Journal of Thermal Analysis and Calorimetry, Vol.
83 (2006) 1, 113–115
DETECTION OF DEGRADATION OF ABS MATERIALS VIA DSC
H. Blom*, R. Yeh, R. Wojnarowski and M. Ling
Baxter Healthcare Corporation, Route 120 and Wilson Road, Round Lake, IL 60060
ABS is a well-known and widely-used rigid engineering polymer. The mechanical properties of ABS are critical to its proper func-
tioning in a given application, such as a medical device. It is therefore important to retain those properties during processing, fabri-
cation, and use. To that end, thermal analysis and mechanical testing were employed to understand the origin of observed mechani-
cal property losses. Oxidation onset temperature (OOT) testing and differential scanning calorimetry (DSC) analysis indicated a
slightly lower onset temperature and a higher glass transition temperature, respectively, for parts which demonstrated a reduction in
mechanical properties. It is also demonstrated that degradation of the butadiene-phase of the ABS is responsible for the mechanical
property reduction, and that this degradation only proceeds at an appreciable rate at elevated temperatures in the presence of
oxygen.
Keywords: ABS, acrylonitrile-butadiene-styrene, differential scanning calorimetry, hot-runner injection molding, polymer degra -
dation, thermal oxidation
Introduction
Table 1 Injection molding conditions
ABS (acrylonitrile/polybutadiene/styrene) graft copoly- Parameter Setting
mers are widely used in many industrial applications.
Barrel temperatures/°C 200, 220, 230, 240
Typically, ABS consists of a styrene/acrylonitrile con-
tinuous phase partially grafted to a dispersed buta- Block temperature/°C 260
diene phase. The latter acts as an impact modifier, and Nozzle temperature/°C 280
imparts excellent mechanical properties to the mate- Mold temperature/°C 40
rial. Preservation of the impact-modifying properties Packing pressure/bar 1000
of ABS during melt-processing (e.g. extrusion, injec-
tion molding) and product use (e.g. stability towards
UV light) center on protecting the polybutadiene injection mold with a screw diameter of 25 and a
phase from degradation. This phase is particularly 7 mm nozzle. Nominal injection molding parameters
susceptible to oxidation due to the presence of resid- are summarized in Table 1.
ual double bonds [1].
Various research papers have appeared address- Oxidation onset temperature
ing detection of degradation, such as color changes
[2], impact testing, dynamic mechanical analysis and The onset of oxidative thermal degradation of the
infrared spectroscopy [3], and derivative thermo- various resins and parts investigated in this study was
gravimety and differential scanning calorimetry [4], determined on a TA Instruments/DuPont 910 Differen-
to mention several. This study correlates DSC, me- tial Scanning Calorimeter controlled by an Instrument
chanical testing, and oxidation onset temperature Specialists control module and software. Samples (~5
analyses, and will be directed to understanding ABS mg) were placed in an open aluminum pan and heated
degradation in a production setting. in the presence of air (100 mL min–1) from room tem-
perature to 280°C at a ramp rate of 20°C min–1. The
onset temperatures were determined manually.
Experimental
Differential scanning calorimetry
Injection molding
Thermal properties were measured with a TA Instru-
Commercial ABS and colorant master batch were
ments® Q1000 Modulated DSC with auto-sampler
gravimetrically fed into a hot-runner-equipped 80-ton
and refrigerated cooling system. In the case of base
1388–6150/$20.00 Akadémiai Kiadó, Budapest, Hungary
© 2006 Akadémiai Kiadó, Budapest Springer, Dordrecht, The Netherlands
BLOM et al.
resin and colorant master batch samples, test samples Table 2 Summary or OOT results
were taken as a slice from the center of the pellet. In Hold time/min OOT/°C
the case of injection-molded parts, samples were
taken from the same location to ensure consistency 0 224
and comparability. Samples, crimped in aluminum 5 212
pans, were heated and cooled at 20°C min–1 from 250 10 212
to –90°C and to 250°C under a 50 mL min–1 N2 flow. 15 210
Sample sizes were 5–10 mg. 25 204
Impact testing
increased, the onset temperature decreased from
Impact testing of injection molded parts was per- about 225 to about 205°C. Clearly, degradation has
formed on an MTS 810 Material Test System, at a occurred in these parts.
speed of 1000 in/min, using a steel blade (blunt tip, A typical DSC curve of the ABS resin used in
thickness 0.05 inch) attached to a load cell. this study is shown in Fig. 1. The rubber-phase glass
transition is apparent at –58.1°C, and the sty-
rene-acrylonitrile glass transition occurred at
Results and discussion 112.2°C. Since the rubber (butadiene) phase of the
ABS is most susceptible to oxidative degradation,
Injection molding equipment shutdowns, while unde-
leading to crosslink formation and decreased chain
sirable, occur from time to time in any production set-
mobility, an increase in the glass transition tempera-
ting. When this happens, molten polymer is trapped in
the barrel of the molding machine, and thermal and
oxidative degradation may accelerate and lead to se-
verely degraded polymer and poor part properties.
While short stoppages may not be an issue for some
polymers (such as polyethylene), they are a concern
for polymers containing residual unsaturation, like
ABS. This is even more important when a hot-runner
injection molding machine is employed.
In an effort to understand how to retain the excel-
lent impact properties of ABS in a hot-runner injection
molding production setting in the event of an equipment
stoppage of moderate duration (up to 25 min), a study
was developed in which a hot-runner injection molding
Fig. 1 Typical second-heat DSC curve of ABS
machine making ABS parts was halted for varying
lengths of time. Specifically, the injection molding pa-
rameters (barrel, hot runner, and melt-drop tempera- ture (Tg) is expected for degraded samples.
tures, injection speed and pressure, hold time and pres- To that end, DSC curves were obtained for the
sure, and mold temperature) were set to nominal operat- injection molded parts at the various hold times. Typi-
ing conditions. Parts were made continuously under cal DSC curves for parts taken from the first shot after
these parameters for 15 min. The machine was then the 2, 5, 15, and 25 min hold times are shown in
stopped for a predetermined period (2, 5, 10, 15, or 25 Fig. 2. An increase in Tg is clearly discernible in these
min, with the mold open), after which time part produc- DSC curves as the hold time increased. It is also ap-
tion was resumed. The first five shots molded after the parent from the DSC curves that as the Tg increases,
stoppage were collected in individual sample bags, the transition also become broader and less distinct.
which were labeled with the hold time and the shot. Af- The rubber phase Tg results of parts from the first
ter these five shots had been collected, the machine was shot molded after the hold time are shown in Fig. 3, in
allowed to run for fifteen minutes to completely purge which Tg is plotted vs. molding machine hold time.
the barrel, hot runners, and melt drops of material, and The circles denote results obtained from samples
return the machine to a ‘steady-state’ condition before taken from the first shot, the squares from the second
the next hold time was run. shot, and the triangles from the fifth shot. It is clear
Oxidative onset temperatures (OOT) of parts from this figure that the glass transition temperature
from the first shot after a stoppage were determined of parts from the first shot increased dramatically as
under an air flow rate of 100 mL min–1. The results, the hold time increased. Surprisingly, however, the
summarized in Table 2, indicate that as the hold time glass transition temperature of parts from the second
shot were typical of parts made during normal pro-
114 J. Therm. Anal. Cal., 83, 2006
DETECTION OF DEGRADATION OF ABS MATERIALS VIA DSC
Fig. 4 Peak load at break as a function of hold time for first
and fifth shot samples
Fig. 2 Second-heat DSC curves of first-shot parts after hold
times of 2, 5, 15 and 25 min tions indicate that enough material for 3–5 shots re-
side in this melt-drop zone. If higher temperatures
alone were sufficient to cause degradation, one might
expect that parts from the first 3–5 shots would show
significant degradation. The fact that only parts from
the first shot undergo any significant degradation in-
dicates the necessity of oxygen for the degradation
process. Only the material at the melt-drop exit is ex-
posed to oxygen, and thus only this material begins to
degrade. Exposing ABS to the temperatures used in
this study for up to 25 min, in the absence of oxygen,
did not lead to measurable degradation. As the length
of the hold time increases, the extent of degradation
increases as oxygen is able to diffuse further into the
Fig. 3 Effect of injection molding hold time on rubber phase melt-drop zone. Thus, material further back from the
glass transition gate begins to degrade as well.
duction, except in the case of the 25 min hold time. In Conclusions
this case, the Tg was between –45 and –50°C. Parts
from the fifth shot had normal Tg’s for all hold times. It has been shown that DSC is a sensitive and useful
Impact testing was also performed on the injec- tool for monitoring the degradation of butadiene-con-
tion molded parts. A portion of the part suitable for im- taining polymers such as ABS. Correlation of the rub-
pact testing was struck at 0.42 m s–1 (1000 in/min), and ber phase Tg to mechanical properties was demon-
the peak load was recorded. The results are summarized strated. Furthermore, degradation of the ABS only
in Fig. 4. Peak load at break for samples from the first proceeds at an appreciable rate at melt-processing
shot after the hold time (diamonds) show a marked de- temperatures in the presence of oxygen.
crease as a function of hold time. Parts from the fifth
shot (squares) exhibited impact strengths only some-
what less than for samples produced under ‘steady-state’ Acknowledgements
injection molding conditions.
Considering the data in Figs 3 and 4 together, it is The authors thank Baxter Healthcare Corporation for permis-
clear that degradation of the butadiene phase of ABS (as sion to publish this work.
indicated by the shift to higher temperatures of the glass
transition) led to losses in impact strength of injection References
molded parts. Extensive cross-linking of the butadiene
during degradation renders the material almost useless, 1 J. B. Adeniyi, Eur. Polym. J., 20 (1984) 291.
with peak loads at break as low at 1 pound. 2 E.M. Borroff and R.E. Spratling, Plastics in Telecom. II,
The observation that only parts from the first 1978, Paper no. 13a.
shot after a mold stoppage are adversely affected in 3 M. Ghaemy and G. Scott, Polym. Degrad. Stab., 3 (1981)
233.
terms of rubber phase Tg and impact strength is impor-
4 M. M. Shapi, Thermochim. Acta, 175 (1991) 25.
tant. The material for this first shot is in the melt-drop
region of the injection molding press, which is the
hottest zone in the process. However, volume calcula- DOI: 10.1007/s10973-005-6977-0
J. Therm. Anal. Cal., 83, 2006 115
