgCl2-
supported high yield catalysts in 1975 (272). Use of these catalyst systems reduced the level
of corrosive catalyst residues to the extent that neutralization or removal from the polymer
was not required. Stereospeci- ficity, however, was insufficient to eliminate the requirement
for removal of the atactic polymer fraction. These catalysts were used in the Montedison
high yield slurry process, which does not contain the sections required for alcohol treatment,
neutralization, and diluent purification in older slurry processes (265).
Current Processes. Introduction of high yield, high stereoregularity cat- alysts
by Montedison and Mitsui in 1983 enabled the development of processes in which removal
of catalyst and atactic polymer is unnecessary. This enabled the widespread use of processes
in which monomer is the polymerization medium replacing slurry processes using an inert
diluent. Investment and operating costs were dramatically reduced because of the
elimination of the sections of the plant required for handling and purifying diluent and
alcohol, removing catalyst and separating atactic polymer. Consequently, many companies
invested in new plants either increasing capacity or replacing plants using the older, now
obsolete pro- cesses. Almost all of the plants built in the past 15 years use one of the simpli-
fied processes. Moreover, the production capacity of a newer plant using these processes is
often many times greater than those of earlier plants. Single line production capacities of
250 kt/y are no longer unusual, and plants with higher production capacities have been
announced. The most widely used processes are Spheripol, licensed by Basell; Unipol,
licensed by Univation; and Novolen, licensed by Novolen Gmbh.
The Spheripol process consists of one or more loop reactors for production of
homopolymer and random copolymer, and one or more fluid bed gas-phase reactors for the
production of the rubber phase for impact-resistant copolymers (Fig. 22).
 33 PROPYLENE Vol.

Fig. 22. Spheripol process

When producing impact-resistant copolymers, monomer and catalyst components are fed to
the loop reactor for homopolymerization. The use of spherical form cat- alyst, with a narrow
particle size distribution, coupled with high liquid velocities, minimizes reactor fouling
maintaining effective heat transfer and enabling spe- cific outputs in excess of 400 kg PP/h
m3. After polymerization in the loop reactors, the polymer is separated from the liquid
monomer by flashing at a pressure suffi- cient to allow condensation and recycle of the
liquid monomer without recompres- sion.· The polymer is then transferred to the gas-phase
reactors for the production of the rubber phase of impact-resistant copolymers. Ethylene and
propylene are fed to the fluid bed reactor to produce ethylene–propylene rubber of the
desired composition. Unreacted monomer is recycled and cooled using an external heat
exchanger. The polymer is then separated from the unreacted monomer at a pres- sure
slightly above 1 atm, and then contacted with steam for complete removal of residual
monomer and termination of polymerization (273).
Liquid monomer is polymerized in continuous stirred tank reactors in a num- ber of
processes. The Hypol process, developed by Mitsui Petrochemical, uses a cascaded series
of stirred reactors for homopolymerization, followed by fluidized bed gas-phase reactors for
copolymerization (274). El Paso (now Huntsman) con- verted the Rexall liquid monomer
process to use high yield catalysts eliminating the sections required for deashing and
removal of atactic material (275). Shell (now Basell) developed the LIPP process to produce
homopolymers and random copolymers, using their high yield catalysts.
The Unipol PP process developed by Union Carbide (now Dow) and licensed by
Univation, uses a large gas phase fluidized bed reactor for the production of homopolymer
and random copolymer. A second, smaller fluidized bed reactor is used in series to produce
the rubber required for impact copolymers. The heat of reaction is removed by cooling the
monomer through an external heat exchanger (Fig. 23). The heat removal capacity of this
heat exchanger and, consequently, the
 Vol. PROPYLENE 33

Fig. 23. The Unipol process.

production capacity of the plant is increased by facilitating condensation of hydro- carbon.

This “condensing mode” technology has enabled this process to be used in very large single
line polymerization plants (patent). Use of high yield catalysts in the Novolen process (Fig.
24), developed by BASF and licensed by Novolen Gmbh, has eliminated the problems
associated with the use of first-generation catalysts. These catalysts enable the plants to
achieve high capacity and improve product quality by minimizing catalyst residues and
atactic polymer. This process uses a single vertical stirred bed reactor for the production of
homopolymer and random copolymer and a second, similar reactor for the production of
impact copolymers. Amoco (now BP) developed a horizontal stirred bed gas-phase reactor
that acts as a series of polymerization stages in a single reactor vessel. This facilitates the
production of homopolymers with broad molecular weight distribution. As in other
processes, a second reactor can be used in series for the production of impact
 34 PROPYLENE Vol.

Fig. 24. The Novolen process.

copolymers. Basell produces specialty propylene copolymers in the multistage gas- phase
Catalloy process (276).
The Borstar PP process developed by Borealis can operate at temperatures above the
critical temperature of the reaction medium. This process uses a loop reactor and gas-phase
reactor in series for the production of homopolymer. Addi- tional gas-phase reactors are
required for the production of impact copolymers. The first commercial scale plant using
this process started in 2000. Basell has announced the development of the Spherizone
process using a recirculating gas- phase reactor (Covezzi paper). The reactor contains two
zones that can be operated under different conditions, enabling the production of multiphase
specialty copoly- mers in a single reactor. This reactor was first used in a commercial scale
plant in 2002.

Processing

PP structure can be tailored for use in most polymer processing technologies. The physical
and mechanical properties of PP in the end use product are a function of both the molecular
structure and the processing conditions. The most commonly used processes for iPP are
discussed in the following.
Injection Molding. In the injection molding (qv) process, molten polymer is
injected into a cold mold cavity. During mold filling, the melt is oriented by a combination
of shear and elongational flow (277). Crystallization partially freezes in this orientation
history. Injection-molded iPP articles are made from homopoly- mers, random and impact
copolymers, and filled polymers. Melt flow rates lower
 Vol. PROPYLENE 34
than 4 dg/min and as high as 100 dg/min in some impact copolymers can be used,
depending on the mold geometry, part thickness, and cycle time desired. Process- ing
conditions vary over a wide range because of the differences in polymer types. Since iPP
melts exhibit shear-thinning properties, high injection pressures and high shear rates are
used to promote the filling of the mold. Higher melt flow (lower molecular weight) polymers
provide more uniform flow and low cycle times in parts with thin sections. Lower melt flow
(higher molecular weight) polymers are employed when toughness is required, and can be
used in parts that have thicker cross sections. Melt temperature varies with the melt-flow
rate of the polymer and the mold shape. Higher temperatures reduce the melt viscosity and
facilitate mold filling; however, the cycle time is increased. Melt temperatures as low as
200◦C can be used with high melt-flow polymers; higher melt temperatures are required
with low melt-flow polymers. Mold temperatures typically range from 20 to 50◦C. Lower
mold temperatures reduce the cycle time, but may produce a rough or low gloss surface.
Orientation is an important determinant of proper- ties and related to the skin layer
thickness (see MORPHOLOGY). Melt temperature, melt-flow rate (MFR), polydispersity, and
proximity to the mold gate influence skin thickness (120,278–286). Lower values result
from higher melt temperature and MFR, and lower polydispersity. Regions far from the gate
also have lower values. Molds should be designed to minimize localized stresses and ensure
mold filling. When nonuniform wall thickness is required, it should decrease gradually in
the flow direction. Like all crystalline thermoplastics, iPP is sensitive to fail- ure at notches,
and smooth radii are recommended at all sharp angles, corners, or ribs. Mold shrinkage
varies with thickness from 1 to 2.5%. Thicker sections shrink more than thinner sections.
Fiber.
Melt Spinning. Melt spinning produces a broad range of iPP fibers, ranging from
short staple fiber to continuous filament (CF) or bulked/textured continuous filaments (BCF)
(see OLEFIN FIBERS). The tex per filament of the as-spun filaments, where tex is the mass of
fiber (g) per 1000 m of length, is typically in the range of 0.14–7.78 tex (1.3–70 dpf). The
lower end of this range corresponds to fine filaments of 15-µm diameter. Noncircular cross
sections can be used to modify fiber appearance.
Melt spinning of iPP typically involves forcing molten polymer through a spinnerette
(a collection of∼small-diameter orifices) and collecting it, typically on a take-up reel, some
distance from the spinnerette at a velocity exceeding the orifice velocity. The tension
provided by the take-up reel (melt drawing) provides partial orientation which greatly
influences the final properties of the fiber. Yarns are collections of individual filaments, and
can range from monofilaments to several thousand filaments depending on the process.
Spinning speeds can approach 3000 m/min or higher in some cases. The iPP melt expands
upon exiting the holes, a phenomenon known as extrudate die swell (287). The diameter of
the extruded fila- ment just after the die plate can typically increase on the order of 30–
100% relative to the spinnerette hole depending on resin structure (melt flow rate,
polydisper- sity). The swelling increases as the size of the die holes is reduced. This factor,
and most often more importantly the maximum sustainable spin speed, determines the
minimum diameter of melt-spun iPP fibers. During the spinning process, a fluid element
experiences an acceleration (increase of velocity), decreasing temperature
 34 PROPYLENE Vol.
due to a high rate of cooling, and decreasing diameter with increasing distance along the
spin line (288). Large vertical air-cooling chambers, or chimneys, as high as 15 m can be
required to cool the molten filament and allow adequate time for crystallization under the
applied extensional force. Short spin processes generally use low spinning speeds to
minimize the space required for quenching.
Fiber properties depend on the complex interrelationship of polymer struc- ture
(polydispersity, molecular weight, tacticity), processing conditions (melt tem- perature, melt
throughput, spinnerette design, spin speed, cooling rate), and equipment design. Both
material and processing variables influence the die swell, extensional melt rheology,
maximum spin speed, crystallization/nucleation char- acteristics along the spin-line, and
fiber orientation. Controlled rheology resins with narrow polydispersity are often used to
improve the balance of spinning performance and fiber properties. Metallocene iPP resins
have been introduced as an alternative technology (289,290). The tenacity (ultimate stress)
of individ- ual as-spun filaments is generally in the range of 0.088–0.353 N/tex (1–4 g/den).
Break elongation decreases and tenacity increases with increasing spin speed due to
increasing orientation. Post-drawing of filaments in the solid state below the polymer
melting point, either via in-line continuous or off-line batch processes, further increases the
tenacity of the fibers by improving fiber orientation. Draw- ing is generally carried out at
temperatures exceeding 70◦C, with draw ratios in the 2–10 range. The drawability is a
function of the starting morphology, polymer structure, draw rate, and draw temperature.
The ultimate tenacity of perfectly oriented iPP fibers has been estimated to be 1.32 N/tex
(15 g/den) (121), though most commercial drawn fibers have tenacities of 0.353–0.794
N/tex (4–9 g/den). A heat setting, or annealing process below the polymer melting point,
can minimize fiber shrinkage.
Melt Blowing. The melt-blowing process uses very high melt-flow rate (low
molecular weight) iPP, sometimes in excess of 400–1500 dg/min. These melt-flow rates are
much higher than for melt-spinning operations. The flow of molten, low viscosity polymer,
extruded through a small die is disrupted by high velocity hot air. A large volume of cooling
air fed near the die exit quenches the fibers and deposits them on a collecting screen as a
mat of entangled fibers. Processing conditions and polymer structure can be varied to alter
the filament diameter, characteristics of the collected mat, and undesirable large “shots,” or
polymer particles. Very fine fibers, less than 5 µm in diameter, can be produced. The fiber
entanglement is sufficient to maintain the integrity of the web, and thermal bond- ing is not
necessary. Fabrics produced by this process are very soft because of the small fiber diameter.
Because the fibers are not highly oriented or bonded, melt- blown fabrics usually have low
tensile strength. Meltblown fabrics have improved barrier properties to aqueous liquids
relative to spun-bonded fabrics.
Spun Bonded Fabrics. Spun bonded fabrics are produced by depositing extruded,
spun filaments onto a collecting belt in a uniform randomized manner, followed by thermal
bonding of the filaments. Polymers with melt-flow rates above 20 dg/min and narrow
polydispersity improve stability during fiber formation and adequate melt throughput. The
fibers are separated during the web-laying process by air jets and the collecting belt is
usually perforated to prevent the air stream from deflecting and carrying the fibers in an
uncontrolled manner. Thermal bonding, using heated embossing rolls or hot needles, imparts
strength to the web
 Vol. PROPYLENE 34
by fusing some of the fibers. This process can be combined with the melt-blowing process
to produce soft, multilayer fabrics with good tensile properties. In these multilayer
structures the meltblown fabric provides barrier resistance and the spun bonded fabric
imparts strength.
Slit and Split Films. Thick industrial-grade yarns are often produced by slitting
films, providing a less expensive alternative to direct extrusion. Cast film is slit in the
machine direction by parallel rotary knives. The resulting tape can then be cold drawn in an
oven below the polymer melting point, in a manner similar to drawn melt spun fibers, to
produce the final fiber. Draw ratios of 4–11 are common. Higher draw ratios produce higher
tenacity. The width of the slit tapes depends on the spacing between knives and the draw
ratio. Knife spacings as low as 1 mm are sometimes used to produce textile fibers, although
spacings of 10–35 mm are more common. Tapes produced by slitting a fully drawn film are
wider than those oriented after slitting because of the physical limitations on minimum knife
spacing. An alternative approach is to directly extrude the tapes prior to drawing. The tapes
are annealed to minimize shrinkage. Fibers from split or fibrillated films are formed by the
drawing of polypropylene film to the degree that it splits into numerous fiber-like
interconnected tapes. In some processes the draw-induced splitting is mechanically
augmented by gears, rollers, or gas jets (291).
Film.
Cast Film. The first commercial iPP films were produced by extrusion cast- ing.
Polymer is extruded through a slit or a tubular die and quenched by cooling on chill rolls or
in a water bath. Cast film is not highly oriented and consequently does not have the stiffness
of oriented films. Resins are typical iPP homopolymer or random copolymers. Random
copolymers have improved clarity, and somewhat improved impact resistance. Rapid
quenching often results in conversion to the mesomorphic form, which can be advantageous
for film clarity. This tendency is greater in random copolymers than in homopolymer. High
chill roll temperatures can result in hazy films, while chill roll temperatures which are too
low (below the dew point of the ambient air) can result in water condensation on the film.
Additives such as antiblock and slip agents are often added for improved handling of the
film rolls.
Biaxially Oriented Polypropylene. Orientation improves the strength of iPP
films. Biaxially oriented polypropylene (BOPP) films have higher strength and stiffness than
cast films and consequently can often be used in much thinner gauges. Homopolymers are
used almost exclusively to provide maximum stiffness and water-vapor barrier. Oriented
films are produced by the tenter frame and tubular blown or double bubble methods. Most
of the newly installed capacity has utilized the tenter frame process, taking advantage of the
economics of the large, high capacity units available. The trend is to ever increasing line
speeds, with 300 m/min not uncommon. In the tenter frame process, iPP is melt extruded
through a slot die to form a sheet after quenching onto a cast roll. The cast sheet is heated to
a temperature below the melting point, and drawn. In the case of sequen- tial orientation, the
softened cast sheet is drawn by a series of rolls to the desired draw ratio in the machine
(longitudinal) direction, and subsequently clamped by a series of clips and conveyed into a
tenter oven where it is subsequently drawn in the transverse direction to the desired draw
ratio by the divergent chain clips. In the bubble process, a tube is extruded, quenched, and
radially expanded by
 34 PROPYLENE Vol.
inflation with air to provide transverse orientation. Axial orientation is provided by
extension in the machine direction through a series of nip rolls. The tube is then flattened
and slit into flat film. Orientation in both processes is provided by stretching below the
melting point. The drawing process is strongly correlated to the crystallinity at the draw
temperature which is closely related to the stereo- regularity of the resin. Generally there is
a trade-off between processability and final film mechanical and barrier properties. Heat
aging under slight tension at a temperature above the orientation temperature but below the
melting point minimizes subsequent shrinkage. Opaque films are produced by introducing
mi- crovoids into the film during the orientation by introducing small particles during
extrustion. During orientation, microvoids are created as the polymer expands from the
inelastic solid. Untreated oriented iPP films are not easily heat sealed. Consequently, lower
melting random copolymers and terpolymers are often coex- truded with the homopolymer
to form a heat sealing layer.
Blow Molding. Low melt flow polymers are used in blow molding (qv) to
provide the melt strength required to maintain stability of the parison, ie, a molten, thick-
walled tube of melt. High density polyethylene has been more commonly used to form large
parts because of its greater stability; however, a number of improved polypropylene grades
are suitable for these applications (292). In extrusion blow molding, the extruded parison
hangs freely before entering the mold, and low melt temperatures (between 205 and 215 ◦C)
are preferred. In injection blow molding, a preform is injection-molded on a steel rod,
transferred to a blow molding mold, and blown. Consequently, the melt strength
requirements of extrusion molding is alleviated to some extent, and higher melt temperatures
can be used. Injection stretch blow molding produces a biaxially oriented part with higher
stiffness, lower temperature impact strength, and greater clarity. The parison is cooled after
it is formed by extrusion or injection molding, reheated to the desired orientation
temperature, mechanically stretched, and then formed into the desired shape by blowing.
Random copolymers of intermediate melt-flow rate ( 10 MFR) have improved processing
characteristics.
Extrusion and Thermoforming. iPP is extruded into sheet, usually for
subsequent stamping or thermoforming (qv), or into pipes and profiles (see EXTRU- SION).
Low melt flow rate resins are used to provide the melt strength required to maintain
uniformity. The choice of resin can often be dictated by subsequent form- ing operations.
High melt strength polymers, produced by post-polymerization modification of
∼
conventional iPP, improve uniformity. Care must be taken for pro- files of complicated
geometry due to dimensional changes on crystallization, and variable cooling rate for
regions differing in thickness. Good mixing of the melt during the extrusion process is
important. Melt temperatures which are too high can promote degradation, leading to loss of
properties, discoloration, and plate out which is the migration of additives and/or low
molecular weight polymer to the surface of the sheet or profile. Lower temperatures reduce
throughput and unifor- mity and lead to higher levels of orientation detrimental to
subsequent forming.
In the thermoforming process, the sheet is extruded either in-line or off-line and
formed, either in melt phase or solid phase, into a part of desired shape. Sheet uniformity is
important. Historically, iPP has not been used in conventional melt- phase thermoforming
equipment because of its narrow forming temperature range and the tendency of the melt to
sag. Controlling the sag is particularly important
 Vol. PROPYLENE 34
for large parts or multiple cavities. The introduction of newer iPP grades with higher melt
strength and improved sag resistance has increased the use of iPP in conventional melt-
phase thermoforming equipment (293). Processes such as Shell’s solid-phase pressure
forming (294) were developed to overcome problems associated with melt forming. The iPP
article is formed at temperatures close to but below the crystalline melting point, by
stretching the sheet into the mold cavity with a shaped plug. The part is forced against the
mold surface by cold air to obtain the desired shape. Manufacturers of thermoforming
equipment have also modified their processes to effectively utilize iPP (295).
Stabilization. Polypropylene is subject to attack by oxygen, radiation, and
excessive heat causing a loss of molecular weight and physical properties. Stabi- lizers are
added to the polymer to minimize these effects. Small quantities of hindered phenolic
antioxidants (qv) are added in the polymerization plant, usu- ally in the drying section, to
protect the polymer against degradation (qv) during short-term storage. Typically 2,6-di-
tert-butyl-p-cresol (BHT) and octadecyl 3,5- di(tert-butyl-4-hydroxy)hydrocinnamate
(Irganox 1076) are used. The bulk of the stabilizer is added during pelletization or
fabrication to protect the polymer during processing or in the final application. Typical
stabilization formulations include a hindered phenolic antioxidant, possibly with a
thiodipropionate synergist, a phos- phite to provide high temperature melt stabilization, and
an acid scavenger such as calcium stearate or dihydrotalcite (296). Hindered phenols limit
the propaga- tion of alkyl radicals and the resulting chain scission. Thiodipropionic acid
esters act to decompose peroxides formed by polypropylene oxidation. More importantly, the
sulfonic acid intermediate of thiodipropionate oxidation acts as a scavenger for the free
radicals formed by the decompositon of phenols, increasing the effective- ness of the
phenolic antioxidant (297). Phosphites also act to decompose peroxides and are most
effective at the temperatures usually encountered in processing. The acid scavenger acts to
prevent the reactions between hindered phenols and metal chlorides that can form colored
titanium phenolates, as well as minimize equip- ment corrosion. Protection against
ultraviolet radiation is usually provided by a hindered amine light stabilizer (HALS), such as
Tinuvin 770 (see UV STABILIZERS). Stabilization (qv) of polypropylene has been reviewed
(157).

Economic Aspects

Polypropylene consumption continues to increase more rapidly than the economy and most
other thermoplastics. Although the comparative growth in polypropy- lene consumption has
slowed from that in the 1970s and 1980s, the relative share of polypropylene use in North
America has increased from 17% to 20% of all thermoplastics during the 1990s (298). The
annual increase in consumption of polypropylene in North America averaged about 7%
through the same period (299). Consumption in Western Europe has also increased at a
similar rate. Growth in Japan has been much slower than in other countries, in part because
of the protracted Japanese recession in the late 1990s, but also because of competition from
the emerging economies in East Asia. The rapid growth of the Asia/Pacific region has been
the major economic story of the past decade, and the growth in polypropylene consumption
and production has been phenomenal. This region is
 34 PROPYLENE Vol.
Table 12. World Consumption of Polypropylene, 10 3 tona,b
Country 1994 1995 1996 1997 1998 1999 2000 2001
United States 4145 4220 4654 5063 5412 6350 6439 7317
Canada 293 358 379 368 361
Mexico 236 219 227 285 310
Brazil 374 527 537 584 612 687
Western Europe 5030 4979 5440 5766 6278 6795 6976 7360
Japan 2080 2219 2365 2439 2268 2298 2756 2734
China 3100 3535 4230
India 763 931 1026 1194
a
Ref. 300.
b
U.S. consumption in 2002 was 7748 × 103 t (301).

now the world’s largest market for polypropylene, accounting for almost 40% of supply and
demand (302). China has developed as one the world’s major markets for polypropylene
and other thermoplastics, and is the leading importer of plas- tics, despite a significant
increase in its production capacity. World consumption data for polypropylene are listed in
Table 12. Malaysia, South Korea, India, and Thailand have joined Taiwan as significant
producers, with the construction of modern, world-scale manufacturing plants. Brazil, the
largest market in Latin America, has also grown at a rapid pace.
This increase in consumption has, however, been more than matched by a larger
increase in production capacity, fostering a situation of oversupply and low capacity
utilization. New plants often have production capacities of 250 kt/y to cap- italize on the
economy of large-scale facilities. In North America, where refinery propylene is the source
for most monomer, the decision to invest in polypropylene capacity is often related to the
need to market propylene. Arco Products (now part of BP) and Tosco (now owned by
Phillips) both decided to produce polypropylene as the most cost-effective way to eliminate
regulatory problems caused by excess propylene production at their refineries. Arco/Tosco
capacity increased by 23% from 1998 to 2001; however, consumption only increased by
17% over the same period. Capacity utilization dropped to the lowest levels since 1989.
Producers’ profit margins have decreased dramatically as the polymer prices have fallen
rel- ative to the cost of production. These profit margins are highly dependent upon the
difference between the price of polymer and the price of propylene monomer. Monomer
prices are related to oil prices; however, polymer prices are related to supply and demand.
This difference has decreased consistently through the past decade, as some producers have
been willing to supply at the lower prices. It has been estimated that a significant number of
producers have been operating at a loss (303). Consequently, producers reduced capacity by
closing or idling plants. Table 13 gives world production data, and Table 14 gives world
capacity infor- mation and Table 15 gives U.S. production and capacity for the years 2000–
2002. The 4% decrease in polypropylene capacity in 2002 is unprecedented in North
America. Excess capacity and low profit margins characterize the polypropylene industry
throughout the world, not just in North America, as the industry has become globalized. The
market situation in Asia can dramatically affect prices in North America and Europe as
imports into that region increase or decrease.
 Vol. PROPYLENE 34
Table 13. World Production of Polypropylene, 10 3 tona
Country 1994 1995 1996 1997 1998 1999
United States 4326 4939 5438 6041 6270 7026
Canada 258 329 320 320 322
Mexico 170 227 210 196
Brazil 535 557 590 636 707 767
Western Europe
Japanb 2248 2488 2683 2788 2597 2678
a
Ref. 300.
b
2001 production in Japan, 2760 × 103 t (302).

Table 14. World Capacity of Polypropylene, 103 tona

Country 1994 1995 1996 1997 1998 1999b
United States 4644 5433 5911 6453 7001 7732
Canada 324 324
Mexico 249 220
Brazil 796 778 755 785 912
Western Europe 8166
Japan 2454 3063
China 841 888 1257 1423 1503
South Korea 1705 1696 2025 2175 2620
India 161 349 490 745 791
Taiwan 460 469 479 489 798
Thailand 310 359 580 800 1088
a
Ref. 300.
b
For countries other than U.S., Ref. 304.

Table 15. United States Production and Capacity of

Polypropylene, 103 ton, for 2000–2002a
2000 2001 2002
Production 7138 7226 7690
Capacity 8117 8620 8241
a
Ref. 301.

To combat the decrease in profitability major producers have chosen to re- duce costs
through mergers. Royal Dutch/Shell and BASF have merged their poly- olefin activities,
formerly Montell, Elenac and Targor, to form Basell, which is the world’s largest producer
of polypropylene. This company contains facilities that were once part of BASF, Hercules,
Hoechst, ICI, Montedison, and Shell. Borealis, one of the largest European producers and
the result of a merger between the polyolefin businesses of Statoil and Neste, acquired PCD
and OMV, two smaller producers. Major mergers of large petroleum companies, such as
Exxon and Mo- bil (ExxonMobil), BP, Amoco and ARCO (BP), and Total, Elf Aquitane and
Fina (TotalFina) have also resulted in the combination of their polymer operations. The Dow
Chemical Co. purchased Union Carbide, combining the polypropylene businesses of the two
companies. Japanese producers affiliated with Mitsui have
 Table 16. Distribution of Polypropylene by Principal North American Market, 10 3ton
Market 1994 1995 1996 1997 1998 1999 2000 2001 (Growth) 2002 (Growth)
Transportation 281 342 329 352 378 334 337 323 (0.051) 345 (0.053)
Packaging 922 1036 1104 1182 1270 1614 1586 1578 0.089 1851 0.096
Building and construction 115 0 0 0 0 171 179 159 0.066 168 0.065
Electrical/electronic 147 162 180 193 207 185 155 156 0.038 164 0.039
Furniture and furnishings 766 866 937 1003 1078 834 913 829 0.034 923 0.041
Consumer and institutional 1091 1215 1283 1373 1476 1538 1497 1510 0.075 1661 0.077
All other 892 904 1200 1285 1381 1784 1771 1784 0.104 1878 0.099
Exports 297 332 463 565 489 539 629 856 0.016 758 0.004
3

Total 4511 4857 5497 5952 6278 7000 7067 7317 0.068 7748 0.068
 Vol. PROPYLENE 34
Table 17. Consumption of Polypropylene by Use in North America, 103 ton
USe 1994 1995 1996 1997 1998 1999 2000 2001
Injection molding 1498 1665 1765 1655 1818 2108 2068 2158
Appliances 120 129 142 116 134 150 113 118
Consumer products 598 539 611 695 768 923 898 938
Rigid packaging 390 438 484 545 609 674 712 795
Transportation 229 281 262 185 215 251 259 244
Other 162 278 266 109 93 110 86 63
Blow molding 75 80 78 78 81 84 66 81
Extrusion 1834 2048 2184 2209 2382 2546 2587 2478
Film 420 488 514 525 548 598 609 601
Sheet 72 108 104 108 116 141 150 196
Fiber 1275 1391 1505 1499 1645 1729 1732 1593
All other 67 60 63 77 73 77 96 88
Other End Use 806 732 1006 1446 1508 1724 1718 1744
Total 4214 4525 5034 5387 5789 6461 6439 6461

merged operations to form Grand Polymers and those affiliated with Mitsubishi have
formed Japan Polychem. Mitsui and Sumituomo have announced a merger of their
chemicals businesses, and included polypropylene. At the end of 2001, the largest producers
of polypropylene in order of capacity are Basell, BP, Atofina, ExxonMobil, and Dow
Chemical (305). In Europe, the largest producers are Basell, Borealis, Atofina, Sabic, and
BP, respectively.
The principal market applications of polypropylene in North America are shown in
Table 16. The use of polypropylene in packaging has grown more rapidly than other areas
because of its increased use in injection-molded containers and packaging films. The
consumer and institutional products sector is the largest market for polypropylene. This
sector is the most diverse and includes nonwo- ven polypropylene fabrics in baby diapers as
well as injection-molded toys and houseware. This market has also grown considerably in
the past decade. The con- sumption of polypropylene in furniture and furnishings, which
includes carpet fiber, continues to be one of the major applications, but is growing more
slowly than overall consumption. The use of polypropylene in transportation, primarily
automobiles, is not growing as fast as other areas. Consumption of polypropylene in
injection-molded transportation applications has declined in recent years, as shown in Table
17. Polypropylene is most commonly extruded into fibers or films, or injection molded.
Other fabrication processes are not used as frequently.

BIBLIOGRAPHY

“Propylene Polymers” in EPST 1st ed., Vol. 11, pp. 597–619, by J. L. Jezl, Avisun Corp., and E. M.
Honeycutt, Patchogue Plymouth Co.; in EPSE 2nd ed., Vol. 13, pp. 464–531, by Richard B.
Lieberman, Himont Research and Development Center, and Pier Camillo Barbe, Himont Italia, Centro
Recerche Guilio Natta, Italy.
1. G. L. Yaws, Physical Properties, McGraw-Hill Inc., New York, (1977).
2. R. W. Gallant, Physical Properties of Hydrocarbons, Vol. 1, Gulf Publishing Co., Hous- ton,
Tex., (1968).
35 PROPYLENE Vol.
3. R. C. Reid, J. M. Prausnitz, and T. K. Sherwood, The Properties of Gases and Liquids, 3rd ed.,
McGraw-Hill Inc., New York, (1977).
4. H. H. Brintzinger, D. Fischer, R. Mulhaupt, B. Rieger, and R. M. Waymouth, Angew. Chem., Int.
Ed. Engl. 34, 1143 (1995).
5. G. H. Llinas, S.-H. Dong, D. T. Mallin, M. D. Rausch, Y.-G. Lin, H. H. Winter, and
J. C. W. Chien, Macromolecules 25, 1242 (1992).
6. E. Hauptman, R. M. Waymouth, and J. W. Ziller, J. Am. Chem. Soc. 117, 11586 (1995).
7. W. J. Gauthier, J. F. Corrigan, N. J. Taylor, and S. Collins, Macromolecules 28, 3771 (1995).
8. E. A. Youngman and J. Boor, Macromol. Rev. 2, 33 (1967).
9. A. Zambelli and C. Tosi, Adv. Polym. Sci. 32, 15 (1974).
10. C. Wolfsgruber, G. Zannoni, E. Rigamonti, and A. Zambelli, Makromol. Chem. 176, 2765
(1975).
11. A. Zambelli, P. Locatelli, G. Zannoni, and F. A. Bovey, Macromolecules 11, 923 (1978).
12. F. C. Shilling and A. E. Tonelli, Macromolecules 13, 270 (1980).
13. Y. Doi, T. Suzuki, and T. Keii, Makromol. Chem. Rapid Commun. 2, 293 (1981).
14. S. N. Zhu, X. Z. Yang, and R. Chuˆ joˆ, Polym. J. 15, 859 (1983). 15.
J. A. Ewen, J. Am. Chem. Soc. 106, 6355 (1984).
16. R. Paukkeri, T. Va¨ a¨ na¨ nen, and A. Lehtinen, Polymer 34, 2488 (1993).
17. M. Farina, G. DiSilvestro, and P. Sozzani, Macromolecules 26, 946 (1993).
18. V. K. Gupta, S. Satish, and I. S. Bhardwaj, J. Macromol. Sci., Rev. Macromol. Chem. Phys. 34,
439 (1994).
19. R. Paukkeri, E. Iiskola, A. Lehtinen, and H. Salminen, Polymer 35, 2636 (1994).
20. V. Busico, P. Corradini, R. DeBiasio, L. Landriani, and A. L. Segre, Macromolecules
27, 4521 (1994).
21. M. Farina, G. DiSilvestro, and A. Terragni, Macromol. Chem. Phys. 196, 353 (1995).
22. M. D. Bruce and R. M. Waymouth, Macromolecules 31, 2707 (1998).
23. R. A. Phillips and M. D. Wolkowicz, in E. P. Moore Jr., ed., Polypropylene Handbook, Hanser,
Munich, 1996.
24. V. Busico, R. Cipullo, and P. Corradini, Makromol. Chem. 194, 1079 (1993).
25. V. Busico, R. Cipullo, and P. Corradini, Makromol. Chem., Rapid Commun. 14, 97 (1993).
26. V. Busico, R. Cipullo, J. C. Chadwick, J. F. Modder, and O. Sudmeijer, Macromolecules
27, 7538 (1994).
27. T. Tsutsui, N. Kashiwa, and A. Mizuno, Makromol. Chem., Rapid Commun. 11, 565 (1990).
28. J. C. Chadwick, A. Miedema, and O. Sudmeijer, Macromol. Chem. Phys. 195, 167 (1994).
29. J. C. Chadwick, G. M. M. van Kessel, and O. Sudmeijer, Macromol. Chem. Phys. 196, 1431
(1995).
30. E. Albizzati, U. Giannini, G. Collina, L. Noristi, and L. Resconi, in E. P. Moore Jr., ed.,
Polypropylene Handbook, Hanser, Munich, 1996.
31. K. Soga, T. Shiono, S. Takemura, and W. Kaminsky, Makromol. Chem. Rapid Commun.
8, 305 (1987).
32. A. Grassi, A. Zambelli, J. L. Resconi, E. Albizzati, and R. Mazzocchi, Macromolecules
21, 617 (1988).
33. A. Toyota, T. Tsutsui, and N. Kashiwa, J. Mol. Catal. 56, 237 (1989).
34. T. Tsutsui, N. Ishimaru, A. Mizumo, A. Toyota, and N. Kashiwa, Polymer 30, 1350 (1989).
35. H. N. Cheng and J. A. Ewen, Makromol. Chem. 190, 1931 (1989).
36. W. Roll, H.-H. Brintzinger, B. Rieger, and R. Zolk, Angew. Chem., Int. Ed. Engl. 29, 279 (1990).
Vol. PROPYLENE 35
37. B. Rieger, X. Mu, D. T. Mallin, M. D. Rausch, and J. C. W. Chien, Macromolecules 23, 3559
(1990).
38. J. C. W. Chien and R. Sugimoto, J. Polym. Sci., Polym. Chem. Ed. 29, 459 (1991).
39. D. Fischer and R. Mulhaupt, Macromol. Chem. Phys. 195, 1433 (1994).
40. U. Stehling, J. Diebold, R. Kirsten, W. Roll, H. H. Brintzinger, S. Jungling, R. Mulhaupt and F.
Langhauser, Organometallics 13, 964 (1994).
41. L. Resconi, A. Fait, F. Piemontesi, M. Colonnesi, H. Rychlicki, and R. Zeigler, Macro-
molecules 28, 6667 (1995).
42. F. A. Bovey, High Resolution NMR of Macromolecules, Academic Press, Inc., New York, 1972.
43. R. Chuˆ joˆ, Y. Kogure, and T. Va¨ a¨ na¨ nen, Polymer 35, 339 (1994).
44. M. Ha¨ rko¨nen, J. V. Seppa¨ la¨ , R. Chuˆ joˆ, and Y. Kogure, Polymer 36, 1499 (1995).
45. Y. Doi, Makromol. Chem., Rapid Commun. 3, 635 (1982).
46. D. R. Burfield and P. S. T. Loi, J. Appl. Polym. Sci. 36, 279 (1988).
47. R. Paukkeri and A. Lehtinen, Polymer 34, 4075 (1993).
48. G. Natta, G. Mazzanti, G. Crespi, and G. Moraglio, Chim. Ind. 39, 275 (1957).
49. J. B. P. Soares and A. E. Hamielec, Polymer 36, 1639 (1995).
50. U. S. Pat. 5,302,897 (Apr. 12, 1994), R. L. Dechene, T. B. Smith, S. A. Marino, J. Tache, and A.
Roy (to Auburn International).
51. M. G. Styring and A. E. Hamielec in A. R. Cooper, ed., Determination of Molecular Weight,
John Wiley & Sons, Inc., New York, 1989.
52. P. J. Flory, Principles of Polymer Chemistry, Cornell University Press, Ithaca, N.Y., 1953.
53. H. Coll, in A. R. Cooper, ed., Determination of Molecular Weight, John Wiley & Sons, Inc.,
New York, 1989.
54. A. R. Cooper, in A. R. Cooper, ed., Determination of Molecular Weight, John Wiley & Sons,
Inc., New York, 1989.
55. B. Chu, in A. R. Cooper, ed., Determination of Molecular Weight, John Wiley & Sons, Inc.,
New York, 1989.
56. M. Kurata and Y. Tsunashima, in J. Brandrup, E. H. Immergut, and E. A. Grulke, eds.,
Polymer Handbook, 4th ed., John Wiley & Sons, Inc., New York, 1999.
57. Th. G. Scholte, N. L. J. Meijerink, H. M. Schoffeleers, and A. M. G. Brands, J. Appl. Polym. Sci.
29, 3763 (1984).
58. I. Pasquon, Pure Appl. Chem. 15, 465 (1967).
59. J. B. Kinsinger and R. E. Hughes, J. Phys. Chem. 63, 2002 (1959).
60. P. Parrini, F. Sebastiano, and G. Messina, Makromol. Chem. 38, 27 (1960).
61. G. Zeichner and P. Patel, 2nd World Congress of Chemical Engineering, Montreal,
P. Q., Canada, 1981.
62. W. W. Graessley, The Entanglement Concept in Polymer Rheology, Advance Polymer Science
Series 16, Springer-Verlag, New York, 1974.
63. G. Natta and P. Corradini, del Nuovo Cimento XV, 40 (1960).
64. B. Lotz, J. C. Wittmann, and A. J. Lovinger, Polymer 37, 4979 (1996).
65. S. Bruckner, S. V. Meille, V. Petraccone, and B. Pirozzi, Prog. Polym. Sci. 16, 361 (1991).
66. A. J. Lovinger, B. Lotz, D. D. Davis, and F. J. Padden Jr., Macromolecules 26, 3494 (1993).
67. C. DeRosa and P. Corradini, Macromolecules 26, 5711 (1993).
68. A. J. Lovinger, B. Lotz, D. D. Davis, and M. Schumacher, Macromolecules 27, 6603 (1994).
69. W. Stocker, M. Schumacher, S. Graff, J. Lang, J. C. Wittmann, A. J. Lovinger, and
B. Lotz, Macromolecules 27, 6948 (1994).
 35 PROPYLENE Vol.
70. P. Corradini, V. Petraccone, C. DeRosa, and G. Guerra, Macromolecules 19, 2699 (1986).
71. P. Corradini, C. DeRosa, G. Guerra, and V. Petraccone, Polym. Commun. 30, 281 (1989).
72. S. V. Meille, D. R. Ferro, S. Bruckner, A. J. Lovinger, and F. J. Padden, Macromolecules
27, 2615 (1994).
73. D. L. Dorset, M. P. McCourt, S. Kopp, M. Schumacher, T. Okihara, and B. Lotz, Polymer
39, 6331 (1998).
74. B. Rieger, X. Mu, D. T. Mallin, M. D. Rausch, and J. C. W. Chien, Macromolecules 23, 3559
(1990).
75. D. Fischer and R. Mulhaupt, Macromol. Chem. Phys. 195, 1433 (1994).
76. R. Thomann, C. Wang, J. Kressler, and R. Mulhaupt, Macromolecules 29, 8425 (1996).
77. K. Mezghani and P. J. Phillips, Polymer 39, 3735 (1998).
78. S. Bruckner and S. V. Meille, Nature 340, 455 (1989).
79. S. V. Meille, S. Bruckner, and W. Porzio, Macromolecules 23, 4114 (1990).
80. B. Lotz, S. Graff, C. Staupe, and J. C. Wittmann, Polymer 32, 2902 (1991).
81. D. R. Ferro, S. Bruckner, S. V. Meille, and M. Ragazzi, Macromolecules 25, 5231 (1992).
82. A. J. Lovinger and B. Lotz, J. Polym. Sci., Part B: Polym. Phys. 35, 2523 (1997).
83. I. Rodriguez-Arnold, Z. Bu, and S. Z. D. Cheng, J. Macromol. Sci., Rev. Macromol. Chem. Phys.
C 35, 117 (1995).
84. P. Sozzani, R. Simonutti, and M. Galimberti, Macromolecules 26, 5782 (1993).
85. A. Marigo, C. Marega, and R. Zannetti, Macromol. Rapid Commun. 15, 225 (1994).
86. C. DeRosa, F. Auriemma, and V. Vinti, Macromolecules 30, 4137 (1997).
87. C. DeRosa, F. Auriemma, and V. Vinti, Macromolecules 31, 7430 (1998).
88. C. DeRosa, F. Auriemma, V. Vinti, and M. Galimberti, Macromolecules 31, 6206 (1998).
89. F. Auriemma, R. H. Lewis, H. W. Spiess, and C. DeRosa, Macromol. Chem. Phys. 196, 4011
(1995).
90. F. Auriemma, R. Born, H. W. Spiess, C. DeRosa, and P. Corradini, Macromolecules 28, 6902
(1995).
91. F. Auriemma, C. DeRosa, O. Ruiz de Ballesteros, and P. Corradini, Macromolecules
30, 6586 (1997).
92. C. DeRosa, F. Auriemma, V. Vinti, A. Grassi, and M. Galimberti, Polymer 39, 6219 (1998).
93. J. Loos, A. M. Schauwienold, S. Yan, J. Petermann, and W. Kaminsky, Polym. Bull.
38, 185 (1997).
94. T. Nakaoki, Y. Ohira, H. Hayashi, and F. Horii, Macromolecules 31, 2705 (1998).
95. Y. Ohira, F. Horii, and T. Nakaoki, Macromolecules 33, 5566 (2000).
96. G. Natta, M. Peraldo, and A. Allegra, Makromol. Chem. 75, 215 (1964).
97. Y. Chatani, H. Maruymama, T. Asanuma, and T. Shiomura, J. Polym. Sci., Polym. Phys. 29,
1649 (1991).
98. D. R. Burfield, P. S. T. Loi, Y. Doi, and J. Majzik, J. Appl. Polym. Sci. 41, 1095 (1990).
99. D. S. Davis, J. Plast. Film Sheeting 8, 101 (1992).
100. J. H. Griffith and B. G. Ranby, J. Polym. Sci. XXXVIII, 107 (1959).
101. F. J. Balta´ -Calleja and C. G. Vonk, X-ray Scattering of Synthetic Polymers, Elsevier,
Amsterdam, 1989.
102. F. J. Padden Jr. and H. D. Keith, J. Appl. Phys. 37, 4013 (1966).
103. F. L. Binsbergen and B. G. M. deLange, Polymer 9, 23 (1968).
104. F. J. Padden Jr. and H. D. Keith, J. Appl. Phys. 44, 1217 (1973).
105. A. J. Lovinger, J. Polym. Sci., Polym. Phys. Ed. 21, 97 (1983).
106. D. R. Norton and A. Keller, Polymer 26, 704 (1985).
107. B. Lotz and J. C. Wittmann, J. Polym. Sci., Polym. Phys. Ed. 24, 1541 (1986).
108. F. J. Padden and H. D. Keith, J. Appl. Phys. 30, 1479 (1959).
 Vol. PROPYLENE 35
109. J. Varga, J. Mater. Sci. 27, 2557 (1992).
110. H. Awaya, Polymer 29, 591 (1988).
111. A. Lustiger, C. N. Marzinsky, and R. R. Mueller, J. Polym. Sci., Part B: Polym. Phys.
36, 2047 (1998).
112. M. Schumacher, A. J. Lovinger, P. Agarwal, J. C. Wittmann, and B. Lotz, Macro- molecules 27,
6956 (1994).
113. A. Galambos, M. Wolkowicz, and R. Zeigler, in E. J. Vandenberg and J. C. Salamone, eds.,
Catalysis in Polymer Synthesis, ACS Symposium Series, 496, 1991.
114. Z-G. Wang, R. A. Phillips, and B. S. Hsiao, J. Polym. Sci., Part B: Polym. Phys. 39, 1876
(2001).
115. J. Schmidtke, G. Strobl, and T. Thurn-Albrecht, Macromolecules 30, 5804 (1997).
116. P. Supaphol, J. E. Spruiell, and J. S. Lin, Polym. Int. 49, 1473 (2000).
117. A. J. Lovinger, D. D. Davis, and B. Lotz, Macromolecules 24, 552 (1991).
118. J. Rodriquez-Arnold, A. Zhang, S. Z. D. Cheng, A. J. Lovinger, E. T. Hsieh, P. Chu,
T. W. Johnson, K. G. Honnell, R. G. Geerts, S. J. Palackal, G. R. Hawley, and M. B. Welch,
Polymer 35, 1884 (1994).
119. J. Kressler, in J. Karger-Kocsis, ed., Polypropylene an A-Z Reference, Kluwer Academic
Publishers, Dordrecht, the Netherlands, 142, 1999.
120. P. Supaphol and J. E. Spruiell, Polymer 41, 1205 (2000).
121. R. Phillips, G. Hebert, J. News, and M. Wolkowicz, Polym. Eng. Sci. 34, 1731 (1994).
122. R. J. Samuels, Structured Polymer Properties: The Identification, Interpretation, and
Application of Crystalline Polymer Structure, John Wiley & Sons, Inc., New York, 1974.
123. A. Peterlin, Polym. Eng. Sci. 17, 183 (1977).
124. A. Peterlin, Colloid Polym. Sci. 265, 357 (1987).
125. M. F. Butler, A. M. Donald, and A. J. Ryan, Polymer 39, 39 (1998).
126. I. Karacan, A. K. Taraiya, D. I. Bower, and I. M. Ward, Polymer 34, 2692 (1993).
127. M. Cakmak, J. E. Spruiell, J. L. White, and J. S. Lin, Polym. Eng. Sci. 27, 893 (1987).
128. P. Galli, Macromol. Symp. 78, 269 (1994).
129. R. Galvan and M. Tirrell, Chem. Eng. Sci. 41, 2385 (1986).
130. W. R. Schmeal and J. R. Street, AIChE J. 17, 1188 (1971).
131. E. J. Nagel, V. A. Kirillov, and W. H. Ray, Ind. Eng. Chem., Prod. Res. Dev. 19, 372 (1980).
132. S. Floyd, K. Y. Choi, T. W. Taylor, and W. H. Ray, J. Appl. Polym. Sci. 32, 2935 (1986).
133. M. A. Ferrero and M. G. Chiovetta, Polym. Eng. Sci. 27, 1436 (1987).
134. M. A. Ferrero and M. G. Chiovetta, Polym. Eng. Sci. 31, 904 (1991).
135. R. A. Hutchinson, C. M. Chen, and W. H. Ray, J. Appl. Polym. Sci. 44, 1389 (1992).
136. E. L. Hoel, C. Cozewith, and G. D. Byrne, AIChE J. 40, 1669 (1994).
137. T. F. McKenna, J. Dupuy, and R. Spitz, J. Appl. Polym. Sci. 57, 3731 (1995).
138. L. Noristi, E. Marchetti, G. Baruzzi, and P. Sgarzi, J. Polym. Sci., Polym. Chem. Ed.
32, 3047 (1994).
139. M. A. Ferrero, E. Koffi, R. Sommer, and W. C. Conner, J. Polym. Sci., Polym. Chem. Ed. 30,
2131 (1992).
140. E. Martuscelli, M. Pracella, and A. Zambelli, J. Polym. Sci., Polym. Phys. Ed. 18, 619 (1980).
141. B. Wunderlich, Macromolecular Physics, Vol. 3, Academic Press, Inc., New York, 1980.
142. E. Martuscelli, M. Pracella, and L. Crispino, Polymer 24, 693 (1983).
143. J. J. Janimak, S. Z. D. Cheng, A. Zhang, and E. T. Hsieh, Polymer 33, 729 (1992).
144. R. Paukkeri and A. Lehtinen, Polymer 34, 4083 (1993).
145. K. D. Hungenberg, J. Kerth, F. Langhauser, B. Marczinke, and R. Schlund, in G. Fink,
R. Mulhaupt, and H. H. Brintzinger, eds., Ziegler Catalysts, Springer-Verlag, Berlin, 1995, p.
363.
 35 PROPYLENE Vol.
146. J. C. Haylock, R. A. Phillips, and M. D. Wolkowicz, in J. Scheirs and W. Kaminsky, eds.,
Metallocene-Based Polyolefins: Preparation, Properties and Technology, Vol. 2, John Wiley &
Sons, Inc., New York, 2000, p. 333.
147. J. Xu, S. Srinivas, and H. Marand, Macromolecules 31, 8230 (1998).
148. U. Gauer and B. Wunderlich, J. Chem. Phys. Ref. Data 10, 1051 (1981).
149. E. B. Bond, J. E. Spruiell, and J. S. Lin, J. Polym. Sci., Part B: Polym. Phys. 37, 3050 (1999).
150. H. S. Bu, S. Z. D. Cheng, and B. Wunderlich, Makromol. Chem. Rapid Commun. 9, 75 (1988).
151. F. Bai, F. Li, B. H. Calhoun, R. P. Quirk, and S. Z. D. Cheng, in J. Brandrup, E. H. Immergut, E.
A. Grulke, eds., Polymer Handbook, 4th ed., John Wiley & Sons, Inc., New York, 1999.
152. C. DeRosa, G. Talarico, L. Caporaso, F. Auriemma, M. Galimberti, and O. Fusco,
Macromolecules 31, 9109 (1998).
153. Y. Z. Wang, W. J. Chia, K. H. Hsieh, and H. C. Tseng, J. Appl. Polym. Sci. 44, 1731 (1992).
154. K. H. Hsieh and Y. Z. Wang, Polym. Eng. Sci. 30, 476 (1990).
155. Pro-fax® Polypropylene Chemical Resistance, Montell Polyolefins Technical Bulletin TL-101,
1996.
156. D. DelDuca and E. P. Moore Jr., in E. P. Moore Jr., ed., Polypropylene Handbook, Hanser,
Munich, 1996.
157. R. F. Becker, L. P. J. Burton, and S. E. Amos, in E. P. Moore Jr., ed., Polypropylene Handbook,
Hanser, Munich, 1996.
158. T. S. Dziemianowicz and W. W. Cox, SPE ANTEC 85 540 (1985).
159. M. Gahleitner, J. Wolfschwenger, C. Bachner, K. Bernreitner, and W. Neißl, J. Appl. Polym. Sci.
61, 649 (1996).
160. T. Simonazzi, A. DeNicola, M. Aglietto, and G. Ruggeri, in S. Laggarwal, and W. Russo, eds.,
Comprehensive Polymer Science, 1st suppl. Pergamon Press, Oxford, 1992.
161. B. Lo¨fgren and J. Seppa¨ la¨ , in J. Scheirs and W. Kaminsky, eds., Metallocene-Based
Polyolefins: Preparation, Properties and Technology, Vol. 2, John Wiley & Sons, Inc., New
York, 2000, p. 143.
162. D. O. Geymer, in M. Dole, ed., The Radiation Chemistry of Macromolecules, Vol. 2, Academic
Press, Inc., New York, 1973.
163. A. J. DeNicola, A. F. Galambos, and M. D. Wolkowicz, ACS PMSE Polym. Prep. 67, 106
(1992).
164. C. E. Ruiz and R. T. LeNoir, in E. P. Moore Jr., ed., Polypropylene Handbook, Hanser, Munich,
1996.
165. Montell Polyolefins Resin Data Sheets, 1996. Adapted.
166. R. B. Lieberman and D. DelDuca, Proc. SPE Retec: Polyolefins VIII, 21 (1993); Plast. World
4, 4 (1989).
167. U. S. Pat. 5,173,540 (Dec. 22, 1992), J. Saito and A. Sampei (to Chisso Corp.).
168. Basell Polyolefins Resin Data Sheets, 2000. Adapted. 169.
U. S. Pat. 3,112,200 (Nov. 26, 1963) (to Montecatini). 170. U. S.
Pat. 3,112,301 (Nov. 26, 1963) (to Montecatini).
171. G. Natta, Atti Accad. Naz. Lincei Rend. Classe Sci. Fis. Mat. Nat. Sez 3a 84, 61 (1955).
172. G. Natta, P. Pino, P. Corradini, F. Danusso, E. Mantica, and G. Moraglio, J. Am. Chem. Soc. 77,
1708 (1955).
173. G. Natta, J. Polym. Sci. 16, 143 (1955).
174. G. Natta, Angew. Chem. 76, 430 (1955).
175. Belg. Pat. 533,362 (Nov. 16, 1953) (to Karl Ziegler).
176. W. Klemm and E. Krose, Z. Anorg. Chem. 253, 209 (1947).
177. G. Natta, P. Corrandini, and G. Allegra, J. Polym. Sci. 51, 399 (1961).
 Vol. PROPYLENE 35
178. M. Sittig, Polyolefin Production Processes, Noyes Data Co., Park Ridge, N.J., 1976.
179. J. Boor, Ziegler–Natta Catalysts and Polymerization, Academic Press, Inc., New York, 1979.
180. U. S. Pat. 3,051,690 (July 29, 1955) (to Hercules).
181. U. S. Pat. 3,128,252 (Aug. 15, 1956) (to Esso Research and Engineering Co.).
182. Brit. Pat. 1,092,390 (Apr. 27, 1966) (to Mitsubishi).
183. Ger. Offen. 2,213,086 (Oct. 5, 1972) (to Solvay & Cie).
184. L. A. M. Rodriguez and H. M. VanLooy, J. Polym. Sci., Part A1 4, 1971 (1966).
185. N. B. Chumaevskii, Makromol. Chem. 177, 747 (1976). 186.
Brit. Pat. 1,286,867 (Nov. 26, 1969) (Montecatini).
187. Belg. Pat. 785,332 (June 23, 1972) (Montedison).
188. Ger. Offen. 2,643,143 (Nov. 21, 1975) (Montedison and Mitsui Petrochemical Ind.).
189. Ger. Offen. 2,735,672 (Aug. 9, 1976) (Montedison).
190. Ger. Offen. 2,822,783 (May 25, 1977) (Montedison).
191. Ger. Offen. 2,828,887 (July 4, 1977) (Montedison).
192. Eur. Pat. 29,232 (Nov. 24, 1979) (Montedison).
193. P. Galli, In IUPAC International Symposium on Macromolecules, Florence, Italy, IU- PAC,
1980.
194. G. Natta, I. Pasquon, and A. Zambelli, J. Am. Chem. Soc. 84, 1488 (1962).
195. A. Zambelli, and co-workers, J. Polym. Sci., Part C 16, 2485 (1967).
196. U. Giannini, U. Zucchini, and E. Albizzati, J. Polym. Sci., Part B 8, 405 (1970).
197. D. G. H. Ballard, Adv. Catal. 23, 263 (1973).
198. U. S. Pat. 3,023,513 (May 1, 1962) (Esso Research and Engineering Co.).
199. U. S. Pat. 3,424,774 (Jan. 28, 1969) (Esso Research and Engineering Co.).
200. E. Tornquist and co–workers, J. Catal. 8, 189 (1967).
201. Z. W. Wilchinsky, R. W. Looney, and E. Tornquist, J. Catal. 28, 351 (1973).
202. Brit. Pat. 2,000,514 (July 3, 1978) (Montedison).
203. P. C. Barbe, G. Cecchin, and L. Noristi, eds. Advances in Polymer Science, Vol. 81, Springer-
Verlag, Berlin, 1–81, 1987.
204. K. Soga and co–workers, Polym. J. 5(2), 128 (1973).
205. E. Suzuki and co–workers, Makromol. Chem. 180, 2275 (1979).
206. U. Giannini, Makromol. Chem. Suppl. 5, 216 (1981).
207. N. Kashiwa, and J. Yoshitake, Makromol. Chem. Rapid Commun. 3, 211–214 (1982).
208. P. C. Barbe and co–workers, Makromol. Chem., Rapid Commun. 4, 249–252 (1983).
209. G. DiDrusco and R. Rinaldi, Hydrocarbon Process 63(11), 113 (1984).
210. N. Kashiwa and J. Yoshitake, Makromol. Chem. 185, 1133–1138 (1984).
211. R. Spitz, J. L. Lacombe, and A. Guyot, J. Polym. Sci., Polym. Chem. Ed. 22, 2641 (1984).
212. R. Spitz, J. L. Lacombe, and A. Guyot, J. Polym. Sci., Chem. Ed. 22, 2625 (1984).
213. T. Keii, Kinetics of Ziegler–Natta Polymerization, Kodanska, Chapman and Hall, Lon- don,
1972.
214. I. Pasquon and U. Giannini, in J. R. Anderson and M. Boudart, eds. Catalysts, Science and
Technology, Springer-Verlag, Berlin, 1984.
215. K. Y. Choi and W. H. Ray, J. Appl. Polym. Sci. 30, 1065 (1985).
216. N. F. Brockmeier, and J. B. Rogan, Ind. Eng. Chem. Prod. Res. Dev. 24, 278 (1985).
217. T. Keii and co–workers, Makromol. Chem. 183, 2285–2304 (1982).
218. Y. Doi and co–workers, Ind. Eng. Chem. Prod. Res. Dev. 21, 580 (1982).
219. V. A. Zakharov, G. D. Bukatov, and Y. I. Yermakov, Polym. Sci. Technol. Coord. Polym.
19, 267 (1983).
220. G. Natta and I. Pasquon, Adv. Catal. 11, 1–65 (1959).
221. N. M. Chirkov, Proceedings of the IUPAC International Macromolecules Symposium,
Bucharest, Romania, IUPAC, 1969.
 35 PROPYLENE Vol.
222. C. D. Nentizescu, C. Huch, and A. Huch, Angew. Chem. 68, 438 (1956).
223. D. B. Ludlum, A. W. Anderson, and C. E. Ashby, J. Am. Chem. Soc. 80, 1380 (1958).
224. G. Natta and co–workers, Chim. Ind. Milan 38, 124 (1956).
225. A. Zambelli and co–workers, Macromolecules 13, 798 (1980).
226. P. Cossee, Tetrahedron Lett. 17, 12 (1960).
227. P. Cossee, Proceedings of the International Congress on Coordination Chemistry, 1961, p. 241.
228. P. Cossee, J. Catal. 3, 80–88 (1964).
229. T. Miyazawa and Y. Ideguchi, J. Polym. Sci. Polym. Lett. 1, 389 (1963); A. Zambelli,
M. G. Giongo, and G. Natta, D. Makromol. Chem. 112, 183 (1968); A. Zambelli and
C. Tosi, Adv. Polym. Sci. 15, 32 (1974).
230. E. J. Arlman and P. Cossee, J. Catal. 3, 99 (1964).
231. A. Zambelli, P. Locatelli, and E. Rigamonti, Macromolecules 12, 156 (1979).
232. A. Zambelli, NMR Basic Principles and Progress, Springer, Berlin, 1971, p. 101.
233. P. C. Barbe´, L. Noristi, and G. Baruzzi, Makromol. Chem. 193, 229–241 (1992).
234. U. S. Pat. 4,971,937 (to HIMONT).
235. P. Pino and co–workers, in J. C. W. Chein, ed., Coordination Polymerization, 1975, Academic Press,
Inc.: New York. p. 25.
236. P. Pino, Adv. Polym. Sci. 4, 393 (1965).
237. P. Pino and G. P. Lorenzi, Chim. Ind. Milan 42, 712 (1960).
238. P. Pino and co–workers, J. Am. Chem. Soc. 84, 1487 (1962).
239. P. Pino, F. Ciardelli, and G. P. Lorenzi, J. Polym. Sci., Part C 4, 21 (1963).
240. P. Pino and co–workers, in Proceedings of the International Symposium Transition Metal
Catalyzed Polymerization—Unsolved Problems, Midland, Mich., 1981.
241. P. Corradini, V. Barone, and R. Fusco, Gazzetta Chimica Italiana 113, 601–607 (1983).
242. P. Corradini, V. Busico, and G. Guerra, in S. G. Allen, ed. Comprehensive Polymer Science,
Pergamon Press, New York, 1988, p. 29–48.
243. A. Zambelli, in J. C. W. Chien, ed., Coordination Polymerization, Academic Press, Inc., New
York. p. 15, 1975.
244. P. Locatelli, A. Immirzi, and A. Zambelli, Makromol. Chem. 176, 1121 (1975).
245. G. Morini and co–workers, Macromolecules 29, 5770–5776 (1996).
246. M. C. Sacchi and co–workers, Macromolecules 29, 3341–3345 (1996).
247. L. Noristi, P. C. Barbe´, and G. Baruzzi, Makromol. Chem. 192, 1115–1127 (1991).
248. E. Iiskola, P. Sormunen, and T. Garoff, W. Kaminsky and H. Sinn, eds., Transition Metals and
Organometallics as Catalysts for Olefin Polymerization, Springer-Verlag, Berlin, 1988, pp.
113–122.
249. T. Okano and co–workers, Stud. Surf. Sci. Catal. Olefin Polym. 1990. p. 177–183.
250. M. Ha¨ rko¨nen, L. Kuutti, and J. V. Seppa¨ la¨ , Makromol. Chem. 193, 1413–1421 (1992).
251. R. Spitz and co–workers, Makromol. Chem. 190, 717–723 (1989).
252. I. W. Parsons and T. M. Al-Turki, Polymer Commun. 30(3), 72–73 (1989).
253. M. Kioka and N. Kashiwa, J. Macromol. Sci. Chem. A 28, 865–873 (1991).
254. K. Imaoka and co–workers, J. Mol. Catal. 82, 37–44 (1993).
255. J. C. Chadwick, A. Miedema, and O. Sudmeijer, Macromol. Chem. Phys. 195, 167–172 (1994).
256. J. C. Chadwick in Polyolefins 2000 International Conference on Polyolefins Society of Plastics
Engineers. Society of Plastics Engineers, Houston, Tex., 2000.
257. E. Albizzati and co–workers, Macromol. Symp. 89, 73–89 (1995).
258. K. Soga, T. Shinon, and Y. Doi, Makromol. Chem. 189, 1531 (1988).
259. J. C. Chadwick and co–workers, Macromol. Chem. Phys. 197, 2501–2510 (1996).
260. J. C. Chadwick, Macromol. Symp., 2001.
 Vol. PROPYLENE 35
261. W. Kaminsky, Rapra Rev. Rep. 10, 1–135 (1999).
262. W. Spaleck and co–workers, Organometallics 13, 954–963 (1994).
263. S. Miyake, Y. Okumura, and S. Inazawa, Macromolecules 28, 3074–3079 (1995).
264. H. R. Blum, in MetCon 2000, Polymers in Transition “Platforms in Single-Site Catal- ysis”,
Houston, Tex., 2000.
265. G. DiDrusco and R. Rinaldi, Hydrocarbon Process. 60(5), 153 (May 1981).
266. U. S. Pat. 3,051,690 (July 29, 1955), E. J. Vandenberg (to Hercules Powder Co.).
267. Brit. Pat. 1,044,811 (Apr. 9, 1962) (to Rexall Drug and Chemical Co.).
268. U.S. Pat. 3,476,729 (Nov. 4, 1969), D. E. Smith, R. M. Keeler, and E. Guenther (to Phillips
Petroleum).
269. J. F. Ross and W. A. Bowles, Ind. Eng. Chem. Prod. Res. Dev. 24, 149 (1985). 270. Oil
Gas J. 68, 64 (1970).
271. Ger. Offen. 2,213,086 (Oct. 5, 1972), J. P. Hermans (to Solvay & Cie SA).
272. Ger. Offen. 2,643,143 (Nov. 21, 1975) (to Montedison and Mitsui Petrochemical); Ger. Offen.
2,735,672 (Aug. 9, 1976), U. Giannini, E. Albizzati, and S. Parodi (to Montedison and Mitsui
Petrochemical); Ger. Offen. 2,822,783 (May 25, 1977), U. Scata, L. Luciani, and P. C. Barbe (to
Montedison); U. S. Pat 4,069,169 (Jan. 17, 1978), N. Kashiwa and
A. Toyota (to Mitsui Petrochemcial).
273. G. DiDrusco and R. Rinaldi, Hydrocarbon Process. 63(11), 116 (Nov. 1984).
274. Hydrocarbon Process 72(3), 204 (1993).
275. C. Cipriani and C. A. Trishman Jr., Chem. Eng. 80, (Apr. 20, 1981). 276.
Plast. World 51(6), 12 (1993).
277. Z. Tadmor and C. G. Gogos, Principles of Polymer Processing, John Wiley & Sons, Inc., New
York, 1979.
278. M. R. Kantz and H. D. Newman Jr., F. H. Stigale, J. Appl. Polym. Sci. 16, 1249 (1972).
279. D. R. Fitchmun and Z. Mencik, J. Polym. Sci., Polym. Phys. Ed. 11, 951 (1973).
280. M. Fujiyama, H. Awaya, and S. Kimura, J. Appl. Polym. Sci. 21, 3291 (1977).
281. M. Fujiyama and S. Kimura, J. Appl. Polym. Sci. 22, 1225 (1978).
282. M. Fujiyama and K. Azuma, J. Appl. Polym. Sci. 23, 2807 (1979).
283. S. S. Katti and J. M. Schultz, Polym. Eng. Sci. 22, 1001 (1982).
284. M. Fujiyama and T. Wakino, J. Appl. Polym. Sci. 43, 57 (1991).
285. M. Fujiyama and T. Wakino, J. Appl. Polym. Sci. 43, 97 (1991).
286. M. Fujiyama, Int. Polym. Proc. 8, 245 (1993).
287. P. Ramanini, Rheol. Acta 21, 699 (1982).
288. V. Bansal and R. L. Shambaugh, Polym. Eng. Sci. 36, 2785 (1996).
289. A. Winter, SPO ’97 Proc. 1 (1997).
290. J. J. McAlpin, C. Y. Cheng, D. A. Plank, and G. A. Stahl, INSIGHT ’95 Proc. (1995).
291. A. Ahmed, Polypropylene Fibers-Science and Technology, Elsevier, Amsterdam, 1982.
292. J. R. Beren and C. Capellman, paper presented at SAE International Congress, 1994.
293. K. E. McHugh and K. Ogale, SPE ANTEC ’90 452 (1990).
294. J. Foster, in Modern Plastics Encyclopedia 1983-1984, Vol. 60 (10A), McGraw-Hill, New York,
1983, p. 328.
295. K. T. Johansson, Proc. SPE Retec: Polyolefins 8, 334 (1993).
296. S. Al-Malaika and G. Scott, in N. S. Allen, ed., Degradation and Stabilization of Poly- olefins,
Applied Science Publishers Ltd., London, 1983, p. 247.
297. C. Armstrong, M. J. Husbands, and G. Scott, Eur. Polym. J. 15, 241 (1979).
298. Society of the Plastics Industry, Facts and Figures 1999.
299. A. H. Tullo, Chem. Eng. News 79, 12–14 (Feb. 5, 2001).
 35 PROPYLENE Vol.
300. Mod. Plast. 22–29 (Feb. 2002); 44–49 (Feb. 2001); 48–79 (Feb. 2000); 72–80 (Jan.
1999); 74–82 (Jan. 1998); 56, 76–84 (Jan. 1997); 54, 70–78 (Jan. 1996); 63–68
(Jan. 1995).
301. American Plastics Council Plastics Industry Producers’ Statistics Group, as compiled by VERIS
Consulting, LLC.
302. Asia Chem. News 14–21 (June 11, 2001).
303. Mod. Plast. 54 (Jan. 1999).
304. Chemical Market Resources Inc., CMR Polym. Tracker 5, 25 (Apr. 2000). 305. Mod.
Plast. 22–29 (Feb. 2002).

RICHARD LIEBERMAN
Basell R&D Center
CONSTANTINE STEWART

PROTEIN FOLDING. See Volume 7.

PSA. See PRESSURE SENSITIVE ADHESIVES.

PULTRUSION. See COMPOSITES, FABRICATIOIN.

PVC. See VINYL CHLORIDE POLYMERS.

PVDC. See VINYLIDENE CHLORIDE POLYMERS.

PVF. See VINYL FLUORIDE POLYMERS.

PVK. See VINYLCARBAZOLE POLYMERS.

PVP. See VINYL AMIDE POLYMERS.