(572b) Characterization Methodologies for Obtaining a Reliable Indicator for the Environmental Stress Cracking Resistance of Polyethylene

Environmental
stress cracking (ESC) is one of the most common failure mechanisms in polymers
where a sudden catastrophic premature brittle fracture occurs. This type of fracture
develops through a slow crack growth mechanism caused by internal molecular
heterogeneities in the polymer structure, promoted during the manufacturing or
production of a specific plastic part. The resistance to such failure is of
critical importance especially for applications where structural integrity is
essential. A complete understanding of the molecular structure-property
relationship of the polymer is required in order to evaluate the performance of
a particular polymer that is subjected to conditions where ESC is prone to occur. Furthermore, reliable
characterization methodologies are needed to predict the environmental stress
cracking resistance (ESCR) of polymers. Currently, ESCR is reported based on
unreliable and extremely time-consuming testing methods such as the notch
constant load test (NCLT) or the bent strip test (BST). In both tests, notched polymer
specimens are subjected to a certain load in the presence of an aggressive
fluid and elevated temperatures. The time of failure is recorded and is
reported as the ESCR. In this study, we aimed to investigate the structural
properties which control the ESCR of polyethylene (PE) resins. The first part
of the investigation involved development and modification of an alternative
testing method to predict the ESCR of various types of PE resins in a more
practical, reproducible, and reliable fashion (related to the strain hardening behaviour of PE under
uniaxial tension). The second part of the investigation was a comprehensive
rheological study, both in shear and extension, in order to identify potential
correlations between rheological properties and findings from the initial stage
of the study (related to the extent of strain hardening in both melt and solid
phases). The third part of the study involved identifying a reliable
crystalline structural indicator, namely, the lamella lateral surface area
(LLSA), which could yield a better prediction for the ESCR. X-ray scattering analyses
were conducted in order to identify the effect of processing and post-processing
temperature on the extent of LLSA. A correlation between the processing factors
and the extent of interlamellar entanglements and, subsequently, ESCR was finally
developed.

Hardening Stiffness (HS) Test as an Indicator of ESCR

A range of
commercially available rotomolding and pipe grades of
LLDPE and HDPE were selected. The interlamellar entanglements (a combination of
long tie-molecules and the physical chain entanglements) are believed to control
the slow crack growth involved in ESC of PE resins. The extent of such
entanglements was previously investigated by monitoring the strain hardening
behaviour of PE resins in the solid state through a uniaxial tensile test ^[1].
A correlation was developed between the slope of the line (referred to as Hardening
Stiffness (HS)) obtained from the load-displacement curve of a tensile test and
ESCR. In order to modify the test for a better prediction of HS, two experimental
designs were adopted: firstly, a D-optimal factorial design to investigate the
significance of specimen dimensions, strain rate, and the molecular weight (MW)
of PE resins, and secondly, a completely randomized central composite design to
investigate the sensitivity of HS to short chain branching content (SCB). Based
on the results, weight average molecular weight (Mw), strain rate, specimen width
and thickness, along with some of their interactions were found to be the
significant factors. The design of the test was based on two main criteria: (1)
to reduce the effect of sample dimensions on HS, and (2) to better reflect the
effect of molecular structure (Mw and SCB) on HS. The developed test ^[2]
was found to provide a more reliable and consistent ESCR picture without the
drawbacks of the subjective notching process and presence of aggressive fluids.
Further, in order to extend the developed methodology to different types of PE
(LLDPE with different comonomer content), a
correction factor was developed (corrected HS or cHS)
^[2]. This cHS is believed to be a better
indicator for ESCR of bulk PE resins as it takes both Mw and branching effects
into account (see Figure 1).

Rheological Behaviour of the Resins

Rheological
studies were conducted to identify a possible relationship between findings
from the melt and solid studies (focus on extent of interlamellar
entanglements). From the shear studies, a correlation between a normalized characteristic
relaxation time (λ_N)
as a measure of network mobility and ESCR (represented by cHS))
was established (see Figure 2) ^[3]. This outcome was critical for
predicting the ESCR of PE samples with similar molecular weight properties, but
different comonomer content (linear low density PE
(LLDPE)). Extensional viscosities were evaluated from entry flow measurements
and Sentmanat extensional rheometry. Cogswell,
Binding, and Gibson methodologies were used to identify the steady state
extensional viscosities from entrance pressure drops using a capillary
rheometer. It was found that the extensional viscosity in the melt is a better
tool to detect differences in molecular structure of PE resins with similar
molecular weight properties. Steady state extensional viscosity obtained from
entry flow measurements was dominated by the effect of molecular weight. A
correlation was further acknowledged between plateau values of extensional
stress and the HS of the LLDPE resins from the extensional measurements. It was
found that the PE that showed higher extent of strain hardening in the melt had
a higher ESCR. This is mainly due to the fact that the increase in molecular
properties such as molecular weight, molecular weight distribution, and
branching content will increase the ESCR of PE. Similarly,
strain hardening is directly a function of these properties and any increase in
such factors will enhance the extent of strain hardening (both in melt and
solid states). Therefore, a direct relation between the degree of strain hardening
(both in solid and melt) and ESCR is useful, given that PE resins exhibit
strain hardening behaviour.

Temperature Effect on Lamella Lateral Surface Area

Inter-lamellar
links, which are critical to ESCR of PE, must ?anchor? lamellae as the term
suggests. Theorization on the ESCR behaviour from the perspective of
interconnectivity between crystalline and amorphous phases, led our research into
the study of the relationship between the lamella lateral surface area (LLSA)
and ESCR. Unlike crystallinity and lamella thickness that predominantly show
SCB effects, LLSA calculations take into account both SCB and MW influences.
Hence this investigation can be applied over resins of a wide range of MW and
MWD. Our work showed that an increasing ESCR is associated with an increasing LLSA
of PE. A larger lamella lateral surface area increases the probability of
inter-lamellar linkage formations, which leads to improved phase
interconnectivity and hence higher ESCR for polyethylene. We carried out an
investigation with four high density polyethylene (HDPE) resins. The effects of
processing (controlled cooling) and post-processing temperature (annealing) on
LLSA at three different cooling rates and two different annealing temperatures
were evaluated. Wide and small angle X-ray scattering (WAXS & SAXS) and
differential scanning calorimetry (DSC) were utilized
in order to characterize the crystal structure of the modified resins. WAXS and
SAXS were used to determine the crystallinity and the lamella thickness of the
resins, respectively. DSC analysis was utilized to verify the results obtained
from the X-Ray diffraction analyses.

References

[1] Cheng, J.; Polak, M.A.; Penlidis, A. J Macromol Sci, Part A, Pure & Appl Chem 2008, 45, 599

[2] Sardashti, P.; Tzoganakis, C.; Polak, M.A.; Penlidis, A. J Macromol Sci, Part A, Pure & Appl Chem 2012, 49, 689.

[3] Sardashti, P.; Tzoganakis, C.; Zatloukal, M.; Polak, M.A.; Penlidis, Adv in Polym Tech (2013, under review)

Figure 1: ESCR vs. cHS

Figure 2: λ_N vs. SCB/ cHS