Thermal degradation of poly(vinyl chloride)

A. Marongiu, T. Faravelli

, G. Bozzano, M. Dente, E. Ranzi

Dipartimento di Chimica, Materiali e Ingegneria Chimica, Politecnico di Milano, Piazza Leonardo da Vinci

32, 20133 Milan, Italy

Received 21 October 2002; accepted 28 January 2003

Abstract

This paper presents an initial attempt at describing poly(vinyl chloride) (PVC) thermal

degradation through a semi-detailed and lumped kinetic model. A mechanism of 40 species
and pseudocomponents (molecules and radicals) involved in about 250 reactions permits quite
a good reproduction of the main characteristics of PVC degradation and volatilization. The
presence of the two step mechanism *

the first step of which corresponds to dehydrochlorina-

tion and the second to the tar release and residue char formation *

are correctly predicted

both in quantitative terms and in the temperature ranges. The model was validated by
comparison with several thermo gravimetric analyses, both dynamic at different heating rates,
and isothermal. When compared with the typical one step global apparent degradation
models, the approach proposed here spans quite large operative ranges, especially when it
comes to predicting product distributions. The initial results of these product predictions, even
though quite preliminary, are encouraging and confirm the validity of the model.
#

Keywords: PVC; Thermal degradation; Pyrolysis; Detailed modeling

1. Introduction

In recent years, plastic waste has been mainly disposed of by landfill or

incineration, but these processes are not fully acceptable under current international
policy which focuses on efficient recovery of raw materials and energy. Pyrolysis and

* Corresponding author. Tel.:

39-02-2399-3282; fax:

39-02-7063-8173.

E-mail address:

tiziano.faravelli@polimi.it

(T. Faravelli).

J. Anal. Appl. Pyrolysis 70 (2003) 519

553

www.elsevier.com/locate/jaap

0165-2370/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0165-2370(03)00024-X

gasification processes are promising routes for optimal upgrading from waste.
Moreover, pyrolysis of plastic mixtures, based on the decomposition of polymers at
different temperatures, allows the treatment of polymers with simultaneous
decomposition and separation

[1]

One of the main plastic components in typical municipal waste is polyvinylchlor-

ide (PVC). Thermal degradation of PVC is a more complex process than with the
other stable plastic wastes, i.e. polyethylene (PE), polypropylene (PP) and
polystyrene (PS). As a matter of fact, whilst PE, PP and PS thermally react at quite
low temperatures, reducing the polymer chain

[2,3]

, PVC pyrolysis involves

significant cross-linking reactions with the formation of polyaromatic structures
(possibly chlorinated) and a carbonaceous residue (char). Moreover, the process
description becomes more complex because chlorine needs to be introduced into the
system in addition to carbon and hydrogen. The complexity of the chemistry of PVC
pyrolysis thus calls for several simplifications and lumping procedures to handle the
huge amount of intermediate and final products.

The literature includes several papers on pyrolysis and gasification of PVC. The

primary goal of the major part of the works reported so far has been the
characterization of the rate of weight loss during the primary thermal degradation

[4]

or the decomposition description by overall kinetic parameters determined on the

basis of the weight loss curves

. Some mechanisms of PVC degradation are

proposed in the literature, but generally the kinetic models refer to global apparent
kinetics

[9]

or to a few step mechanism

[10]

. These models do not take into account

the chemistry of polymer degradation and describe the pyrolysis process by means of
very simplified reaction paths. Each individual reaction step is equivalent to and
representative of a complex network of reactions. This approach can only adequately
describe the apparent kinetics in a narrow range of heating rates and operating
conditions. A single step model is not able to cover a wide range of heating rates,
temperatures and conversion levels with the same kinetic parameters. Therefore,
dynamic and isothermal analysis reveal different decomposition rates

[1]

. Further-

more, the possible presence of mass and heat transfer limitations, which are generally
neglected, spreads the range of variation of these kinetic constants.

As a result of the aforementioned considerations, the interest of a mechanistic

model capable of accounting for the differences in starting material and able to
describe the reaction process over a wide range of heating rates and temperatures is
obvious. This mechanistic model not only allows the prediction of the detail of gas
product distribution (a significant step in the possible upgrading of solid wastes), but
also the characterization of chlorinated pollutant formation.

2. Kinetic mechanism

PVC pyrolysis is a liquid phase chain radical mechanism, at least at temperatures

higher than 200 8C. Several mechanisms are proposed in the literature and various
simplified models with apparent reactions and different levels of simplification are
discussed. Typically the mechanisms refer to single apparent kinetics

[4,11,12]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

520

Anthony proposes a more complete model, which involves five pseudospecies and
five reactions, and which is able to describe the weight loss during the whole
thermogravimetric experiments

[10]

. All the researchers agree on the presence of two

main degradation steps, as clearly shown in

Fig. 1

. Initially dehydrochlorination

forms HCl and polyene structures. During this phase, benzene and some
naphthalene and phenanthrene are also formed through Diels Alder reactions and
successive dealkylation of polyene molecules. Then, when Cl has been quantitatively
released from the melt, the polyene molecules rearrange and through cyclization and
cross-linking reactions, form alkyl aromatic hydrocarbons and char residues.

The PVC degradation process is thus very complex, compared with the

degradation of polymers such as PE, PP or PS. Not only does PVC contain Cl as
a further element, but the complexity of cross linking reactions and the successive

Fig. 1. Predicted dynamic TGA with a heating rate of 10 8C min

: panel (a) residue (wt.%) behavior and

identification of the main thermal decomposition phases panel (b) benzene, PAH and char formation
profiles. The TAR fraction represents the total amount of volatile aromatics.

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

521

formation of polyaromatic hydrocarbons (PAH) and char also play a significant
role. In fact, char and carbonaceous residues up to

15 wt.% can be obtained

depending on the operating conditions. The growth of these polyaromatic species
and char formation involves a large amount of intermediate species and isomers.
Therefore, it is not possible to handle the kinetic modeling of PVC pyrolysis with a
detailed approach similar to the one already adopted for the decomposition of PE,
PP and PS

[2,3]

This very large number of intermediate and final components strongly suggests

that the whole mechanism should be described with a properly selected and limited
set of reference lumped components. The resulting semi-detailed kinetic mechanism
is constituted by about 40 species and/or pseudocomponents, which are assumed to
be representative of the whole system in the proposed model. They characterize the
typical species involved in the different degradation phases and can be classified into
four major groups:

real species (molecules and radicals):

; HCl; C

; Cl

;

. . .

linear reference pseudospecies (molecules and radicals):

P(CH

CHCl)

P; P(CHCH)P; . . .

aromatic and char pseudospecies (molecules and radicals):

P(C

)P; P(C

)P; . . .

chlorinated aromatic and char pseudospecies (molecules and radicals).

The complete detail of these species is reported in

Appendix A

The initial PVC polymer is simply represented by the chlorinated reference unit P

(CH

CHCl)

The successive steps of degradation form polyene molecules and these species, all

of which have different molecular weights, are represented by the alkene reference
unit: P

(CH

CH)

P. The reference species or reacting units (reported in bold

characters in the brackets) are placed inside the polymer chain, represented here by
the Ps at the beginning and end. Intermediate species during the dehydrochlorination
phase are other reference units, which present either Cl or double bonds in the
structure.

H or Cl abstractions from these pseudocomponents form the corresponding

radicals. Cross-linking reactions between polyene molecules lead to alkyl aromatic
intermediates, which can further condensate and grow into char. Reference
pseudocomponents are then introduced according to the different number of
aromatic rings and the possible presence of Cl.

This approach refers to a small number of reference pseudocomponents and

involves the use of equivalent reactions with real stoichiometric coefficients. The
complete kinetic scheme is reported in

Appendix B

. The kinetic parameters of these

liquid phase reactions are directly derived from the analogous gas phase reaction

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

522

ones, properly corrected to take into account the transposition in the liquid phase
where necessary. These corrections become significant mainly for reactions with high
activation energies

[13]

. Corrections are also applied to termination reactions to take

into account the diffusive limitations to radical effective collisions and recombina-
tions. This means that the thermal degradation of PVC is described in terms of
intrinsic kinetics, in which heat and mass transfer limitations are not included. In
principle, no tuning or trial and error procedures have been assumed in the definition
of the rate constants. Starting from analogous reactions and adopting proper
similarity rules, only a very limited tuning activity has been performed when the
literature data were not available. All the rate constants reported in the paper and in

Appendix B

assume the following units: l, mol, s and the activation energy are

expressed in J mol

3. Radical initiation and termination reactions

The reactivity of the system is ruled by the overall radical pool and consequently

the competition between initiation and termination reactions has to be correctly
characterized. It is important to observe that these reactions are strongly influenced
by the liquid phase. In fact, not only is the high activation energy of the initiation
reactions partially reduced by the mobility of molecules in the melt, but the
diffusivity of the radicals partially controls their recombinations

[14,15]

3.1. Radical initiation reactions

The gas phase kinetic parameters of reaction (1):

P(CH

CHCl)

P 0 P(CH

CH)

(1)

can be derived from the unimolecular transition state theory:

gas
i

5:010

exp

339 100

)

and the activation energy corresponds to the bond energy

[16

18]

The correction of the activation energy for the transposition to the liquid phase is

related to the number of carbon atoms (n

) characterizing, as per free volume theory,

the flow unit for the polymer diffusion. The polymer molecules move through the
liquid phase courtesy of the coordinated migration of segments of the polymer chain.
The critical volume (or length) for this migration of a polymer chain is the volume of
this jumping unit, which is only a small segment of the complete chain. This length
(n

) can be estimated from the energy (E

) required for the mobility of the polymer

segment and is experimentally measured as a function of the temperature. At the
reference temperature of 0 K, the E

value is 85 kJ mol

[14,15,19]

The entropy correction is due to the loss of degrees of freedom and can be

estimated on the basis of the translation and rotational contributes:

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

523

liqgas

$9:9 u:e:

Thus, the rate constant for the liquid phase initiation reaction becomes:

liq
i

3:4310

exp

226 000

)

This estimation fits in quite well with literature data

[20]

where activation energies

of about 234 400 J mol

are proposed.

3.2. Radical termination reactions

Termination reactions involve the three families of radicals present in the system:

Cl, alkyl radicals and alkylaromatic radicals. Because of the high reactivity and
mobility of chlorine radicals in the liquid phase, the termination reactions do not
require activation energy, whilst the frequency factors decrease, increasing the
molecular weight of the other terminating radical (reactions 86

137).

The recombination reactions of two alkyl and alkyl aromatic radicals are

estimated by correcting the collisional constant of the gas phase reactions to take
into account the diffusive limitations of the liquid phase. As already discussed and
validated in the case of the visbreaking process as well as PE and PS degradation

[2,3]

, a general expression of the kinetic parameters of these termination reactions in

the condensed phase becomes:

12:8

400

exp

(l mol

)

where F

is a global correction factor, which takes into account the symmetry,

resonance, steric and surface effects

[21]

. V

is the molar volume of the flux unit:

where W is the molecular weight of the flux unit and r is the liquid density, which
can be reasonably considered constant during the process and typically in the range

1100

1300 kg m

. As already discussed in the case of initiation reactions, the

polymer molecules move through the liquid phase via the coordinated migration of
segments of the polymer chain. In this case, Eyring’s free volume theory allows us to
estimate that the monomeric unit of alkyl radicals characterizing the flux unit for the
molecular momentum transfer has a value of about n

17 C atoms

[19]

. Alkyl

aromatic radicals present a polyaromatic structure which inhibits the migration of
small segments and the flux unit is assumed as the whole radical, where the heat of
evaporation is conveniently estimated from the Trouton

Meissner rule

[14,15]

. E

the energy required for the mobility of the flux unit and its value is estimated simply
on the basis of the number of C atoms of the flux unit (n

$ 0:4DH

$ 1210

ﬃﬃﬃﬃﬃﬃ

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

524

The resulting kinetic parameters of the different recombination reactions are

reported in

Appendix B

4. Dehydrochlorination

Dehydrochlorination (DHC) is the first step of the degradation process and

explains the initial weight loss of

Fig. 1

. As clearly reported in the literature

[4,11]

both molecular reactions and radical propagation reactions contribute to this HCl
release. The former is represented by a single reaction while the latter involves a
chain radical reaction mechanism including several intermediate components.

4.1. Molecular dehydrochlorination

At low temperatures ( B

200 8C) molecular dehydrochlorination plays a funda-

mental role in PVC degradation.

P(CH

CHCl)

P 0 P(CHCH)P

HCl

(5)

Kinetic parameters of this four center reaction are well established in the gas phase

[22,23]

. The frequency factors and the activation energies, respectively, range

between 2.75

and 4.37

and 229 400 and 212 100 J mol

, according

to the Cl position (primary or secondary C atom) and/or to the stereo-chemical
configuration. The presence of conjugate double bonds does not significantly affect
the rate constants

[24]

Due to its molecular nature and short range interaction, no large correction are

required to the liquid phase transposition and the following rate constant has been
assumed:

0:510

exp

217 700

)

4.1.1. Chain radical reactions

At temperatures higher than 200 8C the chain radical mechanism becomes

relevant. Initiation reactions can involve either C

C or C

H or C

Cl bond

cleavage. The last is the weakest bond and is consequently the most probable
initiation step. This scission produces a very reactive chlorine radical and a reference
radical:

P(CH

CHCl)

P 0 P(CH

CH)

(1)

As clearly shown in

Fig. 2

, Cl

radicals abstract initially H atoms from the

reference reacting unit P

(CH

CHCl)

P and form HCl and P

(

CHCHCl)

When different radicals R

are present in the reacting system, the propagation

reaction becomes:

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

525

P(CH

CHCl)

P 0 RH

CHCHCl)

(6; 10; 12)

The successive b-scission of P

(

CHCHCl)

P gives rise to a new propagating Cl

radical and to a new reference component with a double bond in the chain structure:

CHCHCl)

P 0 P(CHCH)P

(60)

This leads to the formation of polyene molecules P

(CH

CH)

P which hold up

the system. Their successive condensation and de-alkylation reactions may explain
the formation of benzene and aromatic components. These successive steps are also
shown in

Fig. 2

. Alternatively, a competitive H abstraction reaction on the polymer

chain is:

P(CH

CHCl)

P 0 RH

P(CH

CCl)

The successive fate of P

(CH

CCl)

P could be the dehydrogenation to form P

(CH

CCl)

P or the formation of a primary radical (

CHClCH

)

P. Both these

reactions, with activation energies of

159 000 and

125 600 J mol

, respectively,

are very slow at the usual degradation temperatures. As a consequence, in the
temperatures of interest, the main fate of the radical P

(CH

CCl)

P is to

propagate the reaction chain with an H abstraction reaction from the system and
to return to the initial polymer.

Fig. 2. Sketched mechanisms of the main radical chain propagation steps: panel (a) dehydrochlorination;
panel (b) condensation; panel (c) de-alkylation.

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

526

4.1.2. Abstraction reactions

The different radicals are capable of stabilizing themselves, mainly through H or

Cl abstraction reactions which form a stable molecule (reactions 6

41). The H

abstraction reactions are extremely relevant. In fact, the kinetic parameters reported
in the literature for H and Cl abstractions from dichloroethane:

CHClCHCl 0 HCl

CCl

CHCl

6:310

exp

13 000

(l mol

)

CHClCHCl 0 Cl

CHCl

(l mol

)

aCl

210

exp

89 200

(l mol

)

clearly indicate that H abstractions dominate and Cl abstraction can be disregarded
without significant errors

[25]

. The different reactivity of the intermediate radicals

reflects their different stabilities. Secondary alkyl radicals are more stable than vinyl
radicals and consequently vinyl radicals are more reactive. The kinetic parameters of
these H abstraction reactions have been directly estimated on the basis of the
corresponding gas phase reactions.

With simple analogy and similarity rules, only a small set of reference kinetic data

allows the extension of the parameters to unknown reactions

[26]

. For instance, the

following reference kinetic parameters were selected for the abstraction of a
secondary H atom by Cl, by a secondary radical and by a vinyl radical, respectively:

k 2:410

(l mol

)

Secondary

k 410

exp

56 500

(l mol

)

Vinyl

k 410

exp

20 900

(l mol

)

These values are similar to the ones reported in the literature for the secondary

radicals

[27]

k 1:010

exp

54 100

(l mol

)

and for vinyl radicals (reactions 10

13)

[28]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

527

k 8:9110

exp

25 950

(l mol

)

The chlorine radical is more reactive than all the other radicals and it does not

require activation energy (reactions 14

15)

[29]

4.1.3. b-Decomposition reactions

H abstraction reactions propagate the reaction chain, but do not change the mean

molecular weight of the system. On the contrary, b-decomposition reactions are
initially responsible for the chlorine formation and for the subsequent reduction of
the chain length. The bond energy of C

Cl is lower than that of C

C and for this

reason, the release of Cl radicals is favored in comparison with the formation of
primary radicals. In agreement with the literature data

[30]

, the reference kinetic

parameters of these reactions (reaction 60) were assumed as: k

exp(

83 300/RT) (s

b-Decomposition reactions with a C

C bond cleavage form the primary radicals

(

CH)

P. This reference unit plays a key role in the formation of cross linked

structures and in the molecular weight growth of polycondensed aromatics, as will be
discussed later. The reference kinetic parameters of this reaction k

exp(

129 800/RT) (s

) (reactions 61

62) clearly shows that, whenever possible, Cl release

is the prevailing path. Nevertheless, short and volatile chain fragments leave the
liquid phase and they are responsible for the small amount of chlorinated species
observed experimentally in the gas phase decomposition products.

5. Benzene formation and initial reticulation processes

As already mentioned, polyene molecules formed during DHC, can successively

undergo cyclization reactions. The net result of these reactions is the formation of
aromatic and alkyl aromatic species on the one hand, and the reticulation and cross-
linking of the polymer structure on the other. The formation of benzene and
aromatic products is a function of the residual chlorine in the polymer. As already
mentioned, this complex reacting system, whose molecular weight spans various
orders of magnitude with a large number of intermediates, is described on the basis
of a limited number of lumped or reference components.

These reactions can be both intermolecular and intramolecular and in this case

too, moreover they can proceed via molecular and radical reactions. The relative role
of these unimolecular and bimolecular processes is not completely clarified in the
literature

[12,31,32]

Fig. 3

shows typical examples of these mono- and bimolecular

cyclization reactions with the successive dealkylation reactions and benzene
formation. A very similar benzene formation involves the corresponding aromatic
radicals (reaction 243).

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

528

5.1. Molecular cyclization

5.1.1. Unimolecular reaction

As soon as the DHC process forms polyene structures, molecular cyclization takes

place through six center Diels Alder reactions in the reacting system. Unimolecular
(or intramolecular) reaction is entropically favored but requires a higher activation
energy. Consequently, at low temperatures, bimolecular condensation prevails and
the unimolecular reaction becomes significant only at temperatures higher than
500 8C (reaction 239).

Stereochemical configuration of the polyene influences the rate parameters. The

kinetic

parameters

for

the

condensation

reactions

(E)-

and

(Z)-

Fig. 3. Six (panel a) and four (panel b) center molecular reactions. Cyclization, dealkylation and benzene
formation.

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

529

CHCHCHCHCH

to form 1,3-cyclohexadiene are, respectively, k

2.77

exp(

188 400/RT) (s

) and k

7.14

exp(

121 400/RT) (s

)

[33]

Even if it is difficult to determine the dispersion of the configuration of the polyene

sequences, the prevailing presence of the energetically favored (E) stereo structure is
assumed with the corresponding rate value k

2.0

exp(

173 700/RT) (s

The reduction of the activation energy takes into account the stabilization effect of
the double bonds in the transition state.

The competition between benzene and alkylaromatic formation is ruled out by the

residual Cl in the a-position of the side chains. As a matter of fact, the electronic
delocalization effect of Cl reduces the C

C bond energy. This is the reason why

benzene is formed mainly during the first degradation steps when a reasonable
amount of Cl is present in the chain.

P(CHCH)P 0 0:26C

0:2C

(238)

is a lumped pseudocomponent, which represents several gas species. Gas

species (partially chlorinated C

) are the products of the dealkylation reaction in

this phase. With the progressive Cl depletion from the chain and the loss of its
delocalization activity, the molecular cyclization reactions give rise to alkyl-aromatic
components, whose successive fate is a further condensation and growth.

P(CHCH)P 0 0:2P(C

(239)

The competition between benzene and alkyl aromatic formation is empirically

described by means of an apparent increase of the activation energy with the extent
of the dehydrochlorination process.

DABenzMono

210

exp

169 500 29 300

[HCl]

)

where, [HCl]

is the final and asymptotic concentration of the HCl.

5.1.2. Bimolecular reaction

Diels Alder bimolecular reactions favor the formation of cyclic species as well as a

continuous cross-linking and condensation of the reacting species (

Fig. 3

). When

large amounts of Cl are still present in the chain, these reactions too are followed by
successive dealkylation reactions with the final formation of benzene (reaction 67

68).

P(CHCH)PP(CHCH)P

0 0:5C

0:1C

0:4P(CHCH)P

(67)

P(CHCH)PP(CHCH)P 0 0:4P(C

(68)

Rate constants for these Diels Alder reactions are taken from the literature

[34]

With reference to the gas phase kinetic parameters for the addition of ethylene to
1,3-butadiene to form cyclohexene k

exp(

125 600/RT), a reduction of

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

530

16 700 J mol

in the activation energy is adopted in order to account for the

aforementioned electronic delocalization effect in the first stage.

In this case too, a proper correction accounts for an apparent increase in the

activation energy with the dehydrochlorination process.

DABenz

5:010

exp

104 650 29 300

[HCl]

(l mol

)

A reduction of

12 600 J mol

in the reference value of 125 600 J mol

assumed in the alkylaromatic formation (reaction 68) because of the stabilizing effect
of double bonds.

5:010

exp

108 800

(l mol

)

6. Condensation

As is clear from

Fig. 1

, the thermogravimetric analysis (TGA) curve plateaus

between 350 and 450 8C. During this phase the chemistry of the system is governed
by the significant and progressive formation of poly-condensed aromatic inter-
mediates. These successive cyclization reactions of polyene species take place both
via radical and molecular mechanisms. As already mentioned, the formation of
benzene is mainly observed during the initial DHC phase and is strongly related to
Cl content in the side chains. The progressive cross-linking of the polymer matrix
and the growth of the molecular aromatic structures characterize the formation of
char residues. Condensation reactions of alkylaromatic molecules form PAH,
thereby increasing the molecular weight of the polymer with a continuous reduction
of H content and without significant formation of volatiles. Panel b of

Fig. 1

shows

the predicted evolution of aromatic pseudospecies and the formation of heavy
components along the reaction time. Benzene is mainly released in the initial phase.
When most of the dehydrochlorination has occurred, alkylbenzenes are formed
initially and these progressively condense to form PAH and finally the char residues.
The progressive growth of the aromatic clusters is described on the basis of just three
different aromatic pseudo species: P

)

P, P

)

P and P

)

P. As already reported in

Appendix B

, these, respectively, represent and lump

average structures with one, two and ten unsaturated cycles. The successive growth
of PAH takes place through the following lumped cyclization reactions:

P(CHCH)PP(C

)P 0 0:6P(C

)P0:6H

(69)

P(CHCH)PP(C

)P 0 0:4P(C

)P1:3H

(70)

P(CHCH)PP(C

)P 0 2:04CHAR6:74H

(71)

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

531

These reactions are very similar to reaction (68) already sketched in panel a of

Fig.

Similar radical reactions are also accounted in the scheme (76

84) and they

represent the possible addition reactions of aromatic radicals on aromatic structures
with the corresponding growth of the aromatic cluster. Analogous results are also
obtained through intramolecular reactions with Diels Alder molecular paths
(reactions 241

243). All these reactions are similar to the benzene formation ones

and the same reference kinetic parameters were assumed

[34]

A further path towards the growth of PAH is ruled out by the addition of the

primary radicals on double bonds (reactions 72

75). In agreement with the

corresponding gas phase reactions, a kinetic value k

exp(

25 100/RT) (1

mol

) is adopted for the additions on the first aromatic ring

[35]

. Successive

dehydrogenation reactions increase the stability of the polyaromatic species. Kinetic
constants of addition reactions thus fall as the molecular weight rises and a
progressive increase of activation energy is assumed in order to account for this
higher stability.

Heavy dehydrogenated molecules are also formed from the cyclo-addition

reactions (and successive dealkylation) of alkyaromatic radicals on the double
bonds of the side chain of alkylaromatics (reactions 76

84). These reactions are the

main responsible for the final formation of the char residues:

)PP(C

)P 0 P(

)P1:5H

0:08CHAR

(76)

7. Fragmentation and TAR formation

At temperatures higher than

450 8C, a new change in the TGA slope in

Fig. 1

observed. This is explained by the tar and volatiles released by the liquid phase as a
result of the condensation and dehydrogenation reactions with deep dealkylation.
Most of the resulting aromatic species are no longer linked to the polymeric chains
and can volatilize as secondary tar products, while the char structures’ concentration
increases. The hydrogen content of the char is very low as observed in the typical
residue of PVC pyrolysis. Only a few reference species are assumed to describe this
degradation step also.

H abstraction reactions of different radicals on the alkylaromatic species are

particularly selective on the benzyl-like positions, thus forming resonantly stabilized
radicals (RSRs). This stability increases with the number of polycondensed rings and
the consequent electron delocalization. The main successive fate of these RSRs is a
dealkylation process via b-scission as well as further condensations due to addition
and recombination reactions. PAHs are then assumed with short alkyl side chains
and/or vinyl groups. This assumption is experimentally confirmed by the gas
evolution during tar formation

[36]

. The lumped reactions are the following:

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

532

)P 0 0:6C

(

CH)

0:12CHAR0:5C

0:18C

0:02H

(63)

)P 0 C

(

CH)

0:12CHAR0:75C

0:25C

(64)

)P 0 3C

(

CH)

(65)

)P 0 1:69CHAR(

CH)

4:62C

1:12C

(66)

In the semi-detailed approach adopted, C

and C

represent, respectively, the

classes of hydrogenated and dehydrogenated gas species, lumping together also
larger hydrocarbons. In facts, as experimentally observed

[10]

, the hydrogen

disproportionation can explain the formation of CH

groups in the aliphatic side

chains. In this way, the system dehydrogenation proceeds releasing both directly
hydrogen (during cyclization reactions) and hydrogen rich small hydrocarbons.

Light aromatic components leave the polymer melt as soon as they are formed.

When the molecular weight of the PAH radicals increases, the corresponding
reactivity decreases and both b-decomposition and H abstraction reactions are less
favored. On the contrary, addition and termination reactions become relevant and
explain the char formation.

8. Chlorinated components

As already discussed, the Cl atoms in the polymer chain mainly undergo DHC

reactions, but they may partially remain in the aromatic structures during the
cyclization phase. For this reason, chlorinated pseudospecies have been introduced
into the kinetic model. The first hypothesis is to assume that the reference
pseudospecies contain only a single Cl atom. The detail of these species fully
corresponds to the equivalent aromatic pseudospecies (molecules and radicals) and is
reported in

Appendix A

. Assuming that the Cl atom in the species, close to the

reacting unit, does not significantly influence the reactivity. That is why, the
reactions of these chlorinated species (reactions 138

236) are fully analogous to the

ones already discussed.

Thus, a characterization of chlorine content in tar char components can be

predicted. These preliminary and rough results are in line with the experimental
measurements, even though it is very difficult to find detailed and reliable data. This
possibility of predicting chlorinated aromatics is very important because of their
dangerous impact on the human health.

9. Validation of the model and comparisons with experimental data

The validation of the kinetic scheme relies on several sets of experimental data,

which are taken directly from the literature. No experimental activity was performed

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

533

inside this work. The reference to measures coming from different research groups
allows both a more complete comparison and proper validation and to reduce the
impact of eventual possible experimental deviations. These are mainly based on
TGA, both under isothermal and dynamic conditions, with different heating rates.
The polymer degradation is modeled neglecting the heat and mass transport
resistances inside and outside the polymer melt. This assumption is justified by
considering the very small amount of the sample as well as the main characteristics of
the experimental devices.

The mass balance of each pseudocomponent in the liquid phase is:

where C

is the concentration of ith species and R

is its net formation rate. The

energy balance can be simply expressed as:

const:

where the constant is either the heating rate or zero during isothermal experiments.

The resulting system of stiff ordinary differential equations is solved adopting an

implicit algorithm based on BDF formulas proposed by Gear

[37]

and implemented

classes named BZZODE

[38]

The model assumes that all the fundamental units representative of long chain

molecules remain in the liquid phase with the char residues. All the other lighter
species and pseudospecies are (C

, C

, H

, Cl

, C

, HCl,

CHClCHCl, C

, C

Cl, C

Cl) are considered as evaporat-

ing instantaneously and their successive gas phase reactions are ignored.

As discussed by Bockhorn and co-workers in several papers

[1,39,40]

, the apparent

activation energy of the isothermal decomposition process differs from that of the
dynamic TGA. It is, therefore, important to accurately validate the model through
comparison with both the experiments.

9.1. Dynamic experiments

Fig. 4

shows the comparison of the model results with 11 different measured

dynamic TGA. The experimental results come from different research groups

[4,9,10,12,41,42]

and are carried out at different heating rates, which span from 2 up

to 40 8C min

. The uncertainties in the experiments are due to the different

accuracy of the measures and also to the differences in quality, molecular weight,
and possible additives of the PVC sample.

The characteristic three phases of the polymer decomposition are quite clear and

distinct. Dehydrochlorination starts at about 200

220 8C and proceeds quickly. At

about 350

400 8C, the TGA curve changes its slope and plateaus somewhat up to

420

450 8C, where, condensation reactions occur. At these temperatures the second

important mass loss starts with dealkylation and dehydrogenation reactions and tar

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

534

volatilization. Finally, above 500 8C, the char residue formed remains as fixed
carbon. The model agrees quite well with the experiments in all the different phases
and for all the different heating rates. Char formation too is properly predicted. It is

Fig. 4. Dynamic TGA at different heating rates, predicted data are reported with line, experimental data
with marks: m

Fig. 5. Relative fraction of volatilized during isothermal TGA at low temperatures. Comparison between
predicted data are (lines) and experimental measures (marks)

[4]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

535

important to observe that, in line with the experimental information, the data
obtained at 2 8C min

[4]

are normalized by ignoring the total amount of residual

at 500 8C.

9.2. Isothermal experiments

Bockhon et al.

[4]

presented two set of data.

Fig. 5

shows that the first isothermal

degradation step in the temperature range 270

320 8C. These temperatures are

typical of the dehydrochlorination phase. The model predictions are able to correctly
reproduce the system reactivity across the whole temperature range. A normalized
conversion of 80% is reached in about 2 h at 270 8C, while this takes just 10 min at
320 8C. The normalized conversion refers to the total amount of the residue at the
end of the test.

The second degradation step, corresponding to tar volatilization and char

formation, is usually analyzed on complete dehydrochlorinated samples

[4]

. The

experimental procedure involves heating the PVC sample up to 350 8C at low heating
rate and then keeping the temperature constant for 30 min to complete the
dehydrochlorination step.

Fig. 6

shows the comparison between experimental data

and model predictions for five different isothermal conditions (400

450 8C). In this

case too, the agreement between experimental measurements and model predictions
is very satisfactory. It takes almost 150 min to convert 80% of the sample at 400 8C,
and less than 20 min at 450 8C.

Bockhorn et al.

[1]

discussed the spread of the apparent degradation rates

observed by comparing the two main steps in dynamic and isothermal conditions.
The DHC process presents comparable rate constants, while the estimated activation
energies of the successive condensation and tar formation show some scatter, even
though large uncertainties are reported.

The sensitivity analysis of the four conditions (first and second steps of the

dynamic experiment and low and high temperature isothermal degradations), as
shown in

Fig. 7

, reveals that the typical characterizing reaction paths are very similar

Fig. 6. Relative fraction of volatilized during isothermal TGA at high temperatures. Comparison between
predicted data are (lines) and experimental measures (marks)

[4]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

536

in both the conditions. In the first step, the controlling reactions of polymer
degradation are b-scission forming Cl

and the polyene molecule. H abstraction of

and initiation reactions are of less importance. At high temperatures, the

sensitivity analysis in respect of char tendency shows that H abstraction on
alkylaromatic species are the first critical step toward residue formation. Internal
cyclization toward more condensed compounds with hydrogen release is responsible
for tar formation and competes with the char growth.

The same features can be also observed on the basis of the species evolution shown

Fig. 8

. The profiles are reported as a function of the intrinsic severity parameter,

i.e. the weight loss during the first DHC step and char formation during the second
phase. The first decomposition step is characterized by very similar composition in
terms of reactants and major species, but the first aromatic pseudocomponent is less
important during this phase. At high temperatures, the profiles still remain very
similar. The lower amount of the first aromatic component P

)

P in the

Fig. 7. Sensitivity coefficient calculated in isothermal and dynamic conditions for both the degradation
steps.

Fig. 8. Comparison between isothermal (marks) and dynamic (lines) TG experiments. Predicted species
evolution during the two different degradation steps in function of an intrinsic degree of conversion: the
total weight loss for the first step and the char formation for the second.

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

537

isothermal conditions can partially explain the different behaviors of the two
different conditions as well as the larger scatter in the identification of the apparent
kinetic parameters.

9.3. Comparison with detailed data of product distribution

One of the main features of the approach proposed here is the possibility of

describing the detailed product distribution from PVC thermal degradation.

Fig. 9

shows the comparison between model predictions and the experimental HCl released
during isothermal experiments at 270 and 300 8C

[43]

. The agreement is quite good,

even though an underestimation of up to 8% can be observed in the low temperature
condition. A better agreement is found for the benzene yields from isothermal
thermogravimetric data at 290 and 310 8C

[43]

, as shown in

Fig. 10

Finally,

Table 1

presents a rough comparison between the main predicted

products and the corresponding experimental data collected from a dynamic TGA
carried out at 10 8C min

up to 500 8C

[36,44]

. The general agreement is

reasonable, even though the uncertainty in the experimental data on HCl influence
the overall comparison. The model assumes an ideal polymer, without additives or
branches in the chain and consequently predicts almost total HCl release, quite
different from the experimental data of 52 wt.%. Tar fraction and char formation are
predicted quite well. As already discussed, the model is also able to evaluate the trace
amount of chlorinated species and these predictions too are in line with the
experimental evidences. These comparisons confirm that these preliminary results
are very encouraging. However, in the case of chlorinated species in particular,
further work is required to clarify the product distribution in greater detail.

Nonetheless, it is important to underline that in this respect too, the main interest

of this detailed kinetic model lies in the proper prediction of the relative trends as
functions of the operating conditions, rather than the correct amount of the different
species.

Fig. 9. Normalized HCl released during isothermal TGA. Comparison between predicted data are (lines)
and experimental measures (marks)

[43]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

538

Table 1
Product distribution (wt.%) from dynamic TG up to 500 8C, heating rate 10 8C min

wt.%

HCl

Gas

TAR

CHAR

Chlorinated GAS

Chlorinated TAR

Chlorinated CHAR

Predicted

58.3

5.66

5.67

18.63

11.71

3.4

2.6

2.40

Experimental

[36]

52.6

6.88

6.6

24.32

9.5

0.1

Comparison between predictions and experiments

[36]

A.
Marongiu

et
al.
/

Anal.

Appl.

Pyrolysis

70
(2003)

519

553

539

10. Conclusion

This paper presents an initial attempt at modeling PVC thermal degradation using

a semi-detailed approach. The enormous number of species formed during the
pyrolysis necessitates the introduction of pseudospecies, which represent many
different components. Even in the light of this drastic simplification (only

molecular and radical pseudospecies), the comparison with the experimental data is
satisfactory. Different sets of thermogravimetric data at very different heating rates
and isothermal temperatures were reproduced with a good agreement. The main
features of the degradation are correctly reproduced and the controlling steps
individuated. The adopted approach also allows preliminary considerations regard-
ing the product distribution, especially in relation to the most dangerous and toxic
components, such as chlorobenzenes. Unfortunately measures referring to the
analysis of the emissions from PVC pyrolysis are still quite scarce and further
activity both from the experimental and the modeling point of view are needed to
improve the knowledge of this field and to increase the detail of the kinetic
mechanism.

Acknowledgements

The authors wish to acknowledge Ushi and Andreas Hornung for supplying some

of the data and the helpful work of Francesco Robecchi, Marco Mehl and Marco
Nistler. This work is supported by the EU as part of the Competitive and
Sustainable Growth project, contract no. G1RD-CT-2002-03014 ‘Halocleanapplica-
tion’.

Fig. 10. Normalized benzene released during isothermal TGA. Comparison between predicted data are
(lines) and experimental measures (marks)

[43]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

540

Appendix A

Species and pseudocomponents in modeling thermal decomposition of PVC.

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

541

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

542

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

543

Number

Reaction

att

Source

(CH2CHCl)

P 0

(CH2

CH)

3.43

226 000

[16

18]

(CCl

CCl)

P 0

(

CCl)

6.43

242 800

[16

18]

(CH

CCl)

P 0

(

CH)

3.43

242 800

[16

18]

(CHClCHCl)

P 0

(

CHCHCl)

6.43

226 000

[16

18]

(CH2CHCl)

P 0

(CH

CH)

HCl

0.50

217 700

[22

24]

(CH2CHCl)

(CH2

CH)

P 0

(

CHCHCl)

(CH2CH2)

4.00

56 500

[45]

(CH2CH2)

(

CHCHCl)

P 0

(CH2

CH)

(CH2CHCl)

4.00

60 700

[45]

(CH2CHCl)

(

CH)

P 0

(

CHCHCl)

0.5P

(CH2CH2)

4.00

16 700

[45]

(CH2CH2)

(

CH)

P 0

(CH2

CH)

0.5P

(CH2CH2)

4.00

16 700

[45]

(CH2CHCl)

(

CH)

P 0

(

CHCHCl)

(CH

CH)

4.00

20 900

[28]

(CH2CH2)

(

CCl)

P 0

(CH2

CH)

(CH

CCl)

4.00

20 900

[28]

(CH2CHCl)

(

CCl)

P 0

(

CHCHCl)

(CH

CCl)

4.00

20 900

[28]

(CH2CH2)

(

CH)

P 0

(CH2

CH)

(CH

CH)

4.00

20 900

[28]

(CH2CHCl)

(

CHCHCl)

HCl

2.40

[29]

(CH2CH2)

(CH2

CH)

HCl

2.40

[29]

(CH

CH)

(

CHCHCl)

P 0

(

CH)

(CH2CHCl)

4.00

77 400

[46]

(CH

CCl)

(

CHCHCl)

P 0

(

CCl)

(CH2CHCl)

4.00

77 400

[46]

(CH

CH)

(CH2

CH)

P 0

(

CH)

(CH2CH2)

4.00

73 300

[46]

(CH

CCl)

(CH2

CH)

P 0

(

CCl)

(CH2CH2)

4.00

73 300

[46]

(CH

CH)

(

CH)

P 0

(

CH)

0.5P

(CH2CH2)

4.00

60 700

[46]

(CH

CCl)

(

CH)

P 0

(

CCl)

0.5P

(CH2CH2)

4.00

60 700

[46]

(CH

CH)

(

CCl)

P 0

(

CH)

(CH

CCl)

4.00

60 700

[46]

(CH

CCl)

(

CH)

P 0

(

CCl)

(CH

CH)

4.00

60 700

[46]

(CH

CH)

(

CH)

HCl

3.00

25 100

[46]

(CH

CCl)

(

CCl)

HCl

3.00

25 100

[46]

(CH2CHCl)

(

CHCHCl)

P 0

(CH2

CH)

(CHClCHCl)

l.00

73 300

[25]

(CHClCHCl)

(CH2

CH)

P 0

(

CHCHCl)

(CH2CHCl)

l.00

69 000

[25]

Appendix B

Kinetic scheme of thermal decomposition of PVC. Kinetic rate constant

parameters are expressed in Arrhenius equation type AT

exp(

Att

/RT). (units:

l, mol, s, J).

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

544

(CH2CHCl)

(

CH)

P 0

(CH2

CH)

0.5P

(CHClCHCl)

l.00

50 200

[25]

(CHClCHCl)

(

CH)

P 0

(

CHCHCl)

0.5P

(CHClCHCl)

l.00

50 200

[25]

(CH2CHCl)

(

CH)

P 0

(CH2

CH)

(CH

CCl)

l.00

50 200

[25]

(CH2CHCl)

(

CCl)

P 0

(CH2

CH)

(CCl

CCl)

l.00

50 200

[25]

(CHClCHCl)

(

CH)

P 0

(

CHCHCl)

(CH

CCl)

l.00

50 200

[25]

(CHClCHCl)

(

CCl)

P 0

(

CHCHCl)

(CCl

CCl)

l.00

50 200

[25]

(CH

CCl)

(CH2

CH)

P 0

(

CH)

(CH2CHCl)

1.00

90 000

[25]

(CCl

CCl)

(CH2

CH)

P 0

(

CCl)

(CH2CHCl)

1.00

90 000

[25]

(CCl

CCl)

(

CH)

P 0

(

CCl)

0.5P

(CHClCHCl)

1.00

71 200

[25]

(CH

CCl)

(

CH)

P 0

(

CH)

0.5P

(CHClCHCl)

1.00

71 200

[25]

(CH

CCl)

(

CHCHCl)

P 0

(

CH)

(CHClCHCl)

1.00

94 200

[25]

(CCl

CCl)

(

CHCHCl)

P 0

(

CCl)

(CHClCHCl)

1.00

94 200

[25]

(CCl

CCl)

(

CH)

P 0

(

CCl)

(CH

CCl)

1.00

71 200

[25]

(CH

CCl)

(

CCl)

P 0

(

CH)

(CCl

CCl)

1.00

71 200

[25]

(

CH)

(C10H10)

P 0

(

C10H9)

0.5P

(CH2CH2)

2.00

37 700

(c)

(

CH)

(C18H16)

P 0

(

C18H15)

0.5P

(CH2CH2)

3.00

37 700

(c)

(

CH)

(C47H36)

P 0

(

C47H35)

0.5P

(CH2CH2)

4.00

25 100

(c)

(CH2

CH)

(C10H10)

P 0

(

C10H9)

(CH2CH2)

2.00

64 900

(c)

(CH2

CH)

(C18H16)

P 0

(

C18H15)

(CH2CH2)

3.00

60 700

(c)

(CH2

CH)

(C47H36)

P 0

(

C47H35)

(CH2CH2)

4.00

52 300

(c)

(

CHCHCl)

(C10H10)

P 0

(

C10H9)

(CH2CHCl)

2.00

69 100

(c)

(

CHCHCl)

(C18H16)

P 0

(

C18H15)

(CH2CHCl)

3.00

64 900

(c)

(

CHCHCl)

(C47H36)

P 0

(

C47H35)

(CH2CHCl)

4.00

56 500

(c)

(

CH)

(C10H10)

P 0

(

C10H9)

(CH

CH)

2.00

37 700

(c)

(

CH)

(C18H16)

P 0

(

C18H15)

(CH

CH)

3.00

37 700

(c)

(

CH)

(C47H36)

P 0

(

C47H35)

(CH

CH)

4.00

25 100

(c)

(

CCl)

(C10H10)

P 0

(

C10H9)

(CH

CCl)

2.00

37 700

(c)

(

CCl)

(C18H16)

P 0

(

C18H15)

(CH

CCl)

3.00

37 700

(c)

(

CCl)

(C47H36)

P 0

(

C47H35)

(CH

CCl)

4.00

25 100

(c)

(C10H10)

P 0

(

C10H9)

HCl

2.00

20 900

(c)

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

545

(C18H16)

P 0

(

C18H15)

HCl

3.00

16 700

(c)

(C47H36)

P 0

(

C47H35)

HCl

4.00

16 700

(c)

(

CHCHCl)

P 0

(CH

CH)

6.00

83 300

[30]

(CH2

CH)

P 0

0.5P

(CH

CH)

(

CH)

0.5H2

3.00

129 800

[47]

(

CHCHCl)

P 0

(

CH)

0.5P

(CCl

CCl)

0.5H2

3.00

129 800

[47]

(

C10H9)

P 0

0.6C8H8

(

CH)

0.12CHAR

0.5C2H2

0.18C2H4

0.02H2

1.00

129 800

[47]

(

C18H15)

P 0

C12H10

(

CH)

0.12CHAR

0.75C2H2

0.25C2H4

1.00

129 800

[47]

(

C47H35)

P 0

3C14H10

(

CH)

2C2H2

1.00

134 000

[47]

(

C53H35)

P 0

1.69CHAR

(

CH)

4.62C2H2

1.12C2H4

0.80

138 100

[47]

(CH

CH)

P 0

0.5C6H6

0.1C2H2

0.4P

(CH

CH)

5.00

7(a)

104 650

[34]

(CH

CH)

P 0

0.4P

(C10H10)

5.00

108 800

[34]

(CH

CH)

(C10H10)

P 0

0.6P

(C18H16)

0.6H2

2.00

115 100

[34]

(CH

CH)

(C18H16)

P 0

0.42P

(C47H36)

1.34H2

1.00

115 100

[34]

(CH

CH)

(C47H36)

P 0

2.04CHAR

6.74H2

1.00

115 100

[34]

(

CH)

(CH

CH)

P 0

0.2P

(

C10H9)

0.8(

CH)

0.15H2

1.00

25 100

[35]

(

CH)

(C10H10)

P 0

0.5P

(

C18H15)

0.02CHAR

0.97H2

0.4(

CH)

1.00

29 300

[35]

(

CH)

(C18H16)

P 0

0.38P

(

C47H35)

0.61(

CH)

0.01CHAR

1.39H2

1.00

33 500

[35]

(

CH)

(C47H36)

P 0

0.90P

(

C53H35)

0.09(

CH)

2.6H2

1.00

37 700

[35]

(

C10H9)

(C10H10)

P 0

(

C18H15)

1.5H2

0.08CHAR

1.00

29 300

[35]

(

C10H9)

(C18H16)

P 0

0.57P

(

C47H35)

0.42(

CH)

2.09H2

0.02CHAR

1.00

33 500

[35]

(

C10H9)

(C47H36)

P 0

(

C53H35)

0.16CHAR

3.9H2

1.00

37 700

[35]

(

C18H15)

(C10H10)

P 0

0.57P

(

C47H35)

0.42(

CH)

2.09H2

0.02CHAR

1.00

33 500

[35]

(

C18H15)

(C18H16)

P 0

0.66P

(

C53H35)

3.60H2

0.02CHAR

0.33(

CH)

1.00

37 700

[35]

(

C18H15)

(C47H36)

P 0

(

C53H35)

0.49CHAR

5H2

1.00

37 700

[35]

(CH

CH)

(

C10H9)

P 0

0.61P

(

C18H15)

0.38(

CH)

0.02CHAR

0.56H2

1.00

29 300

[35]

(CH

CH)

(

C18H15)

P 0

0.41P

(

C47H35)

0.58(

CH)

0.97H2

1.00

33 500

[35]

(CH

CH)

(

C47H35)

P 0

0.9P

(

C53H35)

0.09(

CH)

2.3H2

0.03CHAR

1.00

37 700

[35]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

546

(CH2

CH)

(CH2CHCl)

5.97

0.5

13 900

Cl2

2.50

7500

[48]

(

CHCHCl)

P 0

(CHClCHCl)

1.42

0.5

17 300

(c)

(

CH)

P 0

(CH

CCl)

1.42

0.5

17 300

(c)

(

CH)

P 0

0.5P

(CHClCHCl)

1.42

0.5

17 300

(c)

(

CCl)

P 0

(CClCCl)

1.42

0.5

17 300

(c)

(

C10H9)

P 0

(C10H9Cl)

1.55

0.5

13 200

(c)

(

C18H15)

P 0

(C18H15Cl)

2.08

0.5

16 400

(c)

(

C47H35)

P 0

(C47H35Cl)

3.35

0.5

24 300

(c)

(

C53H35)

P 0

CHARC

1.2CHAR

4.74H2

3.55

0.5

25 500

(c)

(

CCl)

(

C10H9)

P 0

0.03H2

0.25P

(C47H35Cl)

0.37Cl2

4.43

22 900

(c)

(

CCl)

(

C18H15)

P 0

0.42P

(C47H35Cl)

0.28Cl2

0.05H2

5.93

26 200

(c)

(

CCl)

(

C47H35)

P 0

CHARC

1.04CHAR

5.75H2

9.55

34 000

(c)

(

CCl)

(

C53H35)

P 0

CHARC

1.29CHAR

4.25H2

1.01

35 200

(c)

(

C10H9)

(

C10H9)

P 0

1.1P

(C18H16)

0.1H2

2.42

18 900

(c)

100

(

C10H9)

(

C18H15)

P 0

0.59P

(C47H36)

1.27H2

6.48

22 100

(c)

101

(

C10H9)

(

C47H35)

P 0

2.37CHAR

7.75H2

1.04

29 900

(c)

102

(

C10H9)

(

C53H35)

P 0

2.62CHAR

6.25H2

1.11

31 200

(c)

103

(

C18H15)

(

C18H15)

P 0

1.5CHAR

6H2

4.34

25 400

(c)

104

(

C18H15)

(

C47H35)

P 0

2.7CHAR

8.75H2

1.40

33 200

(c)

105

(

C18H15)

(

C53H35)

P 0

2.95CHAR

7.25H2

1.48

34 400

(c)

106

(

C47H35)

(

C47H35)

P 0

3.91CHAR

11.5H2

1.13

41 000

(c)

107

(

C47H35)

(

C53H35)

P 0

4.16CHAR

10H2

2.38

42 200

(c)

108

(

C53H35)

(

C53H35)

P 0

4.41CHAR

8.5H2

1.26

42 000

(c)

109

(CH2

CH)

(CH2

CH)

P 0

0.4P

(C10H10)

3.56

20 300

(c)

110

(CH2

CH)

(

CHCHCl)

P 0

0.3Cl2

0.4P

(C10H9Cl)

0.7H2

3.56

20 300

(c)

111

(CH2

CH)

(

CH)

P 0

0.4P

(C10H10)

1.70

23 600

(c)

112

(CH2

CH)

(

CH)

P 0

0.3P

(C10H10)

0.5H2

1.70

23 600

(c)

113

(CH2

CH)

(

CCl)

P 0

0.08P

(C47H35Cl)

0.45Cl2

0.01H2

1.70

23 600

(c)

114

(CH2

CH)

(

C10H9)

P 0

0.66P

(C18H16)

0.66H2

1.86

19 600

(c)

115

(CH2

CH)

(

C18H15)

P 0

0.42P

(C47H36)

1.34H2

2.49

22 800

(c)

116

(CH2

CH)

(

C47H35)

P 0

2.04CHAR

6.75H2

4.00

30 600

(c)

117

(CH2

CH)

(

C53H35)

P 0

2.29CHAR

5.25H2

4.24

31 900

(c)

118

(

CHCHCl)

(

CHCHCl)

P 0

0.4P

(C10H9Cl)

0.8Cl2

0.2H2

3.56

20 300

(c)

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

547

119

(

CHCHCl)

(

CH)

P 0

0.45Cl2

0.08P

(C47H35Cl)

0.01H2

1.70

23 600

(c)

120

(

CHCHCl)

(

CH)

P 0

0.35Cl2

0.3P

(C10H9Cl)

0.15H2

1.70

23 600

(c)

121

(

CHCHCl)

(

CCl)

P 0

0.16CHARC

0.91Cl2

0.08H2

1.70

23 600

(c)

122

(

CHCHCl)

(

C10H9)

P 0

0.66P

(C18H15Cl)

0.16Cl2

0.5H2

1.86

196 00

(c)

123

(

CHCHCl)

(

C18H15)

P 0

0.42P

(C47H35Cl)

0.28Cl2

1.05H2

2.49

22 800

(c)

124

(

CHCHCl)

(

C47H35)

P 0

CHARC

1.04CHAR

6.74H2

4.00

30 600

(c)

125

(

CHCHCl)

(

C53H35)

P 0

CHARC

1.29CHAR

5.24H2

4.24

31 900

(c)

126

(

CH)

(

CH)

P 0

0.16CHAR

2.03

27 000

(c)

127

(

CH)

(

CH)

P 0

0.12CHAR

0.24H2

2.03

27 000

(c)

128

(

CH)

(

C10H9)

P 0

0.5CHAR

2H2

4.43

23 000

(c)

129

(

CH)

(

C18H15)

P 0

0.83CHAR

3H2

5.93

26 100

(c)

130

(

CH)

(

C47H35)

P 0

2.04CHAR

5.75H2

9.55

34 000

(c)

131

(

CH)

(

C53H35)

P 0

2.29CHAR

4.25H2

1.01

35 200

(c)

132

(

CH)

(

CH)

P 0

0.2P

(C10H10)

2.03

27 000

(c)

133

(

CH)

(

CCl)

P 0

0.5P

(CCl

CCl)

0.08CHAR

2.03

27 000

(c)

134

(

CH)

(

C10H9)

P 0

0.61P

(C18H16)

0.11H2

4.43

23 000

(c)

135

(

CH)

(

C18H15)

P 0

0.40P

(C47H36)

0.72H2

5.93

26 100

(c)

136

(

CH)

(

C47H35)

P 0

2CHAR

6H2

9.55

34 000

(c)

137

(

CH)

(

C53H35)

P 0

2.25CHAR

4.5H2

1.01

35 200

(c)

138

(

CH)

(C10H9Cl)

P 0

(

C10H8Cl)

0.5P

(CH2CH2)

2.00

37 700

(c)

139

(

CH)

(C18H15Cl)

P 0

(

C18H14Cl)

0.5P

(CH2CH2)

3.00

37 700

(c)

140

(

CH)

(C47H35Cl)

P 0

(

C47H34Cl)

0.5P

(CH2CH2)

4.00

25 100

(c)

141

(CH2

CH)

(C10H9Cl)

P 0

(

C10H8Cl)

(CH2CH2)

2.00

64 900

(c)

142

(CH2

CH)

(C18H15Cl)

P 0

(

C18H14Cl)

(CH2CH2)

3.00

60 700

(c)

143

(CH2

CH)

(C47H35Cl)

P 0

(

C47H34Cl)

(CH2CH2)

4.00

52 300

(c)

144

(

CHCHCl)

(C10H9Cl)

P 0

(

C10H8Cl)

(CH2CHCl)

2.00

73 300

(c)

145

(

CHCHCl)

(C18H15Cl)

P 0

(

C18H14Cl)

(CH2CHCl)

3.00

69 100

(c)

146

(

CHCHCl)

(C47H35Cl)

P 0

(

C47H34Cl)

(CH2CHCl)

4.00

60 700

(c)

147

(

CH)

(C10H9Cl)

P 0

(

C10H8Cl)

(CH

CH)

2.00

37 700

(c)

148

(

CH)

(C18H15Cl)

P 0

(

C18H14Cl)

(CH

CH)

3.00

37 700

(c)

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

548

149

(

CH)

(C47H35Cl)

P 0

(

C47H34Cl)

(CH

CH)

4.00

25 100

(c)

150

(

CCl)

(C10H9Cl)

P 0

(

C10H8Cl)

(CH

CCl)

2.00

37 700

(c)

151

(

CCl)

(C18H15Cl)

P 0

(

C18H14Cl)

(CH

CCl)

3.00

37 700

(c)

152

(

CCl)

(C47H35Cl)

P 0

(

C47H34Cl)

(CH

CCl)

4.00

25 100

(c)

153

(C10H9Cl)

P 0

(

C10H8Cl)

HCl

2.00

21 000

(c)

154

(C18H15Cl)

P 0

(

C18H14Cl)

HCl

3.00

16 700

(c)

155

(C47H35Cl)

P 0

(

C47H34Cl)

HCl

5.00

16 700

(c)

156

(

C10H8Cl)

P 0

C6H5Cl

0.5C2H2

(

CH)

0.08CHAR

1.00

129 800

[47]

157

(

C18H14Cl)

P 0

C11H9Cl

(

CH)

0.16CHAR

C2H2

1.00

129 800

[47]

158

(

C47H34Cl)

P 0

C15H11Cl

1.12CHAR

(

CH)

0.5C2H2

1.5C2H4

0.75H2

0.80

129 800

[47]

159

(

C53H34Cl)

P 0

CHARC

0.91CHAR

(

CH)

0.5C2H2

2.5C2H4

0.20

138 100

[47]

160

(CHClCHCl)

P 0

CHClCHCl

5.00

4(b)

104 650

(c)

161

(CH

CH)

(CH

CCl)

P 0

0.5C6H5Cl

0.25Cl2

0.04CHAR

5.00

7(a)

104 650

(c)

162

(CH

CH)

(CH

CCl)

P 0

0.2P

(C10H9Cl)

0.08CHAR

0.4Cl2

0.1H2

5.00

108 800

(c)

163

(CH

CH)

(C10H9Cl)

P 0

0.66P

(C18H15Cl)

0.5H2

0.16Cl2

2.00

115 100

(c)

164

(CH

CCl)

(C10H10)

P 0

0.66P

(C18H15Cl)

0.5H2

0.16Cl2

2.00

115 100

(c)

165

(CH

CH)

(C18H15Cl)

P 0

0.42P

(C47H35Cl)

1.05H2

0.28Cl2

2.00

115 100

(c)

166

(CH

CCl)

(C18H16)

P 0

0.42P

(C47H35Cl)

1.05H2

0.28Cl2

2.00

115 100

(c)

167

(CH

CH)

(C47H35Cl)

P 0

CHARC

1.04CHAR

6.75H2

2.00

115 100

(c)

168

(CH

CCl)

(C47H36)

P 0

CHARC

1.04CHAR

6.75H2

2.00

115 100

(c)

169

(

CH)

(CH

CCl)

P 0

0.1P

(

C10H8Cl)

0.9P

(

CCl)

0.54H2

l.00

25 100

[35]

170

(

CH)

(C10H9Cl)

P 0

0.55P

(

C18H14Cl)

0.44(

CH)

0.75H2

0.22Cl2

0.02CHAR

1.00

29 300

[35]

171

(

C10H8Cl)

(C10H10)

P 0

(

C18H14Cl)

0.08CHAR

1.49H2

1.00

29 300

[35]

172

(

C10H9)

(C10H9Cl)

P 0

(

C18H14Cl)

0.08CHAR

1.49H2

1.00

29 300

[35]

173

(

CH)

(C18H15Cl)

P 0

0.38P

(

C47H34Cl)

0.3Cl2

0.61(

CH)

1.08H2

0.01CHAR

1.00

33 500

[35]

174

(

C10H8Cl)

(C18H16)

P 0

0.42P

(

C47H34Cl)

2.62H2

0.28Cl2

0.57(

CH)

0.3CHAR

1.00

33 500

[35]

175

(

C10H9)

(C18H15Cl)

P 0

0.57P

(

C47H34Cl)

0.21Cl2

0.42(

CH)

0.02CHAR

1.87H2

1.00

33 500

[35]

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

549

176

(

C18H14Cl)

(C10H10)

P 0

0.42P

(

C47H34Cl)

2.62H2

0.28Cl2

0.57(

CH)

0.3CHAR

1.00

33 500

[35]

177

(

C18H15)

(C10H9Cl)

P 0

0.42P

(

C47H34Cl)

2.62H2

0.28Cl2

0.57(

CH)

0.3CHAR

1.00

33 500

[35]

178

(

C18H14Cl)

(C18H16)

P 0

0.56P

(

C53H34Cl)

0.21Cl2

3.76H2

0.23CHAR

0.43(

CH)

1.00

37 700

[35]

179

(

C18H15)

(C18H15Cl)

P 0

0.66P

(

C53H34Cl)

0.16Cl2

0.33(

CH)

0.02CHAR

3.43H2

1.00

37 700

[35]

180

(

C10H8Cl)

(C47H36)

P 0

(

C53H34Cl)

0.16CHAR

3.99H2

1.00

37 700

[35]

181

(

C18H14Cl)

(C47H36)

P 0

(

C53H34Cl)

0.49CHAR

5H2

1.00

37 700

[35]

182

(

C10H8Cl)

(CH

CH)

P 0

0.5P

(

C18H14Cl)

0.5(

CH)

0.1CHAR

0.62H2

0.25Cl2

1.00

29 300

[35]

183

(

C18H14Cl)

(CH

CH)

P 0

0.4P

(

C47H34Cl)

0.3Cl2

0.6(

CH)

0.02CHAR

0.74H2

1.00

33 500

[35]

184

(

C10H9)

(C47H35Cl)

P 0

(

C53H34Cl)

0.16CHAR

3.99H2

1.00

37 700

[35]

185

(

C18H15)

(C47H35Cl)

P 0

(

C53H34Cl)

0.49CHAR

5H2

1.00

37 700

[35]

186

(

CH)

(C47H35Cl)

P 0

0.9P

(

C53H34Cl)

0.1(

CH)

0.05Cl2

2.59H2

1.00

37 700

[35]

187

(

C10H9)

(

C10H8Cl)

P 0

(C18H15Cl)

0.08CHAR

0.5H2

2.42

18 900

(c)

188

(

C10H9)

(

C18H14Cl)

P 0

0.59P

(C47H35Cl)

0.2Cl2

1.07H2

6.48

22 100

(c)

189

(

C10H9)

(

C47H34Cl)

P 0

CHARC

1.37CHAR

7.75H2

1.04

29 900

(c)

190

(

C10H9)

(

C53H34Cl)

P 0

CHARC

1.62CHAR

6.25H2

1.11

31 200

(c)

191

(

C18H15)

(

C10H8Cl)

P 0

0.594P

(C47H35Cl)

0.2Cl2

1.07H2

6.48

22 100

(c)

192

(

C18H15)

(

C18H14Cl)

P 0

CHARC

0.5CHAR

6H2

4.34

25 400

(c)

193

(

C18H15)

(

C47H34Cl)

P 0

CHARC

1.7CHAR

8.75H2

1.40

33 200

(c)

194

(

C18H15)

(

C53H34Cl)

P 0

CHARC

1.95CHAR

7.25H2

1.48

34 400

(c)

195

(

C47H35)

(

C10H8Cl)

P 0

CHARC

1.37CHAR

7.75H2

1.04

29 900

(c)

196

(

C47H35)

(

C18H14Cl)

P 0

CHARC

1.7CHAR

8.75H2

1.40

33 200

(c)

197

(

C47H35)

(

C47H34Cl)

P 0

CHARC

2.91CHAR

11.5H2

1.13

41 000

(c)

198

(

C47H35)

(

C53H34Cl)

P 0

CHARC

3.16CHAR

9.99H2

2.38

42 200

(c)

199

(

C53H35)

(

C10H8Cl)

P 0

CHARC

1.62CHAR

6.24H2

1.11

31 200

(c)

200

(

C53H35)

(

C18H14Cl)

P 0

CHARC

1.95CHAR

7.25H2

1.48

34 400

(c)

201

(

C53H35)

(

C47H34Cl)

P 0

CHARC

3.16CHAR

10H2

2.38

42 200

(c)

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

550

202

(

C53H35)

(

C53H34Cl)

P 0

CHARC

3.41CHAR

8.5H2

1.26

42 000

(c)

203

(

C10H8Cl)

(

C10H8Cl)

P 0

(C18H15Cl)

0.08CHAR

0.5Cl2

2.42

18 900

(c)

204

(

C10H8Cl)

(

C18H14Cl)

P 0

0.59P

(C47H35Cl)

0.57H2

0.7Cl2

6.48

22 100

(c)

205

(

C10H8Cl)

(

C47H34Cl)

P 0

2CHARC

0.37CHAR

7.75H2

1.04

29 900

(c)

206

(

C10H8Cl)

(

C53H34Cl)

P 0

2CHARC

0.62CHAR

6.25H2

1.11

31 200

(c)

207

(

C18H14Cl)

(

C18H14Cl)

P 0

1.5CHARC

5.75H2

0.25Cl2

4.34

25 400

(c)

208

(

C18H14Cl)

(

C47H34Cl)

P 0

2CHARC

0.7CHAR

8.75H2

1.40

33 200

(c)

209

(

C18H14Cl)

(

C53H34Cl)

P 0

2CHARC

0.95CHAR

7.25H2

1.48

34 400

(c)

210

(

C47H34Cl)

(

C47H34CI)

P 0

2CHARC

1.91CHAR

11.5H2

1.13

41 000

(c)

211

(

C47H34Cl)

(

C53H34Cl)

P 0

2CHARC

2.16CHAR

10H2

2.38

42 200

(c)

212

(

C53H34Cl)

(

C53H34Cl)

P 0

2CHARC

2.41CHAR

8.5H2

1.26

42 000

(c)

213

(CH2

CH)

(

C10H8Cl)

P 0

0.66P

(C18H15Cl)

0.16C12

0.5H2

1.86

19 600

(c)

214

(CH2

CH)

(

C18H14Cl)

P 0

0.42P

(C47H35Cl)

0.28C12

1.05H2

2.49

22 800

(c)

215

(CH2

CH)

(

C47H34Cl)

P 0

CHARC

1.04CHAR

6.75H2

4.00

30 600

(c)

216

(CH2

CH)

(

C53H34Cl)

P 0

CHARC

1.29CHAR

5.25H2

4.24

31 900

(c)

217

(

C10H8Cl)

P 0

0.41CHARC

1.7H2

0.79CI2

1.56

0.5

13 200

(c)

218

(

C18H14Cl)

P 0

0.75CHARC

2.87H2

0.62C12

2.08

0.5

16 400

(c)

219

(

C47H34Cl)

P 0

CHARC

0.95CHAR

5.75H2

0.5C12

3.35

0.5

24 300

(c)

220

(

C53H34Cl)

P 0

CHARC

1.2CHAR

4.25H2

0.5C12

3.55

0.5

25 500

(c)

221

(

CHCHCl)

(

C10H8Cl)

P 0

0.66P

(C18H15Cl)

0.66C12

1.86

19 600

(c)

222

(

CHCHCl)

(

C18H14Cl)

P 0

0.42P

(C47H35Cl)

0.78C12

0.55H2

2.49

22 800

(c)

223

(

CHCHCl)

(

C47H34Cl)

P 0

2CHARC

0.04CHAR

6.75H2

4.00

30 600

(c)

224

(

CHCHCl)

(

C53H34Cl)

P 0

2CHARC

0.29CHAR

5.25H2

4.24

31 900

(c)

225

(

CH)

(

C10H8Cl)

P 0

0.25P

(C47H35Cl)

0.37C12

0.03H2

4.43

22 900

(c)

226

(

CH)

(

C18H14Cl)

P 0

0.42P

(C47H35Cl)

0.28CI2

0.05H2

5.93

26 200

(c)

227

(

CH)

(

C47H34Cl)

P 0

CHARC

1.04CHAR

5.75H2

9.55

34 000

(c)

228

(

CH)

(

C53H34Cl)

P 0

CHARC

1.29CHAR

4.25H2

1.01

35 200

(c)

229

(

CH)

(

C10H8Cl)

P 0

0.45CHARC

0.27C12

1.97H2

4.43

22 900

(c)

230

(

CH)

(

C18H14Cl)

P 0

0.79CHARC

0.1C12

3.14H2

5.93

26 200

(c)

231

(

CH)

(

C47H34Cl)

P 0

CHARC

CHAR

6H2

9.55

34 000

(c)

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

551

232

(

CH)

(

C53H34Cl)

P 0

CHARC

1.25CHAR

4.5H2

1.01

35 200

(c)

233

(

CCl)

(

C10H8Cl)

P 0

0.5CHARC

0.75C12

1.25H2

4.43

22 900

(c)

234

(

CCl)

(

C18H14Cl)

P 0

0.83CHARC

0.58C12

2.41H2

5.93

26 200

(c)

235

(

CCl)

(

C47H34Cl)

P 0

2CHARC

0.04CHAR

5.75H2

9.55

34 000

(c)

236

(

CCl)

(

C53H34Cl)

P 0

2CHARC

0.29CHAR

4.25H2

1.01

35 200

(c)

237

(CH

CH)

P 0

CH)

1.00

34 3300

(c)

238

(CH

CH)

P 0

0.26C6H6

0.2C2H2

2.00

13(a)

169 500

[33,49]

239

(CH

CH)

P 0

0.2P

(C10H10)

2.00

173 700

[33,49]

240

(C10H10)

P 0

0.55P

(C18H16)

0.55H2

1.00

194 600

[33,49]

241

(C18H16)

P 0

0.38P

(C47H36)

1.1H2

0.50

203 000

[33,49]

242

(C47H36)

P 0

1.95CHAR

6.24H2

0.10

217 700

[33,49]

243

(

C10H9)

P 0

0.66C6H6

(

CH)

0.12CHAR

0.25H2

C2H2

1.00

13(a)

129 800

[47]

(a) Activation energy is considered a function of relative dehydrochlorination (see

Benzene formation).

(b) Pseudo reaction of formation of chlorinated light gas.
(c) Kinetic parameters calculated in this work.

References

[1] H. Bockhorn, A. Hornung, U. Hornung, P. Jakobstroer, M. Kraus, J. Anal. Appl. Pyrol. 49 (1998)

106.

[2] E. Ranzi, M. Dente, T. Faravelli, G. Bozzano, S. Fabini, R. Nava, V. Cozzani, L. Tognotti, J. Anal.

Appl. Pyrol. 40

41 (1999) 305

319.

[3] T. Faravelli, G. Bozzano, C. Scassa, M. Perego, S. Fabini, E. Ranzi, M. Dente, J. Anal. Appl. Pyrol.

52 (1999) 87

103.

[4] H. Bockhorn, A. Hornung, U. Hornung, J. Anal. Appl. Pyrol. 50 (1999) 77

101.

[5] H.J. Borchardt, F. Daniels, J. Am. Chem. Soc. 79 (1957) 41.
[6] A.W. Coats, J.P. Redfern, Nature 201 (1964) 68.
[7] J.H. Flynn, J. Wall, J. Polym. Sci. Polym. Lc. 4 (1966) 323.
[8] H.L. Friedman, J. Polym. Sci. 6 (1963) 183.
[9] M.J.P. Slapak, J.M.N. Van Kasteren, A.A.H. Drinkenburg, Comput. Theor. Polym. Sci. 10 (1999)

481

489.

[10] G.M. Anthony, Polym. Deg. Stab. 64 (1999) 353

357.

[11] R. Bacaloglu, M. Fish, Polym. Deg. Stab. 47 (1995) 33

57.

[12] G. Montaudo, C. Puglisi, Polym. Deg. Stab. 33 (1991) 229

262.

[13] S.W. Benson, Thermochemical Kinetics, second ed, Wiley, New York, 1976.
[14] M. Dente, G. Bozzano, G. Bussani, Comput. Chem. Eng. 21 (1997) 1125

1234.

[15] M. Dente, G. Bozzano, M. Rossi, Proceeding of the First Conference on Chemical and Process

Engineering, Florence, Italy, 1993, pp. 163

172.

[16] M. Weissman, S.W. Benson, Int. J. Chem. Kinet. 16 (1984) 307.
[17] A.E. Shilov, R.D. Sabirova, Zh. Fiz. Khim. 6 (1959) 1365.
[18] K.P. Lim, J.V. Michael, J. Chem. Phys. 98 (1993) 3919

3928.

[19] D.L. Van Krevelen, Properties of Polymers, Elsevier, Amsterdam, 1990.
[20] G.A. Rasuvaev, L.S. Troitskaya, B.B. Troitskii, J. Polym. Sci. 9 (A-1) (1971) 2673

2688.

[21] S.W. Benson, The Foundation of Chemical Kinetics, McGraw-Hill, New York, 1960.

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

552

[22] K. Okada, E. Tschuikov-Roux, P.J. Evans, J. Phys. Chem. 84 (1980) 467

471.

[23] H. Heydtmann, G. Rinck, Z. Phys. Chem. 30 (1961) 250

261.

[24] G. Chuchani, J.A. Hernandez, I. Martin, Int. J. Chem. Kinet. 11 (1979) 1279

1286.

[25] E. Ranzi, M. Dente, M. Rovaglio, T. Faravelli, S.B. Karra, Chem. Eng. Commun. 117 (1992) 17

39.

[26] E. Ranzi, M. Dente, T. Faravelli, G. Pennati, Combustion Sci. Technol. 95 (1994) 1

50.

[27] L. Szirovicza, F. Marta, Int. J. Chem. Kinet. 8 (1976) 897.
[28] K. Scherzer, U. Loser, W. Stiller, Z. Chem. 27 (1987) 300.
[29] J.H. Knox, R.L. Nelson, Trans. Faraday Soc. 55 (1959) 937.
[30] R.B. Barat, J.W. Bozzelli, J. Phys. Chem. 96 (1992) 2494

2501.

[31] R.P. Lattimer, W.J. Kroenke, J. Appl. Polym. Sci. 27 (1982) 1355

1366.

[32] R.P. Lattimer, W.J. Kroenke, J. Appl. Polym. Sci. 27 (1982) 1355

1366.

[33] W. Grimme, L. Schumachers, W.R. Roth, R. Breuckmann, Chem. Ber. 114 (1981) 3197.
[34] D.K. Lewis, J. Bergmann, R. Manjoney, R. Paddock, B.L. Kaira, J. Phys. Chem. 88 (1984) 4112.
[35] A.M. Hogg, P. Kebarle, J. Am. Chem. Soc. 86 (1964) 4558

4562.

[36] I.C. Mcneill, L. Memetea, W.J. Cole, Polym. Deg. Stab. 49 (1995) 181

191.

[37] C.W. Gear, Commun. ACM 14 (3) (1971) 185

190.

[38] G. Buzzi Ferraris, D. Manca, Comput. Chem. Eng. 22 (11) (1998) 1595

1621.

[39] H. Bockhorn, A. Hornung, U. Hornung, P. Jakobstroer, J. Anal. Appl. Pyrol. 49 (1998) 98

106.

[40] H. Bockhorn, A. Hornung, U. Hornung, P. Jakobstroer, Combustion Sci. Technol. 134 (1998) 7

30.

[41] V.P. Gupta, L.E. St Pierre, J. Polym. Sci. 8 (A-1) (1970) 37

48.

[42] S. Ma, J. Lu, J. Gao, Energy Fuels 16 (2002) 338

342.

[43] U. Hornung, Bestimmung der Vergasungskinetik von Kunststoffen mi Hilfe eines isotherm

betriebenen Kreislaufreaktors, 1997.

[44] I.C. Mcneill, L. Memetea, Polym. Deg. Stab. 43 (1994) 9

25.

[45] N.L. Arthur, T.N. Bell, Rev. Chem. Intermed. 2 (1978) 37

74.

[46] S.I. Ahonkhai, X.-H. Lin, M.H. Back, Int. J. Chem. Kinet. 21 (1989) 1.
[47] W. Tsang, J. Phys. Chem. Ref. Data 19 (1990) 1

68.

[48] R.P. Widman, B.A. Degraff, J. Phys. Chem. 77 (1973) 1325

1328.

[49] K.E. Lewis, H. Steiner, J. Chem. Soc. 588 (1964) 3080

3092.

A. Marongiu et al. / J. Anal. Appl. Pyrolysis 70 (2003) 519

553

Document Outline

Thermal degradation of poly(vinyl chloride)

Wyszukiwarka

Podobne podstrony:
article pvc degradation 683c
Degradable Polymers and Plastics in Landfill Sites
Inny świat – hołd złożony człowiekowi i dokument degradacji człowieka, język polski
pvc
PVC
12 DEGRADACJA MATERIAŁÓW
Wykładziny podłogowe z PVC
degradacja
Okna z PVC i aluminiowe
degradacja ekosystemow wodnych
Degradacja i ochrona glebx
DEGRADACJA GLEBY biologiczna, Biologia UŚ !, Ochrona środowiska
cennik dren pvc 01 06 2013
Modelowanie termo chemicznej i mechanicznej degradacji betonu w wysokich temperaturach

więcej podobnych podstron

degradacja pvc

Document Outline