OPTICAL AND MECHANICAL CONSEQUENCES OF

NUCLEATION AND STERILISATION IN PP CAST FILMS

Markus Gahleitner*, Johannes Wolfschwenger & Christelle Grein
Borealis Polyolefine GmbH, St. Peterstr. 25, 4021 Linz, Austria,
email: markus.gahleitner@borealisgroup.com

Rebecca Blell
Université Louis Pasteur, Strasbourg, France

Thomas Koch
Vienna University of Technology, Institute for Material Science and Technology,
Favoritenstraße 9-11, 1040 Vienna, Austria

BIOGRAPHICAL NOTE

Dr. Markus Gahleitner graduated from Johannes Kepler University Linz, Austria,
with a PhD thesis in the field of polymer melt rheology. Since 1992 he works for
Borealis Polyolefine GmbH (formerly PCD Polymere) in research and development,
covering different projects from basic catalyst to application development. Presently
he holds the position of “IPR Group Expert” in the company’s IPR department. He is
author of more than 40 scientific papers in international refereed journals as well as
several book contributions, and he is inventor of more than 15 patents and
applications. His special fields of interest are polymer crystallization, processing-
related rheology problems and sustainability aspects of polymers.

ABSTRACT

An extensive investigation of two series of ethylene/propylene (EP) random copolymers was performed to
understand the factors influencing haze increase in the steam sterilization of extrusion cast films from such
materials. Different analytical methods were employed to elucidate structural changes determining film
optics, and next to the polymer parameters also nucleation and processing effects were studied. The findings
clearly show that a combination of efficient comonomer distribution and nucleation can partly inhibit lamellar
thickening in sterilization, thus preserving high transparency even after a heat treatment.

Introduction

Extrusion cast film, a major application area for polypropylene, requires an excellent combination of
mechanical and optical properties. Next to homopolymers, ethylene-propylene (EP) copolymers are the most
important materials for this segment, taking advantage of their remarkable see-through performance, a result
of their slower crystallization speed compared to standard PP [1]. As a growing fraction of this segment is
subjected to pasteurization or sterilization processes (like in medical applications, see Fig. 1), the
mechanical and optical consequences of such a treatment must be considered in material design already [2].
Numerous studies regarding controlled aging or annealing at different temperatures for various times have
been performed on injection or compression molded thick samples. Mechanical tests along with observation
of the microstructure at different scales are often performed to follow the induced changes and understand
their origin [3,4], but such results are limited in relevance to predict the behaviour of films, especially with
respect to an alteration of the optics after a defined heat treatment.

The present study was therefore directed at understanding both processing and sterilization effects on the
crystalline structures and the resulting film performance for both PP homopolymers and EP random
copolymers, including versions with different nucleating agents frequently used for improving transparency
[5]. Influence of processing parameters (chill-roll temperature. film thickness. co-extrusion line), polymer
parameters (C2-content of EP-copolymers, catalyst systems) and nucleation agents will be discussed, and
the changes in optics will be correlated to crystallinity and morphology variations assessed by differential
scanning calorimetry (DSC), polarizing optical microscopy (POM), wide- and small-angle X-ray diffraction
(WAXD/SAXD) and transmission electron microscopy (TEM).
 Figure 1 – Flexible infusion pouch for medical application produced
from a PP-based multilayer film; typical example for an advanced
packaging application requiring sterilization

Experimental work

Two series of PP materials produced on pilot facilities of Borealis were investigated:

- In the first one, 6 polymers produced with a high yield 4th generation Ziegler-Natta catalyst (C1) having
an ethylene content between 0 and 5 wt% were tested. All grades had a reactor melt flow rate (MFR
230°C / 2,16kg) of 1,5 g/10 min, were visbroken with peroxide to an MFR of 8 g/10min and equipped
with a standard additivation package.
- In the second one, EP-copolymers with a C2-content of about 3.5 wt% were produced using three
different catalyst systems from the same category. The grade made with catalyst C1 had an MFR of
about 1.5 g/10min and was subsequently stabilized, additivated and visbroken to an MFR of 8 g/10min,
while the grades made with catalysts C2 and C3 were neat reactor grades with an MFR of about 6
g/10min. Different nucleating agents (A1 to A3 for the α-modification and B for the β-modification) to
promote either the α- or the β-modification of the EP-copolymers were added during the compounding
step, testing against non-nucleated references.

For all PP materials, extrusion cast films of 50 and in some cases 130 µm thickness were produced on a
PM30 type laboratory extruder with a coathanger slit die 200 mm wide and with a gap range of 0.55 to 0.6
mm. The melt temperature was around 250°C and the chill roll temperature was varied for the second series
(20°C. 55°C and 90°C).

After the extrusion, all films of series 1 and 2 were optically characterized according to ASTM D 1003 using a
BYK-Gardner Hazegard Plus Instrument. The haze was chosen as most relevant parameter to assess
differences in optics before and after sterilization. The measurements were done at least 96h after film
production or sterilization.

Steam sterilization was performed in a Systec D series machine. The samples were heated up at a heat rate
of 5 °C/min starting from 23°C. After having been kept for 30 min at 121°C, there were removed immediately
from the steam sterilizer and stored at room temperature till processed further.

Crystallinity and morphology of the films before and after sterilization were studied by DSC, POM,
WAXD/SAXD and TEM, however not employing every method in each case. DSC measurements were
carried out on a power compensated DSC 2920-Co (TA instrument) on all samples before and after
sterilization. Two heating scans and one cooling scan were done in between -10°C to 210°C at a rate of
10°C/min. The analyzed samples had a weight of about 5 mg. The first heating cycle was considered to be
the most interesting as it did not erase the history of thermal treated samples and was thus believed to give
fairly reliable indications on the morphology of the investigated materials – this is acceptable even though the
test is not as accurate as those provided by non destructive methods (e.g. WAXD). Melting temperature as
the maximum of the considered peak and melting enthalpies as the area under the relevant peak were
assessed.
 Figure 2 – DSC thermograms recorded at
Before Sterilization
10 K/min for EP random copolymer with
Endothermic Heat Flow (W/g)

After Sterilization
MFR 6 and 3,5 wt% C2 processed at a
chill roll temperature of 55°C; scans
before and after sterilization

0 20 40 60 80 100 120 140 160 180

Temperature (°C)

An example of the first heating cycle of a DSC thermogram of a neat and retorted sample is provided in Fig.
2 for an EP-copolymer. The main challenge in this evaluation is to fix in a reproducible way the baseline of
the DSC traces and the split between the two peaks appearing after retorting, which for simplicity was done
by taking the baseline as the line between the first minimum points to the left and the right of the peak and
the split between two peaks as a vertical section at the minimum between the two peaks.

Polarizing optical microscopy was performed on 10 µm thin sections across the film thickness under
transmission mode using a Light Microscope from Olympus with crossed polarizers. The aim was to see the
difference in morphology - before and after sterilization - induced by α and β-nucleation compared to neat
samples and to visualize changes between the structures induced with the different catalyst systems (C1, C2
and C3).

Some selected samples were submitted to WAXD investigations at Vienna University of Technology and to
TEM investigations at the Center for Electron Microscopy Graz, Austria. Wide-angle X-ray was done on the
full films with a Philips X’Pert Pro instrument using CuKα radiation in reflection mode. The peak integration
for determining crystallinity was performed as reported before [6]. For TEM, ultrathin sections of sample
specimens contrasted with ruthenium tetroxide to allow differentiation between regions of high and low
crystallinity were prepared. Images were recorded on a Tecnai G² 12 from FEI, equipped with a CCD camera
(Gatan Bioscan) at 100 kV acceleration voltage.

Results and Discussion

The basic composition and melting points determined on cast films are summarized for both series in tables
1 and 2 below. Two effects are obvious immediately from these tables: The rather linear melting point
reduction with increasing C2 content for series one (presented in Fig. 3), and the fact that both nucleation
and catalyst type co-influence the quenching effect achieved by reducing the chill roll temperature in series
two. Here, the chill-roll temperature (Troll) has been varied between 20 and 90°C to simulate cases of a:
(i) highly amorphous neat structure with moderate stiffness (Troll: 20°C)
(ii) a stiff resin at good starting optics (Troll: 55°C)
(iii) a maximum crystalline grade (Troll: 90°C)

Ethylene
Sample MFR
Content Tm (°C)
number (g/10 min)
(wt %)
1/1 0.0 8.6 162.8
1/2 1.6 7.4 155.9
1/3 2.4 7.6 151.9
1/4 3.3 8.5 148.0
1/5 4.3 7.4 143.0
1/6 5.0 7.2 140.2

Table 1 – Basic properties of the polymers from the first series (melting point from DSC on cast film before
sterilization)
 Ethylene
Sample MFR Reactor
content Catalyst Nucleation Tm(°C) at chill roll temp.
number (g/10min) / CR
(wt%)
20°C 55°C 90°C
2/1 3.5 C1 8.2 CR No 143.1 144.5 146.3
2/2 3.5 C2 5.7 RE No 146.3 144.5 150.0
2/3 3.5 C3 6.0 RE No 142.0 143.3 145.3
2/4 3.5 C1 7.8 CR A1 144.7 144.9 145.2
2/5 3.5 C2 5.8 RE A1 146.7 148.1 149.5
2/6 3.5 C3 6.4 RE A1 143.0 144.0 142.7
2/7 3.5 C1 8.1 CR A2 144.1 145.2 145.8
2/8 3.5 C2 5.6 RE A2 148.4 147.9 149.2
2/9 3.5 C3 6.1 RE A2 143.1 143.8 140.5
2/10 3.5 C1 7.9 CR A3 143.6 144.3 145.5
2/11 3.5 C2 5.7 RE A3 147.7 147.8 149.0
2/12 3.5 C3 6.2 RE A3 142.4 143.4 145.4
2/13 3.5 C1 8.6 CR B 143.2 144.1 145.1
2/14 3.5 C2 5.9 RE B 146.8 147.2 149.4
2/15 3.5 C3 6.1 RE B 142.1 143.4 145.3

Table 2 – Basic properties of the polymers from the second series (visbroken grades indicated by “CR” for
“controlled rheology”; melting point from DSC on cast film processed at different chill roll temperatures before
sterilization)

165 700

160 600
Figure 3 – Changes in melting
500
155 temperature and haze change in
Delta Haze / %
Tm (DSC) / °C

400
sterilization at 121°C depending
y = -4,5715x + 162,95 on the C2-content of a 50 µm
150 2
R = 0,9994 300 thick Cast Film for series 1
145 polymers
200

140 100

135 0
0 1 2 3 4 5
C2 content / wt%

The comonomer influence in the first series is clearly reflected in the optical changes during crystallization.
As Fig. 3 shows, the haze increase remains rather constant up to an ethylene content of about 2,5 wt% and
then rises significantly. For better understanding this phenomenon, the fraction of the melting enthalpy curve
below the sterilization temperature of 121°C was calculated in relation to the total melting enthalpy. The
discontinuity at 2,5 wt% found in figure 3 could not be reproduced here, but the reason for this becomes
obvious when taking a closer look at the DSC curve (heat 1) before sterilization as recorded for material 1/2
(see figure 4).

Three peaks could be distinguished here, corresponding to three transitions [7]:

i. an endothermic peak (T1) with its maximum at around 60°C
ii. an exothermic peak (T2) with its maximum at around 100°C
iii. a large endothermic peak (T3) with its maximum at around 160°C in the case of the homopolymer
(resp. 156°C for the polymer considered here)
Peak T1 is assumed to be the superposition of peak T2 and T3. T1 is therefore assumed to be the beginning
or onset of peak T3. Peak T2 is the result of increased mobility of the crystalline phase starting at 50°C [8].

With this, further crystallization takes place. This is because thick lamellas attract thin ones as well as
crystallisable parts of the amorphous phase becoming mobile. Lamellar thickening takes place and the
 amorphous region is rearranged thus a crystallization peak. Evaluation of the area of peak T2 showed a
linear decrease of this area with the increase in ethylene content. This area evaluation is not exact as the
superposition of the two peaks leads to high error but it is representative of the evolution. The increase in
ethylene content therefore improves the optical properties of the films before sterilization by reducing haze,
increases the difference in the optical properties of the films with sterilization and encourages secondary
crystallization; this increased “quenchability” of PP grades with reduced regularity due to stereodefects
and/or comonomer incorporation has been postulated before [10].

Figure 4 – DSC thermogram (heat 1) recorded at 10

K/min for the 1.6 wt.% ethylene content film before
sterilization
Endothermic Heat Flow (W/g)

T3

T1

T2

0 50 100 150 200

Temperature (°C)
After sterilization, the DSC analysis of the films reveals a double melting peak as already evident from figure
1. Surprisingly, the first peak of these is always at the same temperature of ~ 131°C, indicating a secondary
crystalline structure being generated in the sterilization step which is independent of the ethylene content.
Figure 5 gives the relative areas for the primary (high Tm , depending on C2 content) and secondary (low Tm ,
131°C) crystallization fractions in the films. The qualitative correlation of the latter part to the haze increase in
figure 3 is quite striking – at the highest comonomer content the chain irregularities obviously even reduce
the capacity for post-crystallization, in line with the overall crystallization speed reduction found for such
materials before [1].

100

90

80
Figure 5 – Enthalpy area for primary and
70 secondary crystallization melting for the
Area (mW.°C)

60 sterilized cast films from series 1

50

40

30

20

10

0
0 Primary Crystallization
1.6
2.4
3.3 Secondary Crystallization
4.3
C2-content (%wt) 5

In the second series, where the focus was on processing and nucleation effects, the optical properties were
found to deteriorate with increasing the chill roll temperatures as a result of an increase in crystallinity as
anticipated from literature [10]. When exemplified with material 2/1 (non nucleated sample), this feature
corresponds to an enhancement of the melt enthalpy from 64.0 J/g for a chill roll temperature of 20°C, to
76.1 J/g for a chill roll temperature of 55°C, to 82.7 J/g for a chill roll temperature of 90°C. Two factors play a
parallel role in this effect for high chill roll temperatures: (i) the reduction of the rate of crystallization allowing
the crystallites to grow to large spherulites, in turn increasing the haze (see figure 6) and (ii) a reduction of
the amount of mesomorphic phase in the system [11].
 35

30
Figure 6 – Haze change with chill coll
25
temperature variation for samples 2/1, 2/4 and
2/13 (see also table 1)
20

Haze Values
(%) 15

10

5 90

55 Chill Roll
0
Temperature (°C)
C1 20
C1,A1
C1,B
Samples

Even though the optical performance worsened with the chill roll temperature, this picture was notably
influenced by nucleation as highlighted in figure 6. While the haze increased steadily from 4.9% (Troll 20°C) to
31.3% (Troll 90°C) for the non-nucleated samples, the recorded haze values stood roughly constant at around
6% for the nucleating agent A1. This result – confirmed with the other α-nucleators - but at other haze levels
- suggests that the morphology – and thus the optical performance - of random copolymers can be controlled
by α-nucleation independently of the conversion temperature. This feature can be ascribed to their higher
crystallization rate promoted by heterogeneous nucleation [12].

Another way to visualize the influence of both processing conditions and additivation is provided in figure 7.
There the difference in haze between different chill roll temperatures as a function of the nucleation is
reported. On one hand the low sensitivity of α-nucleated samples to cooling rates is evident as assessed by
a relative change in optics of maximum 80% starting from a very low haze level of around 3 to 6 %. On the
other hand, non-nucleated and β-nucleated samples are influenced to a large extent (up to 670%) by the
processing conditions as already mentioned earlier. The low difference in haze between a chill-roll
temperature of 55 and 90°C suggests that the crystallization speed of the studied polymer is not overruled by
processing parameters, while the opposite holds for chill-roll temperatures of 20°C independently of the
material under investigation (C1, C2 or C3).

650,0

Delta Haze between films Chill Rolled at 55°C and 20 °C

550,0 Delta Haze between films Chill Rolled at 90°C and 55 °C
Delta Haze between films Chill Rolled at 90°C and 20 °C

450,0
Delta Haze (%)

350,0

250,0

150,0

50,0

C1 C2 C3 C1,A1 C2,A1 C3,A1 C1,A2 C2,A2 C3,A3 C1,A3 C2,A3 C3,A3 C1,β C2,β C3,β
-50,0
Film Sample Numbers
Figure 7 – Difference in haze between the cast films extruded at different chill roll temperatures before
sterilization for the 15 samples of series two

Also in this series, steam sterilization induces changes in the optical performance of the studied grades.
However, these variations are more or less pronounced depending on the initial crystalline state of the
grade. Quenched films obtained with a chill roll temperature of 20°C as shown in figure 8 are more sensitive
to a heat treatment than more crystalline samples manufactured at a chill-roll temperature of 90°C. They
undergo the largest reorganization and lamellar thickening with sterilization, as a combination of annealing
and post-crystallization effects as widely documented in the literature [3,4,7,8,9].

The occurrence of lamellar thickening in post-crystallization – as a result of a reduction of the magnitude on

the interfacial free energy [4] – could be confirmed by TEM (see figure 9) and in WAXD/SAXD analysis (see
table 3). On the example of a nucleated sample processed at a chill-roll temperature of 90°C, it becomes
obvious that a heat treatment at 121°C for 30 min promotes the formation of thicker lamella and lead to a
denser network of crystalline structures. The X-ray data moreover show two very interesting phenomena:
i. the crystallinity increase in sterilization is significant only for the non-nucleated sample, providing an
additional explanation for the reduced haze increase in nucleated materials (see again Fig. 8)
ii. in contrast to expectations from theory (which predicts an increase due to the higher crystallization
temperature [13]), nucleation does not affect the lamellar thickness in cast-film processing

400
Haze Difference at 20°C chill roll temperature
Haze Difference at 55°C chill roll temperature
350 Haze Difference at 90°C chill roll temperature

300

250
Delta haze (%)

200

150

100

50

1 2 3 4 5 6 7 8 9 1 11 12 13 14 15
-50
Film Sample Numbers

Figure 8 – Haze difference between sterilized and non-sterilized films for the 15 samples of series two at the
three different chill roll temperatures tested

Figure 9 – TEM images of the lamellar morphology of material 2/6 (catalyst C3, MFR 6, 3.5 wt% C2) from
cast film extruded at chill roll temperature of 90°C before (left) and after (right) sterilization at 121°C; scale
bar dimension 200 nm
 Table 3 – Results of WAXD/SAXD
Sample state Xcr / % L / nm analysis of films from materials 2/3
2/3, non-nucl. extruded 90°C 56 11,3 and 2/6 (catalyst C3, MFR 6, 3.5 wt%
2/6, nucleated sterilized 121°C 64 15,1 C2) before and after sterilization;
2/3, non-nucl. extruded 90°C 61 11,5 overall crystallinity from WAXD and
2/6, nucleated sterilized 121°C 63 15,2 long period L from SAXD given

Also for the second series, an attempt to correlate haze variations with crystallinity changes before and after
sterilization was made using the DSC traces of the first heat after sterilization. As explained before, a double
peak was found with high reproducibility for all samples. In contrast to the results shown in Fig. 5, where a
clear correlation of both primary and secondary peak with the comonomer content could be established, the
situation was found to be far more complex here.

In general, the relative amount (enthalpy) of the secondary crystallization was higher for the 20°C chill roll
temperatures than for the 90°C chill roll temperature, confirming the increased mobility in more strongly
quenched films observed for PP homopolymers before [9]. Another repeated pattern seen in Figure 10 is the
lower area of secondary crystallization in the C2 polymerization method for the quenched films. The
secondary crystallization of the C2 polymerized films showed the lowest area implying the least secondary
crystallization. This could be the effect of the higher melting point mentioned above for these films which
leads to less reorganization due to melting. Another effect could also be the decrease of lamellar thickening
with the C2 polymerization as the lamellae are already thick enough and could not thicken further. The
influence of the nucleating agents on the secondary crystallization is however not visible in the DSC analysis
as the areas of all the films for a given polymerization method are nearly constant and the area of the 15
films are in the same range as seen in Figure 10. This is true for all three chill roll temperatures.

50

45
C1 C2 C3

40
Area (mW.°C)

35

30

25

20
No nucleation A1 A2 A3 Beta
Nucleation

Figure 10 – Area (enthalpy) of secondary crystallization after sterilization for selected materials from series
two extruded at Troll =20°C
Summary and Outlook

We investigated the structure of EP random copolymer films before and after sterilization at 121°C for 30
minutes by DSC, TEM, POM and WAXD/SAXD to explain the changes in optical properties (haze) of
extrusion cast films with changing nucleating agents, extrusion conditions and ethylene content as well as
polymerization conditions. Two series of films were studied, varying the comonomer content in the first one
and catalyst, nucleation and processing conditions in the second one. The findings clearly show that a
combination of efficient comonomer distribution and nucleation can partly inhibit lamellar thickening in
sterilization, thus preserving high transparency even after a heat treatment.

In detail, attention has to be paid to the combined effects of physical ageing and post-crystallization, which
both are affected by the chain regularity. The appearance of a rather constant secondary melting point in
DSC at a temperature of 10 K above the sterilization temperature clearly points to the existence of
secondary crystalline structures postulated before by several authors [4,7,9]. More extensive investigation,
especially using the possibilities of X-ray diffraction, will be required to develop a more complete
understanding.

Acknowlegements

The authors want to thank Dr. Elisabeth Ingolic from the Center of Electron Microscopy (ZFE) Graz, Austria,
for preparing the excellent TEM images.

References

1 – M. Gahleitner, P. Jääskeläinen, E. Ratajski, C. Paulik, J. Reussner, J. Wolfschwenger & W. Neißl,

Propylene–Ethylene Random Copolymers: Comonomer Effects on Crystallinity and Application Properties.
J.Appl.Polym.Sci, 95 (2005) 1073-81
2 – K. Resch, G.M. Wallner, C. Teichert, G. Maier & M. Gahleitner, Optical Properties of Highly Transparent
Polypropylene Cast Films: Influence of Material Structure, Additives, and Processing Conditions.
Polym.Eng.Sci, 46 (2006) 520-31
3 – S. Piccarolo, Ageing of isotactic polypropylene due to morphology evolution, experimental limitations of
realtime density measurements with a gradient column, Polymer 47 (2006) 5610-5622
4 – H. Marand, A. Alizadeh, S. Sohn, J. Xu, R. Farmer, V. Prabhu, S. Cronin, V. Velikov, A Model for the
Physical Aging of Semicrystalline Polymers Above Tg: Secondary Crystallization Induced Constraining
Effects. Proc. SPE ANTEC, 59 Vol. 2 (2001) 1856–1859
5 – N. J. Macauley, E. M. A. Harkin-Jones, W. R. Murphy, The Influence of Nucleating Agents on the
Extrusion and Thermoforming of Polypropylene, Polym.Eng.Sci. 38 (1998) 662-70
6 – T. Koch, S, Seidler, E. Halwax, S. Bernstorff, Microhardness of quenched and annealed isotactic
polypropylene, J.Mater.Sci. 42 (2007) 5318-26
7 – N. Alberola, M. Fugier, D. Petit, B. Fillon, Microstructure of quenched and annealed films of isotactic
polypropylene. Part I, J.Mater.Sci. 30 (1995) 1187-95
8 – J.M.K. Agarwal, J.M. Schultz, The Physical Aging of Isotactic Polypropylene, Polym.Eng.Sci. 21 (1981)
776–781
9 – M. Gahleitner, J. Fiebig, J. Wolfschwenger, G. Dreiling & C. Paulik, Post-Crystallization and Physical
Ageing of Polypropylene: Material and Processing Effects. J.Macromol.Sci.-Phys., B41 (2002) 833-49
10 – K. Resch, G.M. Wallner, C. Teichert & M. Gahleitner, Highly Transparent Polypropylene Cast Films:
Relationships between Optical Properties, Additives and Surface Structure. Polym.Eng.Sci., 47 (2007) 1021-
32
11 – A. Martorana, S. Piccarolo, F. Schichilone, The X-ray determination of the amounts of the phases in
samples of isotactic poly(propylene) quenched from the melt at different cooling rates, Macromol.
Chem.Phys. 198 (1997) 597-604
12 – N. J. Macauley, E. M. A. Harkin-Jones, W. R. Murphy, The influence of Nucleating Agents on the
Extrusion and Thermoforming of Polypropylene, Polym.Engin.Scie. 38 (1998) 662-73
13 – B. Pukánszky, I. Mudra, P. Staniek, Relation of crystalline structure and mechanical properties of
nucleated polypropylene, J.Vinyl.Addit.Technol. 3 (1997) 53-7