ABS (acrylonitrile-butadiene-styrene) is one of the most frequently
used polymers in electrical and electronic equipment, as well as having
widespread applications in automobiles, communication instruments, and
other commodities. A report by the Association of Plastics Manufacturers
in Europe (APME)  has suggested that the use of recycled plastics in
the electrical and electronic sector could be increased if more
polystyrene (PS) and ABS recyclate were to become available. This
emphasizes the importance of studying the recycling of ABS as an aid to
reducing environmental, economic, and energy issues.
Several studies have been carried out on the mechanical recycling
of ABS. For example, Potente and Gao  studied the recyclability of
injection molded lampshade parts based on ABS. They found that, after
regrinding and injection molding, the mechanical properties were not
significantly affected, apart from a slight reduction in the notched
Izod impact strength. Their gel permeation chromatography (GPC)
investigation showed that only slight molecular degradation occurred
during the injection moulding and/or regrinding. Thus, they considered
that the main reason for the reduction in impact strength could be the
volatilization of some of the additives used in the virgin material.
In another study, pieces of ABS from computer monitor casings were
granulated followed by thermal processing in a torque rheometer .
Then the material was regranulated followed by injection molding into
mechanical test specimens. This method was used to simulate a typical
commercial recycling procedure. After this reprocessing, tensile
strength decreased only by a few MPa, tensile modulus increased
slightly, strain to failure decreased from about 11% to 6%, and
un-notched impact strength decreased from 44 to 31 kJ/[m.sup.2]. It was
thought that polymer degradation might have caused the reduction of
impact strength and a slight increased stiffness of the recycled
Some basic studies on multiple recycling of ABS have been carried
out. For example, the effects of reprocessing conditions on the
mechanical properties of ABS have been evaluated by varying temperatures
and dwell times . In this case, the material was ground and remolded
five times. It was found that the mechanical properties varied with
reprocessing temperature. Following the use of higher temperatures, Izod
impact strength decreased but there were increases in tensile strength
and modulus. Also, the variations of these mechanical properties
increased with longer processing dwell times. There was a good
correspondence between the changes in mechanical properties and the
observed structural variations. Degradation of the rubber phase
(evidenced by infrared spectroscopy) was thought to be the main reason
for the decrease in toughness. Following the most severe molding
conditions, toughness reduction was even greater, perhaps because of
degradation of the styrene-acrylonitrile (SAN) phase.
In another study, an ABS material was injection molded and
recovered for five cycles . During this multiple reprocessing, the
strain to failure showed a very slight tendency to decrease. Failure
strength slightly increased after five processing cycles. The toughness
measured by notched impact strength reduced continuously. The changes of
these important properties were probably due to the degradation of the
soft polybutadiene component, which was seen from infrared and dynamic
mechanical analysis results.
Kim and Kang reprocessed three ABS resins five times using an
extruder . After extrusion, the glass transition temperature of the
SAN phase was not changed. The effect of repeated extrusions on
mechanical properties such as tensile strength, strain to failure and
hardness was small. However, impact resistance of all materials
decreased after recycling, especially the impact resistance of ABS with
the highest polybutadiene content. The reason for the decrease in the
impact strength was again thought to be the degradation of the
polybutadiene component in ABS.
In the present study, the effects of reprocessing on both the
polymer and any small molecules (including additives) present in ABS
plastics from waste computer housings are studied. Through the study,
the reasons for the changes in mechanical properties after reprocessing
Several waste computer equipment housings were screened using FTIR
(Fourier transform infrared spectroscopy) and gel permeation
chromatography (GPC) to ensure that (i) they were ABS, (ii) they did not
contain high molecular weight additives, and (iii) the proportion of
additives and oligomers (containing double bonds) was relatively low.
Three individual housings (ABS1, ABS2, and ABS3) were selected for the
subsequent series of recycling experiments.
Material Reprocessing and Sample Preparation
The recycling process was simulated using an appropriate
combination of granulator, torque rheometer, and injection molder.
ABS1 and ABS2 were used to assess the effects on impact and tensile
properties respectively, of reprocessing in the torque rheometer at
different temperatures and rotational speeds (shear rates). Granulated
particles of each individual housing were introduced into a torque
rheometer controlled at a specific temperature (190 or 230 or
270[degrees]C). A fixed time of 3 min was allowed for the charge to
reach the cavity temperature with the blades rotating at 10 rpm. Then,
the rotational speed was increased to 20, 60, or 100 rpm to process for
a further 10 min. Following this dwell time, the blades were stopped,
the torque rheometer was allowed to cool, and the plastic was removed
when the temperature of the processing cavity had reduced to
140[degrees]C. The solid material was granulated again and made into
samples for mechanical testing using a Ray Ran Injection Molder.
ABS3 was used to assess the effects of multiple reprocessing. The
housing was cut into pieces with a saw and granulated. To provide a
control sample (number of cycles, n = 0), some of these granulated
particles were directly injection molded into samples for mechanical
testing. To simulate one recycle (n = 1), the granulated particles were
processed in the torque rheometer at 230[degrees]C for 3 min at 10 rpm,
and 10 min at 60 rpm, cooled, granulated, and injection molded. For
multiple recycling (n = 2, 3, and 4), the material underwent additional
cycles of torque rheometer processing, cooling, and granulating before
the final injection molding.
In all cases, the granulated particles were heated in an oven at
80[degrees]C for about 4 h to remove any moisture immediately before
both thermal processing in the torque rheometer and injection molding.
Dynamic Mechanical Thermal Analysis (DMTA)
A rheometric dynamic mechanical thermal analyzer MK III was used to
test samples in bending geometry using a single cantilever type clamp
configuration at a frequency of 1 Hz, with a heating rate of
Infrared Spectroscopic Analysis by FTIR
Infrared spectra of bulk samples were recorded on a Perkin-Elmer
FTIR spectrometer in ATR (attenuated total reflectance) mode. The
absorbance of trans-2-butene-1,4-diyl moieties at 966 [cm.sup.-1] and
the absorbance of nitrile moieties at 2238 [cm.sup.-1] were compared
with the absorbance of styrene moieties at 1603 [cm.sup.-1], using the
ratios [D.sub.1] and [D.sub.2] as defined below . Values of
absorbance were determined using the baseline method.
[D.sub.1] = Absorbance at 966 [cm.sup.-1]/Absorbance at 1603
[D.sub.2] = Absorbance at 2238 [cm.sup.-1]/Absorbance at 1603
ABS is sensitive to oxidation because of the presence of
polybutadiene components, which act as oxidation sensitizers and lead to
formation of carbonyl groups, which absorb at 1680-1750 [cm.sup.-1] .
In the present study, changes of absorption in this region were also
GPC System for Assessing Smaller Molecules. A sample of plastic
(1.5 g) was placed in a 32 ml long bottle, 25 ml of tetrahydrofuran
(THF) was added and the bottle was sealed. The mixture was stored for 24
h to allow the plastic to dissolve, and then put in an ultrasonic bath
for 3 h. Subsequently, the solution was kept in the dark until solid
particles precipitated completely. GPC samples were drawn from the top
of this solution. The molecular weight distribution was analyzed at room
temperature with a GPC system equipped with four PLgel columns
(dimensions 300 x 7.5 mm, 10 [micro]m particle size, pore sizes 10,
[10.sup.2], [10.sup.3], [10.sup.4] nm) and an ultraviolet detector (at
254 nm). THF was used as the eluent at a flow rate of 1.0 ml/min. The
GPC system was calibrated with polystyrene (PS) standards (of molecular
weights 1,447,000, 401,340, 24,150, 9100, 2700, and 687) obtained from
Aldrich Chemical Company.
GPC System for Assessing Large Polymeric Molecules. GPC analysis of
the larger polymeric molecules was carried out at RAPRA Technology Ltd.
Sample solutions were prepared by adding chloroform (10 ml) to plastic
samples (20 mg), sealing the tube, and leaving the mixture overnight to
dissolve. The solutions were thoroughly mixed and filtered through a
glass fiber prefilter and a 0.2 [micro]m polyamide membrane prior to the
chromatography. The molecular weight distribution was determined at
30[degrees]C using two PLgel columns (mixed bed-B, 300 mm length, 10
[micro]m particle size) with refractive index and differential pressure
detectors. Chloroform was used as the eluent at a flow rate of 1.0
ml/min. The GPC system was calibrated with a mixture of two PS standards
(of molecular weights 3,053,000 and 30,300).
Gas Chromatography/Mass Spectrometry (GC/MS)
Extraction Procedure . Some broken samples of ABS2 and
granulated particles of ABS3, which had undergone thermal reprocessing
0, 1, 2, or 3 times in the torque rheometer, but had not been injection
molded, were used for identification. A sample of plastic (6-10 g) was
placed in a 100 ml round flask and dichloromethane (40-60 ml) was added
as extraction agent. The flask was sealed and the mixture was allowed to
stand for 24 h, then ultrasonicated for 2 h, and finally filtered
through glass wool. The undissolved solid was recovered and extraction
was performed again, this time with dichloromethane (30-40 ml) and for
30 min in the ultrasonic bath. The two extracts were collected together
and methanol (80-160 ml) was added to precipitate any polymer. The
solution was filtered through a glass fiber filter, and then
concentrated to 2-3 ml using a rotary evaporator. After filtering again
through a glass fiber filter, the concentrated extract was analyzed by
Identification of Extracts by GC/MS. GC/MS was performed at the
National Mass Spectrometry Service Centre at University of Wales
Swansea. A Fisons GC 8000 gas chromatograph with Fisons MD 800 mass
spectrometer (Thermoquest UK) was used. The gas chromatograph was
equipped with a fused silica capillary column with 100%
polydimethylsiloxane liquid phase (30 or 15 m x 0.25 mm i.d.; film
thickness 1.0 [micro]m; Chrompack DB-1). The total ion current (TIC)
chromatograms with long retention times were obtained by using the
longer column, while the chromatograms with short retention times were
obtained by using the shorter column. The chromatographic conditions
were as follows: initial temperature 40[degrees]C held for 1 min, then
raised at 8[degrees]C/min up to 300[degrees]C and held for 20 min;
carrier gas, helium (head pressure 31 kPa); injection temperature,
220[degrees]C; injection, splitless; injection volume, 1 [micro]l; Mass
spectrometer: source temperature, 200[degrees]C; electron energy, 70 eV;
electron current, 200 [micro]A. GC/MS interface temp: 300[degrees]C. For
identification of mass spectra, the NIST/EPA/NIH Mass Spectra Library
(version 2.0) was used.
Impact and Tensile Testing
The notched Izod impact tests were performed (on ABS1 and ABS3) at
room temperature using a Ray Ran Universal Pendulum Impact System. All
injection molded samples were 61.0 mm long, 12.0 mm wide, and 4.0 mm
thick with a notch depth of 2.0 mm. Average values and standard
deviations were calculated from at least 5 samples of each material.
Tensile testing was carried out on a Hounsfield H25K-S Benchtop
Testing Machine at room temperature. Dogbone-shaped samples were tested
to failure at an extension rate of 1 mm/min (ABS2) or 5 mm/min (ABS3).
The gauge lengths of the injection molded samples were 36.6 mm (ABS2)
and 30.1 mm (ABS3), and the width and thickness were 4.0 mm. Average
values and standard deviations were calculated from at least 5 samples
of each material.
Analysis of Fracture Surfaces by Scanning Electron Microscopy (SEM)
The fracture surfaces of selected samples were gold coated and
observed in a Philips XL 30 CP scanning electron microscope, normally
operating at 15-20 kV.
RESULTS AND DISCUSSION
Effects of Reprocessing at Different Temperatures and Rotational
Speeds. Some DMTA results for ABS2 are shown in Fig. 1. The peak between
-110 and -60[degrees]C corresponds to the rubber phase . It can be
seen that after reprocessing at 270[degrees]C, the glass-transition
temperature ([T.sub.g]) of the rubber phase increased, suggesting that
crosslinking reactions may have occurred in the rubber phase. There was
negligible variation in [T.sub.g] of the SAN phase, which was 110 [+ or
-] 1[degrees]C for all ABS2 samples. The same trends were observed in
the DMTA spectra of ABS1.
[FIGURE 1 OMITTED]
Absorbance ratios for different bands in the FTIR spectra of ABS1
and ABS2 are shown in Fig. 2. There is a suggestion that the ratio
([D.sub.1]) of trans-2-butene-1,4-diyl moieties to styrene moieties for
ABS2 was slightly higher after reprocessing at 190[degrees]C, compared
with before reprocessing (although it was not so clear for ABS1). Such
an effect may have been caused by some volatile molecules containing
benzene rings being vented out of the material during reprocessing.
Figure 2 also indicates that [D.sub.1] was slightly lower for samples
reprocessed at 270[degrees]C, compared with those reprocessed at
190[degrees]C at the same rotational speed, for both ABS1 and ABS2. This
lower [D.sub.1] may have been mainly caused by crosslinking of the
rubber phase when reprocessing at the higher temperature. No increase in
absorption due to carbonyl groups could be observed in the spectra after
reprocessing, even after reprocessing at 270[degrees]C (see Fig. 3).
There was no significant change in the ratio ([D.sub.2]) of the
absorption of nitrile moieties to the absorption of styrene moieties
after reprocessing at the various temperatures and rotational speeds.
This is consistent with the DMTA results, which also showed that
reprocessing temperature had little effect on the SAN phase. The SAN
phase, therefore, seems to be relatively stable under these processing
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
The results using the GPC column designed for assessing the smaller
molecular species of ABS1 are shown in Fig. 4. The retention time of the
high molecular weight polymer molecules is 18-30 min. The second peak
(30-35 min) is attributed to oligomers of ABS, additives, and their
derivatives. The third peak (35-40 min) is attributed to more small
molecules including for example monomers of ABS. Figure 4 indicates that
the proportion of small size components eluting between 25 and 35 min
increased after reprocessing at high temperature, suggesting that chain
scission reactions of ABS might have occurred, but the degree of such
scission reactions was very low. The third peak became lower after
reprocessing, presumably because of volatilization of the smallest
molecules. Similar trends were observed in the equivalent chromatograms
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
The results from the GPC column used to assess the higher molecular
weight species are shown in Figs. 5 and 6. It can be seen that the
rotational speed had no apparent effect on the size distribution of the
polymeric components (see Fig. 5). The proportion of large size
components increased slightly after reprocessing at the highest
temperature of 270[degrees]C, but the increase in the high molecular
weight component was greater in ABS2 than ABS1 (Figs. 5 and 6). The
major component of soluble polymer molecules is SAN. Some literature has
suggested that the viscosity of SAN reduces after multiple reprocessing
at high temperature [4, 10]. This would suggest that the molecular
weight of SAN tends to reduce during processing. The increased
proportion of high molecular weight components in ABS2 observed in Fig.
6 is not easily interpreted. It was not seen to any great extent in ABS1
(see Fig. 5) and so does not appear to be common feature. Its occurrence
in Fig. 6 may be related to the crosslinking of polybutadiene as
suggested by the DMTA experiments, if small amounts of polybutadiene had
originally been grafted onto the SAN molecules, but this is a
speculative interpretation and needs further investigation.
[FIGURE 6 OMITTED]
[FIGURE 7 OMITTED]
Effects of Multiple Reprocessing. DMTA results for ABS3 show that
after reprocessing at 230[degrees]C for 4 cycles, [T.sub.g] of the
rubber phase increased (see Fig. 7). At the same time, with increasing
number of reprocessing cycles, the value of the maximum tan [delta] of
the rubber phase decreased, and the peak broadened. One explanation for
such observations could be that crosslinking and thermo-oxidative
degradation occurred in the rubber phase . Multiple reprocessing had
no apparent effect on the tan [delta] peak of the SAN phase of ABS3,
[T.sub.g] remaining at about 115[degrees]C.
The FTIR results for ABS3 show that, with an increase in the number
of reprocessing cycles, the ratio ([D.sub.1]) of trans-2-butene-1,4-diyl
moieties to styrene moieties decreased significantly (see Fig. 8). This
shows degradation reactions occurred in the rubber phase while the SAN
phase was relatively stable. Figure 8 shows that the ratio ([D.sub.2])
of nitrile moieties to styrene moieties reduced slightly during multiple
reprocessing. Additionally, FTIR spectra (see Fig. 9) after 4
reprocessing cycles showed evidence for a small increase in intensity of
peaks in the range 1680-1750 [cm.sup.-1], which could be due to
formation of carbonyl groups.
[FIGURE 8 OMITTED]
[FIGURE 9 OMITTED]
The results of the GPC study of the smaller molecular species of
ABS3 are shown in Fig. 10. It may be seen that the proportion of small
size components eluting between 27 and 35 min increased during multiple
reprocessing as also happened during higher temperature single step
reprocessing. This suggests that chain scission reactions occurred.
Also, the third peak (35-40 min) decreased with increasing number of
reprocessing cycles, probably because of volatilization of the smallest
molecules. Changes in the larger polymeric molecular components are
shown more clearly in the GPC data of Fig. 11. Again, this shows some
evidence of an increase in small size polymer molecules after 4
reprocessing cycles, perhaps because of greater oxidative degradation
during multiple reprocessing. However, the degree of chain scission
reactions is still low. Additionally, the GPC results in Fig. 11 also
show that the proportion of large size polymeric components slightly
increased after multiple reprocessing, a similar effect (but to a lesser
extent) to that seen for single recycling of ABS2 at 270[degrees]C.
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
Gas Chromatography/Mass Spectroscopy
The TIC gas chromatograms of extracts from ABS2, before
reprocessing, after reprocessing at 230[degrees]C and 100 rpm, and after
reprocessing at 270[degrees]C and 20 rpm are shown in Figs. 12a-c,
respectively. There are several large peaks in the chromatogram of the
ABS2 'before reprocessing' sample (Fig. 12a). Mass spectra of
these peaks show that most of them might be given by oligomers of ABS
and their derivatives, except the peaks at about 5.7 min (Scan 134) and
about 9.0 min (Scan 330). The mass spectrum of the peak at 5.7 min
suggests that the compound might be 2-phenyl-2-propanol. The peak almost
disappeared following reprocessing. The peak at 9.0 min has been
identified to be 2,4-di-tert-butylphenol, a common antioxidant, by
comparison with a standard. However, this peak increased after
reprocessing, which implies that during reprocessing, some component(s)
in ABS2 decomposed to give 2,4-di-tert-butylphenol. Because of the
increase in the amount of material following reprocessing, it is
possible that all or part of the 2,4-di-tert-butylphenol present before
reprocessing is a result of decomposition of the component(s) over the
lifetime of the material. After reprocessing (Fig. 12b and c), two peaks
(at 11.4 (Scan 474-475) and 12.67 min (Scan 550)) increased
substantially, particularly after reprocessing at 270[degrees]C. The
library search results for the mass spectra of these peaks show that the
peak at 11.4 min might be hexadecanenitrile, and the peak at 12.67 min
might be octadecanenitrile. They are probably decomposition products of
the lubricant EBS . The main component of EBS is
N,N'-ethylenebis-stearamide, but EBS also always contains
ethylene-N-palmitamide-N'-stearamide. When it decomposes,
hexadecanenitrile and octadecanenitrile can be formed.
[FIGURE 12 OMITTED]
[FIGURE 13 OMITTED]
The chromatograms of extracts from multiply reprocessed ABS3 are
shown in Fig. 13. Figures 13a-d are the extracts of ABS3 after
reprocessing at 230[degrees]C and 60 rpm for 0, 1, 2, and 3 cycles,
respectively. During this multiple reprocessing, some small molecules
were lost continuously. For example, the peak at 9.6 min (Scan 368-370)
became smaller as the number of cycles increased, with the largest
reduction occurring during the first reprocessing cycle. The other
apparent changes are that the peaks at 17.5 min (Scan 840-841), 23.2
min, and 25.9 min (Scan 1337-1341) grow with the increasing number of
reprocessing cycles. The peak at 17.5 min, which has been confirmed to
be 2,4-di-tert-butylphenol by comparison with a standard, increased
during reprocessing. This mirrors the situation found during
reprocessing of ABS2. The compounds at 23.2 and 25.9 min could be
hexadecanenitrile and octadecanenitrile. This indicates that ABS3
perhaps contains EBS, like ABS2. Increasing amounts of EBS would
decompose at 230[degrees]C with increasing numbers of reprocessing
cycles. The peak at 24.3 min has been identified to be palmitic acid,
which is another lubricant commonly used in plastics, by comparison with
a standard. After three reprocessing cycles, the peak became smaller
relative to other peaks, which indicates that part of the palmitic acid
[FIGURE 14 OMITTED]
[FIGURE 15 OMITTED]
From the GC/MS results above, it may be seen that some additives
have been identified as decomposing during reprocessing, particularly
during reprocessing at 270[degrees]C and multiple reprocessing. It seems
likely that other similar additives would also be lost by decomposition
under these reprocessing conditions.
[FIGURE 16 OMITTED]
Impact and Tensile Properties
The impact and tensile properties of ABS plastics before
reprocessing and after reprocessing at different temperatures and
rotational speeds are shown in Fig. 14.
From Fig. 14a, it is particularly evident that the impact strength
reduced with increasing reprocessing temperature. Various factors have
been involved in reduction of impact strength, including reduction in
entanglements stabilized by rubber particles , chain scission of the
graft between the SAN matrix and the rubber phase . As well as these
factors, degradation including crosslinking of the rubber phase (as
evidenced from the DMTA data presented above) is likely to be a
significant factor in the reduced impact strength after reprocessing at
higher temperatures. There is also evidence in Fig. 14a that, at the
same reprocessing temperature, the impact strength is slightly reduced
with increasing rotational speed. Oxidation of the polymer, particularly
the rubber component, might have been a significant factor since higher
rotational speeds provide more opportunity for oxygen to diffuse into
the polymer (A small amount of degradation of the rubber phase would not
significantly change the molecular weight distribution ).
The tensile properties of ABS2 after reprocessing are shown in Fig.
14b and c. Compared with the properties after reprocessing at 190 and
230[degrees]C, both the tensile modulus and strength were slightly
higher after reprocessing at 270[degrees]C. Potential explanations for
these higher stiffness and strength values include the loss of small
molecules (including lubricant molecules) , and crosslinking of the
rubber phase. Figure 14 also indicates that rotational speed had little
effect on the tensile properties. The strain to failure in these ABS
samples was typically about 10%, and although there was more scatter in
these data, there were no significant differences between the average
figures for each processing condition. This may have been because the
effects of defects (formed during material recycling and sample
preparation) were more significant than the effects of other factors,
such as processing temperature.
Figure 15 shows how the mechanical properties of ABS3 varied as a
function of the number of reprocessing cycles, and it is clear that the
impact strength is the most sensitive to reprocessing. From Fig. 15a, it
can be seen that the impact strength reduced significantly as ABS3
underwent multiple reprocessing. The greatest reduction was during the
first reprocessing cycle, impact strength dropping by about 44%. With
subsequent reprocessing cycles, the impact strength diminished more
slowly. This could be explained by the major loss in volatile molecules
found during the first cycle since a loss in impact strength has been
associated with loss of these small molecules . Other potential
causes of a reduction in impact strength include crosslinking of the
rubber phase and scission of the ABS polymer chains. This present study
would seem to suggest that the loss of volatiles is a significant factor
since the results indicate that volatiles are most substantially reduced
during the first reprocessing cycle, whereas crosslinking and scission
occur more gradually at each reprocessing cycle.
Figure 15b and c show the tensile properties of ABS3 as a function
of the number of reprocessing cycles. It may be seen that the modulus of
ABS3 did not change significantly with increasing number of reprocessing
cycles, but tensile strength increased slightly. These small increases
in strength may again be related to loss of the small molecules
(including lubricant molecules)  and degradation (including
cross-linking) of the rubber phase .
Morphology of Fracture Surfaces
Figure 16 shows the morphology of the impact fracture surfaces of
ABS1 after reprocessing at different temperatures and rotational speeds.
Compared with the fracture surfaces after reprocessing at 20 rpm (Figs.
16a and c), more holes can be seen in the matrix after reprocessing at
100 rpm (Fig. 16b and d). This suggests that degradation of ABS
increased. On the other hand, on increasing the reprocessing temperature
from 190 to 270[degrees]C, deformation of the matrix reduced. This is
possibly related to crosslinking of the rubber phase. A high degree of
crosslinking of the rubber phase can reduce the ability of the rubber
particles to cavitate. According to recent research, cavitation of the
rubber particles can facilitate deformation of the matrix .
Possibly, due to degradation of ABS, particularly crosslinking and
scission of the rubber phase, the material had become very brittle after
reprocessing at 270[degrees]C and 100 rpm (Fig. 16d).
The fracture surfaces from impact tested samples in the multiple
reprocessing studies of ABS3 are shown in Fig. 17. The micrograph (Fig.
17a) of a sample, which had not been reprocessed in the torque rheometer
(n = 0), exhibits profuse cavitation and extensive matrix shear
yielding. After one reprocessing cycle in the torque rheometer (n = 1),
most of the area shows relatively brittle fracture features (Fig. 17b).
A number of small holes appear on the fracture surface, and the size of
the holes corresponds to the size of rubber particles. Figure 17b
indicates that most rubber particles did not heavily cavitate and
facilitate shearing of the matrix, and that cracking was more generally
at the interface between rubber particles and the matrix. This weakened
interface, which may have resulted from the observed degradation of the
rubber phase, would lead to a reduction in impact strength. It has been
suggested that a small amount of degradation of the rubber phase could
promote a decrease in adhesion with the SAN matrix, causing stress
concentrations and leading to rapid crack propagation between rubber
particles and matrix . The effect of reprocessing was more apparent
after four reprocessing cycles (Fig. 17c). The degree of deformation of
the matrix is noticeably further reduced and the cracks propagated along
the interface between particles and matrix. Many large particles can be
seen on the fracture surface, together with a number of relatively large
voids in the matrix, from which the large particles had been pulled out.
The large particles are possibly due to the agglomeration of the rubber
particles, but this could not be confirmed by SEM. Figure 17c suggests
that, with increasing reprocessing, the original microstructure of the
ABS copolymer is being destroyed. This seems quite likely to be a
significant factor in explaining the reduction in impact strength during
[FIGURE 17 OMITTED]
The results suggest that two of the most important factors
affecting the properties of reprocessed ABS are loss of the smaller more
volatile molecules and degradation (chain scission and particularly
crosslinking) of the rubber (polybutadiene) component. It appears that
loss of the smaller molecules occurs more readily, that is, even at the
lowest reprocessing temperatures and just a single reprocessing cycle,
there is significant reduction in the quantity of small molecules. As
the recycling conditions become more and more severe, such as higher
temperatures or multiple reprocessing, further small volatile molecule
loss occurs but the amount and significance of these further losses is
progressively reduced. On the other hand, degradation of the
polybutadiene phase seems to become progressively more significant with
increasingly severe reprocessing conditions. Thus, with increasing
reprocessing temperature or increasing number of reprocessing cycles,
the significance of the loss of volatile molecules reduces and the
importance of polybutadiene degradation increases. These changes that
occur during reprocessing alter the ABS morphology, in particular
reducing the interfacial bond strength between the SAN matrix and PB
rubber particles. These changes have significant effects on the
Reprocessing of ABS has a much more severe effect on impact
properties than tensile properties. This is consistent with the changes
outlined above, namely loss of small molecules, degradation of the
polybutadiene phase and relatively little change in the SAN phase.
Reduced numbers of small molecules and degradation of the rubber phase
will both contribute to loss of impact resistance. At the same time,
since the load bearing SAN phase is relatively unaffected by the
reprocessing, then the tensile properties also remain relatively
unchanged. Even the slight increase in tensile properties with
increasing temperature might be expected to result from degradation of
the rubber phase and loss of small molecules (including lubricants).
We thank RAPRA for some results of GPC and the EPSRC NMSSC at
University of Wales Swansea for mass spectra. X B thanks the Chinese
Government for a studentship and the University of Wales Swansea for a
bursary to cover her tuition fees.
1. APME, "Assessing the Potential for Post-Use Plastics Waste
Recycling-Predicting Recovery in 2001 and 2006," APME summary
report, APME, Brussel (1998).
2. H. Potente and X. Gao, Conference Proceedings of ANTEC'94,
3. L.B. Brennan. D.H. Isaac, and J.C. Arnold, J. Appl. Polym. Sci.,
86, 572 (2002).
4. A. Casale, O. Salvatore, and G. Pizzigoni, Polym. Eng. Sci., 15,
5. J.I. Eguiazabal and J. Nazabal, Polym. Eng. Sci., 30, 527
6. J.K. Kim and C.K. Kang, Polym. Plast. Technol. Eng., 34, 875
7. J. Haslam and H.A. Willis, Identification and Analysis of
Plastics, 2nd ed., Iliffe, London, 447 (1972).
8. D. Dong, S. Tasaka, S. Aikawa, S. Kamiya, N. Inagaki, and Y.
Inoue, Polym. Degrad. Stab., 73, 319 (2001).
9. C. Nerin, J. Salafranca, C. Rubio, and J. Cacho, Food Addit.
Contam., 15, 842 (1998).
10. S. Bastida, C. Marieta, J.I. Eguiazabal, and J. Nazabal, Eur.
Polym. J., 31, 643 (1995).
11. E.M.S. Sanchez, M.I. Felisberti, C.A.R. Costa, and F.
Galembeck, Polimeros, 13, 166 (2003).
12. F.C-Y. Wang and W.C. Buzanowski, J. Chromatogr. A, 891, 313
13. N.K. Kalfoglou and C.E. Chaffey, Polym. Eng. Sci., 19, 552
14. J.B. Adeniyi, Eur. Polym. J., 20, 291 (1984).
15. R.B. Seymour and C.E. Carraher, Structure-Property
Relationships in Polymers, Plenum, New York, 65 (1984).
16. F. Ramsteiner, W. Heckmann, G.E. Mckee, and M. Breulmann,
Polymer, 43, 5995 (2002).
17. R. Balart, J. Lopez, D. Garcia, and M.D. Salvador, Eur. Polym.
J., 41, 2150 (2005).
Xiaojuan Bai, (1,2) D.H. Isaac, (3) K. Smith (1)
(1) Department of Chemistry, University of Wales Swansea, Singleton
Park, Swansea SA2 8PP, United Kingdom
(2) School of Material Science and Chemical Engineering, China
University of Geosciences, Wuhan 430074, The People's Republic of
(3) Materials Research Centre, School of Engineering, University of
Wales Swansea, Singleton Park, Swansea SA2 8PP, United Kingdom
Correspondence to: X. Bai; e-mail: firstname.lastname@example.org