Polymer Degradation and Stability 91 (2006) 527e534

www.elsevier.com/locate/polydegstab

Kinetic analysis of thermal degradation of recycled

polycarbonate/acrylonitrileebutadieneestyrene mixtures
from waste electric and electronic equipment
R. Balart a, L. Sánchez a, J. López a, A. Jiménez b,*
a
Polytechnic University of Valencia, Department of Mechanical and Materials Engineering, 03801 Alcoy, Spain
b
University of Alicante, Department of Analytical Chemistry, Nutrition and Food Sciences, PO Box 99, 03080 Alicante, Spain
Received 20 October 2004; accepted 4 January 2005
Available online 3 October 2005

Abstract

The study of properties in waste polymeric materials is important for a further recycling and eventual reuse. The determination of thermal
properties of PCeABS mixtures obtained from waste electric and electronic equipment (WEEE) by using thermogravimetric analysis (TGA) has
been carried out in this work. Kinetic parameters, such as apparent activation energies, have been calculated by using the autocatalytic model,
which has shown a good correlation with experimental data. A good agreement between calculated and experimental results was observed for
PC-based mixtures. This is an indication that the addition of a high amount of ABS permits the creation of a network formed by elastomeric
domains which leads to a general increase in heterogeneity of the mixture.
Ó 2005 Elsevier Ltd. All rights reserved.

Keywords: Thermal properties; Kinetic parameters; Polycarbonate; ABS; Thermogravimetric analysis

1. Introduction Production and use of electrical and electronic equipment

(EEE), such as TV sets, computers, mobile phones and
Among the great variety of household materials which need many other daily-life items, are dramatically increasing over
to be recovered after use to increase their value, home electri- the last years. This means that the production of waste electric
cal and electronic appliances are one of the most important for and electronic equipment (WEEE) is also increasing and it is
further recycling. Since not long ago, these products after their a problem that governments and policy makers have to handle.
useful life were simply shredded and landfilled after minor re- In USA, a recent study predicted that over 315 million com-
covery of metals. The residue of the breaking-up process at puters would be at end of their life by the year 2004 [1].
private shredding companies was called shredder dust and des- The situation is quite similar in Europe. Directives 2002/95/
ignated as hazardous waste because of its high heavy metal EC on the restriction of the use of certain hazardous substan-
content. One alternative way to plastics waste disposal is recy- ces in electrical and electronic equipment [2] and 2002/96/EC
cling. It is often necessary to separate mixtures of materials on waste electrical and electronic equipment [3] were de-
into individual plastics by resin in order to produce a useful signed to tackle the fast increasing waste stream of electrical
recycling product as some polymers are not compatible and electronic equipment and complemented European Union
when re-melting. measures on landfill and incineration of waste. Increased recy-
cling of electrical and electronic equipment is supposed to
* Corresponding author. Tel.: C34 96 590 3777; fax: C34 96 590 3781.
limit the total quantity of waste going to final disposal. Pro-
E-mail addresses: [email protected] (J. López), [email protected] ducers are responsible by law for taking back and recycling
(A. Jiménez). electrical and electronic equipment. Therefore, the design

0141-3910/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymdegradstab.2005.01.055
528 R. Balart et al. / Polymer Degradation and Stability 91 (2006) 527e534

and optimisation of recycling processes are important for elec- mechanical, chemical and thermal properties between materi-
tronic industries and a previous complete characterization of als in WEEE and original polymers and blends. Thus, while
materials after a first processing and use is absolutely neces- mechanical and chemical properties have been recently stud-
sary to get the complete information on the input materials ied [5], thermal degradation of WEEE polymer components
in the recycling process. has been only partially studied, in particular for residues of
Many polymeric materials are used in electronic equip- PVC and ABS mixtures [12].
ment, but two important components of these products are pol- Thermal degradation of the target components of WEEE
ycarbonate (PC) and acrylonitrileebutadieneestyrene blends (i.e. PC and ABS) has been the object of recent research. Ther-
(ABS). PC and its mixture with ABS show very effective mal degradation of PC has been studied either individually
flame retardant properties upon the addition of conventional [13e15], or blended with other materials, in particular Bisphe-
halogen and/or non-halogen flame retardant agents, and this nol-A [16e18]. The most significant products evolved in the
is the main reason of their wide use in electrical appliances. pyrolysis of PC have been identified. The main degradation
PC shows several good properties, such as dimensional stabil- pathways of polycarbonate have also been recently proposed
ity, flame resistance, high impact strength and high stability to [18]. On the other hand, there is some recent work on the ther-
different environmental conditions. Nevertheless, PC is rela- mal degradation of ABS terpolymer [12,19,20] in different
tively soft and its mechanical properties should be improved conditions. However, the thermal degradation of PCeABS
by the addition of reinforcement materials [4]. ABS blends blends after their processing and use in EEE has been only
possess good impact and mechanical properties due to the con- partially studied when considering PCeABS blends as
tributing properties of each of their components. Acrylonitrile a base material for montmorillonite-nanocomposites [21].
provides chemical resistance, heat resistance and toughness, Therefore, the thermal behaviour and stability of these materi-
while butadiene provides impact strength and styrene provides als in WEEE is still not well known and need to be studied.
rigidity and easy processing. In addition, their relatively low The main goal of this work is the preparation and optimisa-
cost makes it especially interesting for blending with the tion of physical blends based on PC and ABS obtained from
more expensive high performance engineering plastics. Both WEEE and the further determination of thermal and kinetic
materials can be found either individually or mixed in many parameters. This work is carried out to finally evaluate the in-
commercial blends used in electric and electronic equipment. fluence of each component in thermal degradation of blends,
During the last years some work has been carried out in the considering the complexity of the basic polymers, in particular
development of different methods to give added value to ABS. Kinetic parameters, as apparent activation energies (Ea)
WEEE [5,6]. In most of the cases, plastic residues generated have been calculated by the application of the autocatalytic
after the useful life in EEE, mainly technical and engineering method, which has been successfully applied to other polymer
polymers, still show good properties, for instance adequate systems [22,23].
mechanical properties, high chemical resistance and thermal
stability. Therefore, it is important from the industrial point
2. Experimental
of view that their recycling and eventual reuse, either directly
or more usually, reinforced with materials obtained from other
2.1. Materials
sources, like recycled PET, to improve some of their properties
[7]. Nevertheless, sometimes these processes, even being via-
As a raw material a mixture of PC and ABS obtained from
ble from the technical side, are not economically profitable
WEEE was used. These materials were supplied by Reciclados
[8]. The main difficulty industries have to afford is the separa-
del Plástico S.L. (Alcoy, Spain). As usual in commercial ma-
tion of the different materials in WEEE. This problem leads to
terials, a previous characterization was carried out in order to
several approaches to optimise the process. One of the most
know their mechanical properties (Table 1) and chemical
successful ones is the definition of separation systems based
structure by infrared spectroscopy (Fig. 1a and b) in com-
on the physico-chemical properties of materials to make recy-
parison with commercially pure and virgin polymers. In addi-
cling of materials constituting WEEE economically profitable
tion, ABS showed a Melt Flow Index (MFI) (230
C/
[9,10]. Sometimes feedstock plastic recycling encompasses
5 kg) Z 20.45 g/10 min and MFI for PC was (230
C/
a number of technologies, which can be attractive recovery
5 kg) Z 6056 g/10 min. Glass transitions temperatures were
solutions for plastic waste not best suited to mechanical recy-
104.4
C and 146
C for ABS and PC, respectively.
cling, e.g. mixed packaging waste. Such technologies turn
As can be observed, from IR spectra, recycled ABS shows
plastics into basic chemicals for use, among other things, as
some differences with the virgin material. The presence of
building blocks for new plastics after a sometimes complicated
separation process [9]. However, this approach, which was
successfully applied to some kind of residues [10,11] is not Table 1
Mechanical properties of materials used in this study
easy to adapt to the treatment of WEEE because of the great
amount of materials (not only polymers) present in EEE. Material sR (MPa) A (%) E (MPa) Impact Hardness
energy (kJ/m2) shore D
Before proposing a recycling procedure for WEEE, a previ-
ous complete characterization of each individual component is ABS 37.78 11.9 1760 51.2 80
PC 59.51 60.5 1750 Do not break 76
necessary to get all the information on the differences in
 R. Balart et al. / Polymer Degradation and Stability 91 (2006) 527e534 529

a) 0,5
recycled ABS
0,4

absorbance
0,3

0,2

0,1

4000 3500 3000 2500 2000 1500 1000 500

wave number (cm-1)

1,0 comercial ABS, Polylac PA-714C

0,8
absorbance

0,6

0,4

0,2

0,0

4000 3500 3000 2500 2000 1500 1000 500

wave number (cm-1)

b) 2,5
recycled PC
2,0
absorbance

1,5

1,0

0,5

0,0

4000 3500 3000 2500 2000 1500 1000 500

wave number (cm-1)

1,6
comercial PC, Trirex 3122/3022
1,2
absorbance

0,8

0,4

0,0
4000 3500 3000 2500 2000 1500 1000 500
wave number (cm-1)

Fig. 1. IR spectra of ABS (a) and PC (b) compared to those of virgin materials.

a small but significant peak around 3300 cm1 which is not 950e1000 cm1 which indicates that the butadiene phase
observed in the non-treated polymer is an indication of the for- has decreased and some degradation should be expected.
mation of oxidized species which indicate that some kind of On the other hand, PC spectra for recycled and virgin
thermo-oxidation has been suffered by the polymer. This materials do not show significant differences (Fig. 1b). These
fact is confirmed by the lower intensity of peaks around materials have not suffered significant degradation during
530 R. Balart et al. / Polymer Degradation and Stability 91 (2006) 527e534

processing and use, as expected from the high thermal resis- 1,0
tance of PC.
ABS 100/PC 0
ABS 80/PC 20
0,8
2.2. Equipment and procedure ABS 20/PC 80

degree of conversion, α
ABS 0/PC 100

Dynamic thermal degradation measurements were per- 0,6

formed in a Mettler-Toledo 851e-TGA-SDTA thermal gravi-
metric analyser (Schwarzenbach, Switzerland). Temperature
0,4
was raised from room temperature up to 900
C using differ-
ent heating rates, between 5 and 30
C/min in 5
C/min inter-
vals. TGA tests were performed in alumina crucibles where 0,2
samples were placed without any previous treatment and
experiments were run immediately. TGA tests were carried
out in nitrogen environment using a flow rate 20 cm3/min in 0,0
550 600 650 700 750 800
order to remove all corrosive gases evolved during the degra-
temperature (K)
dation and to avoid thermo-oxidative degradation.
Kinetic parameters were determined by using the autocata- Fig. 2. TGA curves of PCeABS blends (% in weight) (2
C/min).
lytic model which is based on the fundamental kinetic
equation:
also supported when SEM microphotographs are observed,
      like in Fig. 3 for a PCeABS (80e20%) mixture. The dark
da Ea A
ln C Zln Cn lnð1  aÞCm lnðaÞ ð1Þ continuous region corresponds to PC phase and the ABS ap-
dT RT b
pears as clearer spheres over the background. It is observed
where a is the conversion, Ea the apparent activation energy, that while the largest spheres of ABS seem to be just lying
A the Arrhenius pre-exponential factor, b the heating rate for over the PC phase, some of the smallest ABS macromolecules
each individual experiment and n, m kinetic parameters to have established physical links modifying the PC continuous
be determined. structure. This observation is in agreement with another study,
FTIR tests were carried out by using a Satellite-3000 instru- which suggested that ABS can be well dispersed in PC without
ment from Mattson (Fremont, CA, USA) in the middle IR the aid of a compatibiliser [27] and that a co-continuous struc-
region (between 5000 and 600 cm1). Mechanical properties ture in industrial PCeABS blends is usual [28]. The reason of
were determined by using a universal test machine ELIB 30 this high compatibility in industrial blends is still under dis-
from Ibertest (Madrid, Spain) with charges between 5 and cussion, but a recent work proposed that the OH groups on
30 kN. SEM photographs were obtained by using a HITACHI the ends of PC chains might be linked with the anhydride
S-3000N Scanning Electronic Microscope (Hitachi Ltd. groups via esterification as a consequence of the higher polar-
Japan). ity of ABS segments [29].
In the case of ABSePC mixtures where ABS is the main
component, the action of PC is a general decrease in thermal
3. Results and discussion
stability with the overall process taking place at lower temper-
atures (Fig. 2). Nevertheless, this decrease is not enough to
Fig. 2 shows TGA curves at 2
C/min for four different
PCeABS blends. As can be observed, the degradation process
of both materials, either individually or after blending, takes
place in a relatively narrow temperature range. In general,
both degradation processes are quite fast, but some particular-
ities can be drawn. PC thermal degradation is observed at
higher temperatures than those of many other polymers, with
a high thermal stability even after processing and use [4,24].
Meanwhile, ABS shows a degradation process very similar
to other styrene derivatives [25,26].
The analysis of curves for PCeABS blends shows that the
addition of ABS to PC matrixes is not a simple sum of the con-
tributions of both materials. Even with additions of small
amounts of ABS (as low as 20% by weight) the degradation
curves are far from that of pure PC, but are similar to those
of pure ABS. Therefore, some interactions between both
polymers during thermal degradation have to be considered,
in particular the influence of ABS on thermal degradation of Fig. 3. Microphotograph of the structure of a PCeABS (80/20% in weight)
PC. The possibility of interactions between both polymers is (!5000).
 R. Balart et al. / Polymer Degradation and Stability 91 (2006) 527e534 531

determine a significant change in the thermal stability of

materials in real-life conditions. This behaviour is especially
26
evident in thermo-oxidative conditions where it has been re-
ported that the formation of free and peroxy radicals leads

Ln(dα/dT)+Ea/RT
to a fast initiation of the degradation. This observation was
confirmed by IR spectra as was previously indicated. But, 24
this general decrease in thermal properties is also observed
in inert atmosphere and would lead to the formation of
micro-cracks, chain scission and cross-linking which finally β = 20 K/min
β = 10 K/min
result in a general loss of mechanical properties in the 22
β = 5 K/min
PCeABS blend [20,30,31]. β = 2 K/min

The application of the autocatalytic kinetic model to the

TGA results obtained at different heating rates and for the
20
whole range of formulations (between 0 and 100% of ABS 0,0 0,2 0,4 0,6 0,8 1,0
in PC) allows us to calculate all kinetic parameters, such as ap- degree of conversion (α)
parent activation energies and partial reaction orders, n and m
by application of Eq. (1). Results for the calculation of Ea in Fig. 5. Determination of kinetic parameters by non-linear adjustment for re-
cycled PC at different heating rates.
five different mixtures are shown in Fig. 4. Two different be-
haviours can be observed for those mixtures where each poly-
mer is the main component. Thus, in the case of ABS-based energies have been determined, by a plot of ln(da/dT ) vs.
blends Ea values remain practically unchanged (considering 1/T by calculation of the a function and a non-linear fitting
the error in measurements and calculation) between 140 and and iteration process. Therefore, it is possible to determine
160 kJ/mol. These results are in agreement with those reported ln(A/b), n and m (Eq. (1)) simultaneously through the whole
by Zong et al. [21], who calculated kinetic parameters by apply- degradation process, as shown in Figs. 5 and 6 for pure ABS
ing classical Kissinger and Ozawa methods and obtained an av- and PC, respectively. Experimental data obtained directly
erage value of 142 kJ/mol. On the other hand, Ea values show from TGA curves have been plotted for all the PCeABS
a great variability for PC-based blends with a clear increase blends and non-linear fittings in the form of Eq. (2).
at high heating rates, getting values as high as 320 kJ/mol.
These high Ea values are not easy to explain, and can be attrib- yZKCn lnð1  xÞCm ln x ð2Þ
uted to the complexities in the thermal degradation mechanism
of PC which will be further discussed. Moreover, the effect of The fitting of experimental data (dotted points) with the
blending shows that while thermal behaviour is quite similar for calculated results (solid lines) is very good with correlation
all those ABS-based blends, the addition of growing amounts of coefficients (R2) higher than 0.99 in all cases, as observed in
PC (higher than 30% weight) results in a gradual separation of Figs. 5 and 6. This leads us to consider the autocatalytic method
the initial values and the subsequent approximation to the as adequate for the determination of kinetic parameters in
thermal degradation of PC. PCeABS blends.
The adjustment of the proposed kinetic model to these PCe The n and m values obtained by application of the model
ABS blends can be determined, once the apparent activation and Eq. (2) are shown in Fig. 7 for the whole range of

320 40

300 ABS 100/PC 0

ABS 90/PC 10
280 ABS 60/PC 40 35
ABS 20/PC 80
Ln(dα/dT)+Ea/RT

260 ABS 0/PC 100

Ea, (KJ/mol K)

30
240

220
25
200
β = 20 K/min
180 20 β = 10 K/min
β = 5 K/min
160 β = 2 K/min

140 15
0 5 10 15 20 25 30 0,0 0,2 0,4 0,6 0,8 1,0
heating rate, β (K/min) degree of conversion, α

Fig. 4. Determination of kinetic parameters by iterative calculation according Fig. 6. Determination of kinetic parameters by non-linear adjustment for
to the autocatalytic model for ABS (30
C/min). recycled ABS at different heating rates.
532 R. Balart et al. / Polymer Degradation and Stability 91 (2006) 527e534

a) 1,0
ABS 100% [E]
ABS 80% [E]
ABS 20% [E]
0,8 ABS 0% [E]

degree of conversion (α)

ABS 100% [M]
ABS 80% [M]
ABS 20% [M]
0,6 ABS 0% [M]

0,4

0,2

0,0
600 650 700 750 800
temperature (K)

b) 1,0
ABS 100% [E]
ABS 80% [E]
ABS 20% [E]
0,8

degree of conversion (α)

ABS 0% [E]
ABS 100% [M]
ABS 80% [M]
ABS 20% [M]
0,6
ABS 0% [M]

0,4

0,2

0,0
650 700 750 800 850
temperature (K)

Fig. 8. Comparison between the autocatalytic model results and experimental

data for different PCeABS mixtures at 2
C/min (a) and 30
C/min (b).
Fig. 7. Variation of ‘‘n’’ (a) and ‘‘m’’ (b) kinetic coefficients with composition
of mixtures at different heating rates.

heating rate of the TGA experiment. However, some general

mixtures. In this figure, a clear difference for mixtures with conclusions can be obtained from Fig. 8. The fitting of cal-
quantities of PC higher and lower than 40% in weight can culated values (solid lines) with experimental data (dotted
be observed. Thus, while both parameters (n and m) increase points) is better for low heating rates (Fig. 8a). This was ex-
their values for all ABS-based mixtures from 0 to 30% PC, pectable, as TGA tests at high heating rates in activated or
an inflexion is observed at 40% PC blends and a sudden autocatalytic processes show some differences between the
change in tendency is observed for PC-based mixtures by model and experimental results, as was previously observed
reaching a plateau (for n) and even decreasing clearly their for other polymer formulations [22]. The largest differences
values (for m) up to values similar to those of ABS-based are observed in the case of ABS-based blends at low con-
blends (between 0.3 and 0.6). This behaviour is general for version degrees, with a much better fitting for high a values.
the whole range of heating rates used in this study and can This can be due to the heterogeneity of the ABS-based ma-
be attributed to the chemical interaction between both phases, terials and the inherent limitations of the model in explain-
which conditions their thermal degradation, in particular for ing the thermal degradation of ABS, in particular during the
blends with similar quantities of both the components. initial stages of the process.
Once the kinetic parameters have been calculated, it is The use of the autocatalytic model is much more interesting
possible to integrate the differential Eq. (1) and obtain the for PC-based blends. As it can be observed, the superposition
results for the application of the model to every PCeABS of calculated curves and experimental data is good and near
mixture. Results for four different blends from 0 to 100% perfect for a O 0.4 and low heating rates. Therefore, it is re-
ABS at two heating rates in TGA experiments are shown markable that the autocatalytic model can be used for the de-
in Fig. 8a and b. The applicability of the model to each termination of kinetic parameters in recycled PCeABS blends
mixture is variable, depending on the composition and at any proportion and low heating rates (Fig. 8a), while higher
 R. Balart et al. / Polymer Degradation and Stability 91 (2006) 527e534 533

differences are observed at high heating rates in particular for [3] Directive 2002/96/EC of the European Parliament and of the Council of
ABS-based blends. It can be concluded that the autocatalytic 27 January 2003 on waste electrical and electronic equipment (WEEE).
[4] Jiang W, Tjong SC. Thermal stability of polycarbonate composites
model fits especially well for those blends where PC is the reinforced with potassium titanate whiskers: effect of coupling agents
continuous phase. addition. Polym Degrad Stab 1999;66:241e6.
[5] Cui J, Forssberg E. Mechanical recycling of waste electric and electronic
equipment: a review. J Hazard Mater 2003;99:243e63.
[6] Matsuto T, Jung CH, Tanaka N. Material and heavy metal balance in a re-
4. Conclusions
cycling facility for home electrical appliances. Waste Manag 2004;24:
425e36.
Integration through numerical methods of TGA results ob- [7] Evstatiev M, Fakirov S, Krasteva B, Friedrich J, Covas JA, Cunha M. Re-
tained for different PCeABS mixtures shows the validity of cycling of poly(ethylene terephthalate) as polymerepolymer composites.
the autocatalytic model where f(a) Z (1  a)nam. The fitting Polym Eng Sci 2002;42:826e35.
[8] Villalba G, Segarra M, Fernández AI, Chimenos JM, Espiell F. A proposal
between the calculated curves and experimental data is higher
for quantifying the recyclability of materials. Resour Conserv Recycling
for those blends where PC is the continuous phase. This could 2002;37:39e53.
help to elucidate the mechanism of the thermal degradation of [9] Shen H, Forssberg E, Pugh RJ. Selective flotation separation of plastics
these materials obtained from WEEE. by particle control. Resour Conserv Recycling 2001;33:37e50.
Apparent activation energies determined from TGA data [10] Pappa G, Boukouvalas C, Giannaris C, Ntaras N, Zografos V,
Magoulas K, et al. The selective dissolution/precipitation technique for
are practically independent of heating rates, in particular for
polymer recycling: a pilot unit application. Resour Conserv Recycling
ABS-based mixtures, which suggests that the most important 2001;34:33e44.
process in the degradation of these materials corresponds to [11] Poulakis JG, Papaspyrides CD. Recycling of polypropylene by the
ABS, in particular thermo-oxidative processes in the elasto- dissolution/reprecipitation technique 1. A model study. Resour Conserv
meric phase of the blend. This could lead to the proposal of Recycling 1997;20:31e41.
[12] Brebu M, Bhaskar T, Murai K, Muto A, Sakata, Azhar-Uddin MD. The
stabilizing systems for PCeABS blends.
individual and cumulative effect of brominated flame retardant and poly-
On the other hand, those mixtures with high PC content vinylchloride (PVC) on thermal degradation of acrylonitrileebutadienee
show a clear increase in apparent activation energies with styrene (ABS) copolymer. Chemosphere 2004;56:433e40.
heating rate, suggesting that the thermal degradation mecha- [13] McNeill IC, Basan S. Thermal degradation of polycarbonates: reaction
nism of these samples is composed of several complex conditions and reaction mechanisms. Polym Degrad Stab 1993;39:13e9.
[14] Becker L, Lenoir D, Matuschek G, Kettrup A. Thermal degradation of
processes, each predominant during different stages of the
halogen-free flame retardant epoxides and polycarbonate in air. J Anal
overall process. Therefore, the use of complex stabilizing sys- Appl Pyrolysis 2001;60:55e67.
tems will be necessary for an efficient stabilization of these [15] Katajisto J, Pakkanen TT, Pakkanen TA, Hirva P. Ab-initio study on ther-
mixtures and a limitation of the action of the main degradation mal degradation reactions of polycarbonate. J Mol Struct (Theochem)
processes (hydrolysis, thermo and photo-oxidation). 2003;634:305e10.
[16] Montaudo G, Carroccio S, Puglisi C. Thermal and thermoxidative degra-
When comparing our results with mechanisms proposed in
dation processes in poly(bisphenol-A carbonate). J Anal Appl Pyrolysis
the literature [15,16] some comments can be introduced. 2002;64:229e47.
Montaudo et al. [16] proposed a mechanism based on oxida- [17] Montaudo G, Carroccio S, Puglisi C. Thermal oxidation of poly(bisphenol-
tion and decomposition of the polymer with formation of dif- A polycarbonate) investigated by SEC/MALDI. Polym Degrad Stab 2002;
ferent sized oligomers and further cross-linking reactions 77:137e46.
[18] Jang BN, Wilkie CA. A TGA/FTIR and mass spectral study on the
through the oxidative coupling of polymeric units to form
thermal degradation of bisphenol A polycarbonate. Polym Degrad Stab
biphenyl groups. At the end of the degradation many low 2004;86:419e30.
molecular-weight oligomers are formed and acetophenones, [19] Donga D, Tasaka S, Aikawa S, Kamiya S, Inagaki N, Inoue Y. Thermal
benzyl alcohol and phenols are observed by MALDI-TOF degradation of acrylonitrileebutadieneestyrene terpolymer in bean oil.
mass spectrometry. According to this mechanism, which is Polym Degrad Stab 2001;73:319e26.
[20] Tiganis BE, Burn LS, Davis P, Hill AJ. Thermal degradation of
in agreement with observations in the present study, a higher
acrylonitrileebutadieneestyrene (ABS) blends. Polym Degrad Stab
heating rate would lead to a higher number of alternative 2002;76:425e34.
ways to degrade PCeABS mixtures. An increase in the ABS [21] Zong R, Hu Y, Wang S, Song L. Thermogravimetric evaluation of PC/
content would lead to a point where the dispersed ABS do- ABS/montmorillonite nanocomposites. Polym Degrad Stab 2004;83:
mains connect together forming a network, which gives rise 423e8.
[22] Jiménez A, Torre L, Kenny JM. Thermal degradation of poly(vinyl chlo-
to a general stabilization of the system.
ride) plastisols based on low-migration polymeric plasticizers. Polym
Degrad Stab 2001;73:447e53.
[23] Navarro R, Torre L, Kenny JM, Jiménez A. Thermal degradation of re-
cycled polypropylene toughened with elastomers. Polym Degrad Stab
References 2003;82:279e90.
[24] Chiang WY, Hwung DS. Properties of polycarbonate/acrylonitrilee
[1] Silicon valley toxic coalition. Just say no to e-waste: background docu- butadieneestyrene blends. Polym Eng Sci 1987;27:632e9.
ment on hazards and waste from computers. !http://www.svtc.org/ [25] Luda di Cortemiglia MP, Camino G, Costa I, Guaita M. Thermal degra-
cleancc/pubs/sayno.htmO. dation of ABS. Thermochim Acta 1985;93:187e90.
[2] Directive 2002/95/EC of the European Parliament and of the Council of [26] Suzuki M, Wilkie CA. The thermal degradation of acrylonitrilee
27 January 2003 on the restriction of the use of certain hazardous sub- butadieneestyrene terpolymer as studied by TGA/FTIR. Polym Degrad
stances in electrical and electronic equipment. Stab 1995;47:217e21.
534 R. Balart et al. / Polymer Degradation and Stability 91 (2006) 527e534

[27] Wildes G, Keskkula H, Paul DR. Fracture characterization of PC/ABS [30] Motyakin MV, Schlick S. Thermal degradation at 390 K of poly(acrylo-
blends: effect of reactive compatibilization, ABS type and rubber con- nitrileebutadyeneestyrene) (ABS) containing a hindered amine stabilizer:
centration. Polymer 1999;40:7089e107. a study by ID and 2D Electron Spin Resonance Imaging (ESRI) and ATR-
[28] Inberg JPF, Gaymans RJ. Co-continuous polycarbonate/ABS blends. FTIR. Polym Degrad Stab 2002;76:25e36.
Polymer 2002;43:2425e34. [31] Boldizar A, Möller K. Degradation of ABS during repeated
[29] Elmaghor F, Zhang L, Fan R, Li H. Recycling of polycarbonate by blend- processing and accelerated ageing. Polym Degrad Stab 2003;81:
ing with maleic anhydride grafted ABS. Polymer 2004;45:6719e24. 359e66.