Available via license: CC BY
Content may be subject to copyright.
Citation: Bergstra, M.F.; Denifl, P.;
Gahleitner, M.; Jeremic, D.;
Kanellopoulos, V.; Mileva, D.;
Shutov, P.; Touloupidis, V.;
Tranninger, C. Polymerization in the
Borstar Polypropylene Hybrid
Process: Combining Technology and
Catalyst for Optimized Product
Performance. Polymers 2022,14, 4763.
https://doi.org/10.3390/
polym14214763
Academic Editor: Shin-Ichi Yusa
Received: 12 October 2022
Accepted: 1 November 2022
Published: 7 November 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
polymers
Review
Polymerization in the Borstar Polypropylene Hybrid Process:
Combining Technology and Catalyst for Optimized
Product Performance
Michiel F. Bergstra 1, Peter Denifl 2, Markus Gahleitner 2,*, Dusan Jeremic 2, Vasileios Kanellopoulos 2,
Daniela Mileva 2, Pavel Shutov 2, Vasileios Touloupidis 2and Cornelia Tranninger 2
1Borealis Polymers N.V., Industrieweg 148, 3580 Beringen, Belgium
2Borealis Polyolefine GmbH, Innovation Headquarters, Sankt Peterstrasse 25, 4021 Linz, Austria
*Correspondence: markus.gahleitner@borealisgroup.com
Abstract:
Producing isotactic polypropylene (iPP) homo- and copolymers in a wide composition
and property range according to customer demand requires perfect alignment between the process
technology, catalyst system and polymer structure. The present review shows this for the Borstar
®
PP process, a hybrid process employing liquid bulk and gas phase stages, in an exemplary way.
It starts with the process design and continues through two generations of Ziegler–Natta catalyst
development history to the design of advanced multimodal random and multiphase copolymers.
Essential elements of each of the three areas contributing to performance range are highlighted, and
an outlook to future development is given.
Keywords: polypropylene; catalyst; polymerization process; nucleation; application
1. Introduction
The evolution of polypropylene (PP) from an experimental material in 1954 [
1
,
2
] to
a material family, exhibiting a wide variety of possible microstructures, is impressive.
Nowadays, polypropylene grades include multiple copolymers and compounds and attain
a worldwide production volume of more than 60 million tons per year [
3
]. This success story
has been enabled by the parallel development of catalyst technology and polymerization
processes. The various steps along the way have not only made the production cheaper
and less complex through a multifold increase in catalyst productivity, but also widened
the product and property ranges.
Early first- and second-generation “Ziegler–Natta” (ZN) TiCl
3
catalysts typically re-
quired the use of a hydrocarbon slurry and a series of multiple stirred-tank reactors to
achieve acceptable productivity, with the reduction of atactic fraction being the main
progress step between them. Nevertheless, catalyst productivity remained relatively low.
For this reason, after polymerization, the polymer suspension had to be treated with alcohol
to deactivate and solubilize the catalyst components; the suspension was then filtered to sep-
arate the residues and soluble fraction from the polymer. Products of this technology were
limited to homopolymers with rather high molecular weight and a melt flow rate (MFR)
230
◦
C/2.16 kg below 15 g/10 min, random copolymers with low amounts of ethylene
and impact-resistant copolymers of low amorphous phase content, with most limitations
resulting from soluble fractions [
4
]. All following generations combine a MgCl
2
-based
support with TiCl
4
as an active centre. While the high-yield third-generation catalysts
introduced by Montedison and Mitsui Petrochemical in the early 1970s [
5
,
6
] were lim-
ited in stereospecificity and still applied a high-yield slurry process, the introduction of
aromatic monoesters as internal donors changed that situation and allowed the use of
high-productivity bulk and gas-phase processes without atactic fraction removal [7].
Polymers 2022,14, 4763. https://doi.org/10.3390/polym14214763 https://www.mdpi.com/journal/polymers
Polymers 2022,14, 4763 2 of 25
In the 1980s, the use of phthalate esters as internal donors in combination with specifi-
cally designed and highly porous MgCl
2
supports gave the fourth ZN catalyst generation
unprecedented productivity levels and flexibility. Isotacticity was further increased by
the use of alkoxysilanes as external donors. These catalysts can be employed both for
liquid bulk and gas-phase polymerization processes [
8
,
9
]. Modern production units are
either based on pure bulk polymerization in a loop reactor (LR) geometry, pure gas-phase
polymerization in stirred (SBR) or fluidized bed reactors (FBR) or a combination of both,
commonly called hybrid technology [
10
]. The “Spherizone” or multi-zone circulating
gas-phase reactor (MZCR) technology of LyondellBasell [
11
] combines two distinct reaction
zones with the smooth change of process conditions. The growing particle repeatedly
circulates through these zones, which allows for the production of bimodal homo- and
random PP products in one reactor. An overview of the most relevant technology schemes
is given in Figure 1.
Polymers 2022, 14, x FOR PEER REVIEW 2 of 25
catalysts introduced by Montedison and Mitsui Petrochemical in the early 1970s [5,6] were
limited in stereospecificity and still applied a high-yield slurry process, the introduction
of aromatic monoesters as internal donors changed that situation and allowed the use of
high-productivity bulk and gas-phase processes without atactic fraction removal [7].
In the 1980s, the use of phthalate esters as internal donors in combination with spe-
cifically designed and highly porous MgCl2 supports gave the fourth ZN catalyst genera-
tion unprecedented productivity levels and flexibility. Isotacticity was further increased
by the use of alkoxysilanes as external donors. These catalysts can be employed both for
liquid bulk and gas-phase polymerization processes [8,9]. Modern production units are
either based on pure bulk polymerization in a loop reactor (LR) geometry, pure gas-phase
polymerization in stirred (SBR) or fluidized bed reactors (FBR) or a combination of both,
commonly called hybrid technology [10]. The “Spherizone” or multi-zone circulating gas-
phase reactor (MZCR) technology of LyondellBasell [11] combines two distinct reaction
zones with the smooth change of process conditions. The growing particle repeatedly cir-
culates through these zones, which allows for the production of bimodal homo- and ran-
dom PP products in one reactor. An overview of the most relevant technology schemes is
given in Figure 1.
Figure 1. Generalized schemes of the modern polypropylene production processes.
Figure 1. Generalized schemes of the modern polypropylene production processes.
In recent years, increasing concerns about the negative health effects of phthalates [
12
]
have increased the relevance of post-phthalate ZN catalysts. This has again broadened the
Polymers 2022,14, 4763 3 of 25
field, as next to the first developments of LyondellBasell with diethers and succinates as
internal donors [
13
,
14
], a wide range of alternatives is now being explored. An overview
of the developments in the area of post-phthalate ZNCs can be found in the review by
Severn [
15
]. In any case, the alternative use of these catalysts requires a fit with the applied
polymerization technology to reach the same targeted polymer grades [16].
This interlink between process technology and catalyst structure, the perfect fit be-
tween engineering and chemistry, is one of the foundations of the Borstar
®
PP process.
The details of its development and present capability in terms of wide product range and
good economics will be discussed in this review. To the best of our knowledge, this is
the first attempt at drawing direct connections between process and catalyst development
on the one hand and the resulting polymer structure and application performance on the
other hand.
2. Borstar®PP Technology
Borstar
®
polypropylene (PP) is Borealis’ proprietary low-pressure, catalytic technology
for manufacturing a full package of polypropylene materials. The whole process can be
divided in the following areas [17]:
•Feedstock preparation;
•Reactor area;
•Recovery area;
•Dry end (comprising pelletizing);
•Material handling, comprising bagging and storage area.
The reactor part of the Borstar
®
polypropylene modular technology consists of a
series of slurry-loop and gas-phase fluidized bed reactors (GPR), where each reactor can
be independently controlled. Typically, an additional pre-polymerization loop reactor
precedes the reactor cascade. Homopolymers, random copolymers as well as heterophasic
matrix copolymers can be produced by employing this technology that enables tailoring
the molecular weight distribution (MWD), isotacticity and comonomer content (CC) along
the MWD.
The Borstar
®
PP process technology exhibits a number of competitive advantages, including:
•Very high catalyst productivity due to high-temperature operation;
•
Very wide product window and independent reactor control, enabling the tailoring of
the MWD and CC along the MWD in both the matrix and rubber part of the product;
•
High once-through monomer conversion due to propylene conversion from the loop
effluent to the first GPR;
•Competitive monomer factor and energy consumption;
•Robust reactor operability and reliability.
According to the Borstar
®
PP process [
18
], the first reactor of the series is continuously
fed with catalyst, cocatalyst, donor, propylene, comonomer (if desired for the production of
random copolymers) and hydrogen. Typically, heterogeneous Ziegler–Natta or single-site
catalysts are employed. The polymerization reaction takes place at the active centres that
represent the polymerization loci, which are homogeneously dispersed in the catalyst parti-
cles. During the first stages of polymerization, the catalyst particles undergo a controlled
fragmentation process due to the production of polymer. As polymerization proceeds,
the initial catalyst particles remain encapsulated within the gradually growing polymer
particle. The polymer particles produced in the first reactor are then continuously fed
to the next reactor(s) of the series, where the polymerization reaction continues to take
place. The desired mean residence time per reactor defines the production split (weight
percentage of production rate per reactor), as implied by the specific product design. Each
reactor can be controlled separately and operate under desired reaction conditions in terms
of pressure, temperature and reaction species concentrations. This way, each reactor can
vary in production rate as well as the molecular properties of the polymer produced. The
process is designed in such way that the desired final polymer formulation is deconvoluted
Polymers 2022,14, 4763 4 of 25
in different fractions of unimodal product in terms of MWD and CC along MWD, pro-
duced separately in each reactor, fully controlling the polymer microstructure architecture.
Furthermore, the fact that the reaction takes place separately in each particle results in
homogeneity at the intra-particle level, and each particle consists of a ‘chemical’ blend of
the polymer fractions produced in each reactor of the series (see Figures 2–4).
Polymers 2022, 14, x FOR PEER REVIEW 4 of 25
reactor can be controlled separately and operate under desired reaction conditions in
terms of pressure, temperature and reaction species concentrations. This way, each reactor
can vary in production rate as well as the molecular properties of the polymer produced.
The process is designed in such way that the desired final polymer formulation is decon-
voluted in different fractions of unimodal product in terms of MWD and CC along MWD,
produced separately in each reactor, fully controlling the polymer microstructure archi-
tecture. Furthermore, the fact that the reaction takes place separately in each particle re-
sults in homogeneity at the intra-particle level, and each particle consists of a ‘chemical’
blend of the polymer fractions produced in each reactor of the series (see Figures 2–4).
Figure 2. Simplified schematic flow diagram of reactor module 1 and feeding of the Borstar® PP
process.
Figure 3. Simplified schematic flow diagram of reactor module 2 of the Borstar® PP process.
M
PRE-POLYMERIZATION
REACTOR
FEED TANK
PURIFICATION
1st GAS PHASE REACTORLOOP REACTOR
1 2
6 5
3
4
DUMP
ETHYLENE TO MODULE 2
HYDROGEN TO MODULE 2
PROPYLENE TO MODULE 2
PP
TO PURGE
CATALYST
COCATALYST
DONOR
PROPYLENE
ETHYLENE
HYDROGEN
LIGHTS FROM RECOVERY
M
MATRIX PP
TO MODULE 2
Figure 2.
Simplified schematic flow diagram of reactor module 1 and feeding of the Borstar
®
PP process.
Polymers 2022, 14, x FOR PEER REVIEW 4 of 25
reactor can be controlled separately and operate under desired reaction conditions in
terms of pressure, temperature and reaction species concentrations. This way, each reactor
can vary in production rate as well as the molecular properties of the polymer produced.
The process is designed in such way that the desired final polymer formulation is decon-
voluted in different fractions of unimodal product in terms of MWD and CC along MWD,
produced separately in each reactor, fully controlling the polymer microstructure archi-
tecture. Furthermore, the fact that the reaction takes place separately in each particle re-
sults in homogeneity at the intra-particle level, and each particle consists of a ‘chemical’
blend of the polymer fractions produced in each reactor of the series (see Figures 2–4).
Figure 2. Simplified schematic flow diagram of reactor module 1 and feeding of the Borstar® PP
process.
Figure 3. Simplified schematic flow diagram of reactor module 2 of the Borstar® PP process.
M
PRE-POLYMERIZATION
REACTOR
FEED TANK
PURIFICATION
1st GAS PHASE REACTORLOOP REACTOR
1 2
6 5
3
4
DUMP
ETHYLENE TO MODULE 2
HYDROGEN TO MODULE 2
PROPYLENE TO MODULE 2
PP
TO PURGE
CATALYST
COCATALYST
DONOR
PROPYLENE
ETHYLENE
HYDROGEN
LIGHTS FROM RECOVERY
M
MATRIX PP
TO MODULE 2
Figure 3. Simplified schematic flow diagram of reactor module 2 of the Borstar®PP process.
Polymers 2022,14, 4763 5 of 25
Polymers 2022, 14, x FOR PEER REVIEW 5 of 25
Figure 4. Simplified schematic flow diagram of the purge and recovery section of the Borstar® PP process.
2.1. Module 1: Catalyst Preparation and Pre-Polymerization
The employed catalyst feed system can handle catalysts either in dry format or sus-
pended in oil. In the case of an oil-suspended catalyst, the catalyst drum is tumbled in
multiple directions to achieve the desired homogeneity, and afterwards the catalyst sus-
pension is pushed into the catalyst feed vessel. Typically, the catalyst slurry comprises a
hydrocarbon (mineral) oil and the solid catalyst. The slurry is maintained in a homogene-
ous state in the catalyst feed vessel at a controlled, constant temperature within the range
of −30 °C to 80 °C (preferably between 0 °C and 60 °C) and slightly pressurized by an inert
gas (e.g., nitrogen or argon) above atmospheric pressure, while a part of that catalyst
slurry is fed into the pre-polymerization reactor. The typical concentration of the solid
catalyst particles is in the range of 50 to 500 kg/m3 in the slurry. The viscosity of the oil
needs to be in the range of 20 to 3000 mPa s to prevent settling of the catalyst and to ensure
that the catalyst is fed into the reactor with a high level of accuracy and consistency; nor-
mally, hydrocarbon grease is added for this purpose. Downstream, at the metering pump,
the catalyst slurry is diluted with a propylene flow to increase velocity and assure the
smooth flowability of the catalyst slurry to the reactor. Optionally, the catalyst slurry is
mixed with an activator and/or an electron donor upstream of the pre-polymerization re-
actor depending on the selected pre-activation catalyst procedure. Additional compo-
nents, such as antistatic agents and drag-force-reducing agents, can be added to further
improve operability and efficiency [18].
The catalyst is then fed into the pre-polymerization loop reactor, acting as a precon-
ditioning reactor under milder conditions (e.g., the typical operating temperature ranges
from 10 °C to 45 °C). This process step ensures controlled catalyst fragmentation, resulting
in smooth operation in the upcoming reaction stages [19].
Typically, only a small amount of polymer is produced in this reactor, with the pol-
ymer-to-catalyst ratio ranging from 10 to 1000 g PP/g catalyst. Ethylene and hydrogen
may be optionally fed to the reactor. A special advantage can be obtained by adding small
amounts of ethylene in the range of 0.2 to 1.5 mol%, improving the morphology of the
pre-polymerized polypropylene particles [20].
Figure 4.
Simplified schematic flow diagram of the purge and recovery section of the Borstar
®
PP process.
2.1. Module 1: Catalyst Preparation and Pre-Polymerization
The employed catalyst feed system can handle catalysts either in dry format or sus-
pended in oil. In the case of an oil-suspended catalyst, the catalyst drum is tumbled in
multiple directions to achieve the desired homogeneity, and afterwards the catalyst sus-
pension is pushed into the catalyst feed vessel. Typically, the catalyst slurry comprises
a hydrocarbon (mineral) oil and the solid catalyst. The slurry is maintained in a homo-
geneous state in the catalyst feed vessel at a controlled, constant temperature within the
range of
−
30
◦
C to 80
◦
C (preferably between 0
◦
C and 60
◦
C) and slightly pressurized
by an inert gas (e.g., nitrogen or argon) above atmospheric pressure, while a part of that
catalyst slurry is fed into the pre-polymerization reactor. The typical concentration of the
solid catalyst particles is in the range of 50 to 500 kg/m
3
in the slurry. The viscosity of the
oil needs to be in the range of 20 to 3000 mPa s to prevent settling of the catalyst and to
ensure that the catalyst is fed into the reactor with a high level of accuracy and consistency;
normally, hydrocarbon grease is added for this purpose. Downstream, at the metering
pump, the catalyst slurry is diluted with a propylene flow to increase velocity and assure
the smooth flowability of the catalyst slurry to the reactor. Optionally, the catalyst slurry is
mixed with an activator and/or an electron donor upstream of the pre-polymerization re-
actor depending on the selected pre-activation catalyst procedure. Additional components,
such as antistatic agents and drag-force-reducing agents, can be added to further improve
operability and efficiency [18].
The catalyst is then fed into the pre-polymerization loop reactor, acting as a precon-
ditioning reactor under milder conditions (e.g., the typical operating temperature ranges
from 10
◦
C to 45
◦
C). This process step ensures controlled catalyst fragmentation, resulting
in smooth operation in the upcoming reaction stages [19].
Typically, only a small amount of polymer is produced in this reactor, with the polymer-
to-catalyst ratio ranging from 10 to 1000 g PP/g catalyst. Ethylene and hydrogen may
be optionally fed to the reactor. A special advantage can be obtained by adding small
amounts of ethylene in the range of 0.2 to 1.5 mol%, improving the morphology of the
pre-polymerized polypropylene particles [20].
Polymers 2022,14, 4763 6 of 25
2.2. Module 1: Loop Reactor and First GPR
The loop reactor (see Figure 2) is operated at a temperature range of 60–95
◦
C. Fresh
propylene is fed into the loop reactor together with hydrogen to control the molecular
weight, and ethylene is fed into the reactor in the case of copolymer production. The
residence time in the loop reactor is typically less than one hour. Homo-polymers or
random copolymers can be produced in the loop reactor. When producing a copolymer,
the comonomer content typically ranges between 0.1 and 8 mol%. Most copolymers are
produced using ethylene as a comonomer. The melt flow rate for a 2.16 kg load (MFR2)
of a copolymer may vary from 0.1 to 500 dg/min. The operating pressure needs to be
sufficiently high in the loop reactor (typically ranging from 30 to 70 bar) to establish a single
continuous phase of liquid propylene and avoid the presence of a separate gas phase. In
the unfortunate event that this is the case, cavitation effects could potentially result in poor
mixing conditions and flowability issues as well as mechanical damage to the loop pump
and the mixer.
The loop reactor typically has multiple outlets. Special advantages are obtained
when the first outlet is located at a location in the loop reactor where the polymer solids
concentration is higher than or similar to the average solids concentration in the reactor
and when a second outlet is placed at a location in the loop reactor where the solids
concentration is lower than the average polymer solids concentration in the reactor [21].
The loop reactor of Borstar
®
PP is designed to be operated in supercritical conditions.
This feature further extends the operating window for hydrogen and ethylene because gas
bubble formation, which is a limiting factor, is avoided, while increasing at the same time
their solubility in the polymer phase. Moreover, at supercritical conditions the solubility of
the polymer in the liquid propylene decreases, thus resulting in less risk of reactor fouling
due to fouling on the loop reactor walls.
The first GPR (see Figure 2) is a fluidized bed reactor, and the polymerization takes
place in a fluidized bed formed by the growing polymer particles in an upwards-moving
gas stream. The GPR consists of a cylindrical part named the fluidization bed, where
gas–solid fluidization takes place above a distributor plate, and the disengagement zone
used to prevent the entrainment of polymer particles in the outgoing circulation gas stream.
The effluent from the loop reactor is directly fed into the GPR, while the liquid propylene is
vaporized and consumed in the first GPR [
22
]. This enhances the heat-removal capability
of the GPR and improves the propylene conversion per pass, resulting in a lower amount of
propylene that needs to be recycled. In the GPR the polymer particles, still containing the
active catalyst, come into contact with the reaction gases, such as propylene, comonomer(s)
(i.e., ethylene, 1-butene, 1-hexene) and hydrogen, which cause polymer to be produced
within the particles.
The fluidized bed reactor is equipped with a specially designed gas distribution plate
in order to minimize stagnant zones or areas of poor mixing where polymerizing particles
could form sheets or chunks [
23
]. The distribution plate enables part of the fluidization gas
to flow along the inside of the reactor wall in the area where the distribution plate normally
adjoins the reactor wall. The polymer particles coming from the loop reactor are fed into the
GPR via a feed pipe that is usually connected to the GPR at a point above the distribution
plate at a position higher than one-third of the effective diameter of the distribution plate.
The upwards-moving fluidization gas stream is established by withdrawing a circu-
lation gas stream from the top of the reactor, typically at the highest location. The gas
stream withdrawn from the reactor is then compressed, cooled and re-introduced to the
bottom zone of the reactor. Additional propylene and hydrogen are suitably introduced
into the circulation gas line. It is preferred to analyse the composition of the circulation gas,
for instance, by using online gas chromatography and adjusting the addition of the gas
components so that their contents are maintained at desired levels.
The reactor effluent, comprising polymer particles and the gas mixture from the
GPR, is withdrawn continuously or intermittently from the fluidized bed and fed into
a vessel (outlet vessel). Part of the gas mixture is separated in this vessel and returned
Polymers 2022,14, 4763 7 of 25
to the first GPR system to a place where the pressure is lower than the pressure of the
separation vessel [
24
]. The settled polymer is transferred to reactor module 2 or to the
degassing separator. When transferring to reactor module 2, a second gas mixture might
be introduced in the lower part of the vessel or downstream of the vessel to facilitate the
pneumatic transport to the second GPR. The vessel is operated essentially at the same
pressure as the polymer outlet from the first GPR and at a higher pressure than the second
GPR. The second gas flow should be below the minimum fluidization velocity of the
polymer particles to maintain a settled moving bed. A counter-current flow pattern, with
respect to polymer withdrawn from the first GPR, is established so that the reaction mixture,
coming from the first GPR, is replaced.
2.3. Module 2: Additional GPRs
Reactor module 2 comprises a second GPR and optionally a third GPR in series (see
Figure 3). Reactor module 2 enables the production of impact copolymers. Depending on
the required weight fraction of the impact copolymer, one or two additional reactors may
be employed. The second GPR is typically smaller in volume and operated at a slightly
lower pressure (i.e., operating pressure of 1–2 bar lower) compared to the first GPR. This
reactor is responsible for the production of rubber content up to a split level of 25%. The
operating temperature ranges from 75 to 90
◦
C for both the second and the potential third
GPR. The gas composition is controlled independently from the previous reactor due to
the transfer system as well as due to the special design of the product receivers. Typically,
the reaction mixture composition differs for at least one component compared to the first
GPR mixture (referring to hydrogen and/or ethylene). When a third GPR is employed, the
operating pressure of this reactor is slightly lower than that of the second GPR.
The polymer transfer from the second to the third GPR can be similar to the polymer
transfer from module 1 to module 2. More specifically, the polymer material produced
in each GPR is withdrawn from a suitable area, preferably from the middle zone of the
GPR, via the feed pipe into the outlet vessel, through the top part of the outlet vessel.
The polymer material is usually discharged in the form of polymer powder, which can
additionally comprise agglomerates. The polymer material is withdrawn from the GPR
continuously, having a velocity in the range of 5–15 m/s. Flush gas may be used to enhance
the transport of the polymer material from the GPR to the outlet vessel. The outlet vessel
has a main, a bottom and a top part. As a matter of definition, the main part has the highest
effective diameter, whereas the bottom part has a lower effective diameter than the main
part. The top part is merely a closure of the outlet vessel.
The challenge in operating the product outlet vessel is removing the produced polymer
powder with the minimum amount of entrained gas mixture coming from the GPR via the
vessel outlet. An increased amount of gas(es) in the polymer powder results in waste flaring
and can further cause quality problems in the resulting polymer powder. By designing
in detail the outlet vessel, especially regarding the selection of the position of the powder
surface and of the injection point of the second gas mixture—the barrier gas—on the outlet
vessel, the amount of entrained gas mixture in the polymer powder can be significantly
reduced [
25
]. It has to be pointed out that the barrier gas should not disturb the operation
of the gas–solids olefin polymerization reactor.
2.4. Downstream Area
The PP product is withdrawn from the gas-phase reactor and fed into a low-pressure
gas–solid separator (see Figure 4). The vapor is compressed and fed into the recovery area
while the solid PP particles are gravimetrically fed into the purge vessel. In the purge
vessel, the remaining active catalyst is deactivated by steam, while monomers are stripped
of PP particles by nitrogen addition [
17
]. Propylene and nitrogen can be recovered from
the purge gas by a recovery unit utilizing gas-permeable membranes [
26
]. The polymer is
then conveyed to the extrusion and pelletizing area.
Polymers 2022,14, 4763 8 of 25
Pressurized vapor is fed into distillation columns to recover and recycle propylene,
ethylene and hydrogen back to the polymerization reactors. Light components (e.g., inert
gases and impurities), heavy hydrocarbon components (e.g., waxes and oligomers) and
co-catalyst residues are removed by the recovery area [17].
2.5. Process Control
Polymerization reactor control is a key component for achieving stable operation at
an industrial scale. The main challenge is to maintain high product-quality consistency
while maximizing the production rate and minimizing the grade transition times. Borealis
was the first company in the polymer industry to implement a nonlinear model predic-
tive control (NMPC) methodology for reactor control. The proprietary Borealis NMPC
advanced process control technology (BorAPC, OnSpot) has been continuously developed
and optimized since 1994, when it was initially introduced [
27
]. BorAPC allows Borealis to
reach the full potential of the Borstar
®
PP technology by offering the possibility of operating
closer to the process limits (Figure 5).
Polymers 2022, 14, x FOR PEER REVIEW 8 of 25
the purge gas by a recovery unit utilizing gas-permeable membranes [26]. The polymer is
then conveyed to the extrusion and pelletizing area.
Pressurized vapor is fed into distillation columns to recover and recycle propylene,
ethylene and hydrogen back to the polymerization reactors. Light components (e.g., inert
gases and impurities), heavy hydrocarbon components (e.g., waxes and oligomers) and
co-catalyst residues are removed by the recovery area [17].
2.5. Process Control
Polymerization reactor control is a key component for achieving stable operation at
an industrial scale. The main challenge is to maintain high product-quality consistency
while maximizing the production rate and minimizing the grade transition times. Borealis
was the first company in the polymer industry to implement a nonlinear model predictive
control (NMPC) methodology for reactor control. The proprietary Borealis NMPC ad-
vanced process control technology (BorAPC, OnSpot) has been continuously developed
and optimized since 1994, when it was initially introduced [27]. BorAPC allows Borealis
to reach the full potential of the Borstar® PP technology by offering the possibility of op-
erating closer to the process limits (Figure 5).
Figure 5. Temperature–time profile for analysis of structure formation at different cooling rates. The
final, red-coloured heating segment served for analysis of the fraction of crystals in the prior cooling
step.
Nowadays, all Borstar® PP plants operate NMPC in a closed loop, meaning that the
controller is able to decide on the control actions without active input from the operator.
The controlled variables (CVs) are the production rate, solids content, melt flow index
(correlating to the molecular weight of the polymer) and comonomer content. Alterna-
tively, the hydrogen/propylene ratio can substitute for the melt flow index (depending on
the available online equipment of the plant). A separate NMPC controller is applied for
each reactor; however, manipulated variable values are always shared downstream, ena-
bling feed-forward control actions. NMPC technology not only ensures optimized reactor
operation but further simplifies plant control because the operator only needs to provide
the set-point values for the CVs. OnSpot consists of an estimator, a process model and a
control algorithm. The model parameters are continuously adapted based on process
measurements to account for unmeasured disturbances. The control algorithm minimizes
the deviation of the predicted values from the set-point trajectory. The mathematical, non-
linear process model is based on first principles (mass and energy balances), which allows
OnSpot to control the reactor at all operating points using a single model. The controller
sample time is typically 1 min, with a prediction horizon of 3–8 h [28].
Trip limit
Operational limit
Controlled variable
Manual
control
APC
Control
Figure 5.
Temperature–time profile for analysis of structure formation at different cooling rates.
The final, red-coloured heating segment served for analysis of the fraction of crystals in the prior
cooling step.
Nowadays, all Borstar
®
PP plants operate NMPC in a closed loop, meaning that the
controller is able to decide on the control actions without active input from the operator.
The controlled variables (CVs) are the production rate, solids content, melt flow index
(correlating to the molecular weight of the polymer) and comonomer content. Alternatively,
the hydrogen/propylene ratio can substitute for the melt flow index (depending on the
available online equipment of the plant). A separate NMPC controller is applied for each
reactor; however, manipulated variable values are always shared downstream, enabling
feed-forward control actions. NMPC technology not only ensures optimized reactor op-
eration but further simplifies plant control because the operator only needs to provide
the set-point values for the CVs. OnSpot consists of an estimator, a process model and
a control algorithm. The model parameters are continuously adapted based on process
measurements to account for unmeasured disturbances. The control algorithm minimizes
the deviation of the predicted values from the set-point trajectory. The mathematical, non-
linear process model is based on first principles (mass and energy balances), which allows
OnSpot to control the reactor at all operating points using a single model. The controller
sample time is typically 1 min, with a prediction horizon of 3–8 h [28].
2.6. Dry End, Quality Control and Sustainability
The Borstar
®
PP process is supported by highly competent quality control (QC) and
online polymer analysis (OLPA) teams, which are responsible for a wide range of polymer
characterization techniques in an off-line and online format, respectively. Sampling points,
located in appropriate positions in the reactor modules downstream of the deactivation
Polymers 2022,14, 4763 9 of 25
and dry end of the Borstar
®
PP process (e.g., sampling each reactor separately, upstream
and downstream of extrusion), offer the possibility of representative product collection.
The QC lab is fully equipped with state-of-the-art characterization techniques, de-
pending on the plant-specific product portfolio needs. In strong connection with Borealis
Innovation Centres, the QC labs are able to reveal, among others, detailed rheological, me-
chanical, optical, molecular and morphological polymer properties. Typical measurement
times (from sampling until the available characterization result is uploaded to the plant
database) range from 1–3 h.
When required, the Borstar
®
PP process can be further equipped with online char-
acterization capabilities, reducing the measurement time to less than 30 min. The OLPA
lab is responsible for automated sample collection and characterization, using a variety of
potential methodologies, including online NMR (nuclear magnetic resonance), rheometry,
gel analysis and pellet analysis (shape and size distribution) techniques.
The dry end comprises powder transport and buffering, extrusion (including homoge-
nization, additivation and pelletizing), pellet transport and final lot homogenization. Twin
screw extruders are used to extrude and homogenize the product. The final melt flow can
be adjusted by controlled rheology and using a dedicated OLPA rheometer. The pelletizers
are underwater granulators where the molten polymer strand is cut and cooled rapidly,
forming pellets. The pellets are transported by a water stream into a dryer, and downstream
from the dryer, the pellets are classified by size. The pelletized product is conveyed from
the surge tank to the blenders, where the lot is blended and classified. Finally, the product
is transferred to the material handling area.
The material handling area is typically considered outside battery limits. The product
can be conveyed from the blenders to storage silos for bulk transport or to packaging lines
to bag or box the product. An elutriator can be installed upstream of the bulk silos and
upstream of the packaging line.
The Borstar
®
PP process can be further equipped with technologies to minimize the
emission of hydrocarbons to the atmosphere both from the polymer product and during
the process of pellet blending. Extruders can be equipped with vacuum stripping and
condensate injection upstream of the vent port. Moreover, silo aeration can be applied
downstream of the extrusion. Air from silo aeration and off-gas from propylene and
nitrogen recovery membranes are lean in hydrocarbons. These hydrocarbons are then
processed by a catalytic oxidation or regenerative thermal oxidation (RTO) unit to reduce
hydrocarbon emissions to the atmosphere [26].
3. Borealis Catalyst Design
The development of Borealis’ proprietary catalyst started in the late 1980s under the
scope of finding an alternative magnesium halide-based support system allowing better
design of highly porous catalysts. In combination with a specific internal donor, this
should allow the development of a fourth-generation ZN catalyst [
8
,
9
] suitable both for
the production of high-isotacticity iPP homopolymers and high-impact copolymers with
significant amounts of amorphous ethylene–propylene copolymer (EPC).
The standard support type was then micronized MgCl
2
produced by ball-milling,
applying co-milling with TiCl
4
as a titanization process. The role of the internal donor
in generating and stabilizing the active titanium centres and allowing the efficient use
of external donors was becoming clear already [
29
,
30
], and phthalic acid esters were
established as suitable donor compounds. Also, the importance of selecting the right
external donor from the family of alkoxysilanes was supported by respective studies [
31
]
finding a combined effect on stereoregularity, activity and molecular weight. All of these
detail studies also showed the multi-site nature of such catalyst systems [32].
Several ideas were considered for developing the new catalyst type, with spray-
crystallization of a molten magnesium dichloride complex to form the carrier being one of
them, for which also a complex of MgCl
2
-hexahydrate with NH
4
Cl was envisioned [
33
].
Transesterified ZN catalyst types are produced in several stages, starting with the for-
Polymers 2022,14, 4763 10 of 25
mation of an adduct between MgCl
2
and an aliphatic alcohol, like ethanol, to obtain
MgCl
2
* 3 EtOH, as already suggested by Chien et al. [
30
]. This can be spray-crystallized
from a melt of ~100
◦
C into particles of 10–300
µ
m with atomizing nozzles in a nitrogen at-
mosphere [
34
], those being suitable for titanization with an excess of TiCl
4
in a hydrocarbon
solvent (this and all following reaction equations are not to be seen as stoichiometric):
MgCl2×n R1OH + (n + m) TiCl4→MgCl2×n TiCl3OR1 + n HCl + m TiCl4(1)
with n being 1–6 and R1 being, for example, C
2
H
5
. Adding an ester or a carboxylic acid,
preferably an alkyl ester or phthalic acid ester like di(isobutyl phthalate), as an internal
donor to this titanized carrier, an adduct of all components is created:
MgCl2×n TiCl3OR1 + n R3COOR2 →MgCl2×n TiCl3OR1 ×n R3COOR2 (2)
This adduct can be transesterified at a temperature higher than 136
◦
C, i.e., above the
boiling point of TiCl4, in which process the ester groups R1 and R2 exchange places:
MgCl2×n TiCl3OR1 ×n R3COOR2 →MgCl2×n TiCl3OR2 ×n R3COOR1 (3)
The temperature, solvent choice and duration have been found to affect multiple
parameters of the resulting catalyst, like the Ti content, polymerization activity and iso-
tacticity of the resulting PP [
35
]. Finally, undesired residue materials such as titanium
chloro-alkoxides are removed by extraction or washing with a hydrocarbon solvent, receiv-
ing an adduct of the carrier and the ester donor coupled to the active Ti centre:
MgCl2×n TiCl3OR2 ×n R3COOR1 ×m TiCl4→
MgCl2×n R3COOR1 ×m TiCl4+ n TiCl3OR2 (4)
For polymerization, this procatalyst is combined with an aluminium-alkyl or -alkylhalide
and an external donor as usual. By optimizing the components and preparation steps,
catalysts of high porosity, giving the required isotacticity and rather large polymer particles
without a too-high “fines” fraction, could be developed [36,37].
This catalyst type has been both studied systematically and employed in multiple
polymerization setups. The work by Garoff et al. [
38
] highlights the extent to which
isotacticity and molecular weight distribution can be modified by varying the ratio between
the external donor feed and Ti content in the catalyst feed, as shown in Figure 6for
dicyclopentyl dimethoxysilane (donor D). Here, triethylaluminium (TEAL) at a molar
Al/Ti ratio of 250 was employed, resulting in a parallel change of the Al/donor ratio. In
practice, the additional effects on catalyst activity—having a maximum at a donor/Ti ratio
of ~10 in the present case—and the hydrogen response with respect to the molecular weight
of the polymer need to be considered. In any case, the well-known correlation between
isotacticity and modulus [
39
] made the transesterified catalysts suitable for producing
high-stiffness grades, and the improved stability at higher polymerization temperatures fit
well with the Borstar PP technology.
This first generation of Borealis catalysts, commonly designated as BCF types, was
subsequently used in the development of complex grades like multimodal random copoly-
mers and high-impact heterophasic copolymers. In a comparative study by the group of
McKenna [
40
], it was found to be suitable for incorporating high amounts of EPC while
maintaining full powder flowability, proving the porous nature of the matrix polymer.
Polymers 2022,14, 4763 11 of 25
Polymers 2022, 14, x FOR PEER REVIEW 11 of 25
Figure 6. Effect of the molar ratio between donor and Ti on isotacticity (red, pentad regularity by
13C-NMR) and xylene cold soluble (blue, XCS) fraction (data from [38]).
Another basic idea for generating an even particle size in catalyst production was the
origin of the second generation of proprietary catalysts for the Borstar process. The new
emulsion-based catalyst preparation process—commonly called ‘Sirius’—uses liquid pre-
cursors in a three-step process [41–43], starting with a reaction of butyl octyl magnesium
(BOMAG) with 2-ethyl-hexanol to form an Mg-alkoxide:
C4H9MgC8H17 + 2 C6H12(C2H5)OH → Mg(OC6H12(C2H5))2 + C4H10 + C8H18
(5)
This alkoxide, used in excess, is partially allowed to react with phthaloyl dichloride
(PDC) to form a complex between magnesium chloride and di(ethylhexyl)phthalate
(DEHP), acting as internal donor in the catalyst precursor for the later polymerization:
Mg(OC6H12(C2H5))2 + PDC → MgCl2 + DEHP
(6)
Reacting this complex with TiCl4 results in the formation of a two-phase liquid–liquid
system, as shown on the left-hand side of Figure 7. Adding a specific polymeric surfactant
allows the formation of an emulsion with droplets in the range of 10–100 µm diameter,
which can be converted into a dispersion with solid particles by a temperature increase
(details of the reaction are described by Rönkkö et al. [41], with the related conditions in
the patents [42,43]). Washing and drying results in perfectly spherical particles which
have a smooth surface and no apparent porosity, as shown on the right-hand side of Fig-
ure 7.
Figure 7. Process scheme of the ‘Sirius’ emulsion process for ZN catalyst production.
These catalysts were tested in polymerization both alone [44,45] and in direct com-
parison to the earlier generation of transesterified types [46,47]. The first question to be
Figure 6.
Effect of the molar ratio between donor and Ti on isotacticity (red, pentad regularity by
13C-NMR) and xylene cold soluble (blue, XCS) fraction (data from [38]).
Another basic idea for generating an even particle size in catalyst production was
the origin of the second generation of proprietary catalysts for the Borstar process. The
new emulsion-based catalyst preparation process—commonly called ‘Sirius’—uses liquid
precursors in a three-step process [
41
–
43
], starting with a reaction of butyl octyl magnesium
(BOMAG) with 2-ethyl-hexanol to form an Mg-alkoxide:
C4H9MgC8H17 +2C6H12(C2H5)OH →Mg(OC6H12(C2H5))2+ C4H10 + C8H18 (5)
This alkoxide, used in excess, is partially allowed to react with phthaloyl dichloride
(PDC) to form a complex between magnesium chloride and di(ethylhexyl)phthalate (DEHP),
acting as internal donor in the catalyst precursor for the later polymerization:
Mg(OC6H12(C2H5))2+ PDC →MgCl2+ DEHP (6)
Reacting this complex with TiCl
4
results in the formation of a two-phase liquid–liquid
system, as shown on the left-hand side of Figure 7. Adding a specific polymeric surfactant
allows the formation of an emulsion with droplets in the range of 10–100
µ
m diameter,
which can be converted into a dispersion with solid particles by a temperature increase
(details of the reaction are described by Rönkkö et al. [
41
], with the related conditions in
the patents [
42
,
43
]). Washing and drying results in perfectly spherical particles which have
a smooth surface and no apparent porosity, as shown on the right-hand side of Figure 7.
Polymers 2022, 14, x FOR PEER REVIEW 11 of 25
Figure 6. Effect of the molar ratio between donor and Ti on isotacticity (red, pentad regularity by
13C-NMR) and xylene cold soluble (blue, XCS) fraction (data from [38]).
Another basic idea for generating an even particle size in catalyst production was the
origin of the second generation of proprietary catalysts for the Borstar process. The new
emulsion-based catalyst preparation process—commonly called ‘Sirius’—uses liquid pre-
cursors in a three-step process [41–43], starting with a reaction of butyl octyl magnesium
(BOMAG) with 2-ethyl-hexanol to form an Mg-alkoxide:
C4H9MgC8H17 + 2 C6H12(C2H5)OH → Mg(OC6H12(C2H5))2 + C4H10 + C8H18
(5)
This alkoxide, used in excess, is partially allowed to react with phthaloyl dichloride
(PDC) to form a complex between magnesium chloride and di(ethylhexyl)phthalate
(DEHP), acting as internal donor in the catalyst precursor for the later polymerization:
Mg(OC6H12(C2H5))2 + PDC → MgCl2 + DEHP
(6)
Reacting this complex with TiCl4 results in the formation of a two-phase liquid–liquid
system, as shown on the left-hand side of Figure 7. Adding a specific polymeric surfactant
allows the formation of an emulsion with droplets in the range of 10–100 µm diameter,
which can be converted into a dispersion with solid particles by a temperature increase
(details of the reaction are described by Rönkkö et al. [41], with the related conditions in
the patents [42,43]). Washing and drying results in perfectly spherical particles which
have a smooth surface and no apparent porosity, as shown on the right-hand side of Fig-
ure 7.
Figure 7. Process scheme of the ‘Sirius’ emulsion process for ZN catalyst production.
These catalysts were tested in polymerization both alone [44,45] and in direct com-
parison to the earlier generation of transesterified types [46,47]. The first question to be
Figure 7. Process scheme of the ‘Sirius’ emulsion process for ZN catalyst production.
Polymers 2022,14, 4763 12 of 25
These catalysts were tested in polymerization both alone [
44
,
45
] and in direct com-
parison to the earlier generation of transesterified types [
46
,
47
]. The first question to be
answered was whether an apparently non-porous catalyst could yield high productivity at
all, and how fragmentation in polymerization would occur. Most previous polymerization
models, like the one presented by Cecchin et al. [
48
], assume fragmentation around primary
MgCl
2
micro-crystals. No such structure is obvious in the emulsion-based catalyst particles,
but both an X-ray diffraction analysis and the fragment analysis discussed below indicate
its presence. For emulsion-based or self-supported catalysts, different but similar frag-
mentation processes have been observed, however without significant production of ‘fines’
by excessive growth stresses. The rather narrow particle-size distribution and spherical
shape of the original catalyst particles are retained very nicely, analogous to the ‘replica
effect’ for other spherical catalyst types [
44
]. After a single- or two-stage polymerization of
a homopolymer or random copolymer, MgCl
2
fragments in the size range below 100 nm
were observed [45,49].
This first generation of emulsion-type catalysts has, like the transesterified catalyst
based on spray-crystallized support, a phthalate-type internal donor formed in situ from
the precursor pthaloyl dichloride, but the resulting polymers differ significantly. For
homopolymers, a narrower MWD and lower oligomer content but also a greater flexibility
in terms of isotacticity control by the external donor feed were observed [
46
], enabling
high-flow grades for fibre and moulding applications with low emissions. Even more
relevant are the differences for ethylene–propylene (C2C3) random copolymers, where
a higher fraction of isolated C2 units—commonly called ‘improved randomness’—was
found [
47
]. This results in a lower melting point at a given comonomer content and results
in an improved sterilization resistance [
50
]. In line with this performance, this catalyst
type was also found suitable for the production of soft random-heterophasic copolymers
(RAHECOs), the details of which will be discussed in the following chapter.
Limitations in the incorporation of high EPC amounts led to the development of
an emulsion-based catalyst type with dispersed micro- or nanoparticles like Al
2
O
3
or
SiO
2
[
43
,
51
,
52
]. These particles, while inactive for polymerization, are capable of gen-
erating additional cavities in the growing polymer particles during the early stages of
polymerization (matrix stage), allowing a more evenly distributed EPC dispersion in
further stages [53].
The ‘Sirius’ technology was also adopted for developing Borealis’ own line of post-
phthalate ZN catalysts, although first attempts in that direction had been conducted on a
more conventional type of support [
54
]. This general trend of applying alternative internal
donors instead of phthalates was motivated largely by health and environmental concerns
over this substance group. Certain phthalates are known to have estrogenic activity [
55
],
but the respective discussion was rather incited by the high amounts used as plasticizers
in poly(vinyl chloride) (PVC) than by the minute amounts in polyolefin catalysts. The
residual content of internal donors or their decomposition products in iPP is, because of
the high catalyst activity, in the range of less than 0.1 ppm.
Post-phthalate ZN catalyst development was started by LyondellBasell with diethers
and succinates [
13
,
48
,
56
,
57
] as internal donors. The respective catalysts developed at
Borealis use a citraconate as an internal donor [
58
] and have been found to be suitable for
the commercial production of iPP homopolymers [
59
] and copolymers with ethylene [
60
].
The change in internal donor again changes the comonomer incorporation, as observed for
the earlier generations by Vestberg et al. [
61
], resulting in a lower melting point at a given
ethylene concentration (see Figure 8).
Polymers 2022,14, 4763 13 of 25
Polymers 2022, 14, x FOR PEER REVIEW 13 of 25
Figure 8. Melting point as function of ethylene content for monomodal random copolymers based
on emulsion-type catalyst with phthalate (full symbols) and citraconate (open symbols; data from
[47,60]).
This catalyst type is also employed for producing heterophasic ethylene–propylene
copolymers (HECOs) with high impact strength in a wide range of EPC content, achieving
free-flowing polymer powders up to an XCS content of 37 wt% [16]. As in random copol-
ymers, reactivity towards C2 is higher than for ZNC types with phthalate donors, causing
differences in the C2 content of the XCS fraction and Tg(EPC). Further developments
broadening the performance range of copolymers can be expected.
A special catalyst-related technology also developed at Borealis is pre-polymeriza-
tion with a suitable monomer to produce a high-melting-point polymer acting as a nucle-
ating agent for iPP. Due to catalyst fragmentation, this process allows a dispersion of pre-
polymer particles in the final polymer, giving massive nucleation effects at a very low
concentration [62]. One of the best candidates for in-reactor nucleation is isotactic poly(vi-
nyl cyclohexane), PVCH, for which a high degree of lattice-matching to several planes of
the α-phase of iPP has been found by the group of Lotz [63]. Isotactic PVCH is highly
crystalline with a melting point of ~360 °C and can be produced by polymerization with
Ziegler-type catalysts, as already described in the 1960s [64]. The interest in this period
was, however, more on the atactic counterpart produced by the hydrogenation of poly-
styrene and seen as a possible competitor to polycarbonate [64,65]. A patent application
for its use as nucleating agent for iPP was filed by Sumitomo in 1983 [66], already high-
lighting the high activity with an increase in crystallization temperature (Tc) of more than
10 °C at 29 ppm of PVCH at low concentrations. The polymer was marketed by Sumitomo
as masterbatch for nucleation under the trade name CAP, with Kakugo highlighting its
performance in several papers and actually claiming nucleating activity at 1 ppm already
[67].
This process was optimized by Borealis in several respects, including the use of a
transesterified ZN catalyst as described above in combination with a silane-type external
donor for optimizing the PVCH structure and an elevated temperature for shortening the
pre-polymerization time [68]. Combining this with a second pre-polymerization step with
propene and optimization of the isotacticity, stiffness levels of more than 2 GPa could be
reached for homopolymers, and the process was also expanded to random and hetero-
phasic copolymers [69]. Structurally similar but alternative polymeric nucleating agents
like poly(cyclopentene), or PCP, were also found to work at very low concentrations [70]
but did not succeed technically (Lee and Yoon also compared the effects of PVCH and
PCP to other polymeric nucleating agents like PA-66 and PPEK in their study).
More recently, this family of efficient polymeric nucleating agents has been broad-
ened further by poly(trimethylallyl silane), PTMAS [71], and poly(vinyl cyclopentane),
Figure 8.
Melting point as function of ethylene content for monomodal random copolymers
based on emulsion-type catalyst with phthalate (full symbols) and citraconate (open symbols; data
from [47,60]).
This catalyst type is also employed for producing heterophasic ethylene–propylene
copolymers (HECOs) with high impact strength in a wide range of EPC content, achieving
free-flowing polymer powders up to an XCS content of 37 wt% [
16
]. As in random copoly-
mers, reactivity towards C2 is higher than for ZNC types with phthalate donors, causing
differences in the C2 content of the XCS fraction and Tg(EPC). Further developments
broadening the performance range of copolymers can be expected.
A special catalyst-related technology also developed at Borealis is pre-polymerization
with a suitable monomer to produce a high-melting-point polymer acting as a nucleating
agent for iPP. Due to catalyst fragmentation, this process allows a dispersion of pre-polymer
particles in the final polymer, giving massive nucleation effects at a very low concentra-
tion [
62
]. One of the best candidates for in-reactor nucleation is isotactic poly(vinyl cyclo-
hexane), PVCH, for which a high degree of lattice-matching to several planes of the
α
-phase
of iPP has been found by the group of Lotz [
63
]. Isotactic PVCH is highly crystalline with
a melting point of ~360
◦
C and can be produced by polymerization with Ziegler-type
catalysts, as already described in the 1960s [
64
]. The interest in this period was, however,
more on the atactic counterpart produced by the hydrogenation of polystyrene and seen as
a possible competitor to polycarbonate [
64
,
65
]. A patent application for its use as nucleating
agent for iPP was filed by Sumitomo in 1983 [
66
], already highlighting the high activity with
an increase in crystallization temperature (Tc) of more than 10
◦
C at 29 ppm of PVCH at low
concentrations. The polymer was marketed by Sumitomo as masterbatch for nucleation
under the trade name CAP, with Kakugo highlighting its performance in several papers
and actually claiming nucleating activity at 1 ppm already [67].
This process was optimized by Borealis in several respects, including the use of a
transesterified ZN catalyst as described above in combination with a silane-type external
donor for optimizing the PVCH structure and an elevated temperature for shortening the
pre-polymerization time [
68
]. Combining this with a second pre-polymerization step with
propene and optimization of the isotacticity, stiffness levels of more than 2 GPa could be
reached for homopolymers, and the process was also expanded to random and heterophasic
copolymers [
69
]. Structurally similar but alternative polymeric nucleating agents like
poly(cyclopentene), or PCP, were also found to work at very low concentrations [
70
] but
did not succeed technically (Lee and Yoon also compared the effects of PVCH and PCP to
other polymeric nucleating agents like PA-66 and PPEK in their study).
More recently, this family of efficient polymeric nucleating agents has been broad-
ened further by poly(trimethylallyl silane), PTMAS [
71
], and poly(vinyl cyclopentane),
PVCP [
72
]. While all of these are efficient nucleating agents for iPP and also can prevent the
Polymers 2022,14, 4763 14 of 25
formation of the mesomorphic phase at higher cooling rates (i.e., in quenched samples), the
efficiency under the latter conditions is still best for PVCH, as shown in Figure 9, where the
doubling of the points at higher cooling rates indicates the combined formation of the
α
-
and mesomorphic phase. This makes the technology especially suitable for high-isotacticity
film grades combining high modulus and low water vapour permeability even under cast-
film conditions [
73
], offering an economical alternative to biaxially oriented or specially
coated films in the packaging of moisture-sensitive products (an example for such a grade
is HD905CF of Borealis). It works similarly in heterophasic copolymers for applications
requiring high impact strength at sub-zero temperatures, outperforming other nucleating
agents and ensuring stable
α
-phase formation up to more than 1000 K/s [
74
]. High-flow
grades designed in this way are particularly suitable for thin-wall injection moulding
applications, as in the case of BJ368MO. Here, the higher cooling rate still allowing stable
crystallization results in a significant cycle-time reduction.
Polymers 2022, 14, x FOR PEER REVIEW 14 of 25
PVCP [72]. While all of these are efficient nucleating agents for iPP and also can prevent
the formation of the mesomorphic phase at higher cooling rates (i.e., in quenched sam-
ples), the efficiency under the latter conditions is still best for PVCH, as shown in Figure
9, where the doubling of the points at higher cooling rates indicates the combined for-
mation of the α- and mesomorphic phase. This makes the technology especially suitable
for high-isotacticity film grades combining high modulus and low water vapour permea-
bility even under cast-film conditions [73], offering an economical alternative to biaxially
oriented or specially coated films in the packaging of moisture-sensitive products (an ex-
ample for such a grade is HD905CF of Borealis). It works similarly in heterophasic copol-
ymers for applications requiring high impact strength at sub-zero temperatures, outper-
forming other nucleating agents and ensuring stable α-phase formation up to more than
1000 K/s [74]. High-flow grades designed in this way are particularly suitable for thin-
wall injection moulding applications, as in the case of BJ368MO. Here, the higher cooling
rate still allowing stable crystallization results in a significant cycle-time reduction.
Figure 9. Crystallization temperature (Tc) as a function of cooling rate for a non-nucleated (neat)
iPP homopolymer (grey squares), two iPP homopolymers with pre-polymerized nucleating poly-
mers, PVCP (blue stars) and PVCH (black squares), and one conventionally nucleated iPP homo-
polymer with DMDBS (red circles; data are a combination of results in [72] and previously un-
published data).
The in-reactor-produced iPP/PVCH combination can be varied in concentration [75]
and is capable of interacting positively with reinforcing fillers like talc or glass fibres. In
this way, PVCH is also suitable to compensate for the unwanted nucleating effects of cer-
tain pigments. Moreover, the process has also been adapted to work for post-phthalate
ZN catalysts based on an emulsion process, as also detailed above [76]. As a non-migrat-
ing nucleating agent of high thermal stability, it is perfectly suited for high purity require-
ments, a frequent request in the post-phthalate era of iPP design.
4. Borstar PP Composition and Performance Range
One of the main advantages of using the combination of a liquid bulk loop reactor
and a fluidized gas-phase reactor (GPR) in series for producing homopolymers and ran-
dom copolymers with C2 or higher α-olefins like 1-butene (C4) or 1-hexene (C6) is the
higher difference in average molecular weight and comonomer content values that can be
achieved between the two fractions [77,78]. The use of bi- or multimodality for broadening
the MWD with the target of increasing flow-induced crystallization and stiffness [79] is
based on the role of longer molecules in the formation of the stable nuclei and shish-kebab
Figure 9.
Crystallization temperature (Tc) as a function of cooling rate for a non-nucleated (neat) iPP
homopolymer (grey squares), two iPP homopolymers with pre-polymerized nucleating polymers,
PVCP (blue stars) and PVCH (black squares), and one conventionally nucleated iPP homopolymer
with DMDBS (red circles; data are a combination of results in [
72
] and previously unpublished data).
The in-reactor-produced iPP/PVCH combination can be varied in concentration [
75
]
and is capable of interacting positively with reinforcing fillers like talc or glass fibres. In this
way, PVCH is also suitable to compensate for the unwanted nucleating effects of certain
pigments. Moreover, the process has also been adapted to work for post-phthalate ZN
catalysts based on an emulsion process, as also detailed above [
76
]. As a non-migrating
nucleating agent of high thermal stability, it is perfectly suited for high purity requirements,
a frequent request in the post-phthalate era of iPP design.
4. Borstar PP Composition and Performance Range
One of the main advantages of using the combination of a liquid bulk loop reactor and
a fluidized gas-phase reactor (GPR) in series for producing homopolymers and random
copolymers with C2 or higher
α
-olefins like 1-butene (C4) or 1-hexene (C6) is the higher
difference in average molecular weight and comonomer content values that can be achieved
between the two fractions [
77
,
78
]. The use of bi- or multimodality for broadening the
MWD with the target of increasing flow-induced crystallization and stiffness [
79
] is based
on the role of longer molecules in the formation of the stable nuclei and shish-kebab
structures mostly studied in solution or extruder blends [
80
] and found to increase modulus
significantly [
81
]. This approach is, however, not practical at a large scale; reactor-based
Polymers 2022,14, 4763 15 of 25
bimodality is a suitable alternative which has been applied successfully in polyethylene
design [82].
Bimodal homopolymers with a significant MFR difference between the loop reactor
and GPR were among the first innovative products from the Borstar
®
PP process (the
importance of excellent homogenization for utilizing the enhanced rheological performance
has just been highlighted recently again [
78
]). Combining increased polydispersity with
high isotacticity and selective nucleation gave advantages not only in stiffness, but also
in properties otherwise requiring massive orientation in processing like the water vapour
impermeability of films [
73
]. Shortly afterward, this design principle was also applied to
produce advanced random copolymers [
83
–
85
] in the loop/GPR design. PP homopolymers
and C2C3 random copolymers are generally miscible, as long as the comonomer content
does not exceed ~8 wt.% for most ZN catalyst types, and as long as the molecular weight
difference between the two components is not too high [
78
,
86
], resulting in a single-phase
structure allowing high transparency and low haze.
For film applications, this allows a broader melting range, resulting from a wider
distribution of lamellar thickness and leading to a broader sealing window, i.e., a lower
sealing initiation temperature (SIT) and/or a higher sealing end temperature (SET) [
84
].
Likewise, the sterilization behaviour and especially the loss of transparency in steam
sterilization can be influenced by having a low comonomer content fraction in the com-
position [
85
]. Another relevant application area is thin-wall injection moulding requiring
an MFR level of 70–100 g/10 min, for which Figure 10 presents a direct comparison of
mono- and bimodal C2C3 random copolymers. In food packaging, for example, the design
of cups and closures requires materials with a good balance between haze and impact
strength, but also heat resistance and solubles content. The bimodal materials presented
here all have a total C2 content of ~4 wt.%, but with a significant split between the loop and
GPR fractions, allowing an improved property balance [
87
], with a typical example being
BorPure
™
RJ766MO. Combining this design principle with the aforementioned in-reactor
nucleation allows for the production of parts with stable long-term performance at low
cycle times, i.e., high cooling rates [
82
]. This design scheme can be expanded even further
when applying single-site catalysts with their inherently better comonomer incorporation
and lower level of extractables [78,88].
Polymers 2022, 14, x FOR PEER REVIEW 15 of 25
structures mostly studied in solution or extruder blends [80] and found to increase mod-
ulus significantly [81]. This approach is, however, not practical at a large scale; reactor-
based bimodality is a suitable alternative which has been applied successfully in polyeth-
ylene design [82].
Bimodal homopolymers with a significant MFR difference between the loop reactor
and GPR were among the first innovative products from the Borstar® PP process (the im-
portance of excellent homogenization for utilizing the enhanced rheological performance
has just been highlighted recently again [78]). Combining increased polydispersity with
high isotacticity and selective nucleation gave advantages not only in stiffness, but also in
properties otherwise requiring massive orientation in processing like the water vapour
impermeability of films [73]. Shortly afterward, this design principle was also applied to
produce advanced random copolymers [83–85] in the loop/GPR design. PP homopoly-
mers and C2C3 random copolymers are generally miscible, as long as the comonomer
content does not exceed ~8 wt.% for most ZN catalyst types, and as long as the molecular
weight difference between the two components is not too high [78,86], resulting in a sin-
gle-phase structure allowing high transparency and low haze.
For film applications, this allows a broader melting range, resulting from a wider
distribution of lamellar thickness and leading to a broader sealing window, i.e., a lower
sealing initiation temperature (SIT) and/or a higher sealing end temperature (SET) [84].
Likewise, the sterilization behaviour and especially the loss of transparency in steam ster-
ilization can be influenced by having a low comonomer content fraction in the composi-
tion [85]. Another relevant application area is thin-wall injection moulding requiring an
MFR level of 70–100 g/10 min, for which Figure 10 presents a direct comparison of mono-
and bimodal C2C3 random copolymers. In food packaging, for example, the design of
cups and closures requires materials with a good balance between haze and impact
strength, but also heat resistance and solubles content. The bimodal materials presented
here all have a total C2 content of ~4 wt.%, but with a significant split between the loop
and GPR fractions, allowing an improved property balance [87], with a typical example
being BorPure™ RJ766MO. Combining this design principle with the aforementioned in-
reactor nucleation allows for the production of parts with stable long-term performance
at low cycle times, i.e., high cooling rates [82]. This design scheme can be expanded even
further when applying single-site catalysts with their inherently better comonomer incor-
poration and lower level of extractables [78,88].
(a)
(b)
Figure 10. Comparison of high-flow mono- and bimodal C2C3 random copolymers: (a) balance be-
tween toughness (total energy from instrumented puncture test) and haze, and (b) correlation be-
tween total C2 content and melting point (data from [87] and commercial references).
The Borstar® PP Module 2 setup (pre-polymerization, loop and two GPRs in series)
is able to further expand the product properties window by introducing a trimodal
Figure 10.
Comparison of high-flow mono- and bimodal C2C3 random copolymers: (
a
) balance
between toughness (total energy from instrumented puncture test) and haze, and (
b
) correlation
between total C2 content and melting point (data from [87] and commercial references).
The Borstar
®
PP Module 2 setup (pre-polymerization, loop and two GPRs in series) is
able to further expand the product properties window by introducing a trimodal copolymer
design, where both the MWD and CC vary between the three distinct fractions. Composi-
tions for hot-water and pressure pipes are one possible target here [
89
], as the inherently
Polymers 2022,14, 4763 16 of 25
low MFR for this application allows a wide spread of the molecular weights of the fractions,
such as over three orders of magnitude between 0.002 and 2 g/10 min. Trimodal composi-
tions may also include a homopolymer fraction and ultimately a very propylene-(C3)-rich
disperse phase (EPC), similar to the concept of ‘interpolymers’ developed for combining
transparency and impact strength in the area of sterilizable packaging [
90
,
91
] (the Borealis
grade Borpact
™
BC918CF is an example of such a grade). To demonstrate the resulting
phase structures, examples of a HECO with a C3-rich and C2-rich disperse phase are
presented in Figure 11. The difference between the homogeneous EPC particles having a
diffuse interface to the matrix in the case of (a) and the core-shell particles with a smoother
interface to the matrix in the case of (b) is obvious. Transmission-electron micrographs
based on RuO4-contrasted specimens are used here, as this technique gives a very high
resolution down to the lamellar level. A more complete series of images and the related
properties can be found in the work of Grein et al. [
90
]. Nevertheless, a study of the onset
of phase separation in the case of ZN catalysts is still not available in the literature. For
single-site catalysts, this is expected within the range of 10 to 22 wt.% of C2 in the xylene
cold soluble (XCS) fraction, as also indicated by the appearance of a second glass transition
for the EPC [
92
], but the matrix design and molecular weight are likely to play a role as
well in practice.
Polymers 2022, 14, x FOR PEER REVIEW 16 of 25
copolymer design, where both the MWD and CC vary between the three distinct fractions.
Compositions for hot-water and pressure pipes are one possible target here [89], as the
inherently low MFR for this application allows a wide spread of the molecular weights of
the fractions, such as over three orders of magnitude between 0.002 and 2 g/10 min. Tri-
modal compositions may also include a homopolymer fraction and ultimately a very pro-
pylene-(C3)-rich disperse phase (EPC), similar to the concept of ‘interpolymers’ devel-
oped for combining transparency and impact strength in the area of sterilizable packaging
[90,91] (the Borealis grade Borpact™ BC918CF is an example of such a grade). To demon-
strate the resulting phase structures, examples of a HECO with a C3-rich and C2-rich dis-
perse phase are presented in Figure 11. The difference between the homogeneous EPC
particles having a diffuse interface to the matrix in the case of (a) and the core-shell parti-
cles with a smoother interface to the matrix in the case of (b) is obvious. Transmission-
electron micrographs based on RuO4-contrasted specimens are used here, as this tech-
nique gives a very high resolution down to the lamellar level. A more complete series of
images and the related properties can be found in the work of Grein et al. [90]. Neverthe-
less, a study of the onset of phase separation in the case of ZN catalysts is still not available
in the literature. For single-site catalysts, this is expected within the range of 10 to 22 wt.%
of C2 in the xylene cold soluble (XCS) fraction, as also indicated by the appearance of a
second glass transition for the EPC [92], but the matrix design and molecular weight are
likely to play a role as well in practice.
(a)
(b)
Figure 11. Phase morphology overview of heterophasic copolymers (HECO with homopolymer ma-
trix) from the Borstar® PP process: (a) ‘interpolymer’ with C3-rich EPC (12 wt.% XCS with 24 wt.%
C2(XCS)) and (b) conventional type with C2-rich EPC (12 wt.% XCS with 57 wt.% C2(XCS)).
These rather complex designs, clearly beyond the ‘standard toolbox’ of HECO design
[93], bring positive effects in high-flow grades for bottle closure applications. The integra-
tion of hinge caps or tethering elements into soft-drink caps is required to avoid littering
and microplastic formation, and a trimodal copolymer design provides the necessary
combination of stiffness, impact strength and transparency for this application [94].
The availability of a second GPR further allows the production of the highest possible
EPC content value, which is necessary for automotive applications to obtain PP grades of
optimum stiffness–impact balance [95], resulting in grades like EF015AE of Borealis. More
recently, this has also been demonstrated to be possible with post-phthalate catalysts [16].
Maximizing the EPC content may also be combined with bimodal EPC design in order to
manage the manifold requirements of the application, simply because different targets
may require different components. The already mentioned variation of the ethylene con-
tent of the disperse phase, characterized by XCS, is a typical example here [96,97]. A high
amount of ethylene in the rubber phase triggers incompatibility and the development of
Figure 11.
Phase morphology overview of heterophasic copolymers (HECO with homopolymer
matrix) from the Borstar
®
PP process: (
a
) ‘interpolymer’ with C3-rich EPC (12 wt.% XCS with 24 wt.%
C2(XCS)) and (b) conventional type with C2-rich EPC (12 wt.% XCS with 57 wt.% C2(XCS)).
These rather complex designs, clearly beyond the ‘standard toolbox’ of HECO de-
sign [
93
], bring positive effects in high-flow grades for bottle closure applications. The
integration of hinge caps or tethering elements into soft-drink caps is required to avoid lit-
tering and microplastic formation, and a trimodal copolymer design provides the necessary
combination of stiffness, impact strength and transparency for this application [94].
The availability of a second GPR further allows the production of the highest possible
EPC content value, which is necessary for automotive applications to obtain PP grades of
optimum stiffness–impact balance [
95
], resulting in grades like EF015AE of Borealis. More
recently, this has also been demonstrated to be possible with post-phthalate catalysts [
16
].
Maximizing the EPC content may also be combined with bimodal EPC design in order to
manage the manifold requirements of the application, simply because different targets may
require different components. The already mentioned variation of the ethylene content of
the disperse phase, characterized by XCS, is a typical example here [
96
,
97
]. A high amount
of ethylene in the rubber phase triggers incompatibility and the development of high
interfacial tension with the PP matrix. Consequently, the disperse phase tends to reduce the
interface area per volume with the matrix, resulting in bigger particles. Moreover, due to
the statistics of ZN catalyst-based polymerization, the amount of crystalline EPC—tending
Polymers 2022,14, 4763 17 of 25
in structure towards a linear, low-density PE with C3 as a comonomer—increases likewise,
causing enhanced structural diversity inside the EPC particles and additionally affecting
the performance [
98
,
99
]. While this is largely beneficial for impact strength, also owing to
a decrease in the glass transition of the disperse phase, it may cause problems in surface
appearance and paint adhesion.
Heterophasic copolymers (HECOs) with a bimodal EPC design in terms of C2-content
and/or molecular weight, as expressed by the intrinsic viscosity IV(XCS), can be used
advantageously both as pure polymers for moulding applications [
100
] and in compounds
with mineral fillers for automotive applications [
101
]. Figure 12a shows one possible
morphology for such systems. When one of the EPC fractions exhibits sufficiently high
molecular weight, the polymer particles become more stable against deformation in flow
and aggregation, as observed for monomodal EPC systems [
90
,
96
]. Such reactor-based
systems should be distinguished from blends using high-density (HDPE) [
102
] or low-
density polyethylene (LDPE) [
103
] for modifying the overall disperse phase composition,
although such compositions are efficient in reducing stress whitening or improving the
balance between transparency and impact strength.
Polymers 2022, 14, x FOR PEER REVIEW 17 of 25
high interfacial tension with the PP matrix. Consequently, the disperse phase tends to re-
duce the interface area per volume with the matrix, resulting in bigger particles. Moreo-
ver, due to the statistics of ZN catalyst-based polymerization, the amount of crystalline
EPC—tending in structure towards a linear, low-density PE with C3 as a comonomer—
increases likewise, causing enhanced structural diversity inside the EPC particles and ad-
ditionally affecting the performance [98,99]. While this is largely beneficial for impact
strength, also owing to a decrease in the glass transition of the disperse phase, it may cause
problems in surface appearance and paint adhesion.
Heterophasic copolymers (HECOs) with a bimodal EPC design in terms of C2-con-
tent and/or molecular weight, as expressed by the intrinsic viscosity IV(XCS), can be used
advantageously both as pure polymers for moulding applications [100] and in com-
pounds with mineral fillers for automotive applications [101]. Figure 12a shows one pos-
sible morphology for such systems. When one of the EPC fractions exhibits sufficiently
high molecular weight, the polymer particles become more stable against deformation in
flow and aggregation, as observed for monomodal EPC systems [90,96]. Such reactor-
based systems should be distinguished from blends using high-density (HDPE) [102] or
low-density polyethylene (LDPE) [103] for modifying the overall disperse phase compo-
sition, although such compositions are efficient in reducing stress whitening or improving
the balance between transparency and impact strength.
(a)
(b)
Figure 12. Phase morphology overview of special HECOs from the Borstar® PP process: (a) reactor-
based thermoplastic polyolefin (RTPO) with bimodal EPC having significant crystalline fraction
(XCS: 32 wt.-%) and (b) RAHECO with random propylene copolymer matrix (4 wt.-% C2) and 15
wt.% EPC (total XCS: 22 wt.-%).
Aside from designing bimodal EPCs and/or maximizing EPC-content, a Borstar®
four-reactor setup also enables designing HECOs with a trimodal matrix. The matrix de-
livering the stiffness of such materials is built to contain fractions of high, medium and
low molecular weight.
The beneficial effect of the high-molecular-weight part is very likely based on the
orientation of the long chains leading to more flow-induced orientation and higher stiff-
ness as well as on an increase in the nucleation density of the material. The low-molecular-
weight fraction still allows good processing of these polymers, whereas the medium frac-
tion is needed for compatibilization reasons (see Figure 13a). Applying this polymer de-
sign, the stiffness can be further increased, and materials of extraordinary rigidity are ob-
tained (see Figure 13b). Certainly, as in all industrial polymerization processes, the design
freedom is also limited for trimodal matrices from a process point of view by the final
MFR of the product, with process limitations like flash points between reactors and cata-
lyst activity. Nevertheless, currently available fourth+-generation ZN catalysts combined
with a Borstar® PP four-reactor setup allow for the production of HECOs with trimodal
Figure 12. Phase morphology overview of special HECOs from the Borstar®PP process: (a) reactor-
based thermoplastic polyolefin (RTPO) with bimodal EPC having significant crystalline fraction (XCS:
32 wt.-%) and (
b
) RAHECO with random propylene copolymer matrix (4 wt.-% C2) and 15 wt.% EPC
(total XCS: 22 wt.-%).
Aside from designing bimodal EPCs and/or maximizing EPC-content, a Borstar
®
four-reactor setup also enables designing HECOs with a trimodal matrix. The matrix
delivering the stiffness of such materials is built to contain fractions of high, medium and
low molecular weight.
The beneficial effect of the high-molecular-weight part is very likely based on the
orientation of the long chains leading to more flow-induced orientation and higher stiffness
as well as on an increase in the nucleation density of the material. The low-molecular-
weight fraction still allows good processing of these polymers, whereas the medium
fraction is needed for compatibilization reasons (see Figure 13a). Applying this polymer
design, the stiffness can be further increased, and materials of extraordinary rigidity are
obtained (see Figure 13b). Certainly, as in all industrial polymerization processes, the design
freedom is also limited for trimodal matrices from a process point of view by the final
MFR of the product, with process limitations like flash points between reactors and catalyst
activity. Nevertheless, currently available fourth+-generation ZN catalysts combined with
a Borstar
®
PP four-reactor setup allow for the production of HECOs with trimodal matrices
for materials suitable for pipe applications to thin-wall injection moulding [
104
–
106
], like
BH381MO of Borealis.
Polymers 2022,14, 4763 18 of 25
Polymers 2022, 14, x FOR PEER REVIEW 18 of 25
matrices for materials suitable for pipe applications to thin-wall injection moulding [104–
106], like BH381MO of Borealis.
(a)
(b)
Figure 13. Schematic representation of the MWD of a trimodal PP homopolymer and flexural mod-
ulus over MFR for selected trimodal and conventional PP homopolymers of medium flowability
(data from [105]).
Combining the high-stiffness trimodal matrices with a well-designed EPC, copoly-
mers with an outstanding stiffness–impact balance can be achieved (see Figure 14). Not
only is the stiffness superior due to the high-stiffness matrix, but the impact performance
is also significantly improved compared to benchmarks of similar EPC content levels. It is
suggested the trimodal matrix design fosters—also at relatively high end flowability and
therefore an unfavourable viscosity ratio between the disperse and matrix phases
(ηEPC/ηmatrix)—a morphology with fine and well dispersed EPC particles due to the
presence of the high-molecular-weight fraction.
(a)
(b)
Figure 14. Stiffness–impact balance of HECOs with trimodal matrix compared to commercial refer-
ences with unimodal matrix with NIS measured at different temperatures: (a) 23 °C and (b) −20 °C
(data from [106] and commercial grades).
Apart from the EPC design discussed earlier, the matrix can also be modified, sub-
stituting the PP homopolymer with a RACO with C2 or higher α-olefins. This leads to
random-heterophasic copolymers (RAHECOs), the property combination of which results
from the presence of the random propylene copolymer (RACO) matrix and the dispersed
elastomeric phase (EPC). An early example of such grades is the still-existing SD233CF of
Borealis.
The properties of a RAHECO can be varied by different parameters, many of which
are intrinsically linked to each other. In an early development phase [107], the overall
2 4 6 8 20 40110
1800
1900
2000
2100
2200
2300
2400
2500
2600 PP-homopolymer, nucleated with:
trimodal matrix
standard (uni or bimodal) matrix
flexural modulus / MPa
melt flow rate MFR / g 10min-1
1550 1600 1650 1700 1750 1800
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
HECOs with trimodal matrix
HECOs with standard (uni or bimodal) matrix
notched Charpy impact strength NIS (23 °C)
tensile modulus E / MPa
1550 1600 1650 1700 1750 1800
1.5
2.0
2.5
3.0
3.5
HECOs with trimodal matrix
HECOs with standard (uni or bimodal) matrix
notched Charpy impact strength NIS (-20 °C)
tensile modulus E / MPa
Figure 13.
(
a
) Schematic representation of the MWD of a trimodal PP homopolymer (curves with
different colour represent the different fractions) and (
b
) flexural modulus over MFR for selected
tri-modal and conventional PP homopolymers of medium flowability (data from [105]).
Combining the high-stiffness trimodal matrices with a well-designed EPC, copolymers
with an outstanding stiffness–impact balance can be achieved (see Figure 14). Not only
is the stiffness superior due to the high-stiffness matrix, but the impact performance is
also significantly improved compared to benchmarks of similar EPC content levels. It
is suggested the trimodal matrix design fosters—also at relatively high end flowability
and therefore an unfavourable viscosity ratio between the disperse and matrix phases
(
η
EPC/
η
matrix)—a morphology with fine and well dispersed EPC particles due to the
presence of the high-molecular-weight fraction.
Polymers 2022, 14, x FOR PEER REVIEW 18 of 25
matrices for materials suitable for pipe applications to thin-wall injection moulding [104–
106], like BH381MO of Borealis.
(a)
(b)
Figure 13. Schematic representation of the MWD of a trimodal PP homopolymer and flexural mod-
ulus over MFR for selected trimodal and conventional PP homopolymers of medium flowability
(data from [105]).
Combining the high-stiffness trimodal matrices with a well-designed EPC, copoly-
mers with an outstanding stiffness–impact balance can be achieved (see Figure 14). Not
only is the stiffness superior due to the high-stiffness matrix, but the impact performance
is also significantly improved compared to benchmarks of similar EPC content levels. It is
suggested the trimodal matrix design fosters—also at relatively high end flowability and
therefore an unfavourable viscosity ratio between the disperse and matrix phases
(ηEPC/ηmatrix)—a morphology with fine and well dispersed EPC particles due to the
presence of the high-molecular-weight fraction.
(a)
(b)
Figure 14. Stiffness–impact balance of HECOs with trimodal matrix compared to commercial refer-
ences with unimodal matrix with NIS measured at different temperatures: (a) 23 °C and (b) −20 °C
(data from [106] and commercial grades).
Apart from the EPC design discussed earlier, the matrix can also be modified, sub-
stituting the PP homopolymer with a RACO with C2 or higher α-olefins. This leads to
random-heterophasic copolymers (RAHECOs), the property combination of which results
from the presence of the random propylene copolymer (RACO) matrix and the dispersed
elastomeric phase (EPC). An early example of such grades is the still-existing SD233CF of
Borealis.
The properties of a RAHECO can be varied by different parameters, many of which
are intrinsically linked to each other. In an early development phase [107], the overall
2 4 6 8 20 40110
1800
1900
2000
2100
2200
2300
2400
2500
2600 PP-homopolymer, nucleated with:
trimodal matrix
standard (uni or bimodal) matrix
flexural modulus / MPa
melt flow rate MFR / g 10min-1
1550 1600 1650 1700 1750 1800
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
HECOs with trimodal matrix
HECOs with standard (uni or bimodal) matrix
notched Charpy impact strength NIS (23 °C)
tensile modulus E / MPa
1550 1600 1650 1700 1750 1800
1.5
2.0
2.5
3.0
3.5
HECOs with trimodal matrix
HECOs with standard (uni or bimodal) matrix
notched Charpy impact strength NIS (-20 °C)
tensile modulus E / MPa
Figure 14.
Stiffness–impact balance of HECOs with trimodal matrix compared to commercial refer-
ences with unimodal matrix with NIS measured at different temperatures: (
a
) 23
◦
C and (
b
)
−
20
◦
C
(data from [106] and commercial grades).
Apart from the EPC design discussed earlier, the matrix can also be modified, sub-
stituting the PP homopolymer with a RACO with C2 or higher
α
-olefins. This leads to
random-heterophasic copolymers (RAHECOs), the property combination of which results
from the presence of the random propylene copolymer (RACO) matrix and the dispersed
elastomeric phase (EPC). An early example of such grades is the still-existing SD233CF
of Borealis.
The properties of a RAHECO can be varied by different parameters, many of which
are intrinsically linked to each other. In an early development phase [
107
], the overall
modulus was found to result mostly from the comonomer content of the matrix phase and
the EPR content (see Figure 15). Optimizing the toughness and optical performance of such
compositions, however, requires adapting the phases’ compositions and their respective
Polymers 2022,14, 4763 19 of 25
molecular weights, reaching a morphology, as shown in Figure 12b. Optimum transparency
cannot be achieved with a fully amorphous EPC fraction, as shown in Figure 11a, but rather
with a density match between the matrix and disperse phase, which also minimizes the
difference in the refractive index and can equally well be reached with post-phthalate ZN
catalysts [108].
Polymers 2022, 14, x FOR PEER REVIEW 19 of 25
modulus was found to result mostly from the comonomer content of the matrix phase and
the EPR content (see Figure 15). Optimizing the toughness and optical performance of
such compositions, however, requires adapting the phases’ compositions and their re-
spective molecular weights, reaching a morphology, as shown in Figure 12b. Optimum
transparency cannot be achieved with a fully amorphous EPC fraction, as shown in Figure
11a, but rather with a density match between the matrix and disperse phase, which also
minimizes the difference in the refractive index and can equally well be reached with post-
phthalate ZN catalysts [108].
Figure 15. Flexural modulus of random-heterophasic ethylene–propylene copolymers (RAHECOs)
with different C2 contents of the matrix (♢ 4 wt.%, □ 6 wt.%, + 8 wt.%) and EPC content (C2 of EPC
~36 wt.% C2, data from [107]).
Fully ductile behaviour at ambient temperature can be achieved at EPC contents of
15–20 wt.% for a pure RACO matrix, which is related to the transition from crazing to
shear yielding below a certain EPC particle distance [109]. In detail, however, this transi-
tion will also be affected by the matrix stiffness and the molecular weight of the EPC phase
[110]. For producing even softer PP-based materials, targeted at replacing LDPE or even
plasticized poly(vinyl chloride), a careful look at the powder morphology was found to
be necessary. The essential morphology development of HECOs was described already in
the work of Cecchin et al. [111] as the gradual filling of the crystalline and porous particles
from the matrix stage(s) of the process, and it was later studied by the group of Kosek in
more detail both by modelling [112] and by advanced structural characterization methods
[113]. Under the conditions of the gas-phase reactor and the following ‘dry end’ part of
the polymerization plant, i.e., at temperatures of 70–85 °C, the EPC fraction is highly mo-
bile and potentially also plasticized by the monomer mixture, making a fine dispersion
inside the particle without significant surface fractions decisive for maintaining powder
flowability and plant operability. A sufficient porosity of the matrix particle and the ab-
sence of particle breakup are clearly necessary for this, especially at high EPC contents.
Considering this, the Borstar® PP process with a four-reactor configuration allows the
stable production of various ‘Soft-PP’ grades with a bimodal RACO matrix and an EPC
content up to 45 wt.% [114]. The mechanical profile of these and earlier-developed
RAHECO grades is compared to those of RACOs and HECOs in Figure 16. ‘Soft-PP’
grades for medical applications like infusion pouches or blow–fill–seal containers, com-
bining toughness and transparency with a ‘collapsing’ emptying performance and suffi-
cient stability in steam sterilization, are so far the endpoint of development [115], exem-
plified by Bormed™ SB815MO.
Figure 15.
Flexural modulus of random-heterophasic ethylene–propylene copolymers (RAHECOs)
with different C2 contents of the matrix (
♦
4 wt.%,
6 wt.%, + 8 wt.%) and EPC content (C2 of EPC
~36 wt.% C2, data from [107]).
Fully ductile behaviour at ambient temperature can be achieved at EPC contents
of 15–20 wt.% for a pure RACO matrix, which is related to the transition from crazing
to shear yielding below a certain EPC particle distance [
109
]. In detail, however, this
transition will also be affected by the matrix stiffness and the molecular weight of the EPC
phase [
110
]. For producing even softer PP-based materials, targeted at replacing LDPE
or even plasticized poly(vinyl chloride), a careful look at the powder morphology was
found to be necessary. The essential morphology development of HECOs was described
already in the work of Cecchin et al. [
111
] as the gradual filling of the crystalline and porous
particles from the matrix stage(s) of the process, and it was later studied by the group of
Kosek in more detail both by modelling [
112
] and by advanced structural characterization
methods [
113
]. Under the conditions of the gas-phase reactor and the following ‘dry end’
part of the polymerization plant, i.e., at temperatures of 70–85
◦
C, the EPC fraction is highly
mobile and potentially also plasticized by the monomer mixture, making a fine dispersion
inside the particle without significant surface fractions decisive for maintaining powder
flowability and plant operability. A sufficient porosity of the matrix particle and the absence
of particle breakup are clearly necessary for this, especially at high EPC contents.
Considering this, the Borstar
®
PP process with a four-reactor configuration allows
the stable production of various ‘Soft-PP’ grades with a bimodal RACO matrix and an
EPC content up to 45 wt.% [
114
]. The mechanical profile of these and earlier-developed
RAHECO grades is compared to those of RACOs and HECOs in Figure 16. ‘Soft-PP’ grades
for medical applications like infusion pouches or blow–fill–seal containers, combining
toughness and transparency with a ‘collapsing’ emptying performance and sufficient
stability in steam sterilization, are so far the endpoint of development [
115
], exemplified by
Bormed™ SB815MO.
Polymers 2022,14, 4763 20 of 25
Polymers 2022, 14, x FOR PEER REVIEW 20 of 25
Figure 16. Stiffness/impact balance of different standard and advanced PP copolymers (points cor-
respond to commercial or developmental grades, haze range on 1 mm injection-moulded plaques
indicated for each material class).
5. Conclusions and Outlook
This review presents modern polypropylene production as an integrated concept be-
tween the process technology, catalyst system and polymer design for the example of the
Borstar® PP process. Multiple interactions exist between the corners of this ‘performance
triangle’, which have been highlighted in the previous chapters, like between comonomer
response and melting point range or between EPC capacity and impact strength. The con-
stantly changing economic and general environment, however, requires constant adapta-
tion of such processes, especially regarding the considerations of sustainability and reduc-
ing the carbon footprint of production.
At present, an ongoing challenge is the integration of single-site catalyst (SSC) sys-
tems into the Borstar® PP process. While this catalyst type clearly offers advantages in
terms of narrow MWD and low emissions [116,117], in the incorporation of higher α-ole-
fins like 1-butene or 1-hexene [78,118] and also in the design of advanced multiphase sys-
tems [92,119], it also presents new challenges regarding cost and plant operation. The
coming years will tell whether SSC-PP can become a success story, as in the case of PE
[120].
Ultimately even more important will be the need for incorporating monomers from
sustainable and largely carbon-neutral sources like renewable-based [121] and chemical-
recycling processes [122,123]. This is not optional for the polymer industry—it is a clear
requirement, despite the already rather low carbon footprint of PP in comparison to other
polymers. Monomers from such sources present new challenges in terms of purity with
respect to contaminant type, which are likely to require adaptations in catalyst systems
and monomer purification.
Reducing CO2 and other greenhouse gas emissions while generally saving energy in
all stages of the process, but already starting at hydrocarbon cracking, is one more aspect
of this challenge. Improved integration of energy streams, including external use of cool-
ing water for district heating purposes, but also better process control and extended use
of process digitalization [124] are bound to play an elementary role here.
Figure 16.
Stiffness/impact balance of different standard and advanced PP copolymers (points
correspond to commercial or developmental grades, haze range on 1 mm injection-moulded plaques
indicated for each material class).
5. Conclusions and Outlook
This review presents modern polypropylene production as an integrated concept
between the process technology, catalyst system and polymer design for the example of the
Borstar
®
PP process. Multiple interactions exist between the corners of this ‘performance
triangle’, which have been highlighted in the previous chapters, like between comonomer
response and melting point range or between EPC capacity and impact strength. The
constantly changing economic and general environment, however, requires constant adap-
tation of such processes, especially regarding the considerations of sustainability and
reducing the carbon footprint of production.
At present, an ongoing challenge is the integration of single-site catalyst (SSC) systems
into the Borstar
®
PP process. While this catalyst type clearly offers advantages in terms
of narrow MWD and low emissions [
116
,
117
], in the incorporation of higher
α
-olefins
like
1-butene
or 1-hexene [
78
,
118
] and also in the design of advanced multiphase sys-
tems
[92,119]
, it also presents new challenges regarding cost and plant operation. The
coming years will tell whether SSC-PP can become a success story, as in the case of PE [
120
].
Ultimately even more important will be the need for incorporating monomers from
sustainable and largely carbon-neutral sources like renewable-based [
121
] and chemical-
recycling processes [
122
,
123
]. This is not optional for the polymer industry—it is a clear
requirement, despite the already rather low carbon footprint of PP in comparison to other
polymers. Monomers from such sources present new challenges in terms of purity with
respect to contaminant type, which are likely to require adaptations in catalyst systems and
monomer purification.
Reducing CO2and other greenhouse gas emissions while generally saving energy in
all stages of the process, but already starting at hydrocarbon cracking, is one more aspect of
this challenge. Improved integration of energy streams, including external use of cooling
water for district heating purposes, but also better process control and extended use of
process digitalization [124] are bound to play an elementary role here.
Since its original development, PP has solved a large number of technical and societal
problems for humanity. It will be the common task of science and industry to further im-
prove the balance between the beneficial role and the negative side effects of its production.
Polymers 2022,14, 4763 21 of 25
Author Contributions:
Conceptualization, M.G., D.J. and V.T.; writing—original draft preparation,
M.F.B., M.G., P.D., V.K., D.M., P.S. and C.T.; writing—review and editing, M.G., D.J. and V.T. All
authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Data Availability Statement:
Further data can be made available by the corresponding author upon
personal request.
Acknowledgments:
We thank Esa Kokko, Pascal Castro (Borealis Polymers Oy, Muovintie 19, 06850
Porvoo, Finland), Petar Doshev (Borealis Polyolefine GmbH, Linz, Austria) and Bernard Lotz (CNRS
Strasbourg, France) for helpful discussion and the provision of important references.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Natta, G.; Pasquon, I.; Zambelli, A.; Gatti, G. Highly stereospecific catalytic systems for the polymerization of
α
-olefins to isotactic
polymers. J. Polym. Sci. 1961,51, 387–398. [CrossRef]
2.
Natta, G.; Corradini, P.; Allegra, G. The different crystalline modifications of TiCl3, a catalyst component for the polymerization
of α-olefins. I: α-, β-, γ-TiCl3. II: δ-TiCl3.J. Polym. Sci. 1961,51, 399–410. [CrossRef]
3.
Jubinville, D.; Esmizadeh, E.; Saikrishnan, S.; Tzoganakis, C.; Mekonnen, T. A comprehensive review of global production
and recycling methods of polyolefin (PO) based products and their post-recycling applications. Sustain. Mater. Technol.
2020
,
25, e00188. [CrossRef]
4. Moore, E.P. The Rebirth of Polypropylene: Supported Catalysts; Hanser: Munich, Germany, 1998; ISBN 3-446-19587-4.
5. Luciani, L.; Barbe, P.C.; Kashiwa, N.; Toyoda, A. Polymerisation Catalyst. German Patent DE 2643143 A1, 6 February 1977.
6.
Giannini, U.; Albizzati, E.; Parodi, S. Components of Catalysts Useful for the Polymerization of
α
-Olefins, and Catalysts Prepared
Therefrom. German Patent DE 2735672 A1, 16 February 1978.
7.
Albizzati, E.; Galimberti, M.; Giannini, U.; Morini, G. The Chemistry of Magnesium Chloride Supported Catalysts for Polypropy-
lene. Die Makromol. Chem.—Macromol. Symp. 1991,48, 223–238. [CrossRef]
8.
Galli, P.; Vecellio, G. Polyolefins: The Most Promising Large-Volume Materials for the 21st Century. J. Polym. Sci. A Polym. Chem.
2004,42, 396–415. [CrossRef]
9.
Chadwick, J.C. Polyolefins—Catalyst and Process Innovations and their Impact on Polymer Properties. Macromol. React. Eng.
2009,3, 428–432. [CrossRef]
10. Soares, J.B.P.; McKenna, T.F.L. Polyolefin Reaction Engineering; Wiley-VCH: Weinheim, Germany, 2012; ISBN 9783527317103.
11.
Mei, G.; Herben, P.; Cagnani, C.; Mazzucco, A. The Spherizone Process: A New PP Manufacturing Platform. Macromol. Symp.
2006,245, 677–680. [CrossRef]
12. Martino-Andrade, A.J.; Chahoud, I. Reproductive toxicity of phthalate esters. Mol. Nutr. Food Res. 2010,1, 148–157. [CrossRef]
13.
Chadwick, J.C.; Morini, G.; Balbontin, G.; Camurati, I.; Heere, J.J.R.; Mingozzi, I.; Testoni, F. Effects of Internal and External
Donors on the Regio- and Stereoselectivity of Active Species in MgCl
2
-Supported Catalysts for Propene Polymerization. Macromol.
Chem. Phys. 2001,202, 1995–2002. [CrossRef]
14.
Zhang, H.-X.; Lee, Y.-J.; Park, J.-R.; Lee, D.-H.; Yoon, K.-B. Control of Molecular Weight Distribution for Polypropylene Obtained
by Commercial Ziegler-Natta Catalyst: Effect of Electron Donor. Macromol. Res. 2011,19, 622–628. [CrossRef]
15.
Severn, J.R. Recent Developments in Supported Polyolefin Catalysts: A Review. In Multimodal Polymers with Supported Catalysts.
Design and Production; Albunia, A.R., Prades, F., Jeremic, D., Eds.; Springer Nature: Cham, Germany, 2019; pp. 1–53.
16.
Pellecchia, R.; Shutov, P.; Wang, J.; Gahleitner, M. Copolymer Structure and Performance Consequences of High-Impact Ethylene–
Propylene Copolymers Based on a Ziegler–Natta Catalyst with Novel Internal Donor. Macromol. React. Eng.
2020
,14, 2000022.
[CrossRef]
17.
Kivelä, J.; Grande, H.; Korvenoja, T. Borstar Polypropylene technology. In Handbook of Petrochemicals Production Processes; Meyers,
R.A., Ed.; McGraw-Hill: New York, NY, USA, 2005.
18.
Elovainio, E.; Vuorikari, M.; Korhonen, E.; Leskinen, P. Process for Polymerizing Olefins in the Presence of an Olefin Polymeriza-
tion Catalyst. PCT Patent WO 2006063771 A1, 22 June 2006.
19.
Soleimani, A.; Aigner, A.; Touloupidis, V. Single Particle Growth, Fragmentation and Morphology Modelling: A DEM Approach.
Macromol. React. Eng. 2022,16, 202200015. [CrossRef]
20.
Leskinen, P.; Tuominen, O. Process for the Production of Polypropylene Copolymers Using a Prepolymerized Catalyst. PCT
Patent WO 2008151794 A1, 18 December 2008.
21.
Leskinen, P.; Hakola, S.; Alastalo, K.; Nyfors, K. Loop Reactor Providing an Advanced Production Split. PCT Patent
WO 2013092342 A1, 27 June 2013.
22.
Harlin, A.; Alastalo, K.; Korhonen, E.; Kivelä, J. Process for Preparing Propylene Homopolymers and Copolymers. PCT Patent
WO 9858975 A1, 30 December 1998.
Polymers 2022,14, 4763 22 of 25
23. Heino, T.; Karvinen, S. Method and Apparatus for producing Polymers. PCT Patent WO 2005087361 A1, 22 September 2005.
24.
Kokko, T.; Elovainio, E.; Kivelä, J.; Nyfors, K.; Hagane, K. Process for Producing Polyolefins. PCT Patent WO 2011066892 A1,
9 June 2011.
25.
Nyfors, K.; Elovainio, E.; Kanellopoulos, V.; Weickert, G.; Prinsen, E.-J. Process and Apparatus for Removing Polymer Material
from a Gas-Solids Olefin Polymerization Reactor. PCT Patent WO 2018233999 A1, 27 December 2018.
26.
Bergstra, M.F.; Eriksson, E.J.G. Impurities in polyolefin production. In Proceedings of the International Conference on the Reaction
Engineering of Polyolefins, Montreal, QC, Canada, 23–27 June 2008.
27. Hillestad, M.; Andersen, K.S. Model Predictive Control for Grade Transitions of a Polypropylene Reactor. In Proceedings of the
ESCAPE’4, Dublin, Ireland, 28–30 March 1994.
28.
Haugwitz, S.; Wilsher, E.; Hofsten, K.; Andersen, K.S.; Moen, Ø.; Glemmestad, B. Commissioning of Nonlinear Model Predictive
Controllers to a New-Polypropylene Plant. In Proceedings of the Reglermötet, Stockholm, Sweden, 4–5 June 2008.
29. Kashiwa, N. Super Active Catalyst for Olefin Polymerization. Polym. J. 1980,12, 603–608. [CrossRef]
30.
Chien, J.C.W.; Vizzini, J.C. Superactive and Stereospecific Catalysts. IV. Influence of Structure of Esters on MgCl2 Supported
Olefin Polymerization Catalysts. J. Polym. Sci. A Polym. Chem. 1990,28, 273–284. [CrossRef]
31.
Seppälä, J.V.; Härkönen, M.; Luciani, L. Effect of the structure of external alkoxysilane donors on the polymerization of propene
with high activity Ziegler-Natta catalysts. Die Makromol. Chem. 1989,190, 2535–2550. [CrossRef]
32.
Härkönen, M.; Seppälä, J.V.; Salminen, H. External Silane Donors in Ziegler–Natta Catalysis. A Three-Site Model Analysis of
Effects on Catalyst Active Sites. Polym. J. 1995,27, 256–261. [CrossRef]
33.
Garoff, T.; Eloranta, K. A Method for the Preparation of a Carrier for a Ziegler/Natta Polymerization Catalyst, a Carrier Prepared
Using the Method, and its Use in a Polymerization Catalyst System. EP Patent EP 0364098 A2, 18 April 1990.
34.
Iiskola, E.; Koskinen, J. Procedure for Manufacturing Catalyst Components for Polymerizing Olefines. PCT Patent WO 8707620
A1, 17 December 1987.
35.
Garoff, T.; Leinonen, T.; Iiskola, E. A Method for the Modification of Catalysts Intended for the Polymerization of Olefins. EP
Patent EP 0491566 A2, 24 June 1992.
36.
Garoff, T.; Leinonen, T.; Iiskola, E. A Large-Pore Polyolefin, a Method for its Production and a Procatalyst Containing a
Transesterification Product of a Lower Alcohol. EP Patent EP 0586389 A1, 16 March 1994.
37.
Garoff, T.; Leinonen, T.; Iiskola, E. Coarse-Grained Polyolefin, Production Thereof and a Procatalyst Containing a Transesterifica-
tion Product between a Lower Alcohol and Dioctylphthalate Used Therefore. EP Patent EP 0586390 A1, 16 March 1994.
38.
Garoff, T.; Virkkunen, V.; Jäskeläinen, P.; Vestberg, T. A qualitative model for polymerisation of propylene with a MgCl
2
-supported
TiCl4Ziegler–Natta catalyst. Eur. Polym. J. 2003,39, 1679–1685. [CrossRef]
39.
Gahleitner, M.; Bachner, C.; Ratajski, E.; Rohaczek, G.; Neißl, W. Effects of the Catalyst System on the Crystallization of
Polypropylene. J. Appl. Polym. Sci. 1999,73, 2507–2515. [CrossRef]
40. Cancelas, A.J.; Yang, L.; Girod, R.; de Heer, J.; Kleppinger, R.; Delsman, E.; Wang, J.; Gahleitner, M.; Monteil, V.; McKenna, T.F.L.
The Effect of Reactor Conditions on High-Impact Polypropylene Properties and Gas Phase Polymerization Kinetics. Macromol.
React. Eng. 2018,12, 1700063. [CrossRef]
41.
Rönkkö, H.-L.; Knuutila, H.; Denifl, P.; Leinonen, T.; Venäläinen, T. Structural studies on a solid self-supported Ziegler–Natta-type
catalyst for propylene polymerization. J. Mol. Catal. A Chem. 2007,278, 127–134. [CrossRef]
42.
Leinonen, T.; Denifl, P. Catalyst Component Comprising Magnesium, Titanium, a Halogen and an Electron Donor, Its Preparation
and Use. PCT Patent WO 0238631 A1, 16 May 2002.
43. Leinonen, T.; Denifl, P. Preparation of Olefin Polymerization Catalyst Component. EP Patent EP 1270610 A1, 2 January 2003.
44.
Abboud, M.; Denifl, P.; Reichert, K.-H. Study of the morphology and kinetics of novel Ziegler–Natta catalysts for propylene
polymerization. J. Appl. Polym. Sci. 2005,98, 2191–2200. [CrossRef]
45.
Rönkkö, H.-L.; Korpela, T.; Knuutila, H.; Pakkanen, T.T.; Denifl, P.; Leinonen, T.; Kemell, M.; Leskelä, M. Particle growth and
fragmentation of solid self-supported Ziegler-Natta-type catalysts in propylene polymerization. J. Mol. Catal. A Chem.
2009
,
309, 40–49. [CrossRef]
46.
Vestberg, T.; Denifl, P.; Parkinson, M.; Wilén, C.-E. Effects of External Donors and Hydrogen Concentration on Oligomer
Formation and Chain End Distribution in Propylene Polymerization with Ziegler-Natta Catalysts. J. Polym. Sci. A Polym. Chem.
2010,48, 351–358. [CrossRef]
47.
Vestberg, T.; Parkinson, M.; Fonseca, I.; Wilén, C.-E. Poly(propylene-co-ethylene) Produced with a Conventional and a Self-
Supported Ziegler–Natta Catalyst: Effect of Ethylene and Hydrogen Concentration on Activity and Polymer Structure. J. Appl.
Polym. Sci. 2012,124, 4889–4896. [CrossRef]
48.
Cecchin, G.; Morini, G.; Pelliconi, A. Polypropene Product Innovation by Reactor Granule Technology. Macromol. Symp.
2001
,
173, 195–210. [CrossRef]
49.
Abboud, M.; Denifl, P.; Reichert, K.-H. Fragmentation of Ziegler-Natta Catalyst Particles During Propylene Polymerization.
Macromol. Mater. Eng. 2005,290, 558–564. [CrossRef]
50.
Grein, C.; Schedenig, T. Polyolefin Compositions Having Improved Optical and Mechanical Properties. EP Patent EP 2030996 A1,
4 March 2009.
51.
Denifl, P.; Leinonen, T.; Haikarainen, A.; Vestberg, T. Process for the Manufacture of Heterophasic Propylene Copolymer. EP Patent
EP 2065404 A1, 3 June 2009.
Polymers 2022,14, 4763 23 of 25
52.
Haikarainen, A.; Denifl, P.; Leinonen, T. Preparation of Precipitated ZN PP Catalysts with Internal Pore Structure Using
Nanoparticles. PCT Patent WO 2010127997 A2, 11 November 2010.
53.
Vestberg, T.; Denifl, P.; Wilén, C.-E. Increased Rubber Content in High-Impact Polypropylene via a Sirius Ziegler-Natta Catalyst
Containing Nanoparticles. J. Polym. Sci. A Polym. Chem. 2013,51, 2040–2048. [CrossRef]
54.
Iiskola, E.; Mills, K.; Garoff, T.; Leinonen, T. A Procatalyst Composition Containing Substituted Maleic or Fumaric Acid Esters as
an Electron Donor for Olefin Polymerization. PCT Patent WO 9307182 A1, 15 April 1993.
55.
David, R.M.; Gans, G. Summary of mammalian toxicology and health effects of phthalate esters. In The Handbook of Environmental
Chemistry Vol. 3, Part Q; Staples, C., Ed.; Springer Nature: Cham, Switzerland, 2003; pp. 299–316.
56.
Chadwick, J.C.; van der Burgt, F.P.T.J.; Rastogi, S.; Busico, V.; Cipullo, V.; Talarico, G.; Heere, J.J.R. Influence of Ziegler-
Natta Catalyst Regioselectivity on Polypropylene Molecular Weight Distribution and Rheological and Crystallization Behavior.
Macromolecules 2004,37, 9722–9727. [CrossRef]
57.
Yin, X.; Qin, Y.; Dong, J.-Y. Investigation of Chain Microstructure of Polypropylene Polymerized by Ziegler–Natta Catalysts
with Diester and Diether Compound as Internal Donor via Hydrogen Chain Transfer. Ind. Eng. Chem. Res.
2020
,59, 1836–1844.
[CrossRef]
58. Haikarainen, A.; Denifl, P.; Leinonen, P. Catalyst Component. EP Patent EP 2415790 A1, 8 February 2012.
59.
Wang, J.; Lilja, J.; Horill, T.; Gahleitner, M.; Denifl, P.; Leinonen, T. Low Emission Propylene Homopolymer with High Melt Flow.
EP Patent EP 3071606 A1, 28 September 2016.
60. Wang, J.; Lilja, J.; Gahleitner, M. Propylene Copolymer for Thin-Wall Packaging. EP Patent EP 2999721 A1, 30 March 2016.
61.
Menyhárd, A.; Gahleitner, M.; Varga, J.; Bernreitner, K.; Jääskeläinen, P.; Øysæd, H.; Pukánszky, B. The influence of nucleus
density on optical properties in nucleated isotactic polypropylene. Eur. Polym. J. 2009,45, 3138–3148. [CrossRef]
62.
Alczar, D.; Ruan, J.; Thierry, A.; Lotz, B. Structural Matching between the Polymeric Nucleating Agent Isotactic Poly(vinylcyclo-
hexane) and Isotactic Polypropylene. Macromolecules 2006,39, 2832–2840. [CrossRef]
63. Abe, A.; Hama, T. Structure and Properties of Poly(Vinyl cyclohexane). Polym. Lett. 1969,7, 427–435. [CrossRef]
64.
Ishihara, N.; Kuramoto, M. Vinylcyclohexane-Based Polymers and Process for Production Thereof. EP Patent EP 0322731 A2,
5 July 1989.
65.
Wege, V.; Dujardin, R.; Chen, Y.; Rechner, J.; Bruder, F.-K. Amorphous Vinyl Cyclohexane Polymers. US Patent US 6365694 B1,
2 April 2002.
66.
Shiga, A.; Kakugo, M.; Kojima, J.; Wakatsuki, K. Crystalline Propylene Polymer Composition. EP Patent EP 0151883 A1,
21 August 1985.
67. Kakugo, M. Recent progress in polyolefin materials and processing. Macromol. Symp. 2001,174, 295–300. [CrossRef]
68.
Karbasi, A.; Leskinen, P.; Jääskeläinen, P.; Malm, B.; Pitkänen, P.; Härkönen, M.; Haugen, J. Process for Preparing Polypropylene.
PCT Patent WO 9924478 A1, 20 May 1999.
69.
Karbasi, A.; Leskinen, P.; Jääskeläinen, P.; Malm, B.; Pitkänen, P.; Härkönen, M.; Haugen, J. Novel Propylene Polymers and
Products Thereof. PCT Patent WO 9924779 A1, 20 May 1999.
70.
Lee, D.-H.; Yoon, K.-B. Effect of polycyclopentene on Crystallization of Isotactic Polypropylene. J. Appl. Polym. Sci.
1994
,
54, 1507–1511. [CrossRef]
71.
De Rosa, C.; Auriemma, F.; Tarallo, O.; Di Girolamo, R.; Troisi, E.M.; Esposito, S.; Liguori, D.; Piemontesi, F.; Vitale, G.; Morini, G.
Tailoring the properties of polypropylene in the polymerization reactor using polymeric nucleating agents as prepolymers on the
Ziegler-Natta catalyst granule. Polym. Chem. 2017,8, 655–660. [CrossRef]
72.
Mileva, D.; Gahleitner, M.; Shutov, P.; Vestberg, T.; Wang, J.; Androsch, R. Effect of Supercooling on Crystal Structure of Nucleated
Isotactic Polypropylene. Thermochim. Acta 2019,677, 194–197. [CrossRef]
73.
Kirchberger, M.; Jääskeläinen, P.; Gahleitner, M. Non-Oriented Polypropylene Film. PCT Patent WO 03031174 A2, 17 April 2003.
74.
Mileva, D.; Wang, J.; Gahleitner, M.; Doshev, P.; Androsch, R. Crystallization behaviour of heterophasic propylene-ethylene
copolymer at rapid cooling conditions. Polymer 2016,102, 214–220. [CrossRef]
75. Aasetre, S.; Kamfjord, T. Polypropylene Compositions. PCT Patent WO 2007093376 A1, 23 August 2007.
76.
Wang, J.; Leskinen, P.; Vestberg, T.; Gahleitner, M. Nucleated Polypropylene Composition. EP Patent EP 2960279 A1,
30 December 2015.
77.
Gahleitner, M.; Tranninger, C.; Doshev, P. Polypropylene Copolymers. In Polypropylene Handbook: Morphology, Blends and
Composites; Karger-Kocsis, J., Bárány, T., Eds.; Springer International: Berlin, Germany, 2019; pp. 295–355. ISBN 978-3-030-12902-6.
78.
Paulik, C.; Tranninger, C.; Wang, J.; Shutov, P.; Mileva, D.; Gahleitner, M. Catalyst type effects on structure/property relations of
polypropylene random copolymers. Macromol. Chem. Phys. 2021,222, 2100302. [CrossRef]
79.
Pantani, R.; Balzano, L.; Peters, G.W.M. Flow-Induced Morphology of iPP Solidified in a Shear Device. Macromol. Mater. Eng.
2011,296, 60–67. [CrossRef]
80.
Kumaraswamy, G.; Verma, R.K.; Kornfield, J.A.; Yeh, F.; Hsiao, B.S. Shear-Enhanced Crystallization in Isotactic Polypropylene.
In-Situ Synchrotron SAXS and WAXD. Macromolecules 2004,37, 9005–9017. [CrossRef]
81.
Housmans, J.-W.; Steenbakkers, R.J.A.; Roozemond, P.C.; Peters, G.W.M.; Meijer, H.E.H. Saturation of Pointlike Nuclei and the
Transition to Oriented Structures in Flow-Induced Crystallization of Isotactic Polypropylene. Macromolecules
2009
,42, 5728–5740.
[CrossRef]
Polymers 2022,14, 4763 24 of 25
82.
Mileva, D.; Tranchida, D.; Gahleitner, M. Designing polymer crystallinity: An industrial perspective. Polym. Cryst.
2018
,1, e10009.
[CrossRef]
83.
Jääskeläinen, P.; Hafner, N.; Pitkänen, P.; Gahleitner, M.; Tuominen, O.; Töltsch, W. Propylene Random Copolymer. EP Patent
EP 1270628 A1, 2 January 2003.
84.
Hafner, N.; Töltsch, W.; Gahleitner, M.; Pitkänen, P. Propylene Polymer Composition with Improved Balance of Mechanical and
Optical Properties. EP Patent EP 1428853 A1, 16 June 2004.
85.
Gahleitner, M.; Grein, C.; Blell, R.; Wolfschwenger, J.; Koch, T.; Ingolic, E. Sterilization of propylene/ethylene random copolymers:
Annealing effects on crystalline structure and transparency as influenced by polymer structure and nucleation. Express Polym.
Lett. 2011,5, 788–798. [CrossRef]
86.
Doshev, P.; Jääskeläinen, P.; Kheirandish, S.; Leskinen, P.; Johnsen, G.K.; Ding, H. High Flow Polypropylene Composition. EP
Patent EP 2449026 A1, 9 May 2012.
87.
Doshev, P.; Leskinen, P.; Gahleitner, M.; Potter, E.; Kheirandish, S. Random Propylene Copolymer with High Stiffness and Low
Haze. EP Patent EP 2527594 A1, 28 November 2012.
88.
Boragno, L.; Resconi, L.; Lilja, J.; Gahleitner, M. Propylene Copolymer with High Impact Properties. EP Patent EP 2978782 A1,
3 February 2016.
89.
Bergstra, M.; Malm, B.; Leskinen, P.; Kock, C.; Sundholm, T. Random Propylene Copolymers for Pipes. EP Patent EP 2539398 A1,
2 January 2013.
90.
Grein, C.; Gahleitner, M.; Knogler, B.; Nestelberger, S. Melt viscosity effects in ethylene–propylene copolymers. Rheol. Acta
2007
,
46, 1083–1089. [CrossRef]
91. Grein, C.; Schedenig, T. Sterilisable and Tough Impact Polypropylene Composition. EP Patent EP 2176340 A1, 21 April 2010.
92.
Aarnio-Winterhof, M.; Doshev, P.; Seppälä, J.; Gahleitner, M. Structure-property relations of heterophasic ethylene-propylene
copolymers based on a single-site catalyst. Express Polym. Lett. 2017,11, 152–161. [CrossRef]
93.
Gahleitner, M.; Doshev, P.; Tranninger, C. Heterophasic copolymers of polypropylene—Development, design principles and
future challenges. J. Appl. Polym. Sci. 2013,130, 3028–3037. [CrossRef]
94.
Wang, J.; Gahleitner, M.; Potter, E.; Aho, J.; Bernreitner, K.; Monissen, L.; Cigon, M. Process for Preparing a Polypropylene
Composition. PCT Patent WO 2019002346 A1, 3 January 2019.
95.
Amiar, N.; Bouzid, D.; McKenna, T.F.L. Influence of the rubber content and particle morphology on the mechanical properties of
the (hiPP). J. Appl. Polym. Sci. 2016,133, 44197. [CrossRef]
96.
Grein, C.; Bernreitner, K.; Hauer, A.; Gahleitner, M.; Neißl, W. Impact modified isotactic polypropylene with controlled rubber
intrinsic viscosities: Some new aspects about morphology and fracture. J. Appl. Polym. Sci. 2002,87, 1702–1712. [CrossRef]
97.
Doshev, P.; Lohse, G.; Henning, S.; Krumova, M.; Heuvelsland, A.; Michler, G.H.; Radusch, H.-J. Phase interactions and structure
evolution of heterophasic ethylene–propylene copolymers as a function of system composition. J. Appl. Polym. Sci.
2006
,
101, 2825–2837. [CrossRef]
98.
Kim, G.-M.; Michler, G.H.; Gahleitner, M.; Mülhaupt, R. Influence of Morphology on the Toughening Mechanisms of Polypropy-
lene Modified with Core-Shell Particles derived from Thermoplastic Elastomers. Polym. Adv. Technol.
1989
,9, 709–715. [CrossRef]
99.
Rungswang, W.; Saendee, P.; Thitisuk, B.; Pathaweeisariyakul, T.; Cheevasrirungruang, W. Role of Crystalline Ethylene-Propylene
Copolymer on Mechanical Properties of Impact Polypropylene Copolymer. J. Appl. Polym. Sci. 2013,128, 3131–3140. [CrossRef]
100.
Doshev, P.; Kona, B.R.; Bernreitner, K.; Grein, C. High Flowable Heterophasic Polypropylene. EP Patent EP 2174980 A1,
14 April 2010.
101. Tranninger, M. Polyolefin Composition with Reduced Occurrence of Flow Marks. EP Patent EP 2599829 A1, 5 June 2013.
102.
Gahleitner, M.; Grestenberger, G.; Grein, C.; Kock, C. High Flow Polyolefin Composition with Low Shrinkage and CLTE. EP
Patent EP 2731988 A1, 21 May 2014.
103.
Tranninger, C. Transparent Thin Wall Packaging Material with Improved Stiffness and Flowability. EP Patent EP 2936343 A2,
28 October 2015.
104.
Bergstra, M.; Leskinen, P.; Malm, B.; Tranninger, C. Heterophasic Polypropylene with Excellent Creep Performance. EP Patent
EP 2550324 A1, 30 January 2013.
105.
Tranninger, C.; Doshev, P.; Potter, E.; Gsellmann, G.; Leskinen, P. Heterophasic Propylene Copolymer with Excellent Im-
pact/Stiffness Balance. EP Patent EP 2601260 A1, 12 June 2013.
106.
Tranninger, C.; Lummerstorfer, T.; Horill, T.; Gahleitner, M.; Wang, J.; Bernreitner, K.; Berger, F.; Leskinen, P. Multimodal
Polypropylene Composition with High Stiffness and High Flowability and Process for Its Production. EP Patent EP 3945112 A1,
2 February 2022.
107. Paulik, C.; Gahleitner, M.; Neißl, W. Flexible, tough and resilient PP-Copolymers. Kunstst/PlastEurope 1996,86, 16–17.
108.
Wang, J.; Shutov, P.; Lilja, J.; Gahleitner, M. Soft and Transparent Impact Copolymers. EP Patent EP 3102634 A1, 14 December 2016.
109.
Starke, J.U.; Michler, G.H.; Grellmann, W.; Seidler, S.; Gahleitner, M.; Fiebig, J.; Nezbedova, E. Fracture Toughness of Polypropylene
Copolymers: Influence of Interparticle Distance and Temperature. Polymer 1998,39, 75–82. [CrossRef]
110.
Kotter, I.; Grellmann, W.; Koch, T.; Seidler, S. Morphology–Toughness Correlation of Polypropylene/Ethylene–Propylene Rubber
Blends. J. Appl. Polym. Sci. 2006,100, 3364–3371. [CrossRef]
111.
Cecchin, G.; Marchetti, E.; Baruzzi, G. On the Mechanism of Polypropene Growth over MgCl2/TiCl4 Catalyst Systems. Macromol.
Chem. Phys. 2001,202, 1987–1994. [CrossRef]
Polymers 2022,14, 4763 25 of 25
112.
Grof, Z.; Kosek, J.; Marek, M. Modeling of Morphogenesis of Growing Polyolefin Particles. AIChE J.
2005
,51, 2048–2067.
[CrossRef]
113.
Zubov, A.; Pechackova, L.; Seda, L.; Bobak, M.; Kosek, J. Transport and reaction in reconstructed porous polypropylene particles:
Model validation. Chem. Eng. Sci. 2010,65, 2361–2372. [CrossRef]
114.
Sandholzer, M.; Bernreitner, K.; Klimke, K.; Gahleitner, M. Propylene Copolymer for Injection Molded Articles or Films. EP
Patent EP 2794689 A1, 29 October 2014.
115.
Braun, J.; Edler, M.; Emig, J.; Gahleitner, M.; Kahlen, S.; Sandholzer, M.; Scriba, M.; Tranninger, M. Polypropylene (PP): News and
Trends from the Reactor to the Recycled Material. Kunstst. Int. 2017,107, 260596.
116.
Fink, G.; Steinmetz, B.; Zechlin, J.; Przybyla, C.; Tesche, B. Propene Polymerization with Silica-Supported Metallocene/MAO
Catalysts. Chem. Rev. 2000,100, 1377–1390. [CrossRef]
117.
Gahleitner, M.; Kastner, M.; Siebert, H.; Stadlbauer, M. Low Emission Polymer Composition. EP Patent EP 2262858 A1,
8 April 2009.
118.
Paavilainen, J.; Doshev, P.; Reichelt, K. Propylene/1-Hexene Copolymer Composition with Low Sealing Temperature. EP Patent
EP 2561016 A1, 27 February 2013.
119.
Nenseth, S.; Doshev, P. Heterophasic Polypropylene with High Flowability and Excellent Low Temperature Impact Properties. EP
Patent EP 2075284 A1, 1 July 2009.
120.
Chum, P.S.; Swogger, K.W. Olefin polymer technologies—History and recent progress at The Dow Chemical Company. Prog. Polym.
Sci. 2008,33, 797–819. [CrossRef]
121.
Siracusa, V.; Blanco, I. Bio-Polyethylene (Bio-PE), Bio-Polypropylene (Bio-PP) and Bio-Poly(ethylene terephthalate) (Bio-PET):
Recent Developments in Bio-Based Polymers Analogous to Petroleum-Derived Ones for Packaging and Engineering Applications.
Polymers 2020,12, 1641. [CrossRef] [PubMed]
122.
Vollmer, I.; Jenks, M.J.F.; Roelands, M.C.P.; White, R.J.; van Harmelen, T.; de Wild, P.; van der Laan, G.P.; Meirer, F.; Keurentjes, J.T.F.;
Weckhuysen, B.M. Beyond Mechanical Recycling: Giving New Life to Plastic Waste. Angew. Chem.—Int. Ed.
2020
,59, 15402–15423.
[CrossRef] [PubMed]
123.
Coates, G.W.; Getzler, Y.D.Y.L. Chemical recycling to monomer for an ideal, circular polymer economy. Nat. Rev. Mater.
2020
,
5, 501–516. [CrossRef]
124.
Lazzari, S.; Lischewski, A.; Orlov, Y.; Deglmann, P.; Daiss, A.; Schreiner, E.; Vale, H. Chapter Six—Toward a digital polymer
reaction engineering. Adv. Chem. Eng. 2020,56, 187–227.
