Vacuum pyrolysis of commingled plastics containing PVC I. Kinetic study

https://doi.org/10.1016/S0141-3910(01)00048-9Get rights and content

Abstract

Thermal decomposition of commingled plastics comprising high and low density polyethylene, polystyrene, polypropylene and poly(vinyl chloride) was investigated under vacuum conditions by dynamic thermogravimetric analysis in the temperature range of 25–600°C. The mixtures were representative of the major polymeric materials found in a municipal plastic waste (MPW) stream. The study focused on the role of each polymer in the stabilization or destabilization of other interacting plastics, as well as the fate of chlorine produced from PVC. First, the decomposition of each single plastic was investigated. Then the binary combination of all above polymers and finally the multicomponent mixture similar to a MPW sample were studied. Two approaches have been applied to study the interaction between components in the commingled plastics. The first approach involved a comparison of the experimental curve of the decomposition of the mixture with a calculated non-interacting component decomposition curve based on the behavior of the individual polymers. The second approach involved a comparison of the kinetic parameters of each plastic in the mixture as determined by two different methods: (1) fitting the single polymer decomposition curve and (2) fitting the peak corresponding to the pyrolysis of one particular polymer on the mixture pyrolysis curve. The results obtained indicate that some interactions occurred during MPW pyrolysis, mainly at high temperatures (>375°C). PS and PVC appear to be the plastics responsible for the interactions through their intermediate pyrolysis products. The chlorine from PVC in the mixture is released almost completely before 375°C, when the conversion of the MPW to pyrolysis products has reached ca. 13%.

Introduction

Thermal degradation of mixed plastics has recently received much interest because the disposal of municipal solid waste (MSW) has become a growing and costly problem as space available for landfilling decreases. Pyrolysis of waste plastics is of interest because plastic refuses represent an alternative source of energy and chemical raw materials.

In North America, the average composition of the MPW stream is 46 wt.% high and low density polyethylene (HDPE and LDPE), 16 wt.% polypropylene (PP), 16 wt.% polystyrene (PS), 7 wt.% polyvinyl chloride (PVC), 5 wt.% polyethylene terephthalate (PET), 5 wt.% acrylonitrile-butadiene-styrene copolymer (ABS) and 5 wt.% of other polymeric materials [1]. Through pyrolysis, some polymers such as PS can be readily depolymerized, leading to the recovery of a significant proportion of the original monomer. The reported yield of monomer from polystyrene varies between 40 and 60 wt.% [2]. Polyethylene, polypropylene and poly(vinyl chloride) do not yield significant amounts of monomers. However they are transformed into low molecular weight fragments and compounds which can be the source of valuable liquid fuels and new chemicals [3].

In order to recover high yields of useful products from MPW pyrolysis and in order to control any potentially harmful products released during the pyrolysis reactions, the decomposition behavior of the MPW must be known. It is particularly desirable to understand the decomposition behavior of each plastic component in the MPW stream, the interactions involved, and especially the release of chlorine during PVC pyrolysis. The particular problem caused by PVC decomposition is the formation of HCl and chlorinated hydrocarbons. The former can rapidly corrode the pyrolysis equipment. The latter may contaminate valuable pyrolysis products.

In previous studies, the thermal decomposition behavior of individual polymers such as PE [4], [5], [6], [7], [8], [9], [10], PP [11], [12], [13], [14], [15], PS [5], [6], [9], [15], [16], [17], [18], [19], [20] and PVC [21], [22], [23], [24] has been extensively studied. Decomposition mechanisms have been proposed and kinetic parameters were determined under different sets of conditions. The degradation behavior of a number of binary blends has also been studied by several researchers, with the purpose of modifying the thermal and thermo-oxidative stability or some physical properties of the plastic in a desired manner. Several authors have studied the degradation behavior of the following binary blends: PVC–polyacrylonitrile [25], PVC–poly(n-butyl methacrylate) [25], PVC–polyacrylamide [25], PVC–poly-(N-butyl methacrylamide) [25], PVC–(methyl acrylate) [25], PVC–poly(methylmethacrylate) (PMMA) [25], [26], PVC–poly-α-methylstyrene (PAMS) [27], PVC–poly(tetramethylene sebacate) (PTMS) [28], PVC–poly(vinyl acetate) (PVA) [29], PVC–chlorinated rubber [30], PVC–PS [31], PS–PMMA [32], PS–PVA [33], PS–polyisoprene [34], PS–polybutadiene [35] and PE/PS [31]. Other workers have studied the degradation of PP–PMMA [37], [38], PS–PAMS [39], PS–polyoxyethyleneglycol [40], PP in the presence of vinyl polymers [41], PVC–ABS [42], [43], [44], PVC–(polyester elastomer) (PETE) [45], PVC–polyurethane (PU) [43], [44], [46], PVC–(α-methylstyrene-acrylonitrile-methyl-methacrylate) (MS–AN) [47], PVC–PP [21], [43], [44], PVC–PMMA [48], PVC–PE [21], PVC–PS [21], PVC–polyamide 6 (PA6) [21], PVC–PET [49], PP–ABS [43], [44], PP–PU [43], [44], PP–PE [38], [50], [51], [52], PP–PS [50], PS–PE [50], [53], PS–poly(2,6-dimethyl-1,4-phenylene oxide) (PPU) [54], PS–PMMA [48], PE-PMMA [38], [48] and PS-Rubber (NR) [55].

Knümann and Bockhorn [21] have found no observable interaction between components of the PE/PS binary blends during degradation, while Koo and Kim [56], [57] and Mccaffrey et al. [53] for PE/PS mixtures, Csomorová et al. [38] for a PE/PP blend, Day et al. [43], [44] for PP/PVC mixtures, and Dodson and McNeill [31] who studied PVC/PS, have found significant interactions. Goulet and Prud'Homme [58] working with mixtures of PVC–poly(caprolactone) (PCL) and PVC–polyethylene adipate (PEA) and Jachowicz et al. [54] who studied PS–PPU blends using a non-isothermal thermogravimetric method demonstrated that in binary mixtures, the TG curve of the polymer mixture is not a simple superposition of the degradation curves of the component polymers of the blends, due to the interactions between the polymers. Csomorová et al. [38] also concluded that the thermal behavior of polymer blends is related to the miscibility of the respective components of a blend and to their interactions, where immiscible blends show better stability than miscible blends. McNeil et al. [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36] found that the nature of the interaction between different polymers strongly depends on the physical state of the system (the nature of the polymer, the miscibility of the polymer composition or the degree of phase dispersion). In a heterogeneous blend, interactions occur in the bulk of one or both domains and in the phase boundaries. In homogeneous samples, the degradation products of one polymer are directly in contact with the other polymers so that their combined effect on the thermal degradation is greater. McNeil et al. [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36] suggest that the interaction between degradation products of one polymer with others also occur as a result of the diffusion of small mobile molecules or radicals from one to the other through the interfacial layer. In that case, the degradation products of one polymer may stabilize or destabilize the other polymers in the blend.

Despite the fact that MPW is considered to be a source of energy and chemical raw materials, little information has been published until recently on the decomposition behavior of commingled plastics containing more than two polymers. Murata and Akimoto [50] studied binary and tertiary polymer mixtures (PE, PP, PS) by TG and GC analysis. The results indicated that when PS was present in the mixture, the other polymers were destabilized. The authors suggested that the mixing of polymers influence the decomposition rate but not the intra- and inter-molecular reactions. Based on their results, the mixing effect on the rate of decomposition was interpreted in terms of an intermolecular radical transfer between different polymer radicals. A similar result was reported by Dodson and McNeill [31] who mentioned that some radicals produced during PS degradation attack and thus destabilize the PE and PP polymers. Bockhorn et al. [59] studied complex plastic mixtures such as PVC/PS/PE, PS/PA6/PE and PVC/PA6/PS/PE. The authors have calculated the conversion degree of the mixtures, based on the assumption of no interaction between the polymers. Wu et al. [22] reported a kinetic model for a HDPE/LDPE/PP/PS/ABS/PVC mixture, assuming a summation of the degradation curves of each polymer in the mixture. After comparing the calculated results with those from the TG measurements, Wu et al. [22] concluded there was no observable interaction between the components during degradation.

The release of chlorine during the pyrolysis of pure PVC has been carefully studied by the authors in a previous study [23], [60]. That work revealed that 99.84 wt.% of the original chlorine is released between 200 and 360°C, which is lower than the decomposition temperature of most polymers found in the MPW stream. When PVC is mixed with other polymers, the PVC pyrolysis and the release of chlorine may be advanced or delayed, due to interactions with the other plastics. In order to isolate the chlorine released and to control the amount of chlorine in the pyrolysis products [61], the chlorine evolution pattern must be clarified.

Thus the objective of this work is to study the thermal decomposition behavior of commingled plastics with special attention given to the fate of the chlorine released during PVC decomposition. The study has two interconnecting parts: the study of the pyrolysis kinetics (Part I) and the analysis of the pyrolysis products (Part II). The objective of Part I is to study the interactions of polymers during the pyrolysis of a MPW sample, paying special attention to the influence of potential interactions and chlorine release during the PVC pyrolysis. Five principal plastic materials present in MSW, HDPE, LDPE, PP, PS and PVC were studied under vacuum and dynamic nitrogen conditions. These polymers are representative of MPW found in North America [1]. This work includes the following steps:

  • 1.

    The study of the thermal decomposition behavior and determination of the kinetic parameters of the single polymers involved.

  • 2.

    The study of the thermal decomposition behavior and determination of the overall kinetic parameters related to binary and multicomponent mixtures without PVC.

  • 3.

    The study of the thermal decomposition behavior and determination of the overall kinetic parameters related to PVC-containing binary and multicomponent mixtures.

The thermal decomposition behavior was studied by thermogravimetry (TGA) measurement. The kinetic parameters, including the apparent activation energies, the pre-exponential factors and the reaction orders were determined by DTG curve fitting, using an optimization procedure. In order to evaluate the polymer interactions in MPW, the experimental TG/DTG curves of the pyrolysis of a mixture were first simulated by a sum of the single decomposition curves, assuming that each plastic thermally decomposes independently and no interaction takes place. Secondly, kinetic parameters obtained by fitting the DTG peaks of the individual polymers were compared with those estimated by fitting the peak on the commingled plastic-related curve corresponding to that particular plastic.

Section snippets

Plastics investigated

The samples investigated were pure polymers without fillers, stabilizers and colorants including high density polyethylene (HDPE), low density polyethylene (LDPE), polypropylene (PP), polystyrene (PS) and polyvinyl chloride (PVC). The plastic samples used in this study were commercial grade materials representative of plastics commonly found in MSW. The manufacturers supplied the typical properties of the tested materials and their elemental composition, which are presented in Table 1. To

Kinetic study of individual polymers

The mechanism of single plastic pyrolysis is the basis for the understanding of the pyrolysis of commingled plastics. Most previous studies concerning single plastic pyrolysis were carried out under either nitrogen atmosphere or isothermal conditions. Here, investigation of single plastic pyrolysis under both vacuum and nitrogen atmosphere and dynamic conditions was carried out, as the main interest of this work is ultimately the pyrolysis of MPW in industrial vacuum pyrolysis continuous feed

Conclusions

A kinetic study of the pyrolysis of the five main plastics contained in a typical MPW stream revealed that some interactions occur between the plastics during the mixture pyrolysis. A comparison of the TG measured and the calculated curves in the hypothesis of non-interaction, clearly shows that the decomposition rate of most polymers, represented as DTG peaks, was significantly altered in terms of peak shape and positions. However, a comparison of the kinetic parameters shows that the

Acknowledgements

This paper has been carefully reviewed by Dr. Annette Schwerdtfeger. The authors would like to thank the Ministère de l'Éducation du Québec for the financial support (Programme québécois des bourses d'excellence) of one of the co-authors of this work, R.M. Thanks are due also to the polymer suppliers listed in Table 1 of this paper.

References (73)

  • C. Vovelle et al.

    Eur. Polym. J.

    (1983)
  • G.G. Cameron et al.

    Eur. Polym. J.

    (1968)
  • G.G. Cameron et al.

    Eur. Polym. J.

    (1970)
  • A. Marcilla et al.

    Polym. Degrad Stab

    (1995)
  • R. Miranda et al.

    Polym Degrad. Stab

    (1999)
  • A. Marcilla et al.

    Polym. Degrad. Stab

    (1995)
  • I.C. McNeill et al.

    Eur. Polym.J.

    (1970)
  • I.C. McNeill et al.

    Eur. Polym. J.

    (1971)
  • I.C. McNeill et al.

    Polym. Degrad. Stab

    (1991)
  • I.C. McNeill et al.

    Polym. Degrad. Stab

    (1980)
  • I.C. McNeill et al.

    Eur. Polym. J.

    (1970)
  • N. Grassie et al.

    Eur. Polym. J.

    (1975)
  • D.H. Richards et al.

    Polymer

    (1967)
  • L.-P. Blanchard et al.

    Eur. Polym. J.

    (1974)
  • S.V. Kolesov et al.

    Polymer Science USSR

    (1989)
  • T. Kovacic et al.

    Polym. Degrad. Stab

    (1995)
  • D.M. Bate et al.

    Polym. Degrad. Stab

    (1998)
  • Y. Sakata et al.

    Polym. Degrad. Stab

    (1996)
  • C. Albano et al.

    Polym. Degrad. Stab

    (1998)
  • W.R. Waldman et al.

    Polym. Degrad. Stab

    (1998)
  • R. Asaletha et al.

    Polym. Degrad. Stab

    (1998)
  • J.-K. Koo et al.

    Resour. Recovery Conserv

    (1991)
  • J.-K. Koo et al.

    Waste Manage. Res.

    (1993)
  • L. Goulet et al.

    Eur. Polym. J.

    (1986)
  • H. Bockhorn et al.

    J. Anal. Appl. Pyrolysis

    (1998)
  • J.A. Conesa et al.

    J. Anal. Appl. Pyrolysis

    (1996)
  • J.K.Y. Kiang et al.

    Polym. Degrad. Stab

    (1980)
  • G.P. Ravanetti et al.

    Thermochim. Acta

    (1992)
  • D.H. Richards et al.

    Polymer

    (1967)
  • US EPA. Characterization of municipal solid waste in the United States, 1990 update. EPA 530-SW-90-042. Environmental...
  • I.C. McNeill

    Thermal degradation

  • Grassie N, Scott G, editors. In: Polymer degradation and stabilization. London: Cambridge University Press, 1985. p....
  • W.G. Oakes et al.

    J. Chem. Soc. London

    (1949)
  • H.H.G. Jellinek

    J. Polym. Sci.

    (1949)
  • S.L. Madorsky

    J. Polym. Sci.

    (1952)
  • R. Inoue et al.

    Chem. High Polym. Jpn

    (1956)
  • Cited by (118)

    • Kinetic modelling of mixed plastic waste pyrolysis

      2023, Chemical Thermodynamics and Thermal Analysis
    View all citing articles on Scopus
    View full text