Main products and kinetics of the thermal degradation of polyamides

Article abstract of DOI:10.1016/S0045-6535(00)00233-2

The thermal degradation of the polyamides (PA) 6, 12, 66 and 612 was investigated by means of thermal analysis/ mass spectrometry (TA-MS) and pyrolysis in a german standard oven. Sample masses were about 20 and 40 mg. The heating rates used in the dynamic studies were 1, 5 and 10 K min^-1. Both air and nitrogen atmospheres were utilized. The kinetic parameters were calculated from the TA-MS measurements and the main decomposition products were registered online. The evolved products from the pyrolysis oven were captured and analyzed off-line by GC/ MS.

Full text of DOI:10.1016/S0045-6535(00)00233-2

Chemosphere 42 (2001) 601±607
Main products and kinetics of the thermal degradation of
polyamides
a,
a
a,b
M. Herrera *, G. Matuschek , A. Kettrup
a

GSF ± Forschungszentrum f ur Umwelt und Gesundheit, Institut f ur Okologische Chemie, Ingolstadter Landstrasse 1, 85764,
Neuherberg, Germany
b

Technische Universitat Munchen, Institut f ur Okologische Chemie und Umweltanalytik, 85358, Freising, Germany
Abstract
The thermal degradation of the polyamides (PA) 6, 12, 66 and 612 was investigated by means of thermal analysis/
mass spectrometry (TA±MS) and pyrolysis in a german standard oven. Sample masses were about 20 and 40 mg. The
heating rates used in the dynamic studies were 1, 5 and 10 K min 1. Both air and nitrogen atmospheres were utilized.
The kinetic parameters were calculated from the TA±MS measurements and the main decomposition products
were registered online. The evolved products from the pyrolysis oven were captured and analyzed o€-line by GC/
MS. Ó 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Thermal decomposition; Pyrolysis; Dynamic measurements; TA±MS.
1. Introduction
oligomeric products with nitrile and vinyl chain ends has
also been reported (Levchik et al., 1992b; Ballistreri
et al., 1988). PA 66 eliminates as main organic product
cyclopentanone (Ohtani et al.,1982; Ballistreri et al.,
1987) but also some hydrocarbons, nitriles and vinyl
groups (Levchik et al., 1994). For PA 12, Ballistreri et al.
(1988) found the lactams (cyclic monomer and dimer) to
be the major primary products of the thermal degrada-
tion; however, Mailhos-Le®evre et al. (1989) found also
ole®nic nitriles and Levchik et al. (1992a) reported some
a-ole®ns. In the case of PA 612 the formation of cyclic
oligomers has been found to occur as result of its ther-
mal degradation (Ballistreri et al., 1988) but also some
dinitriles have been found (Ohtani et al., 1982).
Polyamides (PA) are an important group of
the thermoplastic polycondensates. The amide group
(±NHCO±) can be obtained by polymerization of lac-
tams (polylactams) or by condensation of diamines with
dicarboxilic acids (nylons). The thermal degradation of
some polyamides has been studied systematically and
intensively in the past (Ohtani et al., 1982; Ballistreri et
al., 1988; Nielsen et al., 1995). The elimination of the
inorganic gases CO2, CO, H2O, NH3 and HCN to a
minor extent has been reported in the literature for the
polyamides (Nielsen et al., 1995). Polynuclear aromatic
hydrocarbons (PAH) and their nitrogen containing de-
rivatives (N-PAH) have been also found in polymers
containing polyamides (Schmaus et al., 1996).
For the analysis of the thermal degradation of plas-
tics, the hyphenated technique TA±MS has proved to be
very helpful in the past (Kettrup et al., 1990; Ohrbach
et al., 1990; Matuschek et al., 1995). Pyrolysis with a
Bayer/ICI/Shell (BIS) combustion device has been used
in several cases (Klusmeier et al., 1986; Matuschek et al.,
1995; Schmaus et al., 1996).
It has been found that the main route of thermal
degradation of PA 6 is the formation of caprolactam
(Levchik et al., 1992b; Bockhorn et al., 1996) with yields
as high as 85% (Czernik et al., 1998); the presence of
The kinetic analysis of a decomposing material can
be done using two di€erent techniques: the dynamic and
* Corresponding author.
0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 2 3 3 - 2

602
M. Herrera et al. / Chemosphere 42 (2001) 601±607
the isothermal thermogravimetric studies. The dynamic
thermogravimetric data can be analyzed using two dif-
ferent methods, the isoconversional and the discrimi-
nation methods (Levchik et al., 1992a,b). The
isoconversional methods use the data from a series of
thermogravimetric (TG) curves obtained at di€erent
heating rates to calculate the energy of activation (Ea)
but do not provide any information about the pre-
exponential factor (k0) and the kinetic function (f(a)) of
the process. The discrimination methods use the solution
of an equation based upon a given model and allow the
simultaneous determination of Ea, k0 and f(a). Detailed
kinetic analysis on plastics (Day et al., 1995; Wey and
Chang, 1995; Bockhorn et al., 1999) and on polyamides
(Levchik et al., 1992a,b; Levchik et al., 1994; Bockhorn
et al., 1996) can be found reported in the literature.
The present paper focuses on the determination of
the major species evolving from the thermal degradation
of polyamides and describes a kinetic study of the
thermal degradation of polyamides under dynamic
conditions using a discrimination method. The main
evolving products from the thermal degradation have
been identi®ed and quanti®ed.
samples were investigated using a modi®ed BIS com-
bustion device. The sample size used was about 40 mg.
Klusmeier et al. (1986) described this device in detail.
The resin Amberlite XAD-4 was used for the enrichment
and adsorption of the evolving degradation products.
The identi®cation and quanti®cation of the products
was done by GC/MS.
2.3. GC/MS analysis
The identi®cation of the products was performed by
means of comparison of their mass spectral data and of
their retention times with reference standards. The GC/
MSD apparatus consisted of a HP 6890 gas chromato-
graph and a HP 6890 mass selective detector. The GC
was equipped with a SGE BPX-5 capillary column
(SGE, 25 m, 0.25 mm i.d., 0.22 lm ®lm); a HP 6890
automatic injector, injection volume ˆ 2 ll; splitless in-
jection at 280°C; transfer line at 280°C; a temperature
program starting isothermally at 40°C for 2 min; 40±
70°C at 10 K min 1; 1, 5 min isothermal; 70±90°C at 3 K
min 1; 90±310°C at 15 K min 1; 5 min isothermal; sol-
vent delay 2.35 min; detector HP MSD 6890 (70 eV),
modus SCAN. As carrier gas used was helium with a
1
controlled constant ¯ow of 1.5 ml min
.
2. Experimental
3. Kinetic law and equations
2.1. Sample materials
Analyzing an ordinary decomposition reaction (1) it
is possible to establish an Eq. (2) to interpret it as a
function of known factors,
All the polyamides used were pure polymers without
any additives. PA 6 was provided by Bayer AG (Dure-
than B30S), PA 66 by BASF AG (Ultramid A3), PA 12

and PA 612 by Huls AG (Vestamid L1600 and Vestamid
Asolid ! Bsolid ‡ Bg0 aseous
ꢀ1†
D16nf). The structures of the polyamides investigated
are shown in Table 1.
dp
ˆ Y ꢀp; r; t; T†;
dt
ꢀ2†
2.2. Thermal degradation/pyrolysis
where p is the concentration of the products, r the
concentration of the reagents, t the time and T is the
temperature. The conversion function Y( p,r,t,T) can be
described through two di€erent separable functions
The studies of thermal degradation were performed
on a STA 429 simultaneous thermal analyzer (Netzsch

Geratebau GmbH) and an on-line coupled QMG 420
quadrupole mass spectrometer (Balzers Hochvakuum
GmbH). The samples were heated from ambient tem-
perature up to 800°C using heating rates of 1, 5 and
Y ꢀt; T; r; p† ˆ kꢀTꢀt††f ꢀr; p†:
ꢀ3†
1
10 K min in nitrogen and synthetic air atmospheres.
For one-step reactions f(r,p) reduces to f(r), where
r ˆ 1)a and p ˆ a, where a is the degree of conversion
The sample size used was about 20 mg. Furthermore, all
Table 1
Structure of the investigated polyamides
PA-type
Formula of repeat unit
Ratio
CH2:CONH
Former product(s)
PA 6
[±NH(CH2)5CO±]n
[±NH(CH2)11CO±]n
5
11
5
e-Caprolactam
PA 12
PA 66
PA 612
x-Lauryl lactam
Hexamethylenediamine and Adipic acid
Hexamethylenediamine &1,12-Dodecanedioic acid
[±NH(CH2)6NHCO(CH2)4CO±]n
[±NH(CH2)6NHCO(CH2)10CO±]n
8

M. Herrera et al. / Chemosphere 42 (2001) 601±607
603
which is de®ned as a ˆ ꢀm0 m†=ꢀm0 m † with m0
overlapping processes taking place during the ®rst de-
gradation step. The weight loss of the polyamides was
about 2% greater in air than in nitrogen.
1
being the mass of the sample, m the actual mass and m
1
the ®nal mass. For complex processes Eq. (2) becomes a
system of di€erential equations and the separation of
variables is not possible any more. Thus, it is not possible
to obtain an analytical solution for multi-level processes.
Expressing the function k(T) through the Arrhenius
Equation (Bockhorn et al., 1999) we obtain kꢀT† ˆ
k0 expꢀ Ea=ꢀRT††, where k0 is the pre-exponential factor
and Ea the apparent activation energy. Using in addition
the de®nition of the heating rate b ˆ dT/dt, Eq. (2) be-
comes
The mass spectra taken at DTGmax in both atmo-
spheres registered common masses for all polyamides at
m/e 15 (NH from NH3), 18 (H2O), 27 (HCN) and 44
(CO2). In the case of the mass m/e 44 (CO2), it could be
observed that it presented a second peak for all poly-
amides during the degradation in air, which lead to the
conclusion that the extra degradation step in air is due to
the decomposition of the char formed in the ®rst step.
The ion current intensity curves of PA 66 in air are
shown in Fig. 2. For PA 6 the masses m/e 30, 41, 55, 67,
85, 96 and 113 are representative of the main product
caprolactam. For both PA 66 and PA612 typical masses
from cyclopentanone at m/e 55, 67 and 84 were mea-
sured. In the case of PA 12 the masses found at m/e 55,
67, 82, 91 and 110 indicate the formation of the cyclic
lactam and some nitriles.
ꢀ
ꢁ
ꢀ
ꢁ
da
dT
k0
b
Ea
ꢀRT†
ˆ
exp
ꢀ1 a†n:
ꢀ4†
The determination of the kinetic parameters k0, Ea and n
(apparent reaction order) was done using a ®fth-order
Runge±Kutta method upon a set of initial parameters
(estimated with an isoconversional model).
4.2. Kinetic analysis
4. Results and discussion
The TG data from the dynamic thermogravimetric
measurements were used for the calculation of the ki-
netic parameters. The kinetic software utilized (Netzsch
thermokinetic analysis) was described in detail by Op-
4.1. TA±MS analysis
All samples showed a similar behaviour in their
thermal degradation curves. The polyamides decompose
in nitrogen apparently in one step but the measurements
at the lowest heating rates (1 K min 1) con®rmed that at
least two overlapping steps are taking place and are thus
dicult to discern. The weight loss takes place in the
range 350±475°C. The decomposition in air occurs in
two major steps with the ®rst one occurring in the same
range of temperature as in nitrogen and the second one
in the range 475±600°C. The TG curves of the four
polyamides in air are shown in Fig. 1. The measurement
at low heating rates showed also in air at least two

fermann and Hadrich (1995) and has been used suc-
cessfully in the past (Opfermann et al., 1998). The core
of the program is the multivariate non-linear regression,
which works with an hybrid Marquardt±Levenberg
process. The kinetic parameters through which the
model is characterized are divided into two groups: the
®rst one is valid for all runs and is therefore model-
speci®c, the other one is speci®c for one run at a time.
The best ®t for all polyamides was found to be a two
step model with competing reactions. The partial equa-
tions describing the model A gives B through two dif-
ferent pathways are:
Fig. 1. TG curves of PA 612, PA 66, PA 12 and PA 6 in air at
1
10 K min
.
Fig. 2. MS ion current intensity curves of PA 66 in air.

604
M. Herrera et al. / Chemosphere 42 (2001) 601±607
f ꢀr† ˆ ꢀ1 a†n;
ꢀ5†
n
f ꢀr† ˆ ꢀ1 a† ‰1 ‡ KcatBŠ;
ꢀ6†
where Eq. (5) represents a nth order reaction and Eq. (6)
a nth order reaction with autocatalysis by B. Introducing
both equations into Eq. (3) we obtain
ꢀ
ꢁ
ꢀ
ꢁ
k0ꢀ1†
b
da
dT
Eaꢀ1†
nꢀ1†
ˆ
exp
ꢀ1 a†
ꢀRT†
ꢀ
ꢁ
ꢀ
ꢁ
k0ꢀ2†
Eaꢀ2†
ꢀRT†
nꢀ2†
‡
exp
ꢀ1 a† ꢀ1 ‡ KcatB†:
ꢀ7†
b
The calculated values for the apparent activation ener-
gies (Ea), the apparent reaction orders (n) and the pre-
exponential factors (K0) are listed in Table 2. These
values have a statistical con®dence of a ˆ 0.05. A com-
parison of the experimental and the calculated points of
Dm vs temperature for PA 6 in nitrogen as well as the
reactionÕs model are shown in Fig. 3.
Fig. 3. Comparison of experimental points (r) and calculated
curves ( ± ) of Dm vs temperature for PA 6 in nitrogen.
tant to achieve good results. It has to be considered that
the inclusion of more parameters (steps) into a model
can result in a better statistical ®t but this alone does not
necessarily assure that the model represents the reactions
actually taking place.
The estimated values di€er somewhat from the ones
found in the literature because the latter were either
estimated with isoconversional methods (Levchik et al.,
1992a,b and 1994) or with a single step model (Bock-
horn et al., 1996). In general it can be seen that all
polyamides behave similarly with the exception of PA
12. The polyamides PA 6, 66 and 612 present similar Ea
4.3. Pyrolysis in BIS-oven
The samples were burned at 800 and 950°C under
nitrogen and air atmospheres. The degradation products
were adsorbed and subsequently analyzed by GC/MS.
The quantities of the most abundant primary products
evolved from the thermal degradation of the samples are
shown in Table 3 (the rest of the products is shown in
the Figs. 4±7). Caprolactam was found to be the major
product from the degradation of PA 6 but was also
detected for PA 66 and PA 612, which agrees with
previous results (Ohtani et al., 1982). PA 66 decomposes
mainly to cyclopentanone and reacts further giving a
wide spectrum of products. For PA 12 it could be ob-
served that its cyclic monomer lauryl lactam was built as
the principal degradation product.
1
of 91±164 kJ mol in the ®rst step and of 310±476 kJ
1
mol in the second step. The apparent reaction orders
for these polyamides are also similar, with that of PA 66
being higher than the other ones. PA 12 present in the
second step values close to the values from the other
polyamides, but in the ®rst step the apparent activation
1
energy of 2208 kJ mol is de®nitively too high as well as
its corresponding apparent reaction order and the pre-
exponential factor. For the contrary, the value of the
catalytic factor Kcat is very low with a value of
log10(Kcat) ˆ )19.68 which helps to minimize the contri-
bution of this step.
The calculation and analysis of kinetic data based
upon a kinetic software has to be done carefully to avoid
incoherent results and interpretations. The consider-
ation of feasible models based on the results from the
TA data as well as the adequate initial values is impor-
Fig. 4 shows the primary degradation products of PA
6, where three major groups can be appreciated: the
nitrile, the ketone and the aromatic group. Nitriles and
ketones show a reduction in yield depending on the at-
mosphere, being higher in nitrogen than in air. An ex-
Table 2
Kinetic data of the thermal degradation of polyamides in nitrogen
Parameter
PA 6
PA 66
PA 12
PA 612
1
Ea1 (kJ mol
log10(k0)1
Order 1
)
162
91
2208
160
164
9.42
2.62
4.57
2.03
310
8.67
1.97
1.35
400
1.34
0.35
8.10
)19.68
260
log10(Kcat)1
Ea2 (kj mol
log10(k0)2
Order 2
1
)
476
32.95
1.35
21.08
1.44
16.47
0.63
27.25
1.08

M. Herrera et al. / Chemosphere 42 (2001) 601±607
605
Table 3
Major products from the pyrolysis in the BIS-apparatus
Product
Capro-
lactam
(mg/g)
Cyclo-
penta-
none
Lauryl
lactam
(mg/g)
(mg/g)
PA 6 in N2, 800°C
PA 6 in air, 800°C
PA 6 in N2, 950°C
PA 6 in air, 950°C
PA 66 in N2, 800°C
PA 66 in air, 800°C
PA 66 in N2, 950°C
PA 66 in air, 950°C
PA 12 in N2, 800°C
PA 12 in air, 800°C
PA 12 in N2, 950°C
PA 12 in air, 950°C
PA 612 in N2, 800°C
PA 612 in air, 800°C
PA 612 in N2, 950°C
PA 612 in air, 950°C
249
126
263
135
30
6
5
5.5
4.6
109
67
15
24
106
64
14
17.31
10.67
14.97
7.83
Fig. 6. Major products from the pyrolysis of PA 12.
73
22
20
15
Fig. 7. Major products from the pyrolysis of PA 612.
ception to this behaviour is shown by hexanedinitrile,
which was built only in air. The aromatic products
(toluene and benzonitrile) show however a strong de-
pendence on the temperature, being higher at 950°C
than at 800°C.
The major degradation products of PA 66 (Fig. 5)
show, as in the case of PA 6, the same three major
groups: nitriles, ketones and aromatics. The behaviour
of the nitrile and the ketone groups is in principle the
same as for PA 6. Pentanenitrile was built only in ni-
trogen and could not be detected in air. Toluene and
benzonitrile show higher concentrations at higher tem-
peratures. The quantity of hexanedinitrile is two times
bigger than for PA 6 because its formation is favoured
due to the hexamethylenediamine fraction in the
polymer.
Fig. 4. Major products from the pyrolysis of PA 6.
Fig. 6 shows the main evolved products from
the decomposition of PA 12. The most abundant com-
1
pound was toluene with 7 mg g in nitrogen at 950°C.
A series of C4±C12 nitriles was detected, being the most
abundant the C10±C12 nitriles. As in the case of PA 6,
pentanenitrile was only registered in nitrogen, which
indicates that this compound undergoes further de-
composition in air atmosphere. Dinitriles could not be
found as degradation products of PA 12. The behaviour
Fig. 5. Major products from the pyrolysis of PA 66.

606
M. Herrera et al. / Chemosphere 42 (2001) 601±607
identi®ed and correlated to the principal degradation
products of each type of polyamide. Based on the dy-
namic thermogravimetric curves the kinetic parameters
Ea, k0 and n were calculated and it was showed that all
polyamides follow the same kinetic model and have
similar values with exception of PA 12, which shows a
particular behaviour in the catalytic competing reaction.
The reason for this peculiar behaviour has still to be
studied. The polyamides consisting of one former
product (polylactams) undergo cyclization to yield
mainly the cyclic monomer. PA 66 eliminates cyclo-
pentanone as the primary decomposition product. PA
612 produce, besides caprolactam, toluene, undecane-
nitrile, lauryl lactam and some traces of the cyclic
monomer as major products. All samples had three main
groups of compounds from the thermal degradation:
nitriles, ketones and aromatics. The nitrile and the ke-
tone groups showed a dependence on the atmosphere
while the aromatic group showed a dependence on the
temperature. The behaviour of the PAHs was in agree-
ment with that of the aromatic group; the major quan-
tity of PAHs was given by PA 12 and the lower one by
PA 6.
Fig. 8. PAHs behaviour by the thermal degradation of poly-
amides.
of the di€erent groups is in agreement with the obser-
vations made for PA 6 and 66.
For PA 612 the primary degradation products, Fig. 7,
show a series of C4±C11 nitriles as well as two aromatic
compounds and one ketone. C12 were not found because
the C12 groups mainly undergo cyclization to form the
cyclic monomer. Similar to the results seen for PA 12,
toluene was the most abundant compound in nitrogen at
950°C. The aromatic compounds show a di€erent be-
haviour than the nitriles as observed in the cases of the
other polyamides.
References
The behaviour of the two most abundant PAHs
evolved during the thermal degradation of the samples,
Fig. 8, shows that, as in the case of the aromatic com-
pounds mentioned above, the formation of the PAHs
depend more on the temperature than on the atmo-
sphere. The major quantity of PAHs was found in PA 12
under air and at 950°C. PA 6 seems to yield the lower
quantity of PAHs and PA 66 and PA 612 yield a similar
amount of PAHs. An explanation for the elevated
quantity of PAHs found in the case of PA 12 can be that
this polyamide undergoes scission of the C±C bonds in b
position to the unsaturation initiating a radical chain
process which leads to formation of some hydrocarbons
(Levchik et al., 1992a). Furthermore, PA 12 has the
highest ratio CH2:CONH (Table 1) which makes it more
similar to the ole®nic polymers.
Ballistreri, A., Garozzo, D., Giu€rida, M., Montaudo, G.,
1987. Mechanism of thermal decomposition of nylon 66.
Macromolecules 20, 2991±2997.
Ballistreri, A., Garozzo, D., Giu€rida, M., Impallomeni, G.,
Montaudo, G., 1988. Primary thermal decomposition pro-
cesses in aliphatic polyamides. Polym. Deg. Stab. 23, 25±41.
Bockhorn, H., Hornung, A., Hornung, U., 1999. Mechanisms
and kinetics of thermal decomposition of plastics from
isothermal and dynamic measurements. J. Anal. Appl.
Pyrolysis 50, 77±101.
Bockhorn, H., Hornung, A., Hornung, U., Teepe, S., Weich-
mann, J., 1996. Investigation of the kinetics of thermal
degradation of commodity plastics. Combust. Sci. Tech.
116±117, 129±151.
Czernik, S., Elam, C.C, Evans, R.J., Meglen, R.R., Moens, L.,
Tatsumoto, K., 1998. Catalytic pyrolysis of nylon-6 to
recover caprolactam. J. Anal. Appl. Pyrolysis 46, 51±64.
Day, M., Cooney, J.D., MacKinnon, M., 1995. Degradation of
contaminated plastics: a kinetic study. Polym. Deg. Stab.
48, 341±349.
Kettrup, A., Ohrbach, K.-H., Matuschek, G., Joachim, A.,
1990. Thermal analysis-mass spectrometry and thermo-
gravimetric adsorption on ®re retardants. Thermochimica
Acta 166, 41±52.
5. Conclusions
The polyamides decompose apparently in one step in
nitrogen and in two steps in air, but a detailed study
con®rmed that the decomposition has two steps in ni-
trogen and at least three steps in air. The extra step in air
is due to the degradation of the char formed in the ®rst
step. CO2, H2O, NH3 and HCN were the major prod-
ucts detected online with TA±MS during the thermal
degradation of the samples. Some other masses could be
Klusmeier, W., Ohrbach, K.-H., Kettrup, A., 1986. Combus-
tion experiments of polystyrene by use of a modi®ed Bayer-
Ici-Shell apparatus. Thermochimica Acta 103, 231±237.
Levchik, S.V., Costa, L., Camino, G., 1992a. E€ect of the ®re-
retardant, ammonium polyphosphate, on the thermal
decomposition of aliphatic polyamides. I. Polyamides 11
and 12. Polym. Deg. Stab 36, 31±41.

M. Herrera et al. / Chemosphere 42 (2001) 601±607
607
Levchik, S.V., Costa, L., Camino, G., 1992b. E€ect of the ®re-
retardant, ammonium polyphosphate, on the thermal
decomposition of aliphatic polyamides. Part II ± Polyamide
6. Polym. Deg. Stab. 36, 229±237.
lubricant systems and additives. Thermochimica Acta 166,
277±289.
Ohtani, H., Nagaya, T., Sugimura, Y., Tsuge, S., 1982. Studies
on thermal degradation of aliphatic polyamides by pyrolysis
Levchik, S.V., Costa, L., Camino, G., 1994. E€ect of the ®re-
retardant, ammonium polyphosphate, on the thermal
decomposition of aliphatic polyamides. Part III ± Poly-
amides 6.6 and 6.10. Polym. Deg. Stab. 43, 43±54.
±
Pyrolysis 4, 117±131.
glass capillary gas chromatography. J. Anal. Appl.

Opfermann, J., Hadrich, W., 1995. Prediction of the thermal
response of hazardous materials during storage using an
improved technique. Thermochimica Acta 263, 29±50.
Opfermann, J., Blumm, J., Emmerich, W.-D., 1998. Simulation
of the sintering behavior of a ceramic green body using
advanced thermokinetic analysis. Thermochimica Acta 318,
213±220.
Mailhos-Le®evre, V., Sallet, D., Martel, B., 1989. Thermal
degradation of pure and ¯ame retarded polyamides 11 and
12. Polym. Deg. Stab. 23, 327±336.
Matuschek, G., Cyrys, J., Lenoir, A., Kettrup, A., 1995.
Thermal degradation of di€erent ®re retardant polyurethane
foams. Thermochimica Acta 263, 59±71.
Schmaus, G., Beck, B., Matuschek, G., Kettrup, A., 1996.

Nielsen, M., Jurasek, P., Hayashi, J., Furimsky, E., 1995.
Formation of toxic gases during pyrolysis of polyacrylonit-
rile and nylons. J. Anal. Appl. Pyrolysis 35, 43±51.
Ohrbach, K.-H., Matuschek, G., Kettrup, A., Joachim, A.,
1990. Simultaneous thermal analysis-mass spectrometry on
Thermolyse neuartiger Kunststo€e und okotoxikologische
Bewertung. J. Thermal Analysis 47, 485±491.
Wey, M.Y., Chang, C.L., 1995. Kinetic study of polymer
incineration. Polym. Deg. Stab. 48, 25±33.