S Baker, 2011. Fundamental Investigations On Petroleum Coke and Polymer Interactions For Recycling Waste Polymers in Aluminium Production

Fundamental investigations on petroleum coke and polymer
interactions for recycling waste polymers in aluminium

production
Author:
Syed Baker, Sharifah Shahnaz
Publication Date:
2011
DOI:
https://doi.org/10.26190/unsworks/4462
License:
https://creativecommons.org/licenses/by-nc-nd/3.0/au/
Link to license to see what you are allowed to do with this resource.
Downloaded from http://hdl.handle.net/1959.4/55284 in https://

unsworks.unsw.edu.au on 2022-06-22
The University of New South Wales
Faculty of Science
School of Materials Science and Engineering
FUNDAMENTAL INVESTIGATIONS ON PETROLEUM COKE

AND POLYMER INTERACTIONS FOR RECYCLING WASTE
POLYMERS IN ALUMINIUM PRODUCTION
A thesis in
Materials Science and Engineering
By
SHARIFAH SHAHNAZ SYED BAKAR
Submitted as Partial Fulfilment

of the requirements for the Degree of
DOCTOR OF PHILOSOPHY
August 2011
ORIGINALITY STATEMENT
‘I hereby declare that this submission is my own work and to the best of
my knowledge it contains no materials previously published or written
by another person, or substantial proportions of material which have
been accepted for the award of any other degree or diploma at UNSW
or any other educational institution, except where due
acknowledgement is made in the thesis. Any contribution made to the
research by others, with whom I have worked at UNSW or elsewhere, is
explicitly acknowledged in the thesis. 1 also declare that the intellectual
content of this thesis is the product of my own work, except to the
extent that assistance from others in the project's design and conception
or in style, presentation and linguistic expression is acknowledged.’
~yo\2
COPYRIGHT STATEMENT
‘I hereby grant the University of New South Wales or its agents the right to
archive and to make available my thesis or dissertation in whole or part in the
University libraries in all forms of media, now or here after known, subject to the
provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent
rights. I also retain the right to use in future works (such as articles or books) all
or part of this thesis or dissertation.
I also authorise University Microfilms to use the 350 word abstract of my thesis
in Dissertation Abstract International (this is applicable to doctoral theses only).
I have either used no substantial portions of copyright material in my thesis or 1
have obtained permission to use copyright material; where permission has not
been granted I have applied/will apply for a partial restriction of the digital copy
of my thesis or dissertation.'
Signed
Date
AUTHENTICITY STATEMENT
‘I certify that the Library deposit digital copy is a direct equivalent of the final
officially approved version of my thesis. No emendation of content has occurred
and if there are any minor variations in formatting, they are the result of the
conversion to digital f<
Signed
Date
ACKNOWLEDGEMENTS
This dissertation would not have been possible without the guidance and the help of
several individuals who have contributed their valuable assistance in the completion of
this study.
First and foremost I offer my sincerest gratitude to my supervisors, Professor Veena

Sahajwalla and Associate Professor Rita Khanna, who have supported me throughout
my research with patience and knowledge whilst allowing me to work independently. I
am thankful for their guidance to overcome the obstacles to complete this research.
I am grateful to Professor Dato' Dr. Kamarudin Hussin, my supervisor in Malaysia, who

helps facilitates this opportunity for me to study at UNSW. 1 am also indebted to
UniMAP and the Ministry of Higher Education Malaysia for providing me the
scholarship, which was critical for my study in Australia.
It is a pleasure to thank those who made this thesis possible, Mr Saha Chaudhury for
helping to maintain the equipment in the laboratories, Dr Yu Wang for help in XRD
analyses, Philip Chatfield and Dr. Rahmat Kartono for helping with the furnace, George
Yang for helping with liquid nitrogen, polishing stuffs and many other areas, Irene
Wainwright for conducting XRF analyses, Dr Dilip Nath for helping with TGA
analyses, Dr Karen Privat and Eugene White for training and advice in SEM
investigations, Barry and Toby for facilitating me in FTIR study, Lana Strizhevsky for
always reminding me that the clock is ticking and other technical staff in the school and
the Analytical Centre for their contributions to improving the quality of my dissertation.
My special thank goes to all my friends and group mates for always being there during
my ups and downs, for brightening my gloomy days and for providing continuous
support throughout the hardships of PhD endeavors. I am grateful to my family for their
love, support and believe, and to God for blessing me with this opportunity, hope and
success.
ABSTRACT
Global demand for plastics has grown significantly over the past decades, and will
continue to expand with rising income levels in emerging economies; a number of
approaches have been used to recycle polymer waste. While chemical recycling is one
of the key methods used as it recovers and reuses the polymer in high-end product; new
avenues for waste recycling need to be developed. Consumable carbon anodes are a
major requirement for the Hall-Heroult process used for producing primary aluminium;
nearly 0.4-0.5 tonnes of carbon is consumed to produce every tonne of aluminium.
Because carbon is a major constituent of waste plastics, which have very low impurity
levels, these clearly have the potential as a cheap readily available auxiliary source of
carbon in carbon anodes. Waste polymers can also be used as alternate resource to
replace/partially substitute pitch. Coal tar pitch, a major by-product produced in
petroleum refining is the binder of choice for carbon anodes. Pitch penetrates the pores
of petroleum coke binding particulates and gets carbonized during the baking process.
However, the usage of pitch is highly risky as it contains a number of known and
potential carcinogens, including benzene, naphthalene and other polycyclic aromatic
hydrocarbons (PAHs).
In-depth wettability and interfacial phenomena investigation was carried out to study
interactions between polymers and petroleum coke (PC). The effect of two main
parameters, temperature and time on the polymer properties and the effect of petroleum
coke presence on the degradation process of polymer have been characterized. The
wettability study of polypropylene (PP) and polyethylene (PE) polymer on PC
substrates at temperature ranging from 150°C to 350°C in 15 min to 60 min has been
carried out. Result shows that PP has a higher degradation rate compared to PE, but has
similar penetration depth.
The wettability behaviour of PP has been further carried out by soaking treatment at
temperatures ranging from 200°C to 350°C and heat treatment up to 600°C on the
soaked samples. The soaking treatment resulted in a negligible loss of polymer,
indicating that PC was able to contain the PP degradation residues in its porous
structures from escaping and devolatilized. However, the FTIR study shows no
chemical bonding formed between PP and PC and all adhesion was physical in nature.
The effect of heat treatment cycles on the blend of 10% to 90% of PP,and 10% to 50%
of PE with PC has also been investigated. No change was detected after Cycle 1 (heat
up to 150°C and dwelled for 30 min); traces of polymer residue has been observed after
Cycle 2 (Cycle 1 - then heat up to 600°C and dwelled for 30 min); and polymer was
completely lost after Cycle 3 (Cycle 2 - then heat up to 1000°C and dwelled for 30
min). A number of analytical investigations including XRD and FT1R analysis and SEM
investigation were carried out on heat treated specimens.
Thermal degradation study on polymers was carried out to understand the behaviour of
polymer during increasing temperature and time of wettability and heat treatment
studies. TGA result showed PP degraded completely after 490°C but the presence of PC
has delayed the process. The effect of polymer properties on the polymer degradation
and PC interactions such as thermoset Bakelite, oxygen-contains PET and pigmented
PP has also been studied. Significant differences were observed between the behaviour
of various polymer types.
The research reported in this thesis provides the basic framework required to recycle
waste polymers in aluminium production. Further studies are however required on
coke/polymer interactions and on the influence of polymers on the carbonisation
characteristics of coal-tar pitch binder.
TABLE OF CONTENTS
ORIGINALITY STATEMENT ii
ACKNOWLEDGEMENTS iii
ABSTRACT iv
TABLE OF CONTENTS vi
LIST OF FIGURES xii
LIST OF TABLES xxii
LIST OF PUBLICATIONS xxvi
CHAPTER 1 INTRODUCTION 1
1.1 Introduction 2
1.2 Contributions of the work in this thesis 3
1.3 Objectives 4
CHAPTER 2 LITERATURE REVIEW 6

2.1 Recycling of Waste Plastics 7
2.1.1 Plastics 7
2.1.2 Waste Plastic Recycling 8
2.1.2.1 Material/mechanical recycling of waste plastics 10
2.1.2.2 Thermal recycling of waste plastics 12
2.1.2.3 Chemical recycling of waste plastics 13
2.1.3 Recycling of Waste Plastics as Mixtures in Pavement 14
Construction
2.1.4 Recycling of Waste Plastic in Steelmaking 16
2.2 Pyrolysis of Waste Plastics 18
2.3 Mechanisms of Reaction Occurring During 24
Thermal Degradation of Polymers
2.3.1 Thermal Degradation of Polyethylene 27
2.4 Polypropylene (PP) 29
2.4.1 Polymerization of Polypropylene 29
vi
2.4.2 Crystallinity of Polypropylene 31
2.4.3 Oxidation of Polypropylene 32
2.4.4 Degradation of Polypropylene 33
2.4.4.1 Thermal activation energy of Polypropylene 34
2.4.5 Pyrolysis of Polypropylene 35
2.4.5.1 Effect of temperature on degradation of PP 35
2.5 Substitution of Polymers in Anode Baking in 39
Aluminium Making Process
2.5.1 Key Components of Carbon Anodes 39
2.5.2 Coking and Calcination of Petroleum Coke 41
2.5.3 Polymer as Filler/Substitute for Coke in Carbon 45
Anode for the Aluminium Industry
2.6 Wettability 46
2.6.1 Contact Angle 49
2.6.2 Wettability Behaviour of Polymers 51
2.6.3 Morphological and Crystalline Studies of Polymer 53
Deformation
2.7 Summary 56
CHAPTER 3 EXPERIMENTAL 57
3.1 Materials Investigated 58
3.1.1 Polypropylene (PP) 58
3.1.2 Polyethylene (PE) 58
3.1.3 Polypropylene Black (PPB) 59
3.1.4 Polyethylene Terephthalate (PET) 60
3.1.5 Bakelite 60
3.1.6 Petroleum Coke (PC) 61
3.2 Experimental 63
3.2.1 Wettability Studies 63
3.2.2 Soaking and Heat Treatment on PP and PC Substrates 67
vii
3.2.3 Heat Treatment on the Mixtures of Polymer and 68
Petroleum Coke Studies
3.2.4 Polymer Thermal Degradation Studies 71
3.3 Analytical Instruments 72
3.3.1 Thermogravimetic Analysis 72
3.3.2 Carbon Analysis 74
3.3.3 X-Ray Diffraction Analysis 74
3.3.4 FTIR Analysis 75
3.3.5 Scanning Electron Microscope 77
CHAPTER 4 RESULTS & DISCUSSION: 78

POLYPROPYLENE
4.1 Interactions of PP with PC Substrates 79
4.1.1 Interfacial Phenomena Between PP and PC 87
4.1.2 Discussion 94
4.2 Interactions Between PP and PC for 97
Temperature Up to 1000°C
4.2.1 Weight Loss Study of the PP and PC Mixtures 98
4.2.2 Effect of PC on the Reactivity of Mixtures 100
4.2.3 Carbon Content of the PP and PC Mixtures 100
4.2.4 Structural Characterization of PP and PC Blends 102
4.2.5 FTIR Analysis of PP and PC Blends 104
4.2.6 SEM Images of PP and PC Mixtures 107
4.3 Thermal Degradation of PP 109
4.3.1 Analysis of Solid Residues of PP Pyrolysis 112
4.3.2 Chemical Analysis of PP Pyrolysis Residues 116
4.3.3 Morphological Study of PP Pyrolysis Residues 121
4.3.4 Structural Characterization of PP Pyrolysis Residues 129
4.3.5 FTIR Reactions Characterization of PP Pyrolysis 134
Residues
4.4 Summary 138
viii
CHAPTER 5 RESULTS & DISCUSSION: 141
INTERACTIONS OF
POLYPROPYLENE WITH
PETROLEUM COKE AND
DEGRADATION MECHANISMS
5.1 Effect of Low Temperature Treatment on the 142
Pyrolysis Yield of PP:PC Blends
5.1.1 Weight Loss after Low Temperature Soaking 142
5.1.2 Heat Treatment to 600°C on Soaked Samples 144
5.2 Nature of Bonding between Polypropylene and 145
Petroleum Coke
5.3 Thermal Degradation of Polymer 149
5.3.1 Reaction Kinetics 149
5.3.2 Calculation of Activation Energy by Arrhenius 155
Equation
5.4 Degradation Mechanisms for Polypropylene 157
5.5 Summary 159
CHAPTER 6 RESULT & DISCUSSION: 160

POLYETHYLENE
6.1 Interactions of PE with PC Substrates 161
6.1.1 Interfacial Phenomena Between PE and PC 171
6.1.2 Discussion 177
6.2 Interactions Between PE and PC for 180
Temperature Up to 1000°C
6.2.1 Weight Loss Study of PE and PC Mixtures 180
6.2.2 Carbon Content of PE and PC Mixtures 182
6.2.3 Structural Characterizations of PE and PC Mixtures 183
6.2.4 FTIR Analysis of PE and PC Blends 186
6.2.5 SEM Images of PE and PC Mixtures 188
6.3 Thermal Degradation of PE 189
IX
6.3.1 Analysis of Solid Residues of PE Pyrolysis 192
6.3.2 Chemical Analysis of PE Pyrolysis Residues 194
6.3.3 Morphological Study of PE Residues 196
6.3.4 Structural Characterization of PE Pyrolysis Residues 200
6.3.5 FTIR Reactions Characterization of PE Residues 204
6.4 Degradation Mechanisms for Polyethylene 209
6.5 Comparison on the Effect of Heat Treatment on 212
PE and PP
6.6 Summary 213
CHAPTER 7 RESULT & DISCUSSION: INFLUENCE 217

OF POLYMER TYPE
7.1 Interactions of PET with PC Substrates 218
7.1.1 Interfacial Phenomena between PET and PC 226
7.1.2 Discussion 231
7.2 Interactions between Bakelite and PC for 233
Temperatures up to 1000°C
7.2.1 Weight Loss Study of Bakelite and PC Mixtures 234
7.2.2 Carbon Content of Bakelite and PC Mixtures 235
7.2.3 Structural Characterizations of Bakelite and PC 236
Mixtures
7.2.4 FTIR Analysis of Bakelite and PC Blends 240
7.2.5 SEM Images of Bakelite and PC Mixtures 242
7.3 Thermal Degradation of Polypropylene Black 244
(PPB)
7.3.1 Analysis of Solid Residues of PPB Pyrolysis 244
7.3.2 Degradation of PPB 245
7.3.3 Morphology Study of PPB Residues 246
7.4 Summary 249
X
CHAPTER 8 SUMMARY AND CONCLUSIONS 254
8.1 Summary and Conclusions 255
8.2 Future Studies 260
CHAPTER 9 REFERENCES 261

LIST OF FIGURES
Figure 2.1 : Plastics material flows in Australia |PAC1A, 2009] 10
Figure 2.2 : Breakdown of mechanically recycled waste by plastic types 11
Figure 2.3 : Reaction mechanism of polymer (Scheirs and Kaminsky, 26

2006]
Figure 2.4 : Random chain scission in polyethylene 27
Figure 2.5 : Degradation mechanisms of polyethylene as suggested by 28

Wampler [1989]
Figure 2.6 : Polymerization of propylene monomer with metal catalyst 30
Figure 2.7 : Stereochemical configurations of polypropylene 31
Figure 2.8 : Oxidative radicals formation of PP 32
Figure 2.9 : The major products of PP by their pathways 37
Figure 2.10 : TGA and DSC curves of thermal degradation of PP in 37

nitrogen (dashed line) and air (solid line)
Figure 2.11 : Product of PP pyrolysis as a function of temperature 38
Figure 2.12 : General reaction scheme of carbonization |Lewis,1982] 41
Figure 2.13 : Carbonization and graphitization of organic materials 43

(Mochida et al., 2006]
Figure 2.14 : Schematic illustration of the solid-liquid wetting principle 47

(Rocha et al., 2005]
Figure 2.15 : Effect of temperature on the droplet on the solid surface in 48

isothermal penetration test (Cao et al., 2002]
Figure 2.16 : Contact angle of liquid droplet and solid surface 49
Figure 3.1 : Raw polypropylene samples 58
Figure 3.2 : Raw polyethylene samples 59
Figure 3.3 : Raw polypropylene black samples 59
Figure 3.4 : PET glass and raw samples after ground 60
Figure 3.5 : Raw bakelite samples 60
xii
Figure 3.6 : Raw petroleum coke samples 62
Figure 3.7 : RockLab ring milling machine 64
Figure 3.8 : Enerpac 10 Tonne Hydraulic Press 64
Figure 3.9 : Petroleum coke substrate 64
Figure 3.10 : Horizontal tube furnace used for pyrolysis 66
Figure 3.11 : Polymer on PC substrate assemblyon alumina powder 66

based graphite holder in horizontal tube furnace
Figure 3.12 : (a) Contact angle measurement and (b) depth of penetration 67
measurement of polymer melt into petroleum coke
substrate
Figure 3.13 : Petroleum coke was screened by Retsch sieving machine to 69

obtain particles in the size range of 2 - 5 mm
Figure 3.14 : Fritsch knife grinder 69
Figure 3.15 : Schematic diagram of horizontal tube furnace set up for 71

pyrolysis of polymers and mixtures samples
Figure 3.16 : Perkin Elmer thermogravimetric analyser 73
Figure 3.17 : LECO CS 444 carbon and sulphur analyzer 74
Figure 3.18 : Philips Xpert Multipurpose X-ray Diffraction System 75

(MPD)
Figure 3.19 : Sample preparation material and instruments (a) KBr 76

powder (b) mould for KBr pellet and (c) hydraulic press for
KBr pellet fabrication
Figure 3.20 : Nicolet Avatar 320 FT-IR instrument 76
Figure 3.21 : Hitachi S3400-X scanning electron microscope 77
Figure 4.1 : Raw PP on petroleum coke substrate before heat treatment 79
Figure 4.2 : PP on petroleum coke substrate after exposure to 150°C for 80

a) 15 min b) 30 min and c) 60 min
Figure 4.3 : PP on petroleum coke substrate and the corresponding 80

SEM micrograph of the interfacial area after 15 min of
reaction at 200°C
xiii
reaction at 200°C

reaction at 200°C

reaction at 250°C

reaction at 250°C

reaction at 250°C

reaction at 300°C

reaction at 300°C

reaction at 300°C

reaction at 350°C

reaction at 350°C

reaction at 350°C
Figure 4.15 : Effect of time and temperature on (a) contact angle and (b) 86
adhesion of PP on PC substrate
Figure 4.16 : EDS analysis of points in the interface of PP-PC system 88

reacted for 60 minutes at 200°C
XIV
Figure 4.17 : Interfacial changes due to heating at 200°C for different 89
time intervals

time intervals

time intervals
Figure 4.20 : Schematic diagram of PP melt penetration into PC 92

substrate

time intervals
Figure 4.22 : Variation of the penetration depth of PP melt into 94

petroleum coke substrates as a function of heating
temperature and time
Figure 4.23 : Residue (%) of PP and PC mixtures after heat treatment 98
Figure 4.24 : TGA profiles of raw PP, raw PC and mixtures their 100
mixture of 10:90
Figure 4.25 : Carbon content (%) of PP and PC mixtures 101
Figure 4.26 : XRD profiles of raw PP and PC 102
Figure 4.27 : XRD profiles of PP and PC mixtures sample after heat 103
treatment
Figure 4.28 : FTIR peaks of raw PP and PC 105
Figure 4.29 : FTIR peaks as observed in the mixtures after various 106
cycles of heat treatment
Figure 4.30 : SEM micrographs and EDS of raw petroleum coke 107
Figure 4.31 : SEM micrographs and EDS of raw PP 108
Figure 4.32 : SEM micrographs and EDS of PP and PC (50:50)mixture 109

samples after heat treatment
Figure 4.33 : Physical changes of PP residues according to time and 110

temperature in the lower temperature range from 400°C-
600°C
Figure 4.34 : Physical changes of PP residues according to time and 111

temperature in the higher temperature range from 650°C-
850°C
xv
Figure 4.35 : Effect of temperature on solid yield of PP pyrolysis low 113
temperature range (400 — 600°C) and high temperature
range (650 - 850°C)
Figure 4.36 : Effect of time on solid yield of PP pyrolysis 115
Figure 4.37 : Effect of temperatures on carbon content (%) of PP 117

residues in the low temperature range (400 - 600°C) and
high temperature range (650 - 850°C)
Figure 4.38 : Effect of time on carbon content (%) of PP residues 117
Figure 4.39 : Ash content (%) of product of PP at low temperature range 119
(400 - 600°C) and high temperature range (650 - 850°C)
Figure 4.40 : Physical and chemical changes during thermal 120

decomposition (National Fire and Society of Fire, 1995]
Figure 4.41 : SEM micrographs of raw PP samples (2mm ground 121

particles) at low magnification and (b) high magnification
Figure 4.42 : SEM micrographs of PP residues after pyrolysis according 125-128

to time and temperature
Figure 4.43 : Effect of temperature on yield samples of PP pyrolysis. 130

XRD spectra of samples at 650°C to 850°C for 5 min
pyrolysis compared to raw PP
Figure 4.44 : Effect of time on yield samples of PP pyrolysis. XRD 133

spectra of samples for 10 to 30 min at 500°C compared to
raw PP
Figure 4.45 : Effect of temperature on residues of PP pyrolysis. FTIR 135

band of samples after 650°C and 750°C for 5 min pyrolysis
compared to raw PP
Figure 4.46 : Effect of time on yield samples of PP pyrolysis. FTIR band 137
of samples for 10 min and 20 min at 550°C pyrolysis
compared to raw PP
Figure 5.1 : Weight loss (%) of 10, 20 and 30% PP in PC blend 143
substrates after soaking for 30 min under 0.5 L/min argon
flow at 200°C, 250°C, 300°C and 350°C
Figure 5.2 : Effect of 30 min soaking at 250°C on the specimen 143

morphology
Figure 5.3 : Weight loss (%) of 10, 20 and 30% PP in PC blend 144
substrates after up to 600°C heat treatment on samples
soaked at 200°C, 250°C, 300°C and 350°C
XVI
Figure 5.4 : Effect of heat treatment up to 600°C on soaked substrate 144
samples
Figure 5.5 : FT1R profiles of raw PP and PC 145
Figure 5.6 : FT1R profiles of 10% PP in PC substrate blend after 250°C 146
heat treatment
Figure 5.7 : FTIR profiles of 10% PP in PC substrate blend after 300°C 147
heat treatment
Figure 5.8 : FTIR profiles of 10% PP in PC substrate blend after 350°C 148
heat treatment
Figure 5.9 : The degradation pattern of PP with different heating rates 151
Figure 5.10 : The difference between programmed temperature and 152

actual measured temperature of TGA profile at 10°C/min
heating rate
Figure 5.11 : Calculation of reaction order in PP decomposition with 154

various heating rates
Figure 5.12 : Activation energy using Arrhenius equation 156
Figure 5.13 : Plausible reaction pathways of PP degradation mechanism 158

as suggested by Yasushi and Kaoru |2003]
Figure 6.1 : Raw PE on petroleum coke substrate surface 161
Figure 6.2 : PE on petroleum coke substrate after 15, 30 and 60 min 162
heat treatment at 150°C
Figure 6.3 : PE on petroleum coke substrate after 15 min of reaction at 162

200°C

200°C

200°C

250°C

250°C

250°C
xvii
300°C

300°C

300°C

350°C
Figure 6.13 : PE on petroleumcoke substrate after 30 min of reaction 167

after at 350°C

350°C
Figure 6.15 : Effect of time and temperature on estimated contact angle 169
and adhesion of PE melt on PC substrate
Figure 6.16 : EDS analysis of the PE/coke interface after heating at 171
350°C for 30 minutes
time intervals
time intervals
time intervals
Figure 6.20 : Penetration and spreading of PE melt into the petroleum 175
coke substrate
time intervals
Figure 6.22 : Variation in the penetration depth of PE melt as a function 177

of time and temperature
Figure 6.23 : Residue left (%) of PE and PC mixtures 182
Figure 6.24 : Carbon content (%) of heat treated PE and PC mixtures 183
Figure 6.25 : XRD profiles of raw PE and PC 183
Figure 6.26 : XRD profiles of raw PE and PC mixtures at 50:50 184
xviii
Figure 6.27 : FTIR peaks of raw PE and PC 186
Figure 6.28 : FTIR peaks of PE and PC mixtures at 50:50 187
Figure 6.29 : SEM micrographs and EDS of raw PE 188
Figure 6.30 : SEM micrographs and EDS of PE and PC (50:50) mixtures 189
samples after treatment
Figure 6.31 : Physical states of pyrolyzed PE residues with different 191

times and temperatures
Figure 6.32 : Effect of temperature on yield % of products of PE 193
Figure 6.33 : Effect of time on yield % of products of PE 193
Figure 6.34 : Effect of temperature on carbon content (%) of products of 195

PE
Figure 6.35 : Effect of time on carbon content (%) of products of PE 195
Figure 6.36 : SEM micrographs of raw PE samples (2 mm ground 196

particles) at lower magnification and (b) higher
magnification
Figure 6.37 : Morphology of PE residues after pyrolysis 198-199
Figure 6.38 : XRD spectra of samples at 650°C to 750°C for 10 min 201
compared to raw PE
Figure 6.39 : XRD spectra of samples for 5 min to 20 min at 750°C 203
compared to raw PE
Figure 6.40 : FTIR band of samples at 650°C, 700°C and 750°C for 10 206
min pyrolysis
Figure 6.41 : Effect of pyrolysis time (5 min to 20min) at 750°C 208

pyrolysis on structure of solid residues
Figure 6.42 : Two possible reaction at elevated temperature (400-600°C) 210
Figure 6.43 : Plausible degradation mechanisms suggested by Scheirs 211

and Kaminsky [2006]
Figure 7.1 : Raw PET on petroleum coke substrate before heat 219
treatment
Figure 7.2 : PET on petroleum coke substrate after 15, 30 and 60min of 219
heating at 150°C
XIX
heating at 200°C
heating at 250°C
Figure 7.5 : PET on petroleum coke substrate after 15 min of reaction 220
at 300°C
at 300°C
at 300°C
at 350°C
at 350°C
at 350°C
at 400°C
at 400°C
at 400°C
Figure 7.14 : Effect of time and temperature on (a) contact angles and 226
(b) adhesion of the PET melt on PC substrate
Figure 7.15 : EDS analysis of points at the PET/PC interface of the 226
samples heated for 15 minutes at 300°C
Figure 7.16 : Interfacial region after heating at 300°C for different time 227
intervals
intervals
intervals
Figure 7.19 : Variations in penetration depth of PET melt with time and 230
temperature of heat treatment
XX
Figure 7.20 : Residue (%) of Bakelite:PC mixtures after heat treatment 235
Figure 7.21 : Carbon content (%) of Bakelite:PC (50:50) residues after 236
heat treatment
Figure 7.22 : XRD profiles of raw PC and Bakelite 236
Figure 7.23 : XRD profiles of Bakelite and PC mixtures sample after 237
heat treatment
Figure 7.24 : FTIR peaks of raw Bakelite and PC 240
Figure 7.25 : FTIR peaks as observed in the mixtures after various 241
cycles of heat treatment
Figure 7.26 : SEM micrographs and EDS of raw Bakelite 242
Figure 7.27 : SEM micrographs and EDS of BakeliteiPC (50:50) 243

mixtures after heat treatment
Figure 7.28 : Effect of temperature on solid yield (%) of products of PPB 245
Figure 7.29 : TGA profiles of PP and PPB 245
Figure 7.30 : SEM micrographs of raw PPB samples (2mm particles) at 246
(a) low magnification and (b) high magnification
Figure 7.31 : SEM micrographs of PPB residues - effect of temperature 247

for 30 min
Figure 7.32 : SEM micrographs of yield products of PP compared to 248

PPB - effect of time on morphology of pyrolysis residues
after 500°C
Figure 8.1 : Variations in contact angles for three different polymers 255
(PP, PE and PET) with petroleum coke as a function of
Figure 8.2 : Average penetration depths of three different polymers 256

(PP, PE and PET) into petroleum coke as a function of
xxi
LIST OF TABLES
Table 2.1 : Three major forms of waste plastic recycling [Plastic 9
Waste Management Institute, 2004J
Table 2.2 : Waste power generation around the world 13
Table 2.3 : Parameters affecting product distribution of plastic 20

pyrolysis [Scheirs and Kaminsky, 2006]
Table 2.4 : Reaction steps and mechanisms of thermal degradation 25

of waste plastics
Table 2.5 : Product of PP pyrolysis according to temperature 38

[Scheirs and Kaminsky, 2006]
Table 2.6 : Wetting contact angle between polymers on solid 52

materials [Luechinger et al., 2008|
Table 3.1 : Carbonaceous materials compositions, (dry wt% basis) 58

ofPP

of PE

ofPPB

of PET
Table 3.5 : Ingredients in Bakelite samples 61

of Bakelite

of PC
Table 3.8 : Physical appearance of polymers used in this study 62
Table 3.9 : Ash compositions (%) of different types of raw 63

materials
Table 3.10 : Wettability studies samples and experimental 65

parameters
Table 3.11 : Experiment conditions of low temperature soaking and 68

600°C heat treatment on PP and PC blend substrates
xxii
Table 3.12 : Heat treatment method of polymer and petroleum coke 70
mixtures
Table 3.13 : Mixtures percentage of sample and experimental 70

conditions of pyrolysis of mixtures of polymer and
petroleum coke
Table 3.14 : Experimental conditions of pyrolysis of polymers 72
Table 4.1 : Dimensional changes in the PP droplet on the PC 86

substrate for different reaction temperatures and times
Table 4.2 : Penetration depth of PP into petroleum coke substrate 93
Table 4.3 : Percentage of solid yield of each sample mixtures 99

(PP:PC) obtained after baking Cycle 1, Cycle 2 and
Cycle 3
Table 4.4 : Carbon content (%) of each sample mixtures (PP:PC) 101
obtained after baking cycles
Table 4.5 : XRD peaks characteristics of raw PP, PC and their 104
mixtures at 50:50*
Table 4.6 : Characterization of FTIR peak profiles of PP, PC and 107

their mixtures at 50:50. The units of data are in (cm'1)
Table 4.7 : Percentage of residue obtained after pyrolysis of PP 113
Table 4.8 : Effect of temperature and time on carbon content (%) of 116
PP residues
Table 4.9 : Ash content (%) of solid residue of PP 118
Table 4.10 : Estimated calculated particle size (agglomerated 122

particles) of PP pyrolysis residues
Table 4.11 : Effect of temperature on XRD peak parameters after 129

650°C to 850°C for 5 min pyrolysis
Table 4.12 : Effect of time on XRD peak parameters for 10 min to 30 132
min at 500 °C
Table 4.13 : Effect of temperature on the peak assignment of PP 134

residues at 650°C and 750°C for 5 min
Table 4.14 : Effect of time on the peak assignment of PP residue for 138
10 min and 20 min at 550°C
xxiii
Table 5.1 : List of peak assignments of 10% PP in PC substrate 147
blend after 250°C heat treatment


Table 5.4 : Pyrolysis kinetic parameters for polypropylene 154
Table 5.5 : Pyrolysis activation energy using Arrhenius equation 156
Table 6.1 : Dimensions of PE melt on PC substrate after reactions 168

at different temperatures and times
Table 6.2 : Variations in the depths of penetration depth of PE into 176

petroleum coke with temperature and time
Table 6.3 : Percentage of solid yield of sample mixtures (PE:PC) 181

obtained after baking Cycle 1, Cycle 2 and Cycle 3
Table 6.4 : Carbon content (%) of PE:PC mixtures 182
Table 6.5 : XRD peaks characteristics of raw PE, PC and their 185
mixtures at 50:50
Table 6.6 : Characterization of FTIR peak profiles of PE, PC and 188

their mixtures at 50:50
Table 6.7 : Percentage of residue obtained after pyrolysis of PE 192
Table 6.8 : Ash content (wt.%) of raw PE 196
Table 6.9 : Effect of temperature on the structural parameters of PE 202

residues XRD peaks for 10 min
Table 6.10 : Effect of time on the crystalline features of PE residues 204

after 750°C pyrolysis
Table 6.11 : Effect of temperature on FTIR peak assignment of PE 207

residue for 10 min
Table 6.12 : Effect of time on FTIR absorption bands of residue at 209

750°C
Table 6.13 : Comparison between PE and PP behavior 212-213
Table 7.1 : Variations in the dimensions of the PET melt on PC 225

substrate with time and temperature
XXIV
Table 7.2 : Depth of penetration of PET melt into PC substrate 230
Table 7.3 : Percentage of solid yield of each sample mixtures 234

(BakeliteiPC) obtained after baking cycle
Table 7.4 : Carbon content (%) of BakeliteiPC mixtures 235
Table 7.5 : XRD peaks characteristics of raw PC, Bakelite and their 239
mixtures at 50:50
Table 7.6 : Characterization of FTIR peak profiles of PC, Bakelite 242

and their mixtures at 50:50
Table 7.7 : Solid residues (wt.%) of PPB 244
Table 7.8 : Comparison of polymers properties during the heat 251-253

treatment
XXV
LIST OF PUBLICATIONS
• Sharifah Shahnaz, Rita Khanna, Kamarudin Hussin, Veena Sahajwalla (2012)

“Wettability and interfacial phenomena investigations on high-density polyethylene
and petroleum coke” Journal of Applied Polymer Science. Volume 125, Issue 3,
pages 2056-2062.
• Shahnaz S., Khanna R., Kamarudin H. and Sahajwalla V. (2011) “High

Temperature Interactions of Polypropylene with Petroleum Coke. ” Journal of
Applied Polymer Science (to be submitted)
xxvi
Chapter 1: Introduction
CHAPTER 1
INTRODUCTION
1
1.1 Introduction
Plastics production and usage has grown significantly in the last 30 years and continued
future growth is expected. Due to the relatively short life spans of plastic goods, there
has been a tremendous growth in the generation of plastic wastes. Post-consumer plastic
wastes are projected to increase to 30 million tonnes per year in the USA alone with the
worldwide waste generation levels being much higher [Curlee, 19861. *n Australia,
more than a million tonnes of plastics are consumed each year, of which only 18.5%
was recycled in 2008 [PACIA, 2009). About 60% of wastes plastics are either
incinerated or disposed off in landfills. There are serious environmental concerns
regarding the use of these methods and this had led to an increased urgency to develop
novel, cost-effective and environmentally sustainable recycling techniques. Although
the waste plastic component in municipal solid waste (MSW) amounts to only ~10wt%,
plastics have a high volume to weight ratio and are generally not biodegradable [Panda
et al., 2010; Lee and Shin, 2007]. With respect to the use of the two main waste plastic
management methods, ~10% of waste is incinerated while the remainder is disposed
landfills. The disposal of waste in landfills requires an active extraction industry,
availability of free land in locations close to waste generation, favourable geology, low-
cost transport facilities and meeting of regulatory requirements (Miskolczi et al. 2006;
Delattre et ah, 2001]. The incineration of plastics on the other hand can lead to waste
destruction and detoxification, with energy recovered from the burning of wastes;
however it also results in the emission of a several environment pollutants |CurIee,
19861.
Recent developments include the commercial utilisation of waste plastics in the power
industry and in blast furnaces in Japan, Korea and Germany [Asanuma et ah 1997].
Through laboratory based research focussed on carbon/slag interactions, UNSW
researchers have established that waste plastics can be utilized in Electric Arc Furnace
steelmaking. A partial replacement of conventional metallurgical coke with high density
polythene (HDPE) waste produced significant levels of slag foaming, and lowered
energy consumption by 8-15 KWhr/tonne [Sahajwalla et ah, 2006].
There is a scope to develop novel uses for polymeric wastes as a carbon source in
consumable carbon anodes for the aluminium industry. The commercial production of
2
aluminium via the Hall-Heroult electrolytic process is an energy intensive process; the
production of one tonne of aluminium requires —13-16 MWhr of direct electric current,
up to half a tonne of carbon, and two tonnes of alumina. Alumina is dissolved in a
molten cryolite bath, and an electric current is passed through the solution, thereby
separating alumina in to aluminium and oxygen. The oxygen immediately reacts with
the carbon anodes to produce carbon dioxide as an off-gas. The process consumes large
amounts of sacrificial carbon from the anodes which are lowered to maintain a constant
distance between the anode and the surface of metal pool (part of the cathode); therefore
prebaked anodes must be replaced regularly. Carbon anode oxidation is responsible for
almost 90% of the on-site CO2 emissions from aluminium production (1.6 metric tonnes
of CO2 per metric tonne of primary aluminium). There is an increasing need for
developing lower cost, robust anodes with performance superior to that of current
anodes, with extended life, decreased CO2 emissions and improved sustainability.
The aim of this research is to develop consumable carbon anodes for the aluminium
industry using waste plastics. This study, for the first time, investigates high
temperature interactions of a number of polymers with petroleum coke. Carbon anodes
are manufactured by baking blends of some varieties of coke with hydrocarbon binder,
which is generally a coal-tar pitch [Couderc et al., 1986]. Since waste plastics are rich
in carbon and have low impurity levels, these materials have a potential to be used as a
cheap, readily available, auxiliary source of carbon. The influence of prolonged
treatment at high temperatures (up to 1000°C) associated changes in properties and the
gas-phase reactivity of petroleum coke/pitch binder/waste plastics blends needs to be
investigated as a function of type/concentration of waste plastics.
1.2 Contributions of the work in this thesis

Polypropylene and polyethylene have been chosen as a major materials studied in this
thesis because these two polymers are considered as the most abundant plastic wastes
available throughout the world. The wettability behavior between polymers and
petroleum coke will be investigated to study the interaction and compatibility between
these two raw materials. By partly replacing polymer as a binder, it is expected to aid
the recycling of wastes polymers and reduce the emission of PAHs from the
conventional binder.
3
The interaction between polypropylene and petroleum coke will be further explored by
mixing both raw materials, then pre-soaking before charging into furnace for high
temperature treatment. The heating cycle treatments on the mixture of petroleum coke
and polymer will be carried out to investigate the effect of petroleum coke on the
degradation steps of polymers. As polymer is known to have low softening/melting
points, the addition of petroleum coke has the potential of increasing the mixtures’
melting temperature. Polymer thermal degradation also will be studied as the behaviour
of polymer is highly influenced by temperature. The understanding of polymer
characteristics during heating is essential to foresee the behaviour of these polymers
when mixed with petroleum coke.
1.3 Objectives
For utilizing waste polymers as a carbon resource in aluminium processing, it is
essential to develop the fundamental knowledge that will decrease the reliance of the
industry on conventional carbonaceous materials (petroleum coke and coal-tar pitch).
The objectives of this research are:
• To investigate the interactions between a wide range of polymers and petroleum

coke with focus on wettability, interfacial phenomena, effects of temperature
(150-400°C) and processing time (15-60 min). These parameters have been
chosen based on typical temperatures/times required for polymer stability and
degradation.
• To investigate the influence of petroleum coke on the carbon yield from

polymers for a number of heat treatment cycles (up to 1000°C). The heat
treatments will be conducted in: Baking cycle 1 (up to 150°C) to investigate
initial changes in polymer characteristics; Baking cycle 2 (up to 600°C) as
thermoplastic polymers are expected to complete their degradation under
temperatures up to 600°C; and Baking cycle 3 (1000°C) as baking of anode is
generally carried out at temperatures ~1000°C.
• To determine thermal degradation mechanisms of polymers in the temperature

range: 400-850°C and time ranges from 2 to 30 min for their optimum utilization
4
as a carbon and binder resource in carbon anodes. The effect of temperature and
time on the pyrolysis residues will be investigated.
• To determine the nature of bonding form between PC and polymer (PP) during
the heat treatment and the effect of binder on the PP and PC blends; to determine
the effect of polymer’s types (thermoplastic and thermoset) and properties
(polymer containing oxygen and pigmented polymer) on the residues.
5
Chapter 2: Literature Review
CHAPTER 2
LITERATURE REVIEW
2.1 Recycling of Waste Plastics
2.1.1 Plastics
Plastic is generally prepared from petroleum by-products and natural gas. These are
composed of high molecular polymeric compounds containing primarily carbon,
hydrogen and a few other elements such as N and O. Crude oil from petroleum is
processed and refined to produce raw materials for plastics. Two main groups of plastics
are thermoplastic and thermoset. Thermoplastics are plastics with properties that soften
when heated and harden when cooled. The main reason is being little bonding between
individual molecular chains. In contrast to thermoplastics, thermoset plastics do not
soften again after they have hardened. The heating of thermoset plastic will cause its
thermal degradation. It is due to the three-dimensional structures and crosslinking
formed during its curing process. Rubber has a structure intermediate between
thermoplastic and thermoset, with molecular chains linked by sulphur bridges during the
vulcanization. A predominant of commodity plastics is made of thermoplastic
materials.
Plastics can also be classified on the basis of criteria such as chemical composition,
chemical structure, stiffness, type of application and processing methods. Chemical
compositions can be described in terms of monomers and polymerization method.
Plastic are subdivided into classes of polyolefins, vinyl polymers, polyamide,
polyesters, epoxy resins, polycarbonates etc. Plastic also can be classified by their
chemical structure, such as linear (polypropylene), branched (low-density
polyethylene), cross-linked and three dimensional networked (thermosets, rubber).
Plastics are further subdivided by their elasticity, flexibility or rigidity, including its
stiffness criteria.
In addition, plastics can also be classified by the various processing methods used
including injection moulding, extrusion, film blowing, blow moulding, thermoforming,
casting, calendaring, and many other techniques. These can be further divided by their
application types, such as commodity, engineering, general purpose and specialty
plastics [Scheirs and Kaminsky, 2006].
7
2.1.2 Waste Plastic Recycling

Statistics have reported total global plastic production has increased by an average of
almost 10% annually since 1950. The total plastic production has grown from around
1.5 million tonnes (MT) in 1950 to 245 MT in 2006 globally. USA is the highest
consumer of plastics of all countries with 27.3 MT against 170 MT of world
consumption in 2000 and expected to reach 39 MT in 2010 [Panda et al., 2010].
Plastics have been one of the materials with the fastest growth because of their wide
range of applications, but the duration of plastic life cycle is relatively small; which
caused the serious environmental problem every year.
According to the data reported, 20-25 wt% of the plastic wastes is disposed by
incineration while up to 65-70 wt% ended up in landfills. Only a small fraction of 10-20
wt% was recycled. This data varies from country to country, however they are
approximately near with some exceptions. Incinerated waste plastics produce energy
recovery; these however generate the highly toxic compounds such as dioxins and
furans. [Panda et al., 2010; Lee and Shin, 2007; Buekens and Huang, 1998|. In
Australia, a total of 1,525,185 tonnes of plastic were consumed, while only 282,032
tonnes (18.5wt%) were recycled in 2008. This was an increase of 8% compared with the
previous year [PACIA, 2009|.
There are three methods of recycling plastic wastes that include material or mechanical
recycling, thermal recycling and chemical recycling. Three major forms of waste plastic
recycling are summarized below in Table 2.1 while Figure 2.1 represents the routes of
plastic flow in Australia.
The system that has been extensively used over a period of time is material recycling,
where the materials are reused without any chemical modifications. This recycling
method may apply to single or mixtures of plastic wastes. Most of the recycled plastics
are used in low grade applications such as fencing, playground equipment, garden
furniture, plastic pallets, pipes, plastic lumber and wood plastic composites. However,
this method may not be the best possible way of recycling because the properties of the
materials often get diminished at their end of life cycle [Williams and Slaney, 2007].
8
Table 2.1: Three major forms of waste plastic recycling [Plastic Waste Management
Institute, 2004]
Category Method of Recycling
Re-used without any chemical modifications

Mechanical recycling
Reprocessed into plastic raw materials
Monomerization
Chemical recycling Blast furnace reducing agent
Coke oven chemical feedstock recycling
Pyrolysis Chemical feedstock

Gasification
Liquefaction Fuel
Thermal recycling
Cement kiln
Waste power generation
High calorific value of plastic materials has enabled them to be recycled for recovery of
energy; heat produced by waste incineration can be used to drive steam turbines and to
generate electricity. In addition, the production of solid fuel also contributes to thermal
recycling of waste plastics. Chemical recycling, also termed as feedstock recycling is
the cracking or depolymerisation of plastic into fuel or chemical substances, which can
then be used as raw materials for producing plastics and chemicals for industries. The
products of feedstock recycling are generally liquids and gaseous products (Murty et
al., 1996; Williams and Slaney, 2007; Siddiqui and Redhwi, 2009].
9
AUSTRALIAN CONSUMPTION
Manufacturing Processes In-house recovery
Post-consumer Industrial Post-consumer Domestic

Pre-consumer Industrial
Material Use
Material
(packaging/durables) (packaging/durables)
DISPOSAL OPTIONS
Landfill Collection & Sorting Waste Treatment
Domestic Reprocessing Export for Reprocessing
Domestic Use Export for Use
Local Use Internal Use
Figure 2.1: Plastics material flows in Australia |PACIA, 2009|
2.1.2.1 Material/mechanical recycling of wastes plastics

Material or mechanical recycling is one of the key processes used to convert and reuse
the waste plastic for low-grade new products. It is a procedure to make new products
out of unmodified plastic waste. This process involves melting and remoulding of used
thermoplastic [Aguado et al., 2007; Di Blassi, 1997]. As this process involves reusing
the same material for the same or down-graded into particular applications, it is limited
to single type of waste plastic to meet the requirement of its end-products. Figure 2.2
shows the breakdown of waste by the plastic type that has been recycled mechanically
in 2002 as reported by the Plastic Waste Management Institute [2004].
10
Figure 2.2: Breakdown of mechanically recycled waste by plastic types
The development of recycling approaches for thermoplastic such as low-density

polyethylene (LDPE), high-density polyethylene (HDPE), polyethylene terephtalate
(PET) and polypropylene (PP) is highly encouraged since these accounts for most of the
total plastic wastes.
In mechanical recycling, the materials that have been collected and sorted are cleaned
and crushed into flakes or granulated pellets. These flakes and pellets are then sent to
the plant where the remelting and remoulding of plastic parts takes place. The original
materials principally remain unchanged in their consistency and their polymeric
structure is maintained [Hegberg, 1992|. The range of recycled products made from
industrial plastic waste come in all kinds and shapes, including containers, benches and
fences, children play equipment, transportation, constructions, railway, parks and other
goods.
The processing methods that are commonly used include extrusion moulding, where the
pellet or flakes were charged into the extruder and melted and continually extruded
through screw mould to form desired products; injection moulding operates by injecting
the melted resin into a mould and it solidifies according to the mould shape, and then air
inflated so the plastic will take the mould shape; blow moulding technique is commonly
used for the making of any types of bottles; others techniques generally used are
vacuum moulding, used to make cups and trays, and inflation moulding is used to make
shopping bags.
11
An investigation on the properties of recycled HDPE from milk bottles was carried out
by Pattanakul et al., [ 1991 ]. It was seen that the tensile strength of different
compositions of virgin and recycled HDPE was quite similar. The incorporation of
recycled HDPE up to about 50% had shown no decrease in impact resistance, whereas a
slight decrease in impact resistance was observed for 80 to 100% recycled materials. It
could be concluded that the properties of recycled HDPE from milk bottles did not show
much degradation and it could be used in a range of applications. Recycled HDPE is
generally much cheaper than virgin HDPE.
Other recycling method that suitable for PP is the use of PP resin as matrix in wood
fiber reinforced plastic composites. A number of manufacturers have substituted
concrete, wood with plastic lumber. PP has shown good potential for use in the process
as it is relatively stiff. [Hegberg, 1992]. Polypropylene and wood fibres were
compounded using machine to produce homogeneous mixture, as the fibres of wood
could be observed uniformly. The use of PP as matrix composites has improved the
moisture absorption of the composites, as compared to wood alone. It has higher
durability, chemical resistance, strength and modulus that would fix for product with
practical mechanical requirements. Most of the fibres used were first treated to improve
the compatibility with polymer matrix |Felix and Gatenholm 1991; Coutinho et al.,
1997; Hornsby etal.,1997].
2.1.2.2 Thermal recycling of waste plastics

Incineration of plastics to convert the waste into energy is known as thermal recycling.
This method is generally used to recycle mixtures of plastic waste. The energy recovery
from waste plastic is promoted due to high calorific values properties. The main focus
nowadays on thermal recycling of waste plastics is on waste power generation, stoker
ovens, gasification and melting furnaces, and gasification reformers, which are likely to
become major resources of generating electricity. The number of waste incineration
facilities is increasing steadily. The incineration process equipped with boilers to
generate steam and hot water use in heating system, healthy spas etc. Table 2.2 shows
the waste power generation around the world reported by Plastic Waste Management
Institute [2004] in 2002.
12
Table 2.2: Waste power generation around the world
Country Number of facilities Power generation capacity (MW)

USA 107 2300
Germany 50 1000
Japan 210 86
Holland 10 50
France 80 30
2.1.2.3 Chemical recycling of waste plastics

Chemical recycling or feedstock recycling is one of the common methods with
promising potential applications for recycling waste products. The process allows the
conversion of used polymer into low molecular weight materials, which consist of their
liquid or gaseous hydrocarbons for chemical industries. Chemical recycling can be
classified into thermal decomposition, depolymerisation (monomerization) and
gasification (partial oxidation) (Iwaya et al., 2006].
As plastics are made from petroleum, it is possible to reproduce oil from them [Di
Blassi, 1999]. The process involves reversing the manufacturing process. The
development of chemical recycling of plastics has been investigated extensively.
Chemical recycling or feedstock recycling has several stepwise advantages; it does not
have to undergo separation process and can be applied to mixed plastic. This type of
recycling generally employs plastics that cannot be recycled by the mechanical route. It
also extends the overall recycling capacities for large waste plastic quantities in the
future |Yaman, 2004; Bhaskar et al., 2003].
Chemical recycling has been proposed as one of the most attractive methods for
sustainable developments in the field. Waste polymers can be converted back into their
original monomers or other valuable chemicals. Pyrolysis, one of the key chemical
recycling techniques has been used to produce a series of refined petrochemical
products and particularly liquid fractions similar to commercial gasoline JAchilias D. S.
et al., 2007].
13
Appropriate combustion procedure is essential in order to obtain the combustion

products of polymeric materials. Large sample combustion in a relatively small area
generates high product concentration. Flameless or limited flaming combustions occur
mostly with incomplete combustion of material forming highly toxic compounds.
The flameless combustion of polyethylene produces hydrocarbons with the number of

carbon atoms ranging from C3 to C26. There is always higher content of olefins with
terminal double bonds. The combustion of polyethylene at 1000°C will generate largely
CO2. The remaining products are olefins, paraffin and ketones and some aromatic
hydrocarbons are produced from the secondary aromatization reactions of primary
products. [Mitera and Michal, 1985].
There are no significant differences between the product of flaming and non-flaming
combustion of polypropylene and polyethylene. During flaming combustion, more
products are formed as the decomposition temperature is increased by the exothermal
combustion reaction. Combustion of PP at 1000°C generates primarily carbon dioxide
products and some products of secondary aromatization reactions, such as benzene and
its alkyl derivatives.
Polystyrene decomposes to monomer, dimer and trimer of styrene under various

conditions. A range of complex compounds and mixtures are formed by thermal non
homogeneity of the medium. The traces of chlorinated hydrocarbons have also been
found indicating the residue of the solvent used in the polymer treatment. Combustion at
1000°C mainly produces CO2. Further reactions lead to benzene and its alkyl
homologues, as well as the dimer and trimer of styrene has been formed in very small
quantity, similar to benzaldehyde [Mitera and Michal, 1985].
2.1.3 Recycling of Waste Plastics as Mixtures in Pavement

Construction
In UK, it is estimated 2.8 million tons of waste plastics are generated annually, most of
which come from industrial and commercial resources and those collected from
domestic sources. The recycling effort that has gained momentum in recent years is the
amalgamation of plastic waste into the mixtures of aggregates in pavement construction.
14
There are two methods of incorporation waste plastics into the raw materials of
pavements, where it can be used as binder modifier or to replace the portion of
aggregates [Huang et al., 2007|.
The replacement of 2.36-5.00 mm aggregates by 30% LDPE has been termed as

‘Plastiphalf has reduced the mix density by 16% and showed 250% increase in stability
and indirect tensile strength. The 15% LDPE in the size range of 0.30 and 0.92mm
aggregates in asphalt surfacing mixtures doubled the stability performance and resulted
in the improvement of rutting and water resistance. Other properties that were reported
to improve by the addition of waste plastic were the stability and aging properties,
flexural behaviour and bending strength. Additional forms of plastic recycling are
generally street furniture, insulation, ducts and pipes etc. [Huang et al., 2007].
The incorporation of recycled polyolefin into modified bitumen has been reported to
improve the properties of neat bitumen. The blend between bitumen and plastic waste
mixtures was carried out using an open mixer aided by a stirring device. According to
these studies, the synthetic polymers appear to produce better results as it can act as
more flexible bituminous binder at low service temperatures and resulting in enhanced
properties, preventing the pavement from being deformed at high in-service
temperatures.
The results showed that the addition of waste plastic has increased the elastic modulus,
which related to the ageing of bitumen due to the oxidation during its processing, thus
decreased the thermal fracture during low temperature service. As for the high
temperature service, it was observed that the shear-thinning behaviour of the modified
bitumens soften as much lower values of the flow index are obtained for the bitumen
with highest polymer content. The addition of scrap tyres was also found to be
favourable as it provides higher dissipation of energy preventing the occurrence of
thermal cracking when stressed by the heavy traffic loadings on the pavement structure
during low temperature service [Garcia-Morales et al., 2006].
15
2.1.4 Recycling of Waste Plastic in Steelmaking

Even though there are more than 13 million tonnes of plastic wastes generated annually,
but the recycling effort of the plastic fraction is still low, only about 10% in Western
Europe as reported by Williams and Slaney [2007]. The Japan Iron & Steel Federation
has set their goal into reducing their industrial energy consumption to 90% based on
their 1990 level by 2010. A voluntary action plan is developing efforts to achieve 1.5%
of energy saving by recycling one million tons of waste plastics annually as raw
material for steelmaking. In order to build sustainable recycling environment, Nippon
Steel Corporation has developed a new technology for waste plastic recycling using
coke ovens. This process was put into practice for the first time in the world in 2000.
The total equipment capacity that operates in these days is 225,000 units per annum
[Kato et al., 2006].
The production of steel from recycled plastic is one of the major discoveries in
recycling waste polymer as a carbon resource. The coke making process is basically
putting the coal into the coke oven chamber and carbonized in a superheated reducing
atmosphere free of oxygen. The plastic mixtures were crushed, and then any metals,
heavy objects and other contaminants were removed as pre-treatment process. The
plastics mixtures were then agglomerated into balls with coals to enhance the
transportation and charging it into coke oven.
The carbonization products are coke (solid) about 20%, coke oven gas (COG) about
40%, tar and light oil (liquid) around 40%. As for plastic wastes, the PE, PS, PET and
PP begin to thermally decompose at about 300°C and completed around 420 to 470°C.
This characteristic behaviour shows that plastic decomposed at temperatures lower than
coal. The decomposition of plastic around 200 to 450°C transformed plastics into
gaseous form and produced carbides at 500°C and over.
The resultant char properties were then evaluated and these showed that the inclusion of
up to 1% of waste plastic mixtures to coal did not affect the coke strength. The
formation of voids by thermal decomposition of plastic at lower temperature than the
softening point of coal has allowed the free expansion of coal that consequently
16
produced a brittle coke structure and the residue of decomposed plastics were observed
inside the voids that occurred after the thermal decomposition of plastics.
Furthermore, the early decomposition of plastic mixtures generated pyrolytic gases in

the temperature range in which coal softens, thus it may affect the coal coking
properties. Therefore, the amount of plastic mixtures added into coal was limited to 1%
only as the coke strength would decrease if the plastic mixture exceeded 1%. Some
parameters such as grain size and bulk density must be adjusted in order to reduce the
area of contact between coal and plastic mixtures if more than 1% of plastic mixtures
need to be added to the coal. The coke produced is generally used as an iron ore
reducing material in a blast furnace, while the light oil is used as raw material for
various plastics and the COG is used for energy generator at thermal stations [Kota et
al., 20061.
Another interesting development that was put into practice was the plastic waste as blast
furnace feed material in Ohgishima. The coke gets gasified into CO, and is used for
reducing oxidized iron (iron ore) to produce iron. Waste plastics are ground and
granulated before being injected through tuyeres into the blast furnace. The
decomposition of plastic produced CO and FC, which both act as reducing iron for
production of iron. About 60% of waste plastic charged into the blast furnace was used
for the reducing the iron ore. The remaining 40% of the plastics was effectively
exploited as fuel in power generating stations and blast oven.
The use of FC in addition of CO as reducing iron ore could reduce CO2 emissions by
30% when blast furnace used only metallurgical coke. Solid plastics, film plastics such
as plastic bottles are pulverized and granulated into specified form and injected directly
into the blast furnace. The only limitation of this process was the incorporation of PVC
as the generation of corrosive hydrochloride tended to erode the furnace equipment
[Hotta, 2003]. The plastics that contain PVCs are shredded and PVCs are separated
before waste being transported to the steelworks.
17
2.2 Pyrolysis of Waste Plastics

The recycling of waste plastic needs to include the range of recycling processes to
convert plastic waste into petrochemical feedstock for use in the production of refined
chemicals and fuels. The technology that can accept a range of plastic waste irrespective
of their types and properties is called thermal cracking. This method of recycling offers
a number of advantages such as decreasing the air pollution that is normally emitted
during the incineration and soil contamination in the landfill. In addition, it also will
reduce the cost and space needed for disposal in a landfill. The main thermal cracking
process discussed and developed, which also produces reliable product yields is
pyrolysis |WilIiams and Williams, 1997; Scheirs and Kaminsky, 2006; Angyal et al.,
2007; Siddiqui and Redhwi, 2009].
Pyrolysis, also termed as thermolysis, comes from the Greek word: pur means fire;
thermos means warm; luo means loosen. Pyrolysis as a process consists of chemical and
thermal reactions, generally leading to the smaller molecules. Pyrolysis may be
performed by using a range of parameters such as temperature, reaction time, pressure,
the presence or absence of reactive gases or liquids, and use of catalysts. Plastic
pyrolysis can be conducted in three ranges of temperatures, which are low (<400°C),
medium (400 - 600°C) and high (>600°C). Key effects of various parameters on the
pyrolysis of waste plastics are summarized in Table 2.4.
Besides, pyrolysis needs no separation of complex plastic wastes, so this key aspect can
save one step of the processing or treatment cost [Lee and Shin, 2007]. However,
thermal decomposition of waste plastic containing PVC is still not favoured as it may
produce chlorine compounds during the process [Bhaskar et al., 2004; Bhaskar et al.,
2005]. Crystallinity can also significantly influence the thermal properties of polymers.
Generally the most known properties affected are the sharp melting point and the
stiffening of thermal mechanical properties. The thermal effect is known to significantly
reduce mechanical properties of polymers, as compared to the other properties such as
electrical characteristic. Amorphous plastics generally have a gradual softening range.
The molecular weight affects the flexibility and brittleness of the material at low
temperatures, while the orientation tends to decrease the dimensional stability at higher
temperatures. Temperature also has a complex effect on the intermolecular bonding,
18
cross linking and copolymerization of polymeric materials. These factors need to be

considered while determining the plastic product design [Shah, 2007|.
Polyethylene, high-density polyethylene (HDPE) and low-density polyethylene (LDPE);

polypropylene (PP); polystyrene (PS); polyethylene terephthalate (PET) and polyvinyl
chloride (PVC) are five types of the most common polymer wastes observed in
municipal solid waste stream. Waste from electric and electronic equipment, such as
computer housings are made of high impact polystyrene (HIPS) and ABS. Recently,
significant amounts of PET have been used widely with PP and PE as laminated sheets
and also in various other applications. These wastes also contain various contaminants,
such as fillers, fire retardants, anti-oxidants, colour pigments or coupling agents
[Bhaskar et al., 2003].
The pyrolysis of model mixed plastics compared to waste plastic mixtures that mainly
contains PE, PP, PVC PET and PS was performed by Bhaskar et al. [2003| in a glass
reactor under atmospheric pressure. In a typical run, the reactor was purged with
nitrogen gas and held at 120°C for 60 min to remove moisture from the waste, then the
nitrogen flow was stopped and the reactor temperature was increased to the degradation
temperature. It was noted that the addition of PET had a great impact in controlling the
amount of chlorine contents emitted during pyrolysis. The presence of PET has
increased the formation of new chlorinated hydrocarbons and at the same time
decreasing the inorganic chlorine compounds in the liquid products.
The analysis of similar waste plastic model pyrolysis also has been performed by
Williams and Slaney [2007]. The selection of PP, PE (HDPE), PS, PET and PVC as a
model is preferable as these are typically found in the majority of European municipal
solid waste. The characterization of the individual plastic present was also carried out to
study the effect of nitrogen or hydrogen atmosphere, pyrolysis conditions, liquefaction
conditions etc. Other samples analysed and compared were mixed plastics obtained
from the green dot recycling scheme in Germany. The experiment was conducted at
5°C/min heating rate to the final temperature of 500°C, with 0.2 MPa nitrogen pressure,
generating a maximum pressure of approximately 10 MPa, held for 1 hour. The
liquefaction reaction was also carried out at the final temperature of 500°C with initial
19
hydrogen pressure of 1 MPa was applied to increase the hydrogen availability as a

reagent for reaction. The final pressure was 18 MPa.
Table 2.3: Parameters affecting product distribution of plastic pyrolysis

[Scheirs and Kaminsky, 2006|
Parameters Effect
The primary pyrolysis products depend directly on
Chemical composition of the chemical structure and composition of resin, and
resins also the mechanism of its de-composition (purely
thermal or catalytic).
Higher operating temperatures and high heating
Pyrolysis temperature and
rates both enhance bond breaking and favour the
heating time
production of smaller molecules.
Longer residence time favours the formation of
secondary primary products, yielding more coke,
Pyrolysis time tar, as well as thermally stable products, thus
gradually obscuring the effect of original polymer
structure.
Determines mainly the quality of heat transfer,
Reactor type mixing, gas and liquid residence times, and the
escape of primary products.
Low pressure reduces the condensation of reactive
Operating pressure
fragments forming coke and heavy chain products.
Presence of reactive gases, Such presence internally generates heat, dilutes the
such as oxygen (air) or products influencing both kinetics and reaction
hydrogen mechanisms.
Their use influences kinetics and mechanisms, and
Use of catalysts
product distributions.
The additives generally either evaporate or
Additives incorporated decompose on heating. Some may influence kinetic
and mechanism.
Liquid phase pyrolysis retards the escape of
Liquid or ‘gas’phase evolving products, thereby enhancing various
interactions.
Both pyrolysis and liquefaction of PE and PP have shown major conversion of plastic
into oil with a low concentration of gas and no solid residue. The pyrolysis under
nitrogen of polystyrene (PS) had a yield 71% of oil product while hydrogen liquefaction
produced 77% of similar product. On the other hand, the resultant solid residues of 27%
20
and 22% also observed in each process. The pyrolysis of PVC could not be completed
due to the corrosion of the reactor were caused by the generation of hydrogen chloride,
while the hydrogen liquefaction of PVC produced 2% of oil, 38% of gas and 52% of
solid residue. PET has shown high yield of solid residue (53%) and oil (15%) under
nitrogen pyrolysis, but liquefaction conditions yielded solid residue of 41% and 27% of
oil.
The emitted hydrocarbon gases observed from the pyrolysis and iiquefaction were
similar, consisting of alkane gases, methane, ethane, propane and butane. In addition,
the generation of high concentration of hydrogen chloride was detected from the PVC,
as well as carbon dioxide and carbon monoxide from PET. The pyrolysis of PS was also
carried out by under vacuum and in a closed vessel for temperatures up to 420°C. It was
observed that the volatilization reached a maximum at 60% of the polystyrene
conversion. The product yield and its composition were directly related to the
conditions of the closed system which prevented the volatile product and intermediates
from leaving the reaction vessel.
Studies on the pyrolysis of waste plastic mixtures have also been reported by Bhaskar
et al. |2003), describing the yield of liquid, gas and residue from municipal plastic
waste as being 59%, 25% and 16% weight respectively. Pyrolysis was performed under
atmospheric conditions at 430°C. There is significant level of liquid yield from the
pyrolysis, similar results have also been reported by Lee and Shin [2007] in their
pyrolysis experiments conducted at 350°C and 400°C. The liquid products obtained
from the pyrolysis of waste plastics mainly consisted of liquid paraffin, liquid olefin,
liquid naphthene and liquid aromatics, with their relative proportions varying with
polymer types, temperature and lapse time.
Williams and Slaney [2007] in their report of simulated plastic waste pyrolysis at
500°C have showed that the yield of the oil component was much higher (48.7%) as
compared to gas and solid residues of 3.7% and 34.6% accordingly. On the other hand,
the hydrogen liquefaction has given 55.8% of oil, 4.4% of gas and 22.7% of solid
residue. The significant yield of solid product suggests that there was significant
interaction between mixed plastics that produced higher residue as calculated
21
production of oil suggested it would be more than 70% yield. As for the yield of
hydrocarbon gases, the results from the simulated plastic waste showed similar results
with their calculated compositions. The calculation was carried out by comparing the
gases produced by the individual plastics.
The effect of pyrolysis temperature and the effect of the addition of chlorinated
polymers on the degradation of PE have also been described. The experiment was
carried out at temperature ranging between 400 to 1000°C under nitrogen flow. The
chlorine-containing PE samples showed lower degradation point than pure PP. LDPE
started to degrade at 390°C and reaching completion at 505°C. The depolymerisation of
PE was observed between 390 to 460°C.
The recycling of PET is easier as compared to other plastic mixtures as it generally

contains lesser contaminants. PET is commonly found as bottle waste, that is easier to
identify and separate, and recycled individually and treated by various methods, to
produce chemicals and monomer feedstock in refinery. PET soft drink bottles amounted
to 290 million kg in USA, contributing large quantity to the solid plastic waste. Three
approaches of recoveries that have been proposed are combustion (heating in the
presence of oxygen condition), pyrolysis (heating in inert condition) and recycling.
Recycling appears to be most favourable as it can be transform into many different
valuable products |Bhaskar et al., 2004J.
Mcllveen-Wright et al. |2006| have reported that the pyrolysis of waste plastic is
important as it can provide oil and wax feedstock for the production of new plastics or
refined fuels. The treatment can also generate a range of gases such as hydrogen,
methane, ethane and propane at higher temperatures. The main products are gas, oil/wax
and char products in some cases as its productions depends to the types of plastic,
reactor type and process conditions [Williams and Slaney, 2007],
Most of plastic wastes have a high calorific value and low moisture content that makes
them suitable for co-combustion or co-gasification with coal in order to preserve fossil
fuel. The degradation process of polymer may be affected by the presence of
contaminants and also by the chemical alterations that occur during its life cycle.
22
Moreover, the accessibility of the catalyst in the recycling course depends on the surface
area, particle size and pore size distribution of the polymer because of the presence of
bulky molecules.
Another significant proportion of waste polymer stream that could be diverted from
landfills into valuable materials is nylon. Nylon is used widely in the textiles and
packaging industries. Over 1.5 million tons of carpets reported ended up in landfills in
US in 1995, which contained a million tons of polymers, including 0.2 million tons of
Nylon 6 [Czernik et al., 1998). Pyrolysis is also known as one of the popular recycling
approaches applicable to mixed polymers, where Nylon 6 is one of its main
constituents. BASF, DuPont, Hoechst-Celanese and Allied Signal are some of the
companies that have developed and patented their own technologies towards
depolymerization of Nylon 6. Their approaches are basically high temperatures, high
pressure, liquid phase cleavage of amide bonds, but these processes proceed at relatively
slow rates and are difficult to apply to mixed materials. Polycondensation polymers
containing ether bond, ester bond or acid amide bond are some of the yield products of
nylon waste [Czernik et al., 1998; Iwaya et al., 2006).
According to the published research, the modification of decomposition kinetics of

mixed polymers and addition of catalysts and co-reactants can yield relatively pure
monomer or target chemical products. During the pyrolysis of polymers, several
reactions can be detected, including depolymerization to monomers, cleavage of
principal chains into different size fragments and elimination of simple inorganic parts
[Czernik et al., 1998J.
However, the pyrolysis of real world plastic mixtures has been demonstrated to be
significantly different as compared to the simulated waste pyrolysis, with the yield of
solid component being notably high (50wt.%). The gases composition generated from
liquefaction was also dominated by alkane gases. Both liquefaction and pyrolysis
processes produced significantly different level of gases as compared to simulated waste
plastics. Observed results were probably affected due to the presence of various
contaminants, such as paper and dirt, the existence of the additives and fillers, and
23
relatively lower levels of purity as compared to simulation wastes produced in the

laboratory.
2.3 Mechanisms of Reaction Occurring During Thermal

Degradation of Polymers
The thermal treatment of polymer materials is generally carried out in the temperature
range between 300 to 500°C. Some studies have also reported experimental pyrolysis up
to 1000°C [Conesa et al., 1994; Blazso et al., 1995]. Extensive studies on thermal
degradation of polymers have been reported in literature. The four key mechanisms
described include end-chain scission or unzipping, random chain scission, chain
stripping and cross-linking [Oakes and Richards, 1949; Wampler, 1989].
In terms of mechanisms, thermal degradation proceeds according to the radical chain

pathway comprising hydrogen transfer steps along with the continuous breaking of the
polymer backbone. The progressive degradation mechanisms can be further described
as three consecutive steps, which are classical initiation, propagation or hydrogen chain
transfer and termination steps.
Initiation step can be described as the initiation reactions of homolytic cleavage of a

carbon-carbon bond by either random or end scission causing increasing break up of
two radicals, depending on the plastic type. End-chain scission occurs when polymer is
broken up into the unsaturated smaller monomers and terminal free radicals starting
from the end of chain, while random scission reaction is the cutting off the polymeric
chain by random fragments, thus producing smaller molecules of varying sizes with
broad molecular weight distributions, saturated or unsaturated bonds.
On the other hand, the chain stripping can be described as the reaction involving the
substitution of monomers on the side branching polymeric chains. This reaction
promotes the elimination of reactive substituents, leaving behind an unsaturated chain.
Cross-linking is more appropriate to describe the thermoset plastic reactions, where the
chain networks are being rearranged to form high strength materials when polymers
were condensed after heated at high temperatures [Blazso et al., 1995; Ashraf].
24
Propagation or intramolecular/intermolecular hydrogen chain transfer reactions occur to

form more stable secondary radicals. Some of these radicals then undergo C-C bond
rupture process, called p-scission to produce olefins (e.g. ethylene and propylene) and
new radicals. This is followed by the final step of degradation mechanism, called
termination reaction. This reaction generally involves disproportionation towards
different olefins and alkanes or bimolecular coupling between radicals. As the thermal
degradation is a non-catalytic process, no radical rearrangement occurs, thus resulting
broad distributions of hydrocarbon products. The general reaction mechanism for plastic
thermal decomposition is illustrated in the Table 2.4 and Figure 2.3.
Table 2.4: Reaction steps and mechanisms of thermal degradation of waste plastics
Reaction steps Mechanism reaction

Polymer broken by - random scission (PP, PE) or end-chain
Initiation
scission (PS, PMMA)
Depropagation Release of monomeric fragments from primary radicals.
Intramolecular
Hydrogen chain
Intermolecular
transfer
P-cleavage
Branches formation Combination between radicals.
Bimolecular coupling
Termination
Disproportionation
25
Initiation
HX h2
Random scission wv'C------C — C -----C wv vw'C------CHX H -C----- C wv
h2 HX h 2 HX
HX h2
End-chain scission wvC------C — C C wv ------- vwc------CHX * H.C------CHX
h 2 HX 2
h
Depropagation
HX H;
wvC------C —C CHX ------ ► vwc------CHX + H_'C=CHX
h 2 H2
Hydrogen Chain Transfer

HX H2 HX H2
Intermolecular vw'C — - CHX wv'C------C • C ---- C —C vw
h2 h2
WV c----CHX *■ VWC — -C —C = CHX 1- HC^vi

h 2 H2 H
HX
uw'C — -CHX wu'C — -C --- C-aaa c------CH X XAA/-C — -C vw.
H2 H- h2 h2 H; H;
HX H, HX HX
Intramolecular uwC — -C —C - -C - -CHj vw'C------C =C- -C —CH.
H: H;, HX H
H2 X h2
P cleavage wv>c —C ---- C - -C vw. vwCHX t H:C: -C VW
HX H;
h2
Formation of branches XHC- C /
h2
vw'C — -CHX wv>C- C w/ H.C------C- -C'Aaa
h2 H2 X h2
X H: X H, H; X H2
wv'C------C------ C wv +■ vw'C------- C------ C wv VWC ---- C------- C WA/
H2 h.
VW'C------ C------ C WVA

H: X H;
Termination
HX H2 HX H2 HX h2
Bmiolecirlar coupling wv'C — CH vw'C —C vw vw'C —C C —C JW
HX
Di spr oporti ona tr on

HX H, H, HX
wv'C------C —CH XHC------C ----- C vw -------- ► wv'C------C —CH- ♦- XHC = C------ C-AAA
H2 ' H2 H h2
Figure 2.3: Reaction mechanism of polymer [Scheirs and Kaminsky, 2006]
26
2.3.1 Thermal Degradation of Polyethylene

The degradation mechanism of polyethylene (PE) is described by the random scission
reaction. Peterson et al. [2001] has explained that a small reaction of unzipping process
occurs at the initiation step of PE degradation resulting in large amounts of
hydrocarbons in the range of 1 to 70 carbon atoms. The propene and 1-hexene were
detected as the most abundant compounds in the reaction products. The radicals formed
then undergo chain scission reaction in two pathways as described in Figure 2.4.
R + h2C= C CH3
H
Figure 2.4: Random chain scission in polyethylene
The reactions begin at the weak sites along the polymer chain once the scission reaction
gets triggered. The possible weak sites detected are peroxides, carbonyls, chain
branches and unsaturated branches. As al! these weak sites get used, the chain scission
process occurs with high activation energy. Besides chain scission, the polymer chain
branching has also been observed. Both scission and polymer branching can take place
simultaneously and lead to increased mass loss of the PE. The cross-linking mechanism
also can result the formation of bond with high molecular weight. Wampler [1989] has
proposed that the degradation of polyethylene may not necessarily be confined via
random-scission, but could also involve H transfer producing internal free radicals that
further undergo P-scission (Figure 2.5).
27
This was verified by the results of carbon number distributed in preferential production of
olefins containing 6. 10 and 10 carbons, which explained as 1-5 shift of hydrogen in the
free radical (I),
h2 i
c
ch2 ch2 ch2
c c c
Hydrogen transfer then produces (II).
• H2 P V H2 • |-) H2 H2 H2 Hi
—j— c —c — C -r-c—c —c —c —c II
ch2 ch2
c c
Which consecutively undergo (5-scissions by pathway 1 or 2. producing 2.4-diethyloctene
(III) or heptyl radicals which abstracts H to produce heptane (IV).
H2 h h2
c=c—c —c—c —ch2 I1T
ch2 ch2 ch2
c c c
h2 h2 h2 h2 h2
h3c—c —c —c —c —c —ch3 I'
At higher temperatures (900°C) the steps proposed are beginning by the random scission
reactions that break C-C bonds at lower temperatures and stronger C-H bonds at higher
temperatures. These reactions produce different hydrocarbon products compositions. The
C-C bonds broken by random scission reaction producing (V).
h2 h2 . h2 h2
c —c —ch2 h2c—c —c ----- V
The radicals then undergo H transfer to produce VI. Besides that, the breaking of C' - H
bonds also takes place according to (VI).
h2 h2 h2
-c ■ -c —c ■ VI
This will undergo (3-scissions to produce alkanes, alkenes and dienes. (5-scissions is major
propagation step in the degradation of polyethylene. In addition, the hydrogen elimination
step has also been detected in the mechanisms as it is explained playing a role to generate
elemental hydrogen and assist in the formation of char.
Figure 2.5: Degradation mechanisms of polyethylene as suggested by Wampler [1989J
28
2.4 Polypropylene (PP)

Polypropylene is one of the largest volume commodities of thermoplastic. Marlex,
Polyfort, Pro-Fax, Vistalon and Vrestolen are some of the popular brand names of
polypropylene [Strong, 2006). Polypropylene has been widely used in various
applications, including food packaging, laboratory equipment, automotive components
textiles, microwave container, reusable containers of various types, loudspeakers,
bathroom products, tableware, and polymer banknotes. The usage of polypropylene
generally offers good economical value with a combination of outstanding physical,
chemical, mechanical, thermal and electrical characteristics which are not found in any
other thermoplastic.
Polypropylene can stand up to 100 — 200°C heat besides having good resistance towards
acid, alkali alcohol and cooking oil. Its properties resemble polyethylenes but PP has
better heat resistance and a glossy look. In addition, PP has a relatively low density of
0.9 g/cm3. Polypropylene consists of 85.6 wt % of carbon and has relatively high energy
bonds of 3440 kJ/mol. Polypropylene provides excellent resistance to organic solvents,
degreasing agents and electrolytic attack. It has lower impact strength that is
comparable to polyethylene and makes it suitable for lower mechanical load
applications, and is also becoming attractive as the matrix composites. Polypropylene
offers a range of benefits including good mechanical properties, low cost, easy handling
and processing in producing fiber reinforced polymer composites [Williams and
Williams, 1999; Achilias et al., 2007; Williams and Slaney, 2007J.
2.4.1 Polymerization of Polypropylene

Polypropylene is an unsaturated hydrocarbon, containing carbon and hydrogen atoms.
Propylene is the gaseous by-product of petroleum refining. Polypropylene
polymerization takes place in the presence of organometalic catalyst under conditions of
controlled heat and pressure. The presence of catalyst is important as it provides the site
for polymerization reaction. In polymerizing propylene, the monomers are joined
together to form one large chain molecules of polypropylene. Figure 2.6 shows the
polymerization reaction of polypropylene. The metal catalysts act as a functional group
on the polymer chain and the unsaturated bond of propylene monomer. Propylene is
reacted with metal catalyst to provide a site for the reaction to occur and propylene
29
molecules are added sequentially on growing polypropylene chain and unsaturated bond
of propylene monomer, forming long, linear polymer chains until the chain reaction is
terminated [Scheirs and Kaminsky, 2006].
H2C = CH
M*
CH3
Metal catalyst Propylene
M------- C ----- CH2 + H2C = CH ---------------►
CH3 ch3
h2 h h2
M------- C ----- C------ C ----- CH2
ch3 ch3
Figure 2.6: Polymerization of propylene monomer with metal catalyst
Propylene molecules attach to polypropylene chain only in a particular direction,

depending on the chemical and crystal structure of the catalyst. The addition of
polypropylene molecules occurs in the main chain, growing the length of the chain. It
does not occur at the methyl groups attached to alternating carbon atoms, which would
result in increased polymer branching [Maier and Calafut, 1998].
Generally, there are three types of stereochemical configurations of polypropylene. The

repeating unit of polypropylene is CH3 group, providing three different types of
polypropylene called stereoisomers. These may be determined by the structural
orientation of the methyl groups attached to the polymer chains. The most conventional
form is the isotactic polypropylene, where all branches are attached to one side of the
main chain. It has regular and repeating arrangement, thus it turns out to have a high
degree of crystallinity. It also exhibits high magnitude of mechanical properties, such as
stiffness and tensile strength (Scheirs and Kaminsky, 2006].
30
On the other hand, the syndiotactic polypropylene has different arrangement of its
methyl groups, where these groups are located on the both sides of polymer backbone.
Syndiotactic polypropylene is less stiff than isotactic but has better impact strength and
clarity. The random orientation of methyl groups on the polymer chain results in the
atactic types of polypropylene. The variable structure of atactic polypropylene results in
low crystallinity and it becomes sticky, amorphous and is mainly used for adhesives and
roofing tars.
2.4.2 Crystallinity of Polypropylene

The relative extent and the orientation of the isotactic, syndiotactic and atactic
components in the polymer formulations are determined by the catalyst and the
polymerization conditions. An increment of atactic polypropylene in formulation can
enhance the impact resistance and stretchability but reduce the stiffness and colour
quality. Figure 2.7 shows the stereochemical orientations of polypropylene.
™---- ch? - ch—ch?—ch—ch2 - CH —ch2 - ch—ch?—CH —™

ch3 ch3 ch3 ch3 ch3
isotactic polypropylene
CH} CH3
I I
---- CH2 - CH—CH2—CH—CH2 - CH —CH2—CH—CH?—CH —^
I I “I
ch3 ch3 ch3
syndiotactic polypropylene
CH3 CH3
™—CH?—CH - CH2—CH - CH?—CH -CH?—CH—CH?—CH ---------vVVAMV
ch3 ch3 ch3
atactic polypropylene
Figure 2.7: Stereochemical configurations of polypropylene
The modification of processing conditions and catalysts can alter the degree of
crystalinity, thus modifying the properties of polypropylene. The crystallization occurs
at the temperature, where the molecules lose their heat and solidifies or evaporated.
Their ability to move freely gets decreased and viscosity increases during crystal
growth. The molecules begin to arrange themselves into crystals, and ordered crystalline
31
phases are formed, along with disordered amorphous regions. The rates of nucleation
and crystal growth have significant impact on the crystallization. As the temperature of
the melted material increases, the nucleation rate increase but the crystal growth
decreases [Maier and Calafut 1998|.
The polymer resin is composed of various chain types containing different lengths and
molecular weight. The toughness of polypropylene is influenced by its molecular
weight. Higher moiecuiar weight provides tougher materials with high impact
resistance, greater elongation capability and less brittleness. The overall mechanical
properties of polypropylene generally depend on its crystallinity, where the stiffness
yield stress and flexural strength increases with increases in crystalinity, but with
decreasing toughness and impact strength [Maier and Calafut 1998J.
2.4.3 Oxidation of Polypropylene

ch3 ch3
■|_| ■ Tertiary hydrogen
Polypropylene Oxygen
ch3 ch3
h2 I h2 I h2
---- c —c—c —c—c — + *o2h
H
Polypropylene free radical (PP )
Figure 2.8: Oxidative radicals formation of PP
Polypropylene has high tendency to undergo oxidation because of the presence of

tertiary hydrogen on the carbon atom bonded to the pendant methyl group [Figure 2.8].
The oxidative chain scission reduces the molecular weight of the polymer chain under
normal processing conditions if the resin is not stabilized. Mechanical stress, heat or the
presence of oxygen or metal catalyst residues result in unstabilized carbon-hydrogen or
carbon-carbon covalent bond in the polypropylene chain [Maier and Calafut, 1998].
32
The chain scission oxidation reaction results in decreased molecular weight which
consequently affects the mechanical properties of the polymer. This oxidation reaction
produces compounds such as carboxylic acids, lactones, aldehydes and esters. Chain
reactions get terminated when two radicals merge to form inactive group |Maier and
Calafut, 1998; Riga et al., 1998J.
2.4.4 Degradation of Polypropylene

The degradation of polymer generally involves several mechanisms, depending on its
structure and exposure conditions upon degradation. It can be divided into two types,
namely mechanical reactions and physical changes. Physical changes lead to the
disruption of polymer morphology, rather than reactions at molecular level. Although
these changes can results in rapid and complete failure, polymer degradation caused by
physical changes has not been investigated extensively [Pinto et al., 1999-a).
Chemical reactions can cause a range of polymer degradation reaction that can be
explained as multiple reactions that occur simultaneously. Therefore, it is essential to
take into account all possible chemical changes that can possibly happen in the chemical
bonds of polymer in order to understand the polymer degradation. Pyrolysis and thermal
oxidation are some of the major modes of polypropylene degradation that involve
chemical reactions. Two most important mechanisms in polypropylene degradation are
depolymerisation and substituent reactions; these two steps might be combined in some
cases.
Depolymerisation is a scission process of bonds along the backbone polypropylene

chains. It is also termed as thermal cracking where the polymer chains were cut in
fragments of various lengths. It can occur randomly (random scission reaction) or start
at the end of polymer chains (end-scission reaction). In most cases, polypropylene
molecules are first broken into large macro-radicals fragments by random chain
scissions as the initial step. Temperature is an important parameter at this point as it
contributes the energy to break the bonds, e.g., 348 kJ/mol is needed to rupture C-C
bond and 413 kJ/mol needed to break C-H bond in polypropylene. Subsequent reactions
of intramolecular chain transfer take place after that, leading to the formation of many
low molecular weight carbons [Zhao et al., 1996[.
33
2.4.4.1 Thermal activation enernv of Polypropylene

Thermal activation energy has been reported by Hsu [2002]. The activation energy of
thermal decomposition process can be calculated from the series of thermal gravimetric
(TG) curves obtained from different heating rates (e.g 5-20°/min). This method can
then directly determine the activation energy from the mass loss vs. temperature data for
a range of heating rates. However, this calculation method is not applicable to a wide
range of decomposition processes. Based on their report, the activation energy of PP
fibre ranged from 71-84 kJ/moi for the mass loss of 5% to 20%, with a standard
deviation of 2.43 that is acceptable in thermal analysis. The decomposition of PP fibre
was reported to start after 200°C [Hsu, 2002[.
The thermal kinetic studies of polypropylene, polystyrene and polyethylene have been
carried out by Peterson et al., [2001] using thermogravimetric analysis (TGA) and
differential scanning calorimetry (DSC). Kinetic analysis is important as it can assist in
determining the degradation mechanism and the thermal stability of the polymers. The
use of variable heating rates is critical for obtaining reliable kinetic data. The kinetics of
polymer degradation is basically determined by the equation given below.
= k (T)/(a) Equation 2.1
Where:
a represents the extent of reaction (a = 0 - 1)
/ time
k (T) rate constant
f (a) reaction model, describes the dependence of the reaction rate on the extent of
reaction
a value determined by TGA as a relative mass loss
k (T) is described by the Arrhenius Equation (Equation 2.2) and can be substituted into
Equation 2.1, yielding (Equation 2.3):
34
Arrhenius Equation
k = A exp(~E(X/RT^ Equation 2.2

Where:
k rate constant
A pre-exponential factor
Ea activation energy
R gas constant
T temperature (in Kelvin)
%=Aexp (f) m Equation 2.3
2.4.5 Pyrolysis of Polypropylene
2.4.5.1 Effect of temperature on degradation of PP

While studies were carried out on single plastics or in simple plastic mixtures, a range
of parameters were tested including temperatures and pressures and the presence of
catalysts and/or hydrogen. From their report on the pyrolysis of polypropylene at 500°C
under nitrogen flow, the reaction has produced an oily product (95%) with lower
concentration of gas (5%) and no solid residue. Obtained gases from the polypropylene
pyrolysis were also analyzed which showing high composition of alkane gases, such as
methane, ethane, propane and butane than alkene gases [Pinto et al., 1999-a; Pinto et
al., 1999-b; William and Slaney, 2007|.
A number of studies of polypropylene pyrolysis have been reported in the literature

[Zhao et al., 1996; Marin et al., 2002; Bhaskar et al., 2004; Kiran et al., 2004; Yu-
Hu and Jou, 2005]. The addition of catalyst has enhanced the process of pyrolysis by
reducing its decomposition temperature, promoting the decomposition speed and also
enabling the modification of the products.
Lee and Shin [2007] in their studies of polymer pyrolysis have reported that the
decomposition of polypropylene started at 380°C and was completed at 500°C. The loss
of 50% of the initial weight was reached at 455°C. Polypropylene tends to form
substituted olefins during its degradation [Marin et al., 2002]. The pyrolysis of
35
polypropylene, which consisting polyolefinic polymer types degrades by the end-chain

scission mechanism that increases gradually with time even during lower temperature
reactions.
A study on the comparison of yield products from catalytic and thermal degradation of
polypropylene was performed by Sakata et al., 11996]. The degradation was carried out
at 380°C at a heating rate of 3°C/min under atmospheric conditions using silica-alumina
cataiyst. The yield of gaseous and liquid products increased during the catalytic
degradation of vapour and liquid phase contact. An increase in gaseous products during
vapour contact phase of catalyst was reported due to the degradation of hydrocarbon
into gases during their encounter with silica-alumina catalyst.
The catalytic degradation produced mainly butene (57%) and propylene (30%), while
thermal degradation produced propylene (70%) and ethane (28%). Similar results were
observed during the liquid phase contact, where the waxy residues (heavier
hydrocarbon) decomposed into liquid products, indicating the contact of acid sites of
silica-alumina with PP melt had accelerated the degradation process extensively. The
wide range of molecular weight distributions found in liquid products of PP degradation
has demonstrated the occurrence of chain scission reaction as the key thermal
degradation mechanism of PP. This reaction breaks the long polymeric chain of PP into
smaller units that have lower molecular weight over a wide range. As a result, the yield
of unsaturated hydrocarbon (olefins) was seen to increase in reaction products [Sakata
et al., 1996|.
The kinetics of thermal degradation of PP can be explained by random scission

followed by radical transfer process as reported by Peterson et al. [2001]. PP sample
was heat treated up to 600°C at heating rates of 0.5-20 K/min under nitrogen and air at
lOOml/min flow rates. The following scheme displays the formation of most abundant
products generated by degradation of PP (Figure 2.9). Pathway A produces major
products such as pentane (24.3%), 2-methyl-1-pentene (15.4%) and 2,4-dimethyl-1-
heptene (18.9%), while the pathway B only produces minor products of propane (1.9%).
36
H H2 H H? h
_q__ p Polypropylene polymeric chain
R—C- -c -c- q
H h2
R—C- C -CH H2C=CH H2C—C—R
ch3 ch3 ch3 CH3
Figure 2.9: The major products of PP by their pathways
The degradation steps described above (Figure 2.9) have been indicated by the TGA
and DSC curves shown in Figure 2.10. The TGA curve has shown the melting step of
PP in air and nitrogen was similar, where the primary degradation step under nitrogen
occurred between 250 to 450°C. Under air, a slight mass increase was detected around
180°C. The same phenomenon was detected in the degradation of PE, where a slight
mass increase occurred around 200°C. The significant mass loss was mostly observed
from 200 to 450°C and followed by minor loss step from 450 to 600°C that showed the
degradation step of random scission was followed by radical transfer process.
Significant mass loss was also observed in DSC curves where the melting endotherms
were observed at 158°C and 460°C under nitrogen flow.
TGA C urves DSC C urves
ra 40
100 200 300 400 500 600 500 600
T/°C T/°C
Figure 2.10: TGA and DSC curves of thermal degradation of PP in

nitrogen (dashed line) and air (solid line)
37
The activation energy shows steady increase from 150 kJ/mol to 250 kJ/mol under
nitrogen flow. The increment has showed the change of rate limiting step from initiating
of 150 kJ/mol activation energy to degradation with 250kJ/mol. On the other hand, a
large exothermic peak has been detected at 316°C under air flow in DSC data scans.
The activation energy remains constant at 85 kJ/mol during initial 40% of degradation,
before increasing to 270 kJ/mol under air. This reaction and thermal kinetic scans show
that the degradation of PP takes place through two-steps mechanisms. These
mechanisms are influenced significantly by temperatures as demonstrated by the TGA
and DSC curves obtained (Figure 2.10) [Peterson et al., 2001].
The effect of temperature on the pyrolysis yield of PP has been summarized in Table
2.5 below and plotted in the Figure 2.11. It was noted that the yield of solid decreased
with increase of temperature. Oil was seen as highest yield product generally around
80%, while gas yield maintained around 11 to 17 %, and at 500°C increased a little and
decreasing back to 13.6% with increasing temperature to 700°C.
Table 2.5: Product of PP pyrolysis according to temperature

[Scheirs and Kaminsky, 2006]
Temperature Product yield (%)

(°C) Gas Liquid Solid
400 14.5 65.7 19.8
410 11.2 85.2 3.6
500 17.5 82.1 0.4
700 13.6 84.4 0.2
q Liquid
Temperature (°C i
Figure 2.11: Product of PP pyrolysis as a function of temperature
38
2.5 Substitution of Polymers in Anode Baking in Aluminium

Making Process
2.5.1 Key Components of Carbon Anodes

Carbon has found application in a number of different fields such as electric and heat
conductions (conductor and semi-conductor), energy storage (battery anode, super
capacitor and gas storage), environmental protection (activated surface), special
materials (mechanical reinforcement, high temperature) etc [Mochida et aL, 2006],
Aluminium is manufactured by electrochemical Hall Heroult process. Carbon is very

essential on the industrial scale in the production of the anode required for these
processes. The manufacturing of anode involves the baking of a blend of some varieties
of coke with hydrocarbon binder, which is generally a coal-tar pitch [Couderc et al.,
1986; Ehrburger, 1994].
The coal-tar pitch is the concentrate part left after the distillation of coke oven coal tar.
Pitches have been used for the preparation of carbon fibers, polygranular graphites or
carbon-carbon composites. Technologies developed to process low toxic emission and
carcinogenic fumes production of carbon pitches, which mainly dominated by coal-tar
pitches [Perez et al., 2002]. Coal-tar pitches have been generally used for producing
anode and electrodes, whereas high density carbons were from coal-tar or petroleum
pitches. These contain organic compounds such as polycyclic aromatic hydrocarbons
(PAH) which are essential components of pitch substances. The aromatic hydrocarbon
enables the formation of anisotropic carbon materials. However, it has high content of
carcinogenic hydrocarbons, such as benzo(a)pyrene that limits the usage of coal-tar
pitch in many areas [Couderc et al., 1986; Ciesinska et al., 2009].
Petroleum pitch is seen as a good alternative for replacing at least part of coal-tar
pitches, as it has low content of metal and heteroatoms (especially sulphur) and also low
in toxic and carcinogenic polycyclic aromatic hydrocarbons. These petroleum pitches
have been recognized as being suitable for the preparation of high-density carbon
precursors and optimum binders for anodes and electrodes [Perez et al., 2002].
39
Pitches are characterized by physico-chemical methods such as selective fractional by

the use of solvents namely quinoline and toluene; specific gravity; coking value;
viscosity (temperature dependent); softening point; residual ash; and carbon and
hydrogen content [Couderc et al., 1986J. One of the important factors in determining a
good pitch is the quantity and nature of quinoline-insolubles. Quinoline-insolubles are
the solid carbon residues found in coal-tar produced during the carbonization process.
The coal related residue may contain char cenospheres, coke fragments, coal particles
and mineral matters [Woinbles and Sadler, 2004].
The interaction between coke and pitch is important in order to determine the final
properties of anode electrode including thermal, electrical, crushing strength and
reactivity. Couderc et al. [1986] have investigated the ability of pitch to penetrate coke
at various temperatures and their end-product characteristics. The interactions were
measured by contact angle (0°) made by pitch on the coke and the height of pitch drop.
The good penetration was defined as the temperature when the drop of pitch has
completely flowed through the bed coke. However, factors such as coke particle size
(100pm<best size<200pm), pitch viscosity, quantity and nature of quinoline and
toluene insolubles. The high content of quinoline insolubles acted like a sieve, thus
preventing good penetration of pitch into coke, even though its high content favours an
increased pitch coke yield in the electrode.
Petroleum coke is a by-product of oil-refining industry. The main factors influencing

output of petroleum coke are the levels of crude oil production, demand of refined
products and the quality of oil extracted. Sponge coke, shot coke and needle coke are
some of green petroleum coke types [Roskill, 2007]. Many considerations need to take
into account for anode-grade carbon facility as consistent quality coke begins with
consistent quality crudes |BP, 2001].
Petroleum coke is used for manufacturing of carbon anode in the aluminium industry. It
was reported global consumption of calcined petroleum coke in aluminium production
is estimated 11-12 million tonnes per year. The demand is expected to rise steadily as
aluminium production has been steady upward trend for many years [Roskill, 2007].
The preparation of petroleum coke for the electrode manufacturing involves two major
40
stages: production of coke by coking and the adjustment of its properties for required
standard by calcination. The process of obtaining the calcined petroleum coke is
involves several steps, which need careful considerations to ensure consistent quality of
coke.
2.5.2 Coking and Calcination of Petroleum Coke

Carbonization can be defined as the thermal conversion of organic materials to carbon.
Characterization techniques are the key of understanding the carbonization process as
the carbonaceous products are highly complex materials, which are the mixtures of
aromatic compounds having alkyl side chain or heteroatoms such as nitrogen and
sulphur. It is well known that the chemical structure of aromatic compounds have
important impact on the carbonization, and controlling the degree of graphitization
| Lewis, 1982].
Petroleum residue known as feed coke formation principally contains resinous asphaltic
material composed of aromatic and hydroaromatic rings [Syunyaev and Voloshin,
1966]. The mechanism of carbonization is generally the growth of aromaticity and
polymerization process, where small aromatic structure polymerize to an aromatic
polymer in order to achieve three dimensional graphitic structures. General reaction
scheme of carbonization is shown in Figure 2.12.
OOO 300°-500°C
AROMATIC HYDROCARBON
2500°-3O00°C CARBON I000°-2500CC
GRAPHITE
Figure 2.12: General reaction scheme of carbonization (Lewis, 1982]
41
The heating of the resin feed results in the polycondensation of weak bonds into larger
molecular units by the recombination of various complex species. As the temperature of
coking increases, the number of chemical bonds between the nuclei of the solid phase
increases, thus reducing the weak Van der Waals reactions. The release of radicals
intensifies and the adsorptions of reactive structural members from the complex
molecules on the surface of the solid phase are sufficient to ensure the subsequent
crosslinking of the mesophase into one large molecule, coke [Syunyaev and Voloshin,
19661.
The increasing of coking temperature to 495-505°C increases the fraction of chemical

bonds between the contact surfaces and enables a massive mass of coke being produced.
As the temperature exceeding 505°C, rapid evaporation leads to binding material being
insufficient for crosslinking the particle into a solid mass. The boundary surface of
contacting fragments formed and coking is practically complete on the surface of
separate particles [Syunyaev and Voloshin, 1966J.
During the calcination process, the disordered crystalline structure of petroleum coke is
rearranged to meet the requirements for anode manufacturing. The dimensions and
ordering of the structure depends on physical properties (thermal conductivity, electrical
conductivity, density etc.) and chemical reactivity (side chain radicals). During
calcination, heating cracks the intermediates into unsaturated hydrocarbons. With an
increasing rate of heating, the rapid yield of intermediate products occurs to some
extent, decrease the cracking, which also leads to an increase of high molecular weight
volatiles. The detachment of side-chains from the molecules containing coke crystal
(mesophase) takes place followed by their subsequent recombination into molecules.
The plausible theory suggested by Syunyaev and Voloshin [1966] was the removal of
volatiles caused molecular tension in coke crystals, and the shrinkage of the external
surface, thus leading to the individual carbon atom interacting with each other and
regrouping.
On the industrial scale, the process of coking and carbonization are linked through pipes
for simultaneous reactions. The crude charge is washed with water in a desalter to
remove solids and salts. The desalting is needed to prevent corrosion in downstream
42
piping, equipment and process units, and lower level of iron in the finished carbon
products favour less corrosion. The crude is then preheated and vaporized, separated
into various fractions according to boiling points. The intermediates (heavy fractions)
advance further downstream where they are removed to another unit, heated and
subjected to vacuum. The vacuum lowers the boiling temperature of the crude charge,
thus enabling the distillation of additional intermediates without thermally decomposing
to carbon/coke |BP, 20011.
The vacuum distillation is then routed to delayed coker unit, where it is thermally
cracked into vapour and liquid. The vapour is returned to fractionation process while the
liquid drops out into the drum and solidified. The full drum is isolated and cooled,
before “green coke” is drilled out with high pressure water and drained. Drained green
coke is best handled by dedicated equipment as repeated handling by mobile equipment
will reduce the size, impacted by crushing and dropping, while its exposure to open will
cause contaminations. Thus, enclosed conveyers and inside storage are the appropriate
equipments. The calcining is the process where the green coke is heated sufficiently to
eliminate the moisture and residual hydrocarbons. The calcined coke is usually oiled for
dust control |BP, 2001].
Heal treatment Phase of reaction Gas Chemical and Molecular

Temperature (°C) Vap(w j; ^ volatilization Physical changes Structures
Organic materials
200 ---
K mol Paraffin or Olefins Mam cham
Radical Aromatization, Condensation
Pyrolyui Arorratizahon ^ I/OW mol Aromatic carbons
Polymerization, Cross-linking
Pelycon-
Coking
dcnsation
i___
500 - 1
CH„ CO, NO, Devolatilization
Carbon* Cokinj? ------► H:S, CO, Crack nucleation
ceous
H, etc Stacking start
materials
600 Loss of viscosity (Inorganic Mat.)
H,
------► Removal of heterogeneous atoms
CO, CO, Dehydrogenation
H,S
1000 Micropore nucleation
etc. La increasing
Carbon H,S
Materials HCN
Removal of heterogeneous atoms
1500 CS, Lc increasing
N, etc
Reducing micro pores
2000
H,
Removal of inorganic materials
Graphites ^2 etc- Formation of 3 D graphitic structure &
3000
Figure 2.13: Carbonization and graphitization of organic materials

[Mochida et al., 2006]
43
Since carbon is generally a porous material, its surface characteristics within pores (pore
size, depth and volume) are very important to determine its performance. The pore
structure is produced through a series of carbonization, graphitization, gasification and
successive modification as shown in Figure 2.13. The transformation of organic
precursor to solid carbon includes graphitization, modification through chemistry
reaction and phase change, which then define the carbon structure and the properties
[Mochida et al., 2006].
Carbonization generally changes the structure of the carbon, which depends on the
composition and characteristics of the raw materials. High quality carbon is defined as
having low thermal coefficient and fibrous texture. The fluidity of carbonization has an
effect on the microscopic structure of the coke obtained since the high fluidity (maintain
in liquid state sufficiently long) enables the nucleation and growth of mesophase, thus
resulted in fibrous structures [Kakuta et al., 1980].
The term ‘kmesophase,■ refers to a range of materials of different viscosities and

macrocrystallinity. It behaves like a plastic material, which can be deformed by
mechanical action, with limited recovery |Marsh et al., 1999], Mesophase is liquid
crystalline formed in pitch stage during the conversion of organic materials to carbon. It
has been seen that during the thermal polymerization, physical alignment and
orientation process occur, forming a liquid crystalline phase. It was suggested that
mesophase is composed of mixture of large and small condensed aromatic molecules
which are linked by aryl-aryl and CH2 bonds. Mesophase development is highly
dependent on the chemical structure and concentration of aromatic components in its
starting material [Lewis, 1982].
Heating rate has a significant influence in fluidity of coals during carbonization, higher
heating rate generally result in increasing overall fluidity. It has also been reported that
higher heating rates give larger sized optical structure of carbon. Thus, the enhanced
fluidity simultaneously increases the development of mesophase into the domain type
optical texture [Ragan and Goodarzi, 1984].
44
2.5.3 Polymer as Filler/Substitute for Coke in Carbon Anode for the

Aluminium Industry
The incorporation of plastics wastes into the production of carbon anodes, graphite
electrodes and sorbent which employ coal-tar pitch was carried out by the Institute of
Chemistry of the Warsaw University of Technology in Plock. The main purpose of
substituting certain amount of coal-tar pitch by polymer was carried out to decrease its
carcinogenic content [Ciesiriska et al., 2009].
Since coal-tar pitch is a multicomponent mixture of medium and higher molecular

condensed aromatic hydrocarbon and heterocyclic compounds, the heating of pitch will
release the atomic hydrogen radicals. Hence, coal-tar pitch can act as a hydrogen donor
solvent at higher temperatures. The co-carbonization of low-rank coking coal and coal
blends with coal-tar pitch as hydrogen-donor additive has enhanced the coking
properties of coals and resulting coke quality [Collin et al., 1997].
An attempt of co-coking coal with waste plastics has also been made since synthetic
polymer may act as hydrogen transfer for replacing the pitch. The disadvantage of
polymer is most thermoplastics decompose below plastic temperature range of coal,
while elastomer and thermoset does not liquefy at all. The modification done by Collin
et al. [1997| on the 1:1 mixture of plastic and coal by thermally autoclaved the mixture
in the liquid phase under atmospheric pressure at reaction temperature between 215-
400°C and reaction times from 1.5 to 6 hours. The TGA analysis of the mixtures has
showed higher residual than its single component alone indicating the co-thermolyisis
of coal-tar pitch with thermoplastics was more stable than the single components. The
co-coking has advantages of improved coking properties, and increment in mechanical
strength and optical anisotropy of yield coke.
The increased of residue yield from the reaction of polymeric additives with pitch by
hydrogen transfer also in the agreement with results reported by Machnikowski et al.
[2002]. He and his co-workers have studied the effect of 10% various polymer
compounds (PVC, PS, PP, PET unsaturated polyester etc.) to coal-tar pitch during
isothermal treatment at 450°C. When the temperature is in the excess of 400°C, the
liquid stage of carbonization led to the creation of liquid crystal system called
45
mesophase, which finally resolidifies into semi-coke. The other possible reaction
scheme occurred was the rearrangement of some polymer fragments, especially as they
comprise of aromatic carbon, to produce mesogenic molecules readily assimilated into
mesophase.
The addition of polymer has significant effect on the mechanism of mesophase

development of the resultant composition. The most noticeable change of optical texture
induced by addition of polymers is an enlargement of mesophase spherules at the early
stage. Different polymers have different impact on the transformation progress of the
mesophase content. The acceleration on the mesophase growth in the early stage of
transformation is attributed by the dehydrogenative activity of polymer degradation
products (reported as hydrogen-donor), which simultaneously increasing the residue
yield [Machnikowski et al., 2002].
The addition of polymer has showed an increase in the heterogeneity of the optical
texture of coke composition. The presence of isotropic regions in the texture indicates
that polymers can act as an agent promoting the concentration of quinoline-insoluble
substances to enhance the agglomeration to isotropic materials (Brzozowska et al.,
1998]. The optimization of binding properties of the pitch is important for the carbon
electrode manufacturing. These characteristic also depends on their wetting and
rheological properties |Ehrburger, 1994].
2.6 Wettability
Wetting process is the interaction between a liquid and a solid when they come into
contact. Wetting behaviour is related to surface tension and viscosity. Low surface
tension and viscosity enable good wetting as the liquid can spread and penetrate into the
solid surface. Both components must interact with each other to generate good bonding
for fine structurally built materials with a high density and good mechanical properties.
The solid/liquid wetting properties are dependent to the characteristics of the liquid such
as chemical properties, softening point, surface tension, viscosity and the texture,
particle size, chemical functional groups at the surface of the solid [Rocha et al., 2005].
46
Wettability is determined by measuring the contact angle of the droplet on the surface.
As a drop of liquid is placed on the solid surface, the liquid adopts a contact angle with
the surface of the solid (Figure 2.14). Such measurement was done in order to
determine the penetration of pitch on the coke bed under controlled temperature [Cao et
al., 2002; Rocha et al., 2005].
The wetting temperature is defined as when the contact angle becomes <90°C. The
penetration temperature is marked as the droplet completely penetrates the bed surface.
As higher temperature is needed for wetting and penetration, therefore it is considered
to have poor wetting. The wetting temperature also can be an indication for the
interaction between droplet/bed surfaces [Cao et al., 2002].
non-wetting
wetting
complete
penetration
Figure 2.14: Schematic illustration of the solid-liquid wetting principle

[Rocha et al., 2005]
As temperature increases, the viscosity of the droplet decreases. At low temperature, the
cohesive forces are greater than adhesive forces. As temperature increases, the adhesion
become higher and overcome cohesion, and thus the droplet height decreases
significantly with temperature until complete penetration is achieved [Rocha et al.,
2005[.
47
The flow of the droplet into the bed mainly depends on the two distinctive parameters:
rheological properties, which determined by the viscosity of the droplet and
thermodynamics, which involves the contact angle. These two parameters must be
emphasized during the characterizing the wetting and penetration properties as the
wetting temperature observed in penetration test basically indicate the onset of the non
equilibrium wetting behaviour [Cao et al., 2002].
Furthermore, the isothermai penetration results showed wetting temperature observed

during the penetration test was significantly different from the true wetting temperature.
Isothermal penetration test was done by observing the heated droplet on the surface at
determined temperature. The wetting has started to happen as the droplet began to
soften and forming ellipsoid (Figure 2.15). This usually happens when the temperature
reach just above the softening point of the droplet. The surface tension of the ellipsoid
droplet has caused it to penetrate into the bed surface by trying to reduce its exposed
volume. The results displayed that the penetration is dependent to the time, where the
contact angle between the ellipsoid and the bed surface change continuously from >90°,
to 90°, to <90°, and finally to 0° [Cao et al., 2002].
Time (min)
10 20
Figure 2.15: Effect of temperature on the droplet on the solid surface in isothermal
penetration test [Cao et al., 2002]
In addition, the functional group present in the liquid may influence the wetting
behaviour. High aliphatic compounds may lead to stabilization effect in the liquid,
which happen when reacted with oxygen. The production of oxygen-containing
functional groups increases the liquid fluidity. Thus, it will effectively prevent the
adhesive forces from overcoming the cohesive ones. This effect can be reduced by using
an inert atmosphere. Moreover, sulphur-containing compounds present may result in
low surface energy and enhancing the interaction between solid and liquid. The low
surface energy allows liquid to spread and flow through the solid surface [Rocha et al.,
20051.
48
According to Owens 11970], the interfacial force across the interface between
immiscible solid and liquid is sharp and adhesion was entirely through the secondary
forces and hydrogen bonding. A liquid of low hydrogen bonding ability cannot interact
fully with high hydrogen bonding ability, and vice versa.
Wetting behaviour can be explained by the combination of two consecutive phenomena

determined by surface tension (spreading of liquid) and viscosity (flowing of liquid
through solid particles). The interaction of soiid-liquid can be increased by blending the
non-wetting component with the wetting ones. Some additives may improve the result
as well. Moreover, the presence of oxygen in the additives (mainly carbonyl) may also
increase the interaction properties [Rocha et al., 2005].
2.6.1 Contact Angle

Contact angle measurement is one of the most common methods to study interaction of
two different materials through surface tension [Saihi et al., 2002]. Contact angle can
be defined as the angle of a liquid droplet in contact with a solid surface in an
equilibrium state. Consider a liquid droplet at rest on a flat and solid surface, where it’s
cross sectional interface can be seen in Figure 2.16.
Figure 2.16: Contact angle of liquid droplet and solid surface
The angle between the solid surface and the tangent line to the upper surface at the end
point is called contact angle. The equilibrium state can be termed when the surface
tension of the liquid droplet, y/, the surface tension of the solid surface, ys, and the
interfacial tension between solid and liquid, ys/, reach a balance, which can be measured
using Young and Dupre equation ]Cao et al., 2002):
COS 6 = V~s..Equation 2.4

yi
49
This equation is however valid for measurement when there are no other forces
involved. The equilibrium state requires no movement of the droplet as the angle is
measured. The external factors such as temperature, humidity, solid surface roughness
and static electricity will affect the results [Busscher et al., 1984). In many cases like
coating or cleaning, the liquid and solid when attached do not maintain their constant
shape. It can be defined as non-equilibrium state, where the contact angle measured at
particular time is not the true contact angle. The phenomenon is called a dynamic state
where it is dominated by a combination of contact angle, fluid mechanics and
rheological properties of the liquid. The dynamic contact angle can be simplified by
Newman equation as below [Cao et al., 2002):
cos 6d = cos 6 [1 — C^expi—C2t)] Equation 2.5

Where:
0d = dynamic contact angle
Ci and C2 = adjustable positive parameters depending mainly on the rheological
properties of the liquid droplet.
The measurement of the above equation becomes relevant only when the solid surface is
flat. If the solid surface is not flat, the contact angle observed were called apparent
contact angle, which provides limited information on the molecular interaction between
the solid and liquid. The influence of the surface roughness on the contact angle has
been studied and presented in the equation by Wenzel [Cao et al., 2002):
cos 9a = ra cos 6 Equation 2.6
Where:
0a = apparent contact angle
ra = roughness factor (defined as the ratio of actual surface area to geometric
surface area, which greater than unity).
The determination of surface energy, ys between liquid and solid is given by the
modified Young's equation below [Saihi et al., 2002; Owens et al., 1970):
50
wsi = K/(l + cos 0) Equation 2.7
Where:
y; = surface tension of the liquid
0 = contact angle between liquid and solid.
The interfacial work of adhesion and the free surface energy are composed of two
components:
wsi — wsid "I" wdi Equation 2.8
y = ynd + yd Equation 2.9
Where:
w™? = non-dispersive component
wdt = dispersive components.
Also, the dispersive attraction can be described by the geometric mean of the dispersive
forces:
wdt = 2(ysY?)1/2 Equation 2.10
While the non-dispersive attraction can be calculated as the geometric mean of the solid
(yf) and the liquid (y;p):
wlY = 2(ysV)1/2 Equation 2.11
2.6.2 Wettability Behaviour of Polymers

When a liquid comes into contact with solid, either it will wet the solid very well, and
thus spread completely across the surface, or not so well with less tendency to spread
across the surface. The contact angle is an important thermodynamic quantity that
characterizes the interaction between a solid and a liquid. The liquid acts as a sensitive
probe of the surface by interacting with functional groups at the surface. Depending on
the chemistry and the molecular volume of the liquid used, the depth sensitivity of a
51
contact angle measurement is typically between 5 and 10 A. This characteristic high-

sensitivity interaction between solid and liquids makes the contact angle an important
key feature for surface characterization [Saihi et al., 2002].
Polymer has low surface energy compared to other classes of materials (Table 2.6). The
physical reason that influences the difficulties of wetting, penetration and dispersion
between different types of material is the adhesion forces that act between particles with
high surface energy. The free energy of adhesion Wa per area between two surfaces of
the same material acting in a liquid form is given below [Luechinger et al., 2008]:
Wa = —2ysi Equation 2.12
Where:
Wa = free energy of adhesion
ysi = interfacial tension between solid and liquid phase
Low interfacial tension results in low adhesion forces between particles, and thus
breaking apart the particles of the nano-sizes agglomerates, and helps the primary
particles to get through. Polymer have low surface-tension in liquid form that facilitates
it to wet most solid materials (0<9O°) from low energy carbon-fibers to high energy-
metals. Some studies have also been carried out by mixing high-surface energy fillers to
the polymers based on their uses to overcome the high surface energies gap. However,
the chemical and processing complexity, thermal stability, costs and health issues of
such additives need to be taken into consideration [Luechinger et al., 2008].
Table 2.6: Wetting contact angle between polymers on solid materials

[Luechinger et al., 2008]
Liquid polymer Solid material Contact Angle

Polypropylene Aluminium 17-28°
Polypropylene Carbon fiber 17°
Polypropylene Carbon nanofiber 18°
Polyethylene Aluminium 30-55°
52
Generally contact angle decrease with decrease in surface tension of the liquid. The
variation in wetting force is highly sensitive to the surface characteristics since it
reflects the effect of functional groups in the surface layer (less than 10A) thick and in
direct contact with the liquid phase. The response of the liquid in contact with the
surface is also related to the chemical nature and surface roughness. The hysteresis
increases with increasing roughness of the surface, in addition the mobility and
reorientation of the solid surface macromolecules can influence the contact angle.
2.6.3 Morphological and Crystalline Studies of Polymer Deformation

The crystallization kinetics and the final morphological character of polymers, whether
it is spherulitic, cylindrite or fibrillar, are deeply influenced by the molecular orientation
induced by flow (in the molten state) and deformation (in the solid state). Plastic
deformation of polymer is a complex process. In the case of semi-crystal line polymers,
high applied pressures can transform the initial orientation of the micro spherulitic
structure completely into a fiber orientation [Peterlin, 1971; Machado et al., 2009].
The deformation mechanism commonly takes place within the crystal, but some of them
act in the amorphous layers between lamellae. The simultaneous activity of several
deformation processes resulting the initial structure to transform continuously into the
final oriented state [Peterlin, 1971].
In the transformation model called “micronecking”, it has been postulated that the
lamellae undergo an abrupt fragmentation and unfolding into much smaller crystalline
blocks, from which the dense-packed microfibrils are subsequently drawn out. Many
semi-crystalline polymers are sufficiently ductile to undergo large plastic strains by
extensional flow in tension or compression, resulting in anisotropic products with
interesting mechanical properties. The structural orientation of polymer is an important
parameter influencing the production of polycrystalline and amorphous material with
specific physical properties [Peterlin, 1971].
The deformation of amorphous component was suggested to occur by two basic modes:
1) interlamellar sliding (shear), believed to be the dominant mechanism of deformation
of the amorphous layers; and 2) lamellar separation, which can happen in layers
53
between specially oriented lamellae that must thicken during crystallographic shear
[Machado et al., 2009|.
The intensity of each deformation mechanism depends on the type of polymers, its
crystallographic unit cell, the initial morphology of the material (orientation of the
crystals with respect to the applied stresses, their dimensions, the short and long range
correlation of crystal arrangement, and the degree of crystallinity) and the conditions of
the deformation test, e.g the rate, the temperature, the pressure, the size and the shape of
the specimen [Machado et al., 2009J.
Devolatilization involves particularly boiling, often called “bubble transport

devolatilization”. Its complicated mechanism includes bubble nucleation, their growth
due to the pressure drop and molecular diffusion of the dissolved gas, coalescence and
breakup, as well as release of bubble content to the ambient atmosphere. Nucleation is
physical phenomenon in which a new stable phase is generated within another
metastable phase. Homogenous nucleation starts due to thermal fluctuations, which the
smaller than a critical size are unstable and disappear, whereas those larger than the
critical size grow. In contrast, the heterogonous one proceeds due to the presence of the
seeds [Yarin et al., 1999|.
The nucleation rate for homogenous bubble nucleation can be expressed as
/ = M(2a/nm.y/2 expj^t-AFcr /kT) Equation 2.13
Where
M = number of molecules per unit volume of the metastable phase
a = surface tension
m = mass of the gas molecule
k = Boltzmann’s constant
T = temperature
AFcr = 1 6tio3 / [3(Pv~Pl)2] is crtical free energy and (pv is the equilibrium pressure and
Pl is the liquid pressure; pv-pL is often referred to as “superheat”.)
54
Equation 2.13 gets modified for the case of heterogonous nucleation, accounting for
different surface areas and the surface energies of the liquid-gas, solid-gas and solid-
liquid interfaces, as follows:
/ ~ expl'ip-FA Fcr/kT) Equation 2.14
Where the factor F depends on the contact angle, geometry and so on. This factor is an
empirical adjustable parameter that permits fitting of the theoretical nucleation rate to
experimental data. However, such a fitting is not always possible and the values of F are
not universal and far from the order of one. However, the attempt of applying the theory
whether classic or modification in bubble nucleation in polymer melts resulted in
temperatures or superheats far from any plausible estimate [Yarin et al., 1999].
Vacuum and supersaturation will cause the gas pocket to grow but not to detach. Forces
that could overcome surface tension can only generated by deformation in the melt, thus
able to free some of the gas entrapped in the cavities and form growing bubble. The
bubble will continue to grow, leaving a gas pocket in the cavity a source for next
bubble. The process is influenced by liquid viscosity and tensile stress in the vicinity of
the primary bubbles can increased drastically the nucleation rate of the secondary
bubbles. These stresses enhance the mechanical degradation which attribute to
nucleation rate (Yarin et al., 1999].
The evidence of boiling, including the craters caused by the volatile eruption can be
seen in PP melt. The minimal observed size of the nuclei of different polymer was 0.2 to
0.4 pm. The impurities (such as dust particles) in the polymer melt may form primary
nuclei for bubble growth. The fast simultaneous growth of the secondary bubbles results
in entire neighbourhood of the primary bubble disappeared and leaving the large voids.
Large number of microblister (secondary nucleated bubble) seen on the surface in the
vicinity of a primary growing bubble. The smooth lateral surface of the strands indicates
that the intensive boiling has been completed (Yarin et al., 1999].
55
2.7 Summary
The recycling of plastic waste has been known and developed predominantly by
mechanical, chemical and energy recycling programmes. The demand of recycled
plastic markets has been improved and consumers are now being aware of the
importance of recycling. However, a limited number of studies have been reported on
the behaviour/characteristics of solid residues after thermal treatment of plastic wastes.
The chemical or feedstock recycling has been carried out for recycling of mixed waste
as the separation of polymer waste is unfavourable due to its difficulty. Carbon content,
calorific value, ash content, structures changes and many other characteristics play
important roles in determining the value and significance of the plastic waste as
individual or in mixtures component.
Polypropylene is an ideal candidate for the research as it has moderately good properties
in mechanical, physical and chemicals aspects. The thermal recycling of polypropylene
has been growing as it has high potential as refinery of chemical stock conversion and
good fuel combustion characteristics. The thermal degradation mechanism of
polypropylene consists mainly of random scission reactions, which result in highly
depolymerisation into small hydrocarbon carbon with low molecular weight.
The utilization of carbon in manufacturing of anodes for aluminium making process is

well established. Coal tar pitch used in the petroleum coke blend acts as a binder of
anode materials for the process. However, the high carcinogenic content brings several
disadvantages, so attempts are being made for replacing it with other materials.
Polymers have been postulated as good candidates for substituting pitch as these display
better optical texture in coke composition and good hydrogen transfer that inhibited by
pitch during co-coking with coke. Moreover, polymer has low surface tension compared
to the other materials that may help wetting the surface to enhance the interaction with
petroleum coke. Therefore, polymers such as polypropylene, polyethylene, Bakelite and
polyethylene terephthalate were chosen in this research to determine their interaction
behaviour with petroleum coke and as a partial replacement for pitch binder.
56
Chapter 3: Experimental
CHAPTER 3
EXPERIMENTAL
57
3.1 Materials Investigated

The following section presents specific details of raw materials used in this study.
3.1.1 Polypropylene (PP)

Polypropylene used in this study was series PPR 2042 of type clear to opaque, white to
off-white solid pellets form of samples purchased from Qenos Pty Ltd (Figure 3.1). It
was a medium flow grade of polypropylene copolymer which has 4.0g/10 min melt-
flow index (230°C/2.16 kg)(based on ASTM D1238) and 0.902 g/cm3specific gravity.
This type of raw PP is suitable for food contact applications. Decomposition products
may contain carbon dioxide (CO2), carbon monoxide (CO), flammable hydrocarbons
and fumes.
Figure 3.1: Raw polypropylene samples
Table 3.1: Carbonaceous materials compositions, (dry wt% basis) of PP
Material Carbon (%) Hydrogen (%) Sulphur (%) Ash (%)

PP 80.60 19.02 0.05 0.33
3.1.2 Polyethylene (PE)

Polyethylene (PE) was in the form of white pellets, and had a good gloss (Figure 3.2).
It was obtained from ExxonMobile Chemical. LL 6201 contains heat stabilizer, is high
flow PE grades of 50g/10min melt-flow index (190°C/2.16 kg) (based on ASTM
D1238) with 0.926 g/cnr density and 123°C melting temperature.
58
(a) Granules (b) Ground

Figure 3.2: Raw polyethylene samples
Table 3.2: Carbonaceous materials compositions, (dry wt% basis) of PE

PE 80.30 19.00 0.03 0.52
3.1.3 Polypropylene Black (PPB)

Black coloured polypropylene flakes size of 1-2 mm samples were used as a
comparison for thermal degradation studies. PPB was purchased from Sigma-Aldrich
(Figure 3.3).
Figure 3.3: Raw polypropylene black samples
Table 3.3: Carbonaceous materials compositions, (dry wt% basis) of PPB

PPB 82.20 16.44 0.01 1.35
59
3.1.4 Polyethylene Terephthalate (PET)

Polyethylene terephthalate was collected from drink bottles and containers and ground
into 1 mm powder size (Figure 3.4).
(a) PET glass (b) PET - ground

Figure 3.4: PET glass and raw samples after ground
Table 3.4: Carbonaceous materials compositions, (dry wt% basis) of PET
Material Carbon (%) Hydrogen (%) Oxygen (%) Ash (%)

PET 62.6 4.19 32.9 0.31
3.1.5 Bakelite
Bakelite also known as phenol-formaldehydepolymer was obtained from Plastic
Products, South Australia as 10 kg bulk black powder odourless samples. However, this
sample has significant additives as listed below in Table 3.5.
Figure 3.5: Raw bakelite samples
60
Table 3.5: Ingredients in Bakelite samples
Proportion
Material Ingredient CAS
(wt%)
Methenamine 100-97-0 < 10
Phenol-
Formaldehyde Phenol 108-95-2 <0.5
polymer Formaldehyde 50-00-0 <0.2
(30-70%)
Moisture 7732-18-5 <3
Pigment Carbon black 1333-86-4 <20
Fillers Mica 12001-26-2 0-60
(30-60%) Softwood flour - 0-30
Kaolin 1332-58-7 0-25
Additives
Calcium carbonate 471-34-1 0-12
Cotton fibre - 0-35
Barium sulphate 7727-43-7 0-30
Talc 14807-96-6 0-12
The thermal decomposition of samples of uncured product above 130°C releases

ammonia and traces of formaldehyde and phenol. Thermal decomposition products of
the cured product above 300°C are expected to include carbon monoxide,
formaldehyde, phenols. Xylenols and methane while above 600°C, the products may
also produce benzene, toluene and benzaldehyde.
Table 3.6: Carbonaceous material compositions, (dry wt% basis) of Bakelite
Material Carbon (%) Hydrogen (%) Oxygen (%) Sulphur (%) Ash (%)
Bakelite 66.87 5.74 16.75 0.0143 10.62
3.1.6 Petroleum Coke (PC)

Petroleum coke clumps was supplied by Rio Tinto Australia. Samples were then sieved
to segregates for particular particle size for further analysis and then ground by ring mill
into fine powder (Figure 3.6).
61
(a) Granules (b) Powder

Figure 3.6: Raw petroleum coke samples
Table 3.7: Carbonaceous materials compositions, (dry wt% basis) of PC
Material Carbon Sulphur Ash Moisture Volatile matter

(%) (%) (%) (%) (%)
PC 96.70 2.60 0.30 0.10 0.30
Table 3.8 summarizes the specific details of polymers used in this study along with
their abbreviations. PC was used to study the possible reaction of polymers during heat
treatment for recycling polymers in carbon anode for aluminium making industries.
Table 3.9 displays ash compositions of each raw material used in this research. The
composition and their magnitudes were obtained from PANalytical PW2400 Sequential
WDXRF Spectrometer at the Analytical Centre, UNSW.
Table 3.8: Physical appearance of polymers used in this study
Chemical
Materials Form Colour Abbreviation
formula
Opaque
Polypropylene (C3H6)n Solid - Pellet PP
white
Polyethylene (C2H4)n Solid - Pellet Glossy white PE
Polypropylene Opaque
(C3H6)n Solid - Pellet PPB
black black
Polyethylene Opaque
(C,0H8O4)n Solid - Glass PET
terephthalate white
Solid - Opaque
Bakelite (C6H60.CH20)n Bakelite
Powder black
Solid -
Petroleum coke C Black PC
Clump
62
Table 3.9: Ash compositions (%) of different types of raw materials
Components PC PP PE PPB PET Bakelite

Fe203 16.50 0.10 0.09 0.28 0.09 0.16
MnO 0.27 0.01 0.01 0.01 - 0.02
TiO, 0.52 0.04 0.03 0.38 0.04 0.04
CaO 10.5 0.01 0.01 0.33 0.03 6.85
k2o 0.66 - - 0.02 - 0.05
so3 3.30 - - 0.06 - 0.06
P2O5 0.98 0.01 0.01 0.01 0.01 0.02
Si02 33.50 0.01 0.09 0.06 0.06 2.16
A1203 8.60 0.12 0.28 0.14 0.21 0.75
MgO 2.70 0.05 0.01 0.01 - 0.55
Na20 3.80 0.01 - 0.05 - -
Zn - - - 0.02 - 0.01
3.2 Experimental
The series of experiments conducted were divided into three main sections, which are:
the wettability studies; heat treatment studies on the mixtures of polymer and PC
studies; and thermal degradation studies. Specific details of each experiment and
analysis conducted are presented in following sections.
3.2.1 Wettability Studies

PC clumps were milled using RockLabs Ring Mill (Figure 3.7) into fine powder. The
PC powder were mixed with 5wt% phenol folmaldehyde binder and put onto roller
milling machine for 24 hours to ensure homogeneous mixing for preparing cylindrical
substrates. Approximately 1.5 grams of petroleum coke and binder mixture were needed
to make a substrate, using mould and die pressed by 8kNxlO Enerpac 10 Tonne Model
PEMA1321 Hydraulic Press machine (Figure 3.8). Figure 3.9 shows the PC substrate
discs. The substrate disc had a 20 mm diameter and 5mm thickness. These substrates
were baked at 180°C for 24 hours to carbonize the binder and harden the substrates.
63
Figure 3.7: RockLab ring milling maehine
Figure 3.8: Enerpac 10 Tonne Hydraulic
Figure 3.9: Petroleum coke substrate
About 0.20-0.30 g of ground polymer was put on the petroleum coke substrate and the
assembly was charged into horizontal tube furnace model HTF 6035 equipped with
program control panel of Eurotherm 2116, using quartz tube under lL/min argon flow
64
(Figure 3.10). The assembly was placed onto graphite rod sample holder which was
firstly filled and flattened by alumina powder (Figure 3.11). Temperatures and holding
times have been shown in Table 3.10. Heat treated assemblies were observed and
recorded by digital camera. Assemblies were then mounted in epoxy resin. Mounted
samples were then sectioned in the middle and polished for optical and SEM
investigations.
Table 3.10: Wettability studies samples and experimental parameters

Polymer Temperature Duration (min)
(°C) 15 30 60
PP 150 X X X
200 X X X
250 X X X
300 X X X
350 X X X
PE 150 X X X
200 X X X
250 X X X
300 X X X
350 X X X
PET 150 X X X
200 X X X
250 X X X
300 X X X
350 X X X
400 X X X
65
Figure 3.10: Horizontal tube furnace used for pyrolysis
Gas outlet
PC substrate with
polymer assembly 3tr •* rFsaLzass
Graphite rod
Graphite holder
A----*-
Hot zone
m
feeder Quartz tube Gas inlet
Figure 3.11: Polymer on PC substrate assembly on alumina powder based graphite

holder in horizontal tube furnace
Samples were polished and coated with gold sputter for SEM investigations.
Measurements of contact angle and depth of penetration of the molten polymer were
carried out with the help of Photoshop CS5 software on SEM micrograph images. The
techniques used for the measurement of contact angle and depth of penetration of the
polymer melt on the petroleum coke are shown in Figure 3.12. Depth of penetration
was determined by measuring the distance up to which of polymer melt had gone
66
through the petroleum coke particles. The measurements were made on at least three
sets of samples; mean values have been reported in this thesis.
Polymer melt
Contact angle, 0°/ \
Petroleum coke substrate
1.00mm
C ' —>
v
| Polymer melt
1
m m &; ,«i
r . ■,: 4
Petroleum coke substrate
Figure 3.12: (a) Contact angle measurement and (b) depth of penetration measurement
of polymer melt into petroleum coke substrate
3.2.2 Soaking and Heat Treatment on PP and PC Substrates

10, 20 and 30 wt% of ground PP were mixed with PC powder, added with 5% of binder
and put onto roll mill for 24 hours. 2 g of the mixtures were weighed and pressed into
substrates, before putting in the oven at 100°C for 24 hour to remove moisture.
The samples were charged for 30 min in the furnace heated to the soaking temperatures
listed in Table 3.11 and weight loss after the soaking was determined. Soaked sample
were then placed in the oven and heated up to 600°C. The experiments were done under
0.5 L/min argon flow.
67
Table 3.11: Experiment conditions of low temperature soaking and 600°C heat
treatment on PP and PC blend substrates
Soaking Temperatures Heat Treatment up to 600°C

pp%
(Dwelled for 30min) (Dwelled for 30 min)
10 200°C X
350°C X
300°C X
350°C X
20 200°C X
350°C X
300°C X
350°C X
30 200°C X
350°C X
300°C X
350°C X
3.2.3 Heat Treatment on the Mixtures of Polymer and Petroleum

Coke Studies
The heat treatment of mixtures between polymer and PC was carried out to investigate
the possible physical or chemical reactions involved. Only PP has been mixed in the full
range, from 0 to 90% of wt ratios, while PE and Bakelite were respectively mixed in 10
to 50 and 40 to 50 wt% to PC. Petroleum coke clumps were crushed into smaller
aggregates with mortar and pestle and segregated by Retsch Seiving Machine type AS
200 Basic (Figure 3.13) using 2-5 mm screen. Polymers were ground using Fritsch
knife grinder (Germany made) as shown in Figure 3.14.
68
Figure 3.13: Petroleum coke was screened by Retsch sieving machine to obtain
particles in the size range of 2 - 5 mm
Figure 3.14: Fritsch knife grinder
69
Table 3.12: Heat treatment method of polymer and petroleum coke mixtures
Heating Cycle Baking method (1 L/min argon flow)

Ambient temperature - heated to 150°C and dwell for 30 minutes
Cycle 1
- cooled to room temperature.
Cycle 2 - heated up to 600°C and dwell for 30 minutes - cooled to room
temperature.
Cycle 3 - heated up to 600°C and dwell for 30 minutes - heated up to
1000°C - cooled to room temperature.
Table 3.13: Mixtures percentage of sample and experimental conditions of pyrolysis of

mixtures of polymer and petroleum coke
Polymer Polymer in Mixtures (%) Cycle 1 Cycle 2 Cycle 3

PP 0 X X X
10 X X X
20 X X X
30 X X X
40 X X X
50 X X X
60 X X X
70 X X X
80 X X X
90 X X X
PE 10 X X X
20 X X X
30 X X X
40 X X X
50 X X X
Bakelite 40 X X X
50 X X X
The mixtures were prepared in the range of petroleum coke/polymer ratios, where
weight percentage was calculated prior to weighing and mixing. The mixed samples of
polymer and petroleum coke were placed on roller milling machine for 24 hours to
obtain homogeneous mixtures.The mixture was put into an alumina crucible, and placed
onto graphite rod feeder and held in the cold zone for 10 minutes to allow lL/min argon
gas to purgethe horizontal tube furnace (Figure 3.12) before pushing it into the hot
zone. The heating cycles and experimental conditions used for various heat treatments
are detailed in Table 3.11. Table 3.12 shows the percentages of each polymer used and
their heating cycles.
70
3.2.4 Polymer Thermal Degradation Studies
Gas outlet Alumina crucible containing

polymer samples
Graphite rod
feeder Quartz tube Gas inlet
Figure 3.15: Schematic diagram of horizontal tube furnace set up for pyrolysis of
polymers and mixtures samples
The studies of thermal degradation of polymers were carried out to develop fundamental
understanding as temperature and time were found to have a significant impact on the
polymer behaviour and morphology as indicated in wettability and mixtures studies
investigated. About lOg of powdered polymer sample was put into alumina crucible and
weighed using using Precisa 1212 MSCS Weighing Instrument. The pyrolysis was
carried out in a horizontal tube furnace model (Figure 3.10) under lL/min argon flow.
Figure 3.15 shows the schematic diagram of sample assembly and horizontal tube
furnace that has been used in this experiment.
The furnace was heated up according to the temperatures listed in Table 3.13. Weighed
samples in the alumina crucible was placed on the graphite feeder and inserted into cold
zone. The sample was held for 10 min while argon was purged at lL/min in order to
purge out the excessive air in the furnace and completely filled it with inert gas to
optimize the pyrolysis. The sample was then pushed into the hot zone and held
according to the time as listed in Table 3.13. The exhaust pipe was checked to make
sure the gas emitted from the experiments was channelled efficiently to the exhaust
system.
71
Table 3.14: Experimental conditions of pyrolysis of polymers
Temperature Time (min)

Polymers
(°C) 2 5 10 20 30 60
400 - - X X X -
450 - - X X X -
500 - - X X X -
550 - - X X - -
600 - - X - - -
PP
650 - X - - - -
700 - X - - - -
750 - X - - - -
800 X - - - - -
850 X - - - - -
600 - - - - X -
650 - - X X X -
PE
700 - X X X X -
750 - X X X - -
300 - - X X X X
320 - - X X X X
340 - - X X X X
360 - - X X X X
380 - - X X X X
PPB 400 - - X X X X
420 - - X X X X
440 - - X X X X
460 - - X X X X
480 - - X X X X
500 - - X X X X
3.3 Analytical Instruments
3.3.1 Thermogravimetic Analysis

The PP sample was subjected to thermogravimetric Analysis (TGA) to determine the
decomposition mechanism and its kinetic studies. Thermogravimetric study was carried
out using Perkin Elmer Thermogravimetric Analyzer - Pyris 1 TGA and analysed by its
individual Pyris 1 TGA software (Figure 3.16). The vertical design of TGA was
equipped with high sensitivity balance and quick response microfumace. The
microbalance is located above the furnace and is thermally isolated from it. A precision
72
hang-down wire is suspended from the balance down into the furnace. The sample pan
is located at the end of the hang-down wire. The sample pan’s position is reproducible.
Figure 3.16: Perkin Elmer thermogravimetric analyzer
About 5-10mg samples were loaded into the pan and placed onto the hang-down wire
very carefully. Nitrogen gas was purged into the samples area to remove 99% of oxygen
and replaced the volume of ambient gas. The furnace was programmed by heating from
25°C to 100°C at 40°C/min, then hold for lOmin at 100°C to remove the sample’s
moisture. The pyrolysis was carried out by heating from 100°C to 600°C at 10°C/min.
The same procedure was repeated for 15°C/min and 20°C/min heating rates. The
furnace was controlled by integrated Pyris 1 TGA software. The software has been
optimized for better than 0.02% accuracy.
The raw PC and PP, and their 10% PP mixture were also subjected to
thermogravimetric analysis (TGA) studies, to determine the effect of PC addition into
the decomposition of PP polymer. The sample were heated from 25°C to 100°C at
40°C/min, and then held for 10 min at 100°C to remove moisture before continue
ramping up from 100°C to 600°C at 10°C/min. The mass change was determined in
0.03min step.
73
3.3.2 Carbon Analysis

The treated samples were analyzed to determine the carbon content using LECO CS 444
(Figure 3.17). LECO CS 444 machine is a carbon and sulphur analyser of non-
dispersive, infrared and digitally controlled instrument which is designated to measure
the carbon and sulphur content in variety of samples. Sample weights were around 0.5
grams in the crucible and placed into instrument which had been preheated to 1250°C.
The decarbonisation occurs at 1250°C and was held at that temperature until the
estimation of carbon and sulphur content of the sample charged completed.
Figure 3.17: LECO CS 444 carbon and sulphur analyzer
3.3.3 X-Ray Diffraction Analysis

The structural changes of the heat treated residues have also been investigated by the X-
Ray Diffraction (XRD) analysis. The crystal structure of each sample was determined
by Philips Xpert Multipurpose X-ray Diffraction System (MPD) at the Analytical
Centre, UNSW (Figure 3.18). X-ray diffraction is a non-destructive technique of
analyzing the crystallographic structure and its chemical compositions. The samples
were powdered and filled into 10mm diameter and 5mm depth sample holder prior to
experimentation.
74
The samples were scanned from 10° to 100° 20 diffraction angles, under the step size
used of 0.02° and the scan rate of 1° per minute. The operating conditions of the
instrument were 45kV and 40mA. The phase identification was done using Xpert
Highscore software, which it has built in library data to match up possible peak
occurred from the sample.
Figure 3.18: Philips Xpert Multipurpose X-ray Diffraction System (MPD)
3.3.4 FT1R Analysis

In addition to analyzing the structural changes, the chemicals reactions of the samples
also have been analyzed by Fourier Transform Infra-red Analysis (FT-IR).FT-IR is a
chemically-specific analysis technique that can be used to identify chemical
compounds, and substituent groups of the samples. It is an absorption technique, where
IR radiation is passed through the sample. Some of the infrared radiation is absorbed by
the sample and some of it passes through. The resulting spectrum represents the
molecular absorption and transmission, creating a molecular fingerprint of the sample.
The fingerprints of the sample correspond to the frequencies of vibrations between the
bonds of the atoms making up the material. Each specific material is a unique
combination of atoms; no two compounds produce the exact same infrared spectrum. In
addition, the size of the peaks in the spectrum is a direct indication of the amount of
material present.
75
KBr
' ussium Brum*
For IK
(a) (b) (c)

Figure 3.19: Sample preparation material and instruments (a) KBr powder (b) mould
for KBr pellet and (c) hydraulic press for KBr pellet fabrication
Figure 3.20: Nicolet Avatar 320 FT-IR instrument
A small amount of sample was mixed with potassiumbromide (KBr) and ground in
agate mortar and pestle for several minutes before pouring into a mould and assembly
for hydraulic pressing at 15kPa. The KBr, disc mould and hydraulic press used in this
sample preparation is shown in Figure 3.19. The 7mm diameter KBr disc is a
transparent disc, then placedin the plate of V-shaped sample holder inside the FTIR
analyzer for analysis. The analysis was carried out by integrated OMN1C software
connected to Nicolet Avatar 320 FT-IR instrument (Figure 3.20).
76
3.3.5 Scanning Electron Microscope

The effect of heat treatment on the morphology of residues was also explored. The
samples were subjected to scanning electron microscope (SEM) analysis. Hitachi
S3400-X (for imaging and X-ray microanalysis) was used for scanning electron
microscope studies (Figure 3.21). This instrument is fitted with secondary and
backscattered electron detectors that allow for topographic and compositional (atomic
number contrast) surface imaging of samples. These microscopes are routinely used for
imaging from 20x to 20,000x magnification. The S3400-X is fitted with a Thermo
Fisher Scientific energy-dispersive X-ray microanalysis (EDS) system utilising
NORAN System SIX software. Users can undertake qualitative and semi-quantitative
elemental characterisation of their sample surface using whole-screen scans, points,
linescans and maps.Sample preparation involves as simple as putting sample on carbon
tape and then coating with gold sputter. Sample was put onto the sample holder prior
imaging.
Figure 3.21: Hitachi S3400-X scanning electron microscope
77
Chapter 4: Results & Discussion - Polypropylene
CHAPTER 4
RESULTS & DISCUSSION:

POLYPROPYLENE
In depth investigations on polypropylene (PP) are presented in this chapter; these have
been divided in three sub-sections to focus on various aspects of carbon anode baking
process. The first section investigates the interaction on PP melt with petroleum coke
(PC) focused specifically on wettability and interfacial phenomena. The second section
investigates the effect of temperature and time on the mixtures of PP and PC as these
are processed through a number of heating cycles. In-depth studies on the degradation
of PP are presented in the third section as a function of time and temperature.
4.1 Interactions of PP with PC Substrates

Wettability investigations were carried out in the temperature range 200°C to 350°C
focussed on the wettability and interfacial phenomena between PP and PC.
Understanding interactions of these materials is important for assessing the
compatibility of different waste polymers for use in the carbon anode baking process in
aluminium making. Anode properties and characteristics depend on the raw materials
used (carbon/polymers), their individual properties and interactions at high
temperatures.
The contact angle and the depth of penetration are reported as a measure of wettability
and interfacial phenomena. The images of the cross-sectional region of the reacted
sample were captured by a digital camera, and these were used in the computation of
contact angles. SEM micrographs of the interface were used to calculate the depth of
penetration of the polymer into the coke. The interfacial behaviour of waste polymers
will depend on the material properties as well as the temperature and the time of
contact. The influence of these factors has been determined. Figure 4.1 shows the
image of raw PP placed on the surface of the petroleum coke substrate prior to high
temperature experiment.
Figure 4.1: Raw PP on petroleum coke substrate before heat treatment

79
Chapter 4: Results & Discussion — Polypropylene
a) Wetting behaviour at 150°C

Figure 4.2 shows the images of the polymer on the coke substrate after different time
intervals at 150°C. The experiment was carried out for time ranging between 15 to 60
minutes. No change was observed since the polymer did not melt even after 60 minutes.
Q a) 15 min
Hk
b) 30 min
SR
c) 60 min
Figure 4.2: PP on petroleum coke substrate after exposure to 150°C for a) 15 min
b) 30 min and c) 60 min
b) Wetting behaviour at 200°C

Figure 4.3 to 4.5 show the variation in the interfacial behaviour at 200°C for different
reaction times. Minor changes were observed with increasing reaction time.
Figure 4.3: PP on petroleum coke substrate and the corresponding SEM micrograph of
the interfacial area after 15 min of reaction at 200°C
PP was observed to have started melting after 15 minutes. However, during this time,
PP had only softened and then re-solidified on cooling. The SEM micrograph shows an
uneven top surface of PP, indicating that it had not melted completely. PP was still
fairly intact in its raw form. A number of voids were also observed. The melting of PP
had drawn its particles together towards the centre of the substrate.
80
When the reaction time was increased to 30 min, PP appeared to fully melt as the
surface was much smoother as compared to that observed after 15 minutes. Three big
voids were also noted in the PP region from the SEM micrographs (Figure 4.4). The re-
solidification of the polymer upon cooling resulted in shrinkage resulting in a crack
being formed underneath the melt contact area between PP and coke.
After 60 min, the PP had melted completely forming a semi spherical dome on the coke
surface as shown in Figure 4.5. In addition, the fracture of the petroleum coke substrate
below the interface was greater than previously seen. Small bubbles within the PP melts
were also visible at this stage.
c) Wetting behaviour at 250°C

Figures 4.6-4.8 shows the variation in the wetting behaviour with time at 250°C. The
wetting appeared to be similar to that observed at 200°C for the similar time period. The
melt was still opaque, exhibiting raw PP properties. PP melt was uneven in shape and
had holes of various sizes in the polymer region. The fracture of the coke surface
underneath the interfacial region of melted PP was also observed in this sample.
81
When the sample was reacted for 30 minutes, some small voids were visible within the
PP melt (Figure 4.7). These bubbles may be caused by the devolatilization of PP during
the heat treatment. The colour of the PP melt was lighter than that seen for the
corresponding sample exposed to 250°C for 15 minutes. Only one big void was
observed in this sample, together with a few smaller cavities on the PP region. The
cracking of the substrate was also detected in this sample.
. .
••
Figure 4.8 shows the wetting behaviour of the polymer after 60 minutes at 250°C. The
PP melt had become yellowish with even lower transparency. Bubble formation was
82
less compared to the sample after 30 minutes of reactions. This indicates the reduced
intensity of PP devolatilization after 60 minutes in the hot zone.
d) Wetting behaviour at 300°C
Figure 4.9 shows the image of the droplet of PP on petroleum coke substrate at 300°C
after 15 minutes of reaction. Bubbles formation was clearly seen from the top of the
droplet while the melt colour was yellowish and less transparent. The contact angle
between PP droplet and petroleum coke surface was 48.75° and the dimensions of the
melt were 6.45mm in length and 1.83mm in height. Large voids were observed along
with small cavities throughout the droplet.
Figure 4.10 shows the image of PP on PC substrate after reaction at 300°C for 30min.
The yellowish of the droplet seen in Figure 4.10 has started to fade and the melt has
become more translucent. The droplet had a contact angle of 31.00° and the length of
melt was 7.33mm, which was 13% higher than that seen for corresponding droplet after
15min of contact at 300°C. Moreover, the height of the droplet had also decreased by
more than half compared to the previous sample, resulting in a thinning out of the PP
after 30min of reaction. Large voids were also spotted in the PP region.
83
The length of the PP droplet was observed to increase up to 9.96 in length as the
reaction time was increased to 60 min, while the height decreased to to 0.81mm (Figure
4.11). As expected, the contact angle was lower (24.3°) compared to sample earlier (15
minutes and 30 minutes). Reduced amount of bubbles were seen at this stage and these
appeared to have accumulated close to each other in one region of the droplet. The
cracking of the coke surface was also observed in this sample.
e) Wetting behaviour at 350°C
Figure 4.12 shows the image of the melted PP on coke after reactions at 350°C for
15min. The droplet had a contact angle of 27.6°, with a length of 7.33mm across the
substrate and height of 0.97mm. A large number of small bubbles could be seen and in
addition, one large cavity was seen in PP area, showing the unmelted PP in some region
as 15 minutes was too short for PP to melt completely.
84
Figure 4.13 shows the wetting of the polymer after 30 minutes of reaction at 350°C.
The length of the spread was 9.07mm while the height was 0.94mm, and this was
similar to that observed for samples after 15minutes. Similar results were also seen in
terms of the contact angle measurement (26.75°). As viewed from the top of the PP
melt, the sizes of bubbles were noticeably larger and these were present all over the
droplet. A large void was also noted along with few cavities from the cross-sectional
view.
When reaction time was further extended up to 60 minutes (Figure 4.14), the PP was
seen to have fully melted and complete wetting of the petroleum coke substrate was
seen. PP formed a thin layer with a contact angle of 12.55°, length of 9.67mm and
thickness of 0.48 mm on the coke substrate. The solidification of PP upon cooling
resulted in the lifting up the attached coke substrate surface.
Observed experimental results on contact angle measurements and dimensional change

of the PP melt calculated have been summarized in Table 4.1.
85
Table 4.1: Dimensional changes in the PP droplet on the PC substrate for different
reaction temperatures and times
Experimental conditions Horizontal

Height Contact
Temperature Time cross-section
(±0.05 mm) angle (±5°)
(°C) (min) (±0.05 mm)
200 15 6.83 1.82 137.80
30 6.56 1.69 75.65
60 7.74 1.47 69.05
250 15 10.35 1.60 87.60
30 8.12 1.34 50.30
60 10.55 1.34 45.30
300 15 6.45 1.83 48.75
30 7.33 0.98 31.00
60 9.96 0.81 24.30
350 15 7.33 0.97 27.60
30 9.07 0.94 26.75
60 9.67 0.48 12.55
The work of adhesion has been calculated from the data according to the Equation 4.1
derived by Schroder (1999].
fVs(V)L= Yl (1+COS0 ) Equation 4.1
Where:
R^S(V)l = work of adhesion of the liquid drop to the vapour-equilibrated solid surface
Yl = free energy of liquid
Contact angle results as function of time and temperature and time have been plotted in
Figure 4.15.
2.50
(b)
§ 2.00 . --------- •
•------
1 A—
! 1.50 *
O
N
-‘5 i.oo : ^
§
o
* 0.50
0.00
15 30 60
Time (min)
A 300 c • 350°C
Figure 4.15: Effect of time and temperature on (a) contact angle and (b) adhesion of PP
on PC substrate
86
When PP is heated, it was first contracts and then swells to a much larger volume than
its original size. This phenomenon can be seen at initial stage of heating (low reaction
temperature and time, e.g. at 200°C after 15 min) where high contact angle was
measured. The heating of polymer melt increased the devolatilization rate, which
precede the transition from liquid to gas phase. Devolatilization occurs due to bubble
transport of gases to the surface and when part of the gas escapes, thus the viscosity of
the melt reduced significantly and thus allowing some of the melt flow through
petroleum coke particles. Continuous bubbling and blistering plays important roles in
pyrolysis process as they serve as mechanism of products and yield of pyrolysis reaction
[Attar, 1978J. The reduction of melt viscosities due to bubbling and blistering
subsequently condensed the molten PP and its contact angle. The eruption of
devolitilized gas forms voids and crusty blisters, which tend to contract due to
interfacial surface tension upon cooling.
After rapid devolatilization, the reduction of the interfacial tension reduced the
contraction during the melt solidification and consequently decreased its contact angle.
As can be seen in Figure 4.15, the adhesion increased as the contact angle decreased.
The reduction in contact angle shows that the melt of PP has bonded with petroleum
coke particles. The increase of temperature has also decreased the melt viscosity and
allowed PP to flow through and get trapped in between petroleum coke particles, thus
form the bond inside the substrate. The interfacial phenomena and the adhesion between
PP and petroleum coke is further explained and demonstrated in Section 4.11.
4.1.1 Interfacial Phenomena Between PP and PC

The melting of PP and its subsequent penetration into petroleum coke substrate was
investigated through SEM micrographs of the interface. EDS scanning was also carried
out but did not produce significant distinction of PP polymeric region and petroleum
coke substrate (Figure 4.16). The visual differentiation between the polymeric and coke
regions was done using micrographs, as both of these regions had a distinct appearance.
From the micrograph, it can be clearly seen that the upper part of the micrograph has a
ragged surface, which represent plasticity while the lower part is smoother with
assembly of petroleum coke particles. This identification is important as it will be used
to distinguish these two materials throughout this study.
87
Point 1
20.0 kV x2.00k
Point 2 Point 3
Figure 4.16: EDS analysis of points in the interface of PP-PC system reacted for 60
minutes at 200°C
Since the PP had not melted sufficiently at 150°C, SEM micrographs are presented for
the temperature range of 200°C to 350°C. High carbon peak at points 1, 2 and 3
represents C from PP and PC, while observed Au peaks are from the gold coating used
for SEM analysis of the sample (Figure 4.16).
a) Reactions at 200°C
Figure 4.17 shows the penetration of PP into petroleum coke substrate at 200°C. After
15 minutes, the penetration depth was measured to be 8.59 pm (Table 4.2). The
penetration was lower than that observed at higher temperatures and times. At some
locations, PP did not even come into contact with the petroleum coke substrate. At this
temperature and time, PP had just melted and did not penetrate through the petroleum
coke substrate, possibly due to the high viscosity of the semi-molten polymer.
Reactions for 30 min resulted in a change in PP surface morphology (Figure 4.17). The
polymeric strands appeared to be pulled-out from its plane, and starting to deform as
bubbles. Polymeric materials are generally soft and can be deformed easily. The PP
melt started to penetrate into the petroleum coke substrate and went as deep as 21.59
pm (83% greater than seen for the 15 min sample).
88
After 15 min After 30 min After 60 min
ffffpp 8 5k -
4c k_ *rmm- •70
M' ^ %
mm $ f
(a) (b) (C)
Figure 4.17: Interfacial changes due to heating at 200°C for different time intervals
The increase in the heating time to 60 min resulted in further degradation of the PP
surface, with increased levels of pulling observed. More voids were observed in the PP
region. The penetration of PP into petroleum coke surface was also seen to be higher.
Some petroleum coke particles were seen to be lifted from the surface into polymer
during shrinkage associated with re-solidification.
b) Reactions at 250°C
Figure 4.18 shows the effect of reactions at 250°C on the interfacial between PP and
petroleum coke. After 15 min, a low penetration of PP into the substrate was observed
(15.58 pm). This value is higher than the corresponding value for the sample reacted at
200°C for 15 minutes (8.59 pm). The pulling of polymer strands also damaged the PP
surface.
After 30 min of reaction, the penetration of PP melt into the coke substrate increased to
22.87 pm, which is higher than seen for the samples reacted at lower temperatures for
89
the same time. The penetration was more extensive in this case. Molten PP had
penetrated into the substrate creating cracks between the coke particles which resulted
in further penetration.
This effect was even more pronounced after 60 minutes [Figure 4.18 (c)j. This is
probably due to the lowering of the viscosity of the melt, which helps in penetration.
After 60 min of exposure to 250°C, long polymeric chains of PP are expected to
degrade into smaller molecular chains, also accompanied by loss of volatiles. The depth
of penetration at this stage was 28.67 pm. The PP region showed an increase in the
number and size of voids with increasing period of heat treatment, which also points to
an increase in the volatiles released with time.
90
c) Reactions at 300°C
The increase in the heat treatment temperature to 300°C clearly resulted in the
formation of cracks between the coke particles due to increased penetration [Figure
4.19 (a)|. Even though a low penetration of melt was detected (21.45 pm) after 15 min,
the penetrated areas were broader and deeper compared to the samples exposed to lower
temperatures for the same period of time. Polymeric region also showed a smoother
surface (less fibres), but it contained several small voids. These small voids indicate that
the initiation of PP degradation was resulting in the release of volatiles as gaseous
phase.
However, PP strands started to form after 30 min [Figure 4.19 (b)[. In addition, the
number of voids formed was seen to increase. The penetration was almost 30 pm deep
into the PC substrate. Some PC particles were found to be present within the PP melt.
Shrinkage of PP surface with solidification resulted in detachment of the coke layer
close to the interface.
91
Figure 4.20: Schematic diagram of PP melt penetration into PC substrate
Deeper penetration also resulted in a greater binding of PP and PC particles from the
substrate surface. The schematic diagram of the event is further demonstrated in Figure
4.20. The depth of penetration also increased up to 41.17 pm, which was significantly
higher than that observed at lower times for the same temperature. Large voids were
spotted over the surface of molten PP.
d) Reactions at 350°C
92
Figure 4.21 shows the effect of increasing time on the interactions of PP melt with
petroleum coke. The polymeric surface of PP appeared to be less fibrous, thereby
revealing significant numbers of voids. Several cracks were observed in the coke
substrate possibly caused by the upward force of polymeric shrinkage.
As temperature was increased to 350°C, the penetration of PP increased further as

compared to the samples exposed to the same time at lower temperatures (Figure 4.21).
The penetration depths at 15 minutes, 30 minutes and 60 minutes of heat treatment were
24.10 pm, 31.10 pm and 73.30 pm, respectively. The increase in the time resulted in
increased penetration depth. Unlike the other samples treated for 15 min period, the PP
sample at 350°C was seen to deform after 15 min of heat treatment. PP melt was seen to
penetrate through narrow gaps in petroleum coke substrate.
Table 4.2 summarizes our experimental results on the depths of penetration of PP into
petroleum coke substrate. These results have also been plotted in Figure 4.22. The
penetration values are the average of 10 points taken along the penetrated region
between PP and coke.
Table 4.2: Penetration depth of PP into petroleum coke substrate
Experimental conditions Average depth of penetration

Temperature (°C) Time (min) (pm)
15 8.59
200 30 21.59
60 27.82
15 15.58
250 30 22.87
60 28.67
15 21.45
300 30 30.61
60 41.17
15 24.10
350 30 31.10
60 73.30
93
80
S
60
Q.
<L>
C
O
'§
i-.
a>
C
(L>
D-i
30 60
l ime (min)
200°C X 250°C A 300°C 350°C
Figure 4.22: Variation of the penetration depth of PP melt into petroleum coke
substrates as a function of heating temperature and time
Figure 4.22 shows that increasing time and temperature of the experiment generally
increased the penetration depth of the PP melt into PC substrate. The increasing heat
treatment time also accelerated the degradation of PP, resulting in the rapid production
of gaseous and liquid products. The degradation process produced gases, as indicated by
bubble formation. Similar effect was observed with increase in temperature. The
degradation of the polymer decreased its viscosity allowing it to penetrate deeper into
the petroleum coke substrate.
4.1.2 Discussion
Increasing temperature of heat treatment increased the ability of PP to melt, spread and
penetrate into the substrate. The heat treatment was started at temperature as low as
150°C, however, nothing happened to PP at this temperature. The heat treatment
temperature was then increased to 200°C, and PP had started to melt and agglomerate.
The agglomeration was seen to increase as time was increased to 30 min, and was seen
to totally melt after 60 min. During heat treatment at 250°C, the agglomeration was
observed only after 15 minutes, while the prolonged treatments (30 and 60 min) resulted
in a total melting of PP. As temperature was increased further to 300°C and 350°C, PP
started to melt even before 15 min. Bubble formation was observed inside the PP melt
94
Chapter 4: Results & Discussion — Polypropylene
during these temperature regimes and time range. The amount of bubbles formed
increased rapidly as the temperature and time were increased. The formation of bubbles
indicates possible decomposition of PP, as the PP polymer is known to decompose into
solid, liquid and gas. The bubbles signify gaseous devolatization from PP
decomposition.
The devolatization of polymers through foaming and blistering mechanisms has been
studied by a number of researchers [Yarin et ai., 1999; Shuiman and Levitskiy, 1996].
The heating of polymer resulted in boiling of polymer and foaming as bubbles that later
vaporized as a gas. The growth of the gas pocket is determined by the ability of the
polymer liquid to supply heat for the phase transition [Shuiman and Levitskiy, 1996].
The extent of heat transfer is enhanced by mechanical means, while taking into account
the generation of normal stresses and elasticity of liquid [Levitskiy et a!., 1996]. During
this heat treatment, heat has been conveyed throughout the entire polymer melt while its
surface was the most affected area with highest temperature point.
The devolatization was indicated by the nucleation of bubbles, their growth,

coalescence and rupture of the foamed up bubbles on the surface of polymer. Other
phenomenon observed was the appearance of randomly scattered voids inside the
polymer termed as macrobubbles (size 100pm and above). The inner surface of these
macrobubbles was sometimes inhabited by blisters, which are thin, dome-shaped
vapour-filled pockets. It has been suggested that a microblister (l-3pm diameter)
emerges through the soft molten surface of a macrobuble after having been formed as
tiny microbubles under the surface [Albalak et al., 1990].
These vapour-filled microblisters can grow to a maximum diameter of 3pm, after which
they reach a state where its skin is too thin and therefore are too weak to withstand the
pressure; bursting and releasing the vapours contained within the microbuble. In some
cases, the microblisters close to each other can merge to form a larger structure called
miniblister (10-15pm). At some stage, miniblister also burst, collapsing the nearly
empty skin and entrapping small vapour pockets [Albalak et al., 1990]. These
mechanisms help explain the presence of various voids observed within the polymer
melt in the SEM micrographs.
95
Temperature and time were seen to have an important impact on the PP melting,
spreading and its penetration into petroleum coke substrate. The exposure of PP to
150°C heat treatment did not affect PP, while at 200°C, PP had begun to show the
melting behaviour after 60 min. This observation is in agreement with the degradation
range reported in the literature, where polypropylene starts to degrade in the
temperature range from 250°C to 450°C [Peterson et al., 2001],
Polypropylene begins to fully melt after being heated for 30 min at 250°C of heat
treatment. The voids observed in 15 min sample are probably due to the non-uniform
melting of PP, therefore creating gaps between the PP which had not yet melted. Unlike
the sample after 30 and 60 min, the PP had fully melted and had begun to show initial
signs of devolatization. Some bubble nucleation was observed inside the melt while the
cross-section area has showed voids within the PP melt. This effect becomes obvious
after the increment of temperature to 300°C. There were various sizes of voids, from the
gas pocket occurrence in the 15 min samples.
After 30 min, the effect of bubble foaming was intensified (Figure 4.10) where there
were more bubbles foaming and growth observed in the PP melt. The degradation rate
was seen to increase with increasing temperature and longer time enabled broken
polymer chains to further depolymerise into smaller chain. However, the amount of
bubbles was less after 60 min as compared to 30 min sample. It indicates that the
foaming activity has now reduced and can be assumed the devolatization was nearly
complete. The degradation rate reduces as the volatile content gets reduced [Albalak et
al., 1990]. The reduction of the devolatization activity also been indicated by the contact
angle measurement of the sample, where the angle was seen to reduce after 60 min of
treatment, with lesser number of bubbles forming in the melt.
The phenomenon was even more apparent in the sample treated at 350°C. After 15 min,
the melting of PP appeared to be complete from the outer surface, but there was still
some unmelted PP granules seen in the cross-section area. The increased time had
allowed sufficient heat for devolatilization to take place, as seen in the sample treated
for 30 min. Lots of bubbles foamed with voids of blistering showing intensive
devolatilization activities. However, the completely flattened melts after 60 min with
significant slim layer of PP show the PP had completely devolatilized. The blister
96
indentations were seen on the trace of PP on the substrate. At this stage, the penetration
of PP into the petroleum coke substrate also reached its maximum depth of 73.30 pm.
The increasing temperature had also impacted the viscosity of the polymer, as the
heating had allowed devolatilization through nucleation, growth and coalescence of the
bubbles involved in releasing gas of by-product of the polymer degradation process.
Cranmer et al. [1983] in their investigation on the pitch interactions with carbon
substrate have reported that rate of formation and coalescence depends on the rate of
flow of molten material, which was generated by a temperature gradient. Two scenarios
need to be considered in the wettability study, i.e., whether the melts adsorbed onto the
surface or only “sweeping action” occurs. The flow of molten polymer can enhance the
formation, growth and coalescence of the melt, in addition to their spreadability to align
near the surface of the substrate. Results have reported similar phenomena as seen in the
polymer melt on petroleum coke substrate, where the temperature had a major influence
on the wettability behaviour.
Good wetting is critical to mechanically interlock and to create adhesion between the
binder and coke in order to produce strong carbon anodes used in aluminium smelting.
Strong bonding between these two materials is essential to develop bond strength,
depending strongly on the wettability and penetration characteristics [Cao et al., 2002].
Penetration of polymer melt into coke surface also depends on the viscosity of the melt
and its contact angle. These properties were seen to be connected to the temperature and
time of exposure as indicated by our experimental observations.
4.2 Interactions Between PP and PC for Temperature Up to

1000°C
Blends of PP and PC were prepared with the concentration of polymer in the blend
ranging from 10% to 90%. The baking treatment of sample was carried out to
investigate the effect of slow baking (2°C/min heating rates) in inert conditions
(pyrolysis) on the polymer and petroleum coke. The experiments were carried out in
three temporal cycles, and the maximum temperature of each cycles were raised from
150°C (Cycle 1) to 600°C (Cycle 2) and 1000°C (Cycle 3). The heating rate of 2°C/min
was maintained under lL/min argon flow. The experiment was conducted under
97
isothermal conditions to allow slow softening and melting. This also enabled the
polymer to stay in the liquid phase to wet petroleum coke particles.
4.2.1 Weight Loss Study of the PP and PC Mixtures

Collected experimental data after heat treatment have been presented in Table 4.3 and
the % residue from the mixtures collected after the baking cycles have been plotted in
Figure 4.23. The baking of cycle 1, where the mixture was heated from ambient to
150°C and allowed to dwell for 30 min, showed negligible weight loss. The normalized
weight loss of about 0.21 to 0.72% in each sample indicates primarily the evaporation of
moisture.
In Cycle 2, the baking of mixtures (heating up to 600°C and dwell time of 30 min) has
showed mass loss, which was seen to be proportional to the percentage of PP in the
mixtures. The data in Table 4.3 show that after heating and dwelling at 600°C, there
were some traces of PP still remaining in the samples, except for the 10% PP sample
mixtures where there was 10 percent loss of PP after the heating of Cycle 2.
However, after Cycle 3, the weight loss increased proportional to the increase in the
percentage of PP in the mixtures. The mixtures of 20-50% PP has showed some traces
of PP portion left in the residues. However, the sample with 60-90% of PP had
completely decomposed, leaving no residues of the polymer. The mixtures with 90% PP
has the highest rate of loss as it has the highest PP percentage, which has been totally
decomposed in the high temperature heating cycle.
PP(%)
OCycIc 1 □ Cycle 2 A Cycle 3
Figure 4.23: Residue (%) of PP and PC mixtures after heat treatment

98
Table 4.3: Percentage of solid yield of each sample mixtures (PP:PC) obtained after
baking Cycle 1, Cycle 2 and Cycle 3
PP (wt%) Initial Weight (g) Final Weight (g) Loss (%)

0 7.36 7.34 0.21
10 6.28 6.26 0.25
20 6.00 5.98 0.28
30 4.96 4.95 0.24
Cycle 1 40 4.00 3.99 0.25
50 3.50 3.49 0.36
60 3.50 3.49 0.28
70 2.77 2.76 0.34
80 2.88 2.87 0.42
90 2.67 2.65 0.72
0 7.38 7.36 0.23
10 6.27 5.65 9.92
20 6.02 5.02 16.58
30 4.99 3.92 21.54
40 4.01 2.62 34.59
Cycle 2
50 3.53 1.85 47.49
60 3.50 1.57 55.12
70 2.76 0.92 66.77
80 2.89 0.62 78.44
90 2.66 0.30 88.89
0 7.39 7.37 0.27
10 6.28 5.67 9.70
20 6.04 4.95 18.04
30 5.00 3.67 26.61
40 4.05 2.51 38.01
Cycle 3
50 3.52 1.77 49.62
60 3.51 1.36 61.18
70 2.76 0.83 70.02
80 2.86 0.56 80.59
90 2.68 0.27 90.06
99
4.2.2 Effect of PC on the Reactivity of Mixtures
Raw PC
10 PP:90 PC
Raw PP
Temperature ( C)
Figure 4.24: TGA profiles of raw PP, raw PC and mixtures their mixture of 10:90
Figure 4.24 shows the TGA profiles of raw PC, raw PP and their mixtures with 10% PP
(90% PC). The graph shows that no residue of PP was left after 500° while PC shows
no changes throughout the heating. As for the mixture sample, the weight started to
reduce around 280°C and the reduction continued till 430°C where the graph became
almost a plateau around 12 to 9% weight residues.
4.2.3 Carbon Content of the PP and PC Mixtures

The percentage of carbon content of the raw PP and PC and their mixtures were
measured by LECO and are recorded in Table 4.4. The graph of carbon content (%)
according to the mixtures percentage is presented in Figure 4.25. According to Table
4.4, the carbon percentage of raw PP, PC and their mixtures are in the range of 97.6%
(raw PC) to 89.1% (90:10 mixtures after Cycle 1). Both petroleum coke and
polypropylene had a high carbon percentage. The mixtures of both components were
therefore expected to have high carbon percentages.
The carbon in mixture sample of 50:50 at the third cycle of heat treatment has the
highest percentage, followed by samples treated in Cycle 2 and Cycle 1 (Figure 4.25).
The third cycle has generated more carbon since high temperature treatment has
decomposed PP and resulted in more carbon residue, while the volatiles have been
100
released. As compared to Cycle 1 and 2, where volatiles due to incomplete pyrolysis

still exist, the weight percentage of carbon had increased in the case of Cycle 3.
Table 4.4: Carbon content (%) of each sample mixtures (PP:PC) obtained after baking
cycles
Heat treatment
PP (wt%)
Cycle 1 (%) Cycle 2 (&) Cycle 3 (%)
0 97.5 97.1 97.6
10 96.0 96.7 97.3
20 95.0 96.2 97.0
30 94.2 95.4 96.5
40 92.0 93.5 95.4
50 93.9 94.0 96.1
60 92.0 93.5 95.4
70 90.8 93.2 95.1
80 90.0 91.6 95.0
90 89.1 92.8 94.7
The lowering trend of carbon content (%) by the increasing of raw PP ratio indicates
that the increase of PP has simultaneously decreased the total carbon available in the
mixtures. This is due to the fact that the more PP ratio, the more volatiles in the
mixtures. These volatiles were released in the heat treatment. Higher heating as
conducted in Cycle 3 released more volatiles, thus increasing the carbon content in the
blend as compared with Cycle 2 (incomplete devolatilazation) and Cycle 1 (no reaction
due to low temperature).
0 10 20 30 40 50 60 70 80 90
PP(%)
A Cycle I x Cycle 2 O Cycle 3
Figure 4.25: Carbon content (%) of PP and PC mixtures

101
4.2.4 Structural Characterization of PP and PC Blends

The XRD profiles of raw PP and PC and their 50:50 mixtures have been plotted in
Figure 4.26 and Figure 4.27, respectively and characterization presented in Table 4.5.
It was observed that the intensities of raw PP peaks were reduced after mixing with raw
PC. The mixing with petroleum coke has marginally shifted the (110) peak in the
mixtures during Cycle 1. Additionally, new small peak (017) was also observed around
20.96° in Cycle 1.
As temperature was increased for baking in Cycle 2 and Cycle 3, the PP peaks had
diminished in intensity and only PC is visible, including a new peak (015) at 43.16° and
43.43°, for both Cycle 2 and Cycle 3, respectively. The XRD profile of mixture sample
after these cycles was quite similar to that for raw PC, as at this stage most of raw PP
has decomposed and was gone. Additional peaks have been bolded in the table.
50000
Raw PC
40000
30000
20000
Raw PP
j= 40000
30000
20000
10000
10 20 30 40 50 60
Scattering angle (20)
Figure 4.26: XRD profiles of raw PP and PC

102
- Polypropylene
50000
Cycle 1 - up to 150 °C
40000
30000
20000
10000
50000
40000
30000
20000
10000
50000
©
40000
30000
20000
0
30 40
XRD profiles of PP and PC mixtures sample after heat treatment
103
Table 4.5: XRD peaks characteristics of raw PP, PC and their mixtures at 50:50*
Miller Possible
Peak
Heating d-spacing indices Compound
Sample Position
Cycle
(20)
(A) (chemical
h k 1 formula)
Raw PC - 26.03 3.37 0 0 2 C (graphite)
44.64 2.03 1 0 1 C (graphite)
Raw PP 14.13 6.27 1 1 0 (C3H6)n
16.80 5.27 0 0 8 (C3H6)x
18.49 4.79 1 3 0 (C3H6)n
21.82 4.07 0 2 6 (C3H6)x
25.45 3.50 1 5 0 (C3H6)n
28.52 3.13 0 1 2 (C3H6)n
42.52 2.11 1 0 0 C (graphite)
Mixtures Cycle 1 13.87 6.40 1 1 0 (C3H6)x
50 : 50 16.75 5.27 0 0 8 (C3H6)x
18.33 4.79 1 3 0 (C3H6)n
20.96 4.42 1 1 7 (C3H6)x
21.59 4.07 0 2 6 (C3H6)x
25.33 3.50 1 5 0 (C3H6)n
28.10 3.13 0 1 2 (C3H6)n
42.54 2.11 1 0 0 C (graphite)
Cycle 2 25.91 3.37 0 0 2 C (graphite)
42.56 2.11 1 0 0 C (graphite)
43.16 2.09 0 1 5 C (graphite)
Cycle 3 26.16 3.37 0 0 2 C (graphite)
42.68 2.11 1 0 0 C (graphite)
43.43 2.09 0 1 5 C (graphite)
* Additional peaks have been bolded
4.2.5 FTIR Analysis of PP and PC Blends

The FTIR results on the raw PP, raw PC and their 50:50 mixtures are presented in
Figure 4.28 and Figure 4.29, respectively. The possible functional groups of various
peaks were identified and have been presented in Table 4.6. As indicated in Table 4.6,
the C=C stretching in raw PC at 1685cm'1 was gone after the mixing with raw PP. The
loss of C=C stretching after mixing may due to the reaction of PC with PP, thus
releasing the unstable double bond in PC. Similar observation was noticed after Cycle 2
and 3, where the C-CH3 bending (1377cm-1) from raw PP was gone. These vibration
bands were only found in the raw samples and after Cycle 1 of mixture sample.
104
The other significant changes identified were the C=C stretching vibrated at 973cm’1
from raw PP. This band was only observed after Cycle 1 but had disappeared after
Cycle 2 and Cycle 3. The aromatic ring band visible at 893cm'1 (raw PC) and 842 cm'1
(Cycle 1) has gone after Cycle 2 and 3 heat treatment. The C-CH3 bending around 1060
cm'1 of raw PC had remained up till 1000°C heat treatment.
f-
4000 3500 3000 2500 2000 1500 1000 500

Wavenumbers (cm1)
Figure 4.28: FTIR peaks of raw PP and PC
105
Cycle 1 - up to 150°C
Cycle 2 - up to 600°C
Cycle3-up to I000°C
Wavenumbers (cm1)
Figure 4.29: FTIR peaks as observed in the mixtures after various cycles of heat
treatment
106
Table 4.6: Characterization of FT1R peak profiles of PP, PC and their mixtures at
50:50. The units of data are in (cm1)
Raw PC Raw PP Cycle 1 Cycle 2 Cycle 3 Possible Functional

Groups
3484.12 3473.35 3477.47 3790.21 3471.11 Moisture
2913.98 2917.04 2920.32 2920.32 - CH2 stretching
1685.49 C=C stretching
H 3
C s.
^
. CH 3
J r — r>
C C
1653.83 1637.94 1637.99 1637.99 1637.21 C=C (alkene absorption)
1457.52 1459.14 1459.61 1457.52 1456.64 CH3 bending
- 1377.05 1377.33 - - C-CH3 bending
1060.47 - 1070.81 1068.07 1068.92 C-CH3 bending

- 973.18 972.74 - - C=C Alkenes
893.93 - 842.12 - - Aromatic ring
4.2.6 SEM Images of PP and PC Mixtures
Figure 4.30: SEM micrographs and EDS of raw petroleum coke
Figure 4.30 shows the SEM micrographs and EDS of raw petroleum coke used in this
study. Raw PC has a coarse layered structure with smaller particle pieces scattered all
over big particles. EDS analysis illustrates high intensity of sulphur, carbon and silica
with small traces of aluminum.
The SEM micrographs and EDS of raw PP are presented in Figure 4.31. The raw PP
has a plastic smoothness on its surface, while EDS has indicated high intensity of
carbon resulted from the overall scan. The SEM and EDS scans of the 50:50 raw PC
and PP mixture samples after being heat treated are displayed in Figure 4.32.
107
Figure 4.31: SEM micrographs and EDS of raw PP
Figure 4.32 (a) shows the SEM of mixture samples after Cycle 1, where nothing
happened to the mixtures as 150°C was too low for any reaction as raw PP melting
starts at about 170°C. The micrographs show raw PP particles scattered on petroleum
coke surface (both identified by EDS). After Cycle 2, the raw PP has started to melt and
decompose. Figure 4.32 (b) shows the traces of raw PP coating part of raw PC
particles. However, PP levels in the mixtures are significantly reduced, which is in
agreement with the result of the mass loss where the heating in Cycle 2 has reduced
almost 50% of the original weight sample. The high temperature treatment of Cycle 3
(up to 1000°C) has significantly reduced raw PP by decomposing it. No trace of raw PP
was found in the samples of mixtures |Figure 4.32(c)].
The co-pyrolysis of polymers with pitch has been studied before, and is known to have a
significant influence on the physical properties such as softening point, dripping point,
penetration and toluene insoluble component. It has been reported that the carbonization
residues yield of the blend of pitch and some polymers is higher than the single
component. PP has been reported to increase the solid residue yield by 3-5% during the
co-carbonization with coal tar pitch. However, it contributed to the deterioration of the
optical texture of resultant cokes as well [Brzozowska et al., 1998]. The co
carbonization of coal blend with polymers resulted in an increase in the micro-strength.
Additionally, coke reactivity and porosity increased with increasing addition ratio of
polymers JMin et al., 2009],
108
(a) Cycle 1 - up to 150°C
(b) Cycle 2 - up to 600°C
(c) Cycle 3 - up to 1000°C

Figure 4.32: SEM micrographs and EDS of PP and PC (50:50) mixture samples after
heat treatment
4.3 Thermal Degradation of PP

Investigations were carried out on PP over a wide range of temperatures from 300 to
850°C with a variable time interval (5 - 30 min). Temperature and time were selected
based on the melting point of polymers used and the physical state of the residues.
Pyrolysis and comprehensive analysis of solid resides was carried out to understand the
influence of polymer structure on kinetics and decomposition patterns.
Figure 4.33 - 4.34 show the physical changes of solid residues with temperature and
time. Figure 4.33 displays the physical features of solid residues of PP after pyrolysis in
the temperature range (400 - 600°C), which was conducted up to 30 min in most cases.
The plasticity of samples got reduced, and the sample became more brittle as the
temperature and time was increased. The lowest temperature pyrolysis was done at
400°C, and the sample was hard to break while still retaining the ductility of plastic at
109
10 and 20 min. However, as the holding time was prolonged to 30 min, the samples had
become brittle and could be ground easily to a fine powder.
PP products after 400 C pyrolysis
Polypropylene
450V I Omni 11. min N; 450T 20min II. min N; 4S0V JOmin II min N
PP products after 450 C pyrolysis
,||||
i# 1
Polypropylene Polypropylene Polypropylene
500V lOmin ILmin N, 500V 20tiun 11 min N, 500*C JOmin 11 mm N-
•
Effect of pyrolysis at 500 C
"H
: ____ j
Polypropylene Polypropylene
550V lOmin IL'min Nj 20nun 11 min Nj
Polypropylene
000‘C lOmin IL'min N>
Figure 4.33: Physical changes of PP residues according to time and temperature in the
lower temperature range from 400°C-600°C
110
The molecular weight of PP has an effect on the flexibility and brittleness of the
material at low temperature, while the orientation tends to decrease the dimensional
stability at higher temperatures. Temperature also has a complex effect on
intermolecular bonding, cross linking and copolymerization of polymeric material. The
relative abundance of the yield produced is greatly affected by the actual temperature
used, with higher temperature producing more small molecules and fewer large ones
[Kim et al., 2008; Wampler, 1989].
The changes in the colour of the samples were also notable with the increment of time
and temperature. The effect can be seen clearly at 450°C, as the powdered samples had
become darker towards higher residence times. The samples at 550°C showed white
colour for 10 minutes after which it turns to brownish black for 20 minutes pyrolysis
time. This change was caused by the progression in PP depolymerisation, as the longer
residence time has allowed more PP branches to detach from their main polymer bone,
causing samples to become more brittle and easy to break. In this regime, the kinetics of
scissoring action of depolymerisation of PP has resulted in the production of smaller
monomers towards carbonization.
34
- >\:*m
* ^' W. IP

MO'C 5min 11. min Nj 700*C 5niin IL/min Nj 750*C5minll min N;
Effect of pyrolysis of PPat 650 C, 700 C and 750 C after 5min
\ ,
1i#- 'jnnp
m tJr
^ ..........._
*■

K00*C 2min IIVminN, 850*C 2min II.'min N, X50*C 5min 1 IVinin N,
liffect of pyrolysis of PP at 800 C and 850 C after 2 and 5 min

Figure 4.34: Physical changes of PP residues according to time and temperature in the
higher temperature range from 650°C-850°C
111
The decomposition products are comprised of very complex mixtures of hydrocarbons

ranging from one carbon atom to molecules with 70 carbon atoms |Wall and Straus,
I960]. These results in general agree well with the decomposition mechanisms that
involve bond ruptures, which lead to the change of molecular weight on pyrolysis as
reported in the solid yield residue and carbon content in this chapter.
Figure 4.34 displays the effect of higher temperature pyrolysis on PP samples. As can
be seen, higher temperature pyrolysis needs shorter reaction time to cause the
degradation of PP. Similar effect which was observed after 30 min at 450°C was
observed after 5 min at 650°C. The depolymerization and degradation process of PP
was accelerated by higher temperature during pyrolysis. The decomposition of PP
appears to be complete after 850°C after 5 min, as the sample was completely charred
as displayed in Figure 4.34.
4.3.1 Analysis of Solid Residues of PP Pyrolysis

The pyrolysis results are presented in Table 4.7. The percentages of yields are plotted
against temperature and time in Figure 4.35 and Figure 4.36, respectively. Figure 4.35
shows the effect of temperature on the pyrolysis residue of PP, for low temperatures
range of pyrolysis (400 to 600°C) and high temperatures range of pyrolysis (650 -
850°C). As for lower temperatures range, holding times of 10, 20 and 30 minutes were
applied.
The yield of solid after 10 to 30 minutes pyrolysis at 400°C shows similar value,
indicating no chain had been broken and these samples were still exhibiting the initial
chain structure and form of original polypropylene. The significant loss started to occur
after 450°C for 20 min pyrolysis as more than 20% of weight had been reduced. Further
increment of temperatures decreased the weight of PP samples considerably, leaving
less than 10% of the original sample weight at 500°C and 550°C pyrolysis.
112
Table 4.7: Percentage of residue obtained after pyrolysis of PP
Temperature Time Initial weight Final weight Yield

(°C) (min) (g) (g) (wt.%)
400 10 9.60 9.58 99.80
20 9.63 9.55 99.13
30 9.66 9.45 97.80
450 10 9.47 8.86 93.59
20 9.45 6.72 71.14
30 9.52 3.15 33.10
500 10 9.52 4.17 43.85
20 9.54 1.76 18.41
30 9.60 0.54 5.62
550 10 9.42 3.54 37.55
20 9.51 0.67 7.04
600 10 9.67 2.12 21.89
650 5 9.52 2.82 29.64
700 5 9.55 2.53 26.49
750 5 9.64 2.13 22.12
800 2 9.67 4.79 49.56
850 2 9.45 2.99 31.64
850 5 9.41 0.00 0.03
temperature (°C)
O 10 min X 20 min ▲ 30 min O 5 min
Figure 4.35: Effect of temperature on solid yield of PP pyrolysis low temperature range
(400 - 600°C) and high temperature range (650 - 850°C)
The pyrolysis done at lower temperature also generated the by-products such as waxy
materials and thick smoke that condensed into white oily solid at room temperature.
Similar observations were also reported by Achillas et al, [2007]. These waxy and oily
materials consist of mainly paraffin with a carbonized char, in the case of complete
113
pyrolysis. According to their report, pyrolysis of PP at 450°C has a yield of 6.2 wt% of
gas, 67.3wt% of liquid and 26.5wt% of residue.
The solid residues obtained during this range of pyrolysis temperatures were seen to
loose their hardness and simultaneously become brittle with increasing temperature and
residence times. Figures 4.33 and Figure 4.34 show with the increase in temperature,
the colour of samples residue became darker, and these could be ground easily by
mortar and pastel, indicating the PP chains had been broken by heat due to higher
temperature. Panda et al. [2010] have stated the macromolecular structures of polymers
are broken down into smaller molecules or oligomers or sometimes monomeric units
during pyrolysis.
Low temperature pyrolysis (400 - 500°C), also known as thermolysis, is generally used
to obtain three fractions of products: high calorific value gas, condensable hydrocarbon
oil and waxes. Lower temperature pyrolysis produces relatively small gaseous fractions;
the amount of gas and liquid are dependent on the type of polymer used. The process
conditions including temperature and residence times also play major roles [Panda et
al, 2010].
As the temperature was further increased, the yield of char produced was seen to
decrease with increasing pyrolysis temperature and time. A similar trend was observed
at higher temperature pyrolysis (650 to 850°C) as shown in Figure 4.35. Shorter
residence time was applied at these higher temperature ranges as longer times tended to
completely eliminate all the residues. At higher temperatures, more drastic changes
have been detected.
The phenomenon at higher temperature can be explained in term of higher rates for the
breaking of PP particles. Temperature has a significant effect on molecular weight
changes of polymer. Increasing temperature reduces the average life-time of the longer
chains, thus higher temperature appears to favour the generation of lower molecular
weight chains [Ven, 1990]. This aspect has been taken into consideration in determining
the pyrolysis conditions.
In addition, the time period of pyrolysis was also seen to play an important role in
determining the product yield (Figure 4.36). Longer residence time also helped to
weaken the chemical bonds in polymer structure. Longer polymeric chains, which are
heavier, decomposed into smaller molecular chains, which are more volatile causing
loss in the residue’s weight. Generally the decomposition of polyolefins can be
explained by four stepwise mechanisms: initiation, depropagation, inter- or intra
molecular hydrogen transfer followed by p-scission and completed by termination
[Achiiias et ai., 2007].
Panda et ai. [2010] have suggested the thermal decomposition mode for PP is random
chain rupture and olefins are produced during lower temperature pyrolysis, while
products are gases and light oils at higher temperatures. These occurrences are also
detected during this study. Lower temperature pyrolysis has generated heavy oil,
whereas the products become lighter and more gaseous as temperatures are increased.
Thermal degradation of PP was strongly promoted by increasing temperature and time.
Time (min)
□ 400°C X 450°C A 500°C O 550°C O 850°C
Figure 4.36: Effect of time on solid yield of PP pyrolysis
The result of solid yield (%) obtained after pyrolysis of PP also has been compared to
the yield produced in the mixtures between PP and PC in Section 4.2.1. Heat treatment
of Cycle 2 (mixture samples heated to 600°C and let dwell for 30 min) had relatively
higher yield compared to the yield obtained after pyrolysis of PP alone. As can be seen
from the data (Table 4.7), only 5.62% of solid residue left after 30 min of 500°C
pyrolysis, while up to 8.46% (30% PP mixture) of residues were observed even after
115
being heated up and dwelled for 30min at 600°C in Cycle 2 (Table 4.3). This result
indicates that the presence of PC helped contain PP and thereby yielding higher
residues.
4.3.2 Chemical Analysis of PP Pyrolysis Residues

The carbon content of the residues of PP pyrolysis was determined by LECO SC 444
analyzer and presented in Table 4.8. The data is then plotted as a function of
temperature in Figure 4.37. The percentage of carbon content was obtained from each
of the pyrolysis residues (as in Figure 4.33 - 4.34). Since the residue comprises of
incomplete decomposition of PP, it contains hydrocarbon polymeric chain, which was
accounted for in carbon determination via LECO analysis. The carbon percentages of
residues reduced moderately with an increase in temperature. However, the carbon
content of residues has significantly reduced after pyrolysis at 850°C for 5 min, where
at this stage PP has completely charred and lost all its volatile matter.
Table 4.8: Effect of temperature and time on carbon content (%) of PP residues
Temperature (°C) Time (min) Carbon content (%)

400 10 74.5
20 72.7
30 71.8
450 10 73.7
20 68.6
30 66.1
500 10 71.5
20 67.2
30 61.1
550 10 70.4
20 66.4
600 10 68.7
650 5 70.5
700 5 66.8
750 5 65.4
800 2 70.4
850 2 67.7
5 10.6
116
80
2 75
5
c 70
o
5
-E
« 65
CJ
60
400 450 500 550 600 650 700 750 800 850
Time (min)
♦ 10 min □ 20 min ▲ 30 mm X 2 min O 5 min
Figure 4.37: Effect of temperatures on carbon content (%) of PP residues in the low
temperature range (400 - 600°C) and high temperature range (650 - 850°C)
The effect of time on the carbon content of PP residues is presented in Figure 4.38.
Increasing time from 10 to 30 minutes has noticeably decreased the carbon content,
which was expected as the solid yield of residue also decreased with increasing time.
This simultaneously decreased the amount of carbon measured. From the data, 74.5%
of carbon content in 400°C for 10 min pyrolysis has reduced to 71.8% after 30 min. The
percentage of reduction had increased at elevated temperature, such as 71.5% of carbon
at 500°C for 10 min, but the amount was reduced to 61.1% after 30 min.
Time (min)
□ 400°C X 450°C A 500°C O 550°C O 850°C
Figure 4.38: Effect of time on carbon content (%) of PP residues
The carbon content of overall residues was in the range 61.1 to 74.5%, at lower
temperatures. The pyrolysis done at 850°C caused no significant change to carbon
content after 2 min, due to the short residence time; but the residue sample obtained
117
after 850°C was charred after 5 min where only 10.0% carbon was obtained, while the
rest were incomplete pyrolyzed hydrocarbon and ashes. This amount of carbon in the
charred sample is possibly the amount of fixed carbon left in the pyrolysed sample after
complete devolatization.
The samples (residues in Figure 4.33 - 4.34) were then devolatized and ashed before
being analysed by XRF to determine their ash content. The corresponding data is
presented in Table 4.8. The composition and stereo-specificity of polymer can also have
a major role in altering the decomposition rates and kinetics, thus varying the activation
energy of polymers. Besides that, the molecular weight, and method of measuring the
weight loss behaviour and methodology can also influence the quantitative values of
kinetic parameters [Yang et al, 2001]. Table 4.9 shows the main compounds present in
PP ash, for each residue obtained from the pyrolysis after being completely devolatized.
The ash content calculated during ashing of PP residues is plotted in Figure 4.39.
Table 4.9: Ash content (%) of solid residue of PP
Pyrolysis conditions Ash (%) Total

Temperature Time ash
Fe2C>3 Si02 AI2O3 MgO Others
(°C) (min) (%)
Raw PP - 0.09 0.01 0.12 0.05 0.04 0.32
400 30 0.09 0.06 0.28 0.03 0.05 0.42
450 10 0.10 0.11 0.24 0.05 0.01 0.54
20 0.11 0.14 0.33 0.04 0.00 0.64
30 0.10 0.09 0.36 0.14 0.10 0.81
500 10 0.08 0.02 0.22 0.04 0.23 0.61
20 0.11 0.10 0.49 0.73 0.06 1.51
30 0.16 0.12 0.27 0.41 0.31 1.30
550 10 0.17 0.09 0.31 0.07 0.10 0.75
20 0.13 0.11 0.31 0.60 0.73 1.90
600 10 0.10 0.00 0.12 0.00 0.70 0.94
650 5 0.0 0.34 0.42 0.09 0.70 0.23
700 5 0.08 0.05 0.28 0.09 0.00 0.52
750 5 0.08 0.35 0.71 0.08 0.04 1.19
800 2 0.09 0.16 0.33 0.13 0.08 0.82
850 2 0.09 0.70 0.33 0.13 0.26 1.53
5 0.15 0.75 0.37 0.14 0.45 1.87
118
Generally, PP is composed of a stabilizer, catalyst, fire retardant and colorant, that

impart impurities into PP. Aluminium is present in PP and acts as a catalyst, and
contributes towards controlling the molecular weight distributions. A1 and Ti (also
present in small amounts is included in category “others”) are catalysts used to modify
the molecular weight by either changing sites in reactivity (propagation/termination
rate) or increasing level of chain transfer agent. The presence of Mg as MgCl as support
catalyst showing just how rapid the rates of propagation and termination are in
propylene polymerization [Ven, 1990].
O'
I
0.0
♦ 10 min X 20 min O 30 min ▲ 2 min □ 5 min
Figure 4.39: Ash content (%) of product of PP at low temperature range (400 - 600°C)
and high temperature range (650 - 850°C)
The ash content of the residues increased as pyrolysis temperature increased. Raw PP
has a 0.326% of ash content, and the value increased to 0.425% after 400°C. The
magnitude of ash content was raised with temperature and time, as can be seen in
pyrolysis at 450°C. At 450°C for 10, 20 and 30 min pyrolysis has resulted in 0.54, 0.64
and 0.81% of ash, respectively. Apart from the time factor, temperature also has an
impact in increasing the ash content of PP residues. Residues of 450, 500, 550 and
600°C for 10 min pyrolysis have ash contents of 0.54, 0.61, 0.75 and 0.94%,
respectively. The increasing in ash content with increasing of temperature was expected
as the increasing temperature increases the burnt-off of polymeric chains, which
enhanced the devolatization activity leaving behind oxides in the residues.
According the data reported in Table 4.8 and Table 4.9, the total values of carbon and
ash only accounted for around 60 - 75% of mass of pyrolysis residues (from 400°C/10
min pyrolysis up to 850°C/2min pyrolysis). The highest temperature and time (at 850°C
119
and 5 min) of pyrolysis resulted in approximately 12% carbon and ash of its mass solid
residue. This value is expected as PP polymer decomposes into short polymer chains
that may contain dienes, alkanes and major products of alkenes which vary as light
liquid to gaseous form jBockhorn et al., 1999-aj. The physical changes as observed in
Figure 4.33 and Figure 4.34 are important as these can significantly influence the
chemical decomposition behaviour.
Temperature has played a key role in affecting the physical form of residues. The
principal physical change is the transformation from a glass or solid to a fluid state, and
further on to a gaseous phase. If heated at temperatures below the decomposition
temperature, it becomes more likely that the product will drip and/or flow [National
Fire and Society of Fire, 1995J. This phenomenon can explain the high yield residues
collected at lower temperature pyrolysis. Thermoplastic can melt without undergoing a
chemical reaction to form a viscous state (polymer melt). A representative example of
the type of chemical and/or physical changes that occur during the heating of
thermoplastics is presented in Figure 4.40.
Melting
Solid _______ „ Liquid * Gas
Thermoplastic Char
...............► Physical change

----------- ► Physical/Chemical change
Figure 4.40: Physical and chemical changes during thermal decomposition

[National Fire and Society of Fire, 1995]
As observed in this pyrolysis, PP melt decomposes into smaller liquid fragments, and
then the liquid fragments decompose further until they are too, are sufficiently volatile
to vaporize. This process evolved with increasing temperature. During the pyrolysis, it
was noted that apart from the residue collected in the crucible, a liquid-form substance
was observed coming out from the furnace, and some sticky stuffs deposited in on the
furnace wall. Bockhorn et al. [1999-aj in their PP decomposition studies reported that
the key product yields at 410 - 460°C up to 30 minutes pyrolysis are alkenes, alkane
and dienes (Cg-Cn) that are present in semi-solid and liquid form .
120
The pyrolysis at elevated temperatures (>700°C) favours the production of highly

gaseous compounds, that consists mainly of ethene, propene and C4 olefins (up to about
53 wt.%) [Kaminsky et al., 1995]. Moreover, the rate of physical and chemical
conversion of polymer is significantly influenced by temperature and time, where higher
temperature and time promote higher conversion [Bockhorn et al., 1999-b]. These two
phenomena are in agreement with the observations in the 850°C pyrolysis. During
850°C pyrolysis, white smoke was observed coming from the furnace through the
exhaust system in the first two minutes of experiment, and then it reduced gradually and
stopped towards the end of the experiment (5 min). As the rate of conversion is very
high at this high temperature (850°C), PP had rapidly decomposed into gaseous
products (observed smoke) that had been purged out by argon, and leaving only
0.03wt.% of solid residue.
4.3.3 Morphological Study of PP Pyrolysis Residues
(a) (b)
Figure 4.41: SEM micrographs of raw PP samples (2mm ground particles) at low
magnification and (b) high magnification
The surface morphology of raw PP is presented in Figure 4.41 (a) and (b) in form of
ground particle and its surface structure, respectively. The glossy and smooth plasticity
at the surface of raw PP is one of the typical characteristics of its morphology. Figure
4.42 represent SEM images on the residues of PP after pyrolysis with time and
temperature. At low temperature such as 400°C, the surface of PP residue remained
similar to its initial plastic-state, but the breakage started to occur after 30 minutes of
pyrolysis. Under these conditions, the particles were breaking from each other. Similar
121
behaviour was observed for residues of 450°C after 10 min. However, at this stage,
some of the PP was still intact while other parts started to break apart.
With further increases in time and temperature, the physical structure of PP had
significantly deteriorated and the particle sizes became smaller. As the temperature was
further increased, this effect was seen to become more significant. The sizes of the
particles were relatively smaller and the separation became obvious as the time of the
pyrolysis was increased. These phenomena were observed clearly at 850°C, where
particle size of residue was reduced to 1.53|am after 5 min of pyrolysis. The estimated
particle size of residues was measured and has been summarized in Table 4.10.
Table 4.10: Estimated calculated particle size (agglomerated particles) of PP pyrolysis

residues
Temperature (°C) Time (min) Particle size (pm)

400 10 NA
20 NA
30 15.28
450 10 NA
20 10.73
30 9.42
500 10 11.12
20 10.18
30 9.25
550 10 9.78
20 7.48
600 10 9.17
650 5 32.07
700 5 32.61
750 5 32.13
800 2 NA
850 2 NA
850 5 1.53
The residue samples obtained after 5 min pyrolysis, at temperatures of 650, 700 and
750°C were similar in appearance. It can be assumed in this duration of time, influence
of temperature within this range was similar. As referring to Figure 4.34, the
experiment conducted only for 5 min in these temperatures showed powdery yet sticky
122
residues. This indicates that the polymer has melted and some volatiles were able to
rapidly escape during this period of time. The remaining samples turned into brittle but
sticky particles.
Identical appearance of residues obtained at 800 and 850°C for 2 minutes suggests
similar occurrence compared with other samples carried out in short time pyrolysis.
However, after heating up to 5 min of pyrolysis had transformed both the sample
surface and structure. The size of particles had dramatically reduced and the surface
features observed after 5 min has changed significantly. The sample was found to be
charred at this stage [Figure 4.34].
The increased time has contributed to the physical and morphological changes of
pyrolysis residues. However, times less than 5 minutes in the temperature range from
650 to 750°C and 2 minutes for temperature range of 800 to 850°C caused no major
changes in the samples. An abrupt change was seen, where PP charring happened soon
after 5 min time. It is believed that the heat transfer in PP being low delayed the
occurrence of significant changes.
Kim et al. [2008| have reported that during isothermal decomposition kinetics, PP
experienced a retardation period in the initial stages of reaction where the temperature
of PP samples might increase to the set temperature. The retardation period is likely to
be developed before PP reaches a critical temperature and begins to melt. It is also
expected that PP may melt to yield viscous liquids during this retardation period. This
behaviour may explain why PP had not decomposed during 2 min of pyrolysis in both
800 and 850°C pyrolysis.
PP may have experienced a similar phenomenon, where at this stage (2 min period),
retardation had taken place. However, the critical temperature was reached after 5 min
period, thus PP vigorously devolatized leaving behind a char residue. It is interesting to
note that the residue obtained at 550°C pyrolysis after 30 min reaction has a similar
morphology to the residue obtained at 850°C pyrolysis, after only 5 min. This
observation indicates that higher temperature reactions require much shorter time for PP
to degrade, while lower temperature requires much longer time for the similar result.
123
It is quite likely that the kinetic degradation of PP could have caused such behaviour,
where the rate of reaction varied with temperature. At lower temperatures, volatiles
produced in melt and the fragment sizes of major pyrolysis products were larger than
the minimum molecules that are able to vaporize; and the evolution of volatiles
accomplished predominantly through the surface of the melt. Thus it would require
longer time to undergo secondary reactions and eventually evaporate. On the other
hand, higher temperature enables bubble nucleation as a mechanism of devolatization.
The rapid nucleation of bubbles results in a critical concentration, and once critical
temperature is reached; the highest molecular size is small enough to be vaporized, thus
triggering the explosive generation of volatiles. This reaction is very vigorous and rapid
devolatization caused charring of PP to happen quickly [Kim et al., 2005].
Alternative mechanism could be that PP degrades by random scission reaction; higher

temperature reaction favours the degradation of larger molecules, to form more of the
smaller products in a short period of time (Wampler, 1989]. Thermal conductivity on
the other hand does not have a significant effect on the increasing temperature since PP
has low segmental mobility due to complex polymer chains. Sato and his co-workers
(2004] have reported the average flow velocities of 0.5-2.7 mm/sec flow and
temperatures of 200-240°C, the heat transfer coefficient of PP is 160-220 W/m°C. Heat
transfer coefficient of PP does not show significant temperature dependence, thereby
indicating that the increasing temperature had no significant effect on the heat transfer
ofPP.
124
400°C Pyrolysis
igure 4.42: SEM micrographs of PP residues after pyrolysis according to time and temper;
500°C Pyrolysis
igure 4.42: SEM micrographs of PP residues after pyrolysis according to time and temperature (cont.)
600°C Pyrolysis
Figure 4.42:
800°C Pyrolysis
4.3.4 Structural Characterization of PP Pyrolysis Residues

The XRD patterns of PP residues are presented in Figure 4.43, showing the effect of
temperature on the crystal structure of PP. The peak parameters and crystal planes of PP
residues were recorded in Table 4.11.
Table 4.11: Effect of temperature on XRD peak parameters after 650°C to 850°C for 5
min pyrolysis*
Miller indices Possible

Peak
Temperature d-spacing compound
Sample position
(°C)
(0°)
(A) h k 1 (chemical
formula)
Raw PP - 14.13 6.27 1 1 0 (C3H6)n
16.80 5.27 0 0 8 (C3H6)x
18.49 4.79 1 3 0 (C3H6)n
21.82 4.07 0 2 6 (C3H6)x
25.45 3.50 1 5 0 (C3H6)n
28.52 3.13 0 1 2 (C3H6)n
42.52 2.11 1 0 0 C
Residues 650 14.06 6.30 1 1 0 (C3H6)n
16.84 5.26 0 0 8 (C3H6)x
18.56 4.77 1 3 0 (C3H6)n
19.85 4.47 1 1 7 (C3H6)x
21.05 4.22 2 0 2 (C3H6)x
21.82 4.07 0 2 6 (C3H6)x
25.34 3.51 1 5 0 (C3H6)n
28.35 3.15 0 1 2 (C3H6)n
42.47 2.11 1 0 0 c
750 14.03 6.31 1 1 0 (C3H6)n
16.82 5.27 0 0 8 (C3H6)x
18.53 4.88 1 3 0 (C3H6)n
20.02 4.44 1 1 7 (C3H6)x
21.11 4.21 2 0 2 (C3H6)x
21.81 4.07 0 2 6 (C3H6)x
25.29 3.52 1 5 0 (C3H6)n
28.42 3.14 0 1 2 (C3H6)n
42.63 2.11 1 0 0 C
850 42.39 2.13 - - - Copper
129
50000
Raw PP
40000
30000
20000
10000
50000
40000
30000
20000
10000
2
I
8
50000
1 5 O 750 X'
40000
91 i
,
30000
20000
4 i
\J O
10000
■J HA
Figure 4.43: Effect of temperature on yield samples of PP pyrolysis. XRD spectra of

samples at 650°C to 850°C for 5 min pyrolysis compared to raw PP
130
PP shows strong diffraction peaks at 14.13, 16.80, 18.49 and 21.82°. Peaks at 14.13 and
18.49° correspond to (110) and (130) diffraction planes of a monoclinic crystal
configuration, while diffraction peaks at 16.80 and 21.82° correspond to (008) and (026)
diffraction planes of the orthorhombic crystal. Weak peaks at (150) and (012) are planes
of monoclinic phase detected at 25.45 and 28.52° [Meille et al, 1990].
PP subjected to heating at 650°C for 5min has slightly changed the orientation of crystal
structure, where new peaks appeared at 19.85 (weak) and 21.05°, which represents
(117) plane of orthorhombic and (202) plane of orthorhombic, respectively. The (202)
plane of orthorhombic generated as division from (026) plane structure. Weak
diffraction peaks of (150) and (012) monoclinic planes were also observed after 5 min at
650°C.
The increase of pyrolysis temperature to 750°C again has showed similar crystalline
structures to the sample obtained after 650°C. The small weak peak (117) plane of
orthorhombic phase and (202) plane of orthorhombic remain at 20.02° and 21.11°.
Similar results were observed for diffraction of (150) and (012), where both of these
peaks were seen at this stage. Pyrolysis carried out at 850°C had completely destroyed
the crystalline phase of PP, leaving amorphous structure in the residue. Temperature had
a significant effect on altering the crystal structure of PP. The small trace of Cu around
42° was found in each pyrolysis residue was picked up from copper crucible used
during pyrolysis.
The effect of time on the crystalline structure of solid residues of PP is presented in

Figure 4.44 and the data has been summarized in Table 4.12. After 10 min of pyrolysis,
PP residues still has strong PP peaks at 14.13, 16.80 and 18.49°, respectively, denoted
by (110), (008) and (130) planes. However, crystallinity rearrangement due to heat
treatment has resulted in new peaks at 20.97° corresponding to (117) plane of
orthorhombic phase structures [Xie et al., 2002]. The increase of pyrolysis time from 10
to 20 min altered the peak profile of PP residue. The intensity of [130] peak got
significantly reduced, while the (117) became sharper and more crystalline. The new
peak at 21.28 signifying (202) was emerged after 20 min of heating. This peak (202) is
131
highly crystalline and was located on the shoulder of (026) at 21.95°, and both peaks
belong to an orthorhombic structure.
Table 4.12: Effect of time on XRD peak parameters for 10 min to 30 min at 500°C*
Miller Indices Possible

Peak
Time d-spacing compound
Sample position
(min)
(0°)
(A) h k 1 (chemical
formula)
Raw PP - 14.13 6.27 1 1 0 (C3H6)n
16.80 5.27 0 0 8 (C3H6)X
18.49 4.79 1 0 (C3H6)n
21.82 4.07 0 2 6 (C3H6)x
25.45 3.50 1 5 0 (C3H6)n
28.52 3.13 0 1 2 (C3H6)n
42.52 2.11 1 0 0 c
Residues 10 13.99 6.33 1 1 0 (C6H]2)n
16.83 5.27 0 0 8 (C3H6)x
18.51 4.79 1 3 0 (C3H6)n
20.97 4.24 1 1 7 (C3H6)x
21.78 4.08 0 2 6 (C3H6)x
25.49 3.49 1 5 0 (C3H6)n
28.37 3.15 0 1 2 (C3H6)n
42.50 2.11 1 0 0 c
20 13.79 6.42 1 1 0 (C6H12)n
15.09 5.87 1 1 3 (C3H6)x
16.91 5.24 0 0 8 (C3H6)x
18.58 4.77 1 3 0 (C3H6)n
20.09 4.42 1 1 7 (C3H6)X
21.28 4.17 2 0 2 (C3H6)x
21.95 4.05 0 2 6 (C3H6)x
25.45 3.50 1 5 0 (C3H6)n
28.94 3.08 1 3 1 (C3H6)x
42.43 2.11 1 0 0 c
30 14.07 6.29 1 1 0 (C3H6)n
16.79 5.28 0 0 8 (C3H6)x
18.54 4.78 1 3 0 (C3H6)n
19.91 4.46 1 1 7 (C3H6)X
21.10 4.21 2 0 2 (C3H6)x
21.84 4.07 0 2 6 (C3H6)x
25.25 3.53 1 5 0 (C3H6)n
28.68 3.11 0 1 2 (C3H6)n
42.45 2.11 1 0 0 c
132
50000
40000
30000
20000
10000
50000
40000
30000
20000
| 10000
g
>.
1 50000
3
20 min
40000
30000
20000
10000
50000
40000
30000
20000
10000
0
10 20 30 40 50
Diffraction angle (20)
Figure 4.44: Effect of time on yield samples of PP pyrolysis. XRD spectra of samples
for 10 to 30 min at 500 °C compared to raw PP
133
There were no significant changes detected between XRD pattern of 20 and 30 min in
residues, except the change in the relative intensities of (130), which was increased,
while (117) became less after 30min. The planes of (150) and (012) remained
unchanged throughout the pyrolysis up to 30 min.
As discussed earlier, it has became quite clear that specific details including the
formation of gas, liquid and solid phases play a critical role in determining the kinetics
of reaction and pyroiysis behaviour during charring solids. It has also been reported that
different volatile products are released in different temperature ranges. Production of tar
is dominant for a significant part of the pyrolysis, so that prediction of total mass loss
would allow prediction of tar release rates [Milosavljevic and Suuberg, 1995].
4.3.5 FT1R Reactions Characterization of PP Residues

The structural features observed in the degradation of PP were further investigated by
FTIR analysis. Figure 4.45 and Table 4.13 represent the FTIR spectra of PP residues
and its peak configuration, respectively.
Table 4.13: Effect of temperature on the peak assignment of PP residues at 650°C and
750°C for 5 min
Raw PP After 650°C After 750°C Possible Functional Groups

3473.35 3477.85 3471.42 Moisture
- 2950.96 2951.42 CH3 stretching
2917.04 2919.03 2918.82 CH2 stretching
- 2863.79 2863.79 CH3 stretching
- 2839.32 2839.89 CH2 symmetric
1459.14 1455.82 1455.76 CH3 bending
1377.05 1376.51 1376.48 C-CH3 bending
1167.92 1167.97 1167.20 C-C stretching
- 1067.97 1067.20 C—CH3 bending
997.33 972.84 997.71 C=C Alkenes
- 841.15 840.22 Aromatic ring
Jhc piajor absorption bands were selected in Figure 4.45 and display^ in "fable 443.
The FTIR peak vibrations of PP and its residue after 650 and 750°C for 5 min pyrolysis
were observed at similar positions, but different in their intensities. For example, CH3
stretching, positioned around 2950 and 2860 cm'1 and CH2 asymmetric and CFf
134
symmetric around 2919 and 2840 cm'1, respectively were four distinctive peaks of PP
and residues, which increased significantly with increasing temperature.
Raw PP
" g
w V a
I
3500 3000 2500 2000 1500 1000 500

Wavenumbers (cm1)
Figure 4.45: Effect of temperature on residues of PP pyrolysis. FTIR band of samples

after 650°C and 750°C for 5 min pyrolysis compared to raw PP
Decomposition of PP also generated several new peaks such as C-CH3 bending and
aromatic compounds around 1067 cm'1 and 840 cm'1, respectively, besides increasing
the intensity of peaks of CH3, C-CH3 and C-C stretching which were quite low in the
raw PP sample. These new peaks indicates that PP has polymer chain had degraded and
135
broken into smaller chains that contains of these aromatic compounds and also
corresponding bonds as many smaller chains have been formed at higher temperature.
The results showed only small changes in the absorption bands at elevated temperature.
In this case, pyrolysis duration also has a considerable impact as 5 minutes was too
short for a complete decomposition of the sample. The effect of time on the
decomposition of PP was reported by Yasushi and Kaoru [2003], indicating 30 min
was not sufficient to initiate thermal cracking of PP polymers at 400°C.
The effect of time of the FT1R absorption is discussed based on results presented in
Figure 4.46 and Table 4.14. Greater the pyrolysis time, more the number of products
generated. This is because molecular chains were able to undergo the secondary
reactions and produce lower molecular chain products [Milosavljevic and Suuberg,
1995]. After 10 min pyrolysis, the intensities of CH3 stretching at 2950 and 2860 cm1,
and CH2 asymmetric and CFh symmetric around 2919 and 2840 cm"1, correspondingly,
were increased. However, these peaks decreased as pyrolysis was prolonged to 20 min.
Pyrolysis for 10 and 20 min also generated several new peaks of C=C=C stretching
around 1720cm'1, CH3 bending at 1461cm'1, CH bending around 1280cm'1, C-CH3
around 1067cm'1 and aromatic rings around 841cm"1. In addition, alkene absorption
band and aromatic band were detected only after 20 min pyrolysis, showing the effect of
secondary reaction has taken place [Gambiroza and Cunko, 1992]. The formations of
double bonds (C=C) for alkene compounds are considered as stabilization reaction at
the end of chain scission reaction [William and Slaney, 2007]. This reduction may
have been caused by the depolymerization of PP chains into volatiles which escaped as
output gaseous products.
136
Raw PP
gj 10 min
20 min
1 1 IJx
3500 3000 2500 2000 1500 1000 500

Wavenumbers (cm"1)
Figure 4.46: Effect of time on yield samples of PP pyrolysis. FTIR band of samples for
10 min and 20 min at 550°C pyrolysis compared to raw PP
137
Table 4.14: Effect of time on the peak assignment of PP residue for 10 min and 20 min
at 550°C
Raw PP 10 min 20 min Possible Functional Groups

3473.35 3471.60 34718.32 Moisture
- 2951.60 2951.32 CH3 stretching
2917.04 2920.68 2919.56 CH2 stretching
- 2866.83 2869.87 CH3 stretching
- 2838.93 2848.59 CH2 symmetric
- 1730.94 1723.75 Asymmetric C=C=C stretching
- - 1550.97 C=C (alkene absorption)
- - 1542.10 C=C (alkene absorption)
1459.14 1461.49 - CH3 bending
- - 1410.53 CH stretching
1377.05 1376.63 1388.21 C-CH3 bending
- 1285.90 1282.93 CH bending
1169.22 1167.15 - C-C stretching
- 1067.15 1060.64 C-CH3 bending
973.18 972.97 - C=C Alkenes
- 841.10 860.08 Aromatic ring
- - 757.19 Aromatic ring
4.4 Summary
An in depth investigation on PP and its blends with PC has been presented in this
chapter. The primary analysis was carried out by investigating the wettability and
interfacial phenomena of PP and PC, where the increasing time and temperature (200 to
350°C) resulted in an improved wettability and penetration of PP melt into PC substrate.
The highest penetration depth was obtained at 350°C temperature after 60 min of heat
treatment with a penetration depth of 73pm. The formation of gaseous bubbles and
blisters indicate devolatizatilization of PP at these low temperatures. The flow of melt
into PC particles had limited the heat transfer and decomposition of PP.
The influence of PC on the pyrolysis yield from PP was investigated in three heating
cycles. PP was mixed with PC in the concentration range of 10 to 90 wt% in 10% steps.
During Cycle 1 (up to 150°C), no reaction was observed probably due to low
temperatures, as PP begins to melt only around 160-170°C. Reaction started to take
place during Cycle 2 (up to 600°C), where the extent of solid residue was in proportion
138
to the PC% content in the blends; some traces of PP were observed in most of the
sample blends. The data from the heat treatment of Cycle 3 (up to 1000°C) shows that
some traces of PP were still remaining in composition up to 50%. No PP was observed
in higher PP mixtures. The TGA profile on 90% PC: 10% PP mixture indicates that PP
started to decompose around 280°C till 430°C, and the weight loss became a plateau
around 9-12% of weight residue.
As C and H are the key constituents of PP and PC, high carbon content (89-97%) was
observed in carbon content analysis. The residues of the mixture were also analyzed by
XRD to determine structural changes if any. The increase of temperature in Cycle 2 heat
treatment resulted in the loss of PP peaks, and only PC peaks were visible. Similar
results were observed after Cycle 3. These results indicate that PP did not survive in its
original form after heat cycles 2 and 3.
The FTIR results showed decreasing intensity of CH2 stretching and a missing aromatic
peak with increasing temperature (Cycle l<Cycle 2<Cycle 3); it is likely to be caused
by the decomposition due to high temperatures. The SEM images have shown that while
some of PP still remained after Cycle 2 but it had completely decomposed after Cycle 3.
Pyrolysis of PP was carried out to understand the effect of time and temperature on PP
degradation. The pyrolysis residue color changed from white to brown and became
much darker with increasing time and temperature, until eventually it became a black
powder after 850°C and 5 min of pyrolysis, thereby indicating PP had completely
decomposed and charred. PP started to loose around 30% of its mass after 20 min in
450°C pyrolysis, and then this loss increased to 70% after 30min. The increasing time
and temperature had enhanced the decomposition rate of PP, where PP lost about 95%
of its mass after 30min of 500°C pyrolysis.
The pyrolysis at 850°C was complete in 5 min producing a charred PP with a yield of
only about 0.03%. The carbon content obtained from the residues measured the C from
hydrocarbon chain of PP. The carbon percentages reduced moderately with an increase
of temperature and time, and the lowest carbon content was detected in residue after
139
850°C at 5 min pyrolysis with only 10% carbon. Four main ash impurities found in the
residues were Fe2C>3, SiC>2, AI2O3 and MgO.
The pyrolysis yield from PP alone was somewhat lower than the yield from PP and PC
blends after the heat treatment. As the data of Cycle 2 (samples heated up to 600°C and
let dwelled for 30 min) being compared to pyrolysis of PP at 500°C after 30 min (the
closest conditions), the mixture samples have showed a 8.5% of solid residues (30% PP)
while pyrolysis of PP alone resulted in about 5.6% of solid residues. This value
indicates that PC has helped retaining the product of PP decomposition in the mixture,
as PC particles may suppress the fluid phase from getting evaporated. More solid
residues obtained means higher carbon can be attained in heat treated coke/polymer
blends.
Increasing time and temperature of pyrolysis also had a significant effect on the
morphology of PP, where the PP particles were broken down into smaller sizes, and
agglomerated as PP has a tendency of sticking together. The charred residues after
850°C at 5 min pyrolysis had produced 1.53pm powder particles. The pyrolysis also
changed the crystal structure of PP, where the heating generated additional peaks at
(117) and (202), both are orthorhombic as noted in 650°C and 750°C pyrolysis. Both
peaks also can be observed as a function of time, where in 500°C pyrolysis, they
emerged after 10 to 30 min of pyrolysis. However, the increasing temperature to 850°C
completely destroyed crystalline PP peaks. The FTIR analysis showed a small increase
in the intensity of several functional group peaks as pyrolysis time and temperature
were increased. This was attributed to the generation of several smaller molecules
containing associated groups.
140
Chapter 5: Results & Discussion - Interactions of PP with PC and Degradation Mechanisms
CHAPTER 5

INTERACTIONS OF POLYPROPYLENE
WITH PETROLEUM COKE AND
DEGRADATION MECHANISMS
141
According to thermal degradation study of PP reported in Chapter 4, PP decomposed

and charred after being exposed for 30 min at 500°C, and similar result was observed
after 5 min heating at 850°C. These results indicate that the decomposition of PP is
highly influenced by temperature and time. This characteristic has reflexes on the
polymer interaction behaviour with petroleum coke. In Chapter 4, the wettability of PP
and petroleum coke was only observed across their interfacial/contact area, while the
effect of heat treatment cycles on the PP and petroleum coke mixtures provided
qualitative information. On the other hand, the nature of bonding was yet unclear.
Therefore in this chapter, PP was soaked with petroleum coke at low temperature before
subjected to higher temperature treatment. The main purpose of this experiment is to
determine the effect of low temperature soaking and higher temperature heating on the
pressed sample of PP and petroleum coke. FTIR study was carried out to qualitatively
define the nature of bonding occurred between PP and petroleum coke during the heat
treatment. The degradation mechanisms were also investigated to determine the effect
of different heating rates on PP and identify its reaction kinetics.
5.1 Effect of Low Temperature Treatment on the Pyrolysis

Yield of PP:PC Blends
In this section, substrates of PP mixed in the proportion of 10, 20 and 30 wt% with
petroleum coke and 5 wt% of binder were subjected to soaking carried out at
temperatures ranging from 200°C to 350°C for 30 min, before heat treatment at 600°C
for 30 min.
5.1.1 Weight Loss after Low Temperature Soaking

Figure 5.1 shows the weight loss of PC substrate samples (10, 20 and 30% PP) after
soaking treatment conducted at 200, 250, 300 and 350°C. All samples displayed almost
negligible weight loss during the soaking treatment. The result indicates that the PP has
been confined between PC particles and glued by the binder in the substrate.
142
Temperature (°C)
Figure 5.1: Weight loss (%) of 10, 20 and 30% PP in PC blend substrates after soaking
for 30 min under 0.5 L/min argon flow at 200°C, 250°C, 300°C and 350°C
10% 20% 30%

Figure 5.2: Effect of 30 min soaking at 250°C on the specimen morphology
Figure 5.2 displays the samples soaked at 250°C for 30 min under 0.5 L/min argon
flow. As can be observed, samples remained fairly intact in substrates form after the
soaking. However, 10% sample was more intact as compared to 20% PP, while 30%
sample showed some degradation. This is because 10% sample contained smallest
amount of PP that minimized the impact of devolatilization. Even though the weight
loss was small in these three samples, 30% PP has relatively highest loss compared to
10% and 20% PP mixtures. Each of the substrate samples at every stage showed similar
phenomenon as displayed in Figure 5.2.
143
5.1.2 Heat Treatment to 600°C on Soaked Samples
10% PP 20% PP 30% PP
Temperature (°C)
Figure 5.3: Weight loss (%) of 10, 20 and 30% PP in PC blend substrates after up to
600°C heat treatment on samples soaked at 200°C, 250°C, 300°C and 350°C
The soaked samples were then charged into the furnace for heat treatment up to 600°C
and the weight loss was calculated and plotted in Figure 5.3. The loss was varied and
increased with increasing PP % in the sample and also the soaking temperatures. The
percentages of loss relative to the PP content in the blends, where almost 11%, up to
23% and almost 30% of wt loss was determined in 10%, 20% and 30% PP samples
respectively. Assuming the loss was caused by the loss of polymer, however, 20%
samples showed loss more than 20%. This increasing loss may be affected by the
binder, where it started to redistribute and impregnate PC particles, and at the same time
started to release its light volatiles. In carbon anode manufacturing, pitch has been used
as a binder as it easily wets the carbon surface, penetrates the micropores of coke and
glues them together in unbaked anodes [Ehrburger and Lahaye, 1984; Sadler and
Welch, 2001J.
10% 20% 30%

Figure 5.4: Effect of heat treatment up to 600°C on soaked substrate samples
144
It is however interesting to note that the sample containing 30% PP, nearly 2% polymer
was still remaining for soaking temperature up to 300°C, but it was lost after 350°C
soaking. This composition showed a trend different from the other two compositions.
Figure 5.4 shows the effect of 600°C heat treatment on the substrate samples (soaked at
300°C). The substrate with highest PP content (30%) has been most degraded as
compared to samples with 10% and 20% PP (10% PP <20% PP <30% PP).
5.2 Nature of Bonding between Polypropylene and

Petroleum Coke
As discussed in Chapter 4, the decomposition of PP polymer is highly influenced by
temperature and time of the pyrolysis. This section investigates the nature of bonding if
any, formed during the low temperature heating from 200°C to 350°C treated for 15, 30
and 60 min on the substrate containing 10% PP blended with PC, without any binder.
120
- Raw PP
100
V yf
- /
80 -
if szL .£
C=C-
\ / cij o 1 OiJ
Ol] c
60 J^
40
■ •
:
•
:
V--- y--- 1
*13
2 i£
o •S =3
1I 1
|
<L>
O Moisture
tS
£ Cl " X
CJ
1
G 5
20 U Cl ^ u u
B u u1
’p 1 1 —1_____
£ 120
cs Raw PC
£ 100
£
80
60
40
20
Moisture
0
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm1)
Figure 5.5: FTIR profiles of raw PP and PC
145
Figure 5.5 shows the FTIR profiles of raw PP and PC while the following Figure 5.6 to
Figure 5.8 show the effect of temperature (250°C, 300°C and 350°C) and time (15 min,
30 min and 60 min) on the chemical bonding between PP and PC. The list of peaks and
their corresponding possible functional groups of each sample have been listed and
compared with the raw PP and PC in Table 5.1 to Table 5.3.
120
After 15 min
100
J
f
80
'j
lil* vu |
60 .E l CJ
c -C
■S
£
1/
JD ~
r*~,
t/S
40 £O
cj I CJ
20 CJ
120
After 30 min
0 100 :
r—
1 80 ; \x/ Cl)
| 60 : ^: O 5 %
.O
£ V---- y---- J u r*',
O
f— 40 Moisture
CJ
C/5
- CJ
u u
ii i i i i 1 1 1 1 ■
120
100
80
60
40
20
0
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm"1)
Figure 5.6: FTIR profiles of 10% PP in PC substrate blend after 250°C heat treatment
146
Table 5.1: List of peak assignments of 10% PP in PC substrate blend after 250°C heat
treatment
Raw PC Raw PP 15 min 30 min 60 min Possible Functional Groups

3476.76 3472.45 3465.41 3466.55 3466.55 Moisture
- 2960.58 2960.06 - - CH3 stretching
- 2921.70 2926.37 - - CH2 stretching
- 2841.32 - - - CH2 symmetric
- 1460.01 1463.70 - - CH3 bending
1382.89 1378.86 1379.45 1383.38 1383.36 C-CH3 bending
1163.24 1165.06 1164.33 1168.50 1163.70 C-C stretching
120
After 15 min
100
80
60
40
20
120
After 30 min
<u 100
0
1 80
I 60
£ 40
^ 20 Moisture
120
After 60 min
100
80
60
40
20
0
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm1)
Figure 5.7: FT1R profiles of 10% PP in PC substrate blend after 300°C heat treatment
147
treatment

3476.76 3472.45 3466.55 3469.50 3463.26 Moisture
- 2960.58 - - - CH3 stretching
- 2841.32 - - - CH2 symmetric
- 1460.01 - - - CH3 bending
1382.89 1378.86 1383.77 1382.89 1383.81 C-CH3 bending
1163.24 1165.06 1165.89 1164.60 1167.24 C-C stretching
120
After 15 min
100
80
60
40
20
J_ _ _ _ _ _ _ _ I_______ I_ _ _ _ _ _ _ _ I_ _ _ _ _ _ _ _ I_ _ _ _ _ _ _ _ I_ _ _ _ _ _ _ _ I_______ I_ _ _ _ _ _ _ _ L
120
- After 30 min
<L> 100
CJ
r—
B 80 - Oil
01}
| 60 -
O J*
1
C3 II
u
0
U*
H 40 v----- V----- 1
u
:
Moisture Y
^ 20 u u
-
_l_________ 1_________ 1__________1 1_________ L.________ 1_________ 1_________ 1_________ 1 1 L
120
After 60 min
100
80
60
40
Moisture
20
0
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm1)
Figure 5.8: FTIR profiles of 10% PP in PC substrate blend after 350°C heat treatment
148
treatment

3476.76 3472.45 3475.48 3473.64 3472.45 Moisture
- 2841.32 - - - CH2 symmetric
- 1460.01 - - - CH3 bending
1382.89 1378.86 1383.57 1383.31 1383.62 C-CH3 bending
1163.24 1165.06 1169.19 1166.37 1163.46 C-C stretching
Almost every sample showed similar results as each of them had similar peak positions,
which originated from PC: around 1620cm'1 of C=C, 1380cm'1 of C-CH3 absorption
and around 1165cm-1 of C-C stretching and PP peaks have been lost during the
heating. However, the heating of sample at 250°C for 15 min (Figure 5.6) shows some
traces of PP at 2960cm 1 (CH3 stretching), 2926.37 cm'1 (CH2 stretching) and 1463 cm'1
(CH3 bending). This result indicates PP was not affected when heated for 15 min at
temperature as low as 200°C. However, the mixture of PP and PC only involves a
physical bonding, as can be seen no new functional groups were observed in the sample
other than peaks originated from both raw materials.
These results have clearly shown that no new chemical bonding was formed during the
heat treatment of PP and PC substrate blend. Also, from the visual observation, the
substrate samples have completely broken and did not retain their original form. This
indicates that PP polymer had devolatilized during the heating, leaving gaps and voids
in PC substrate causing the substrate to crumble.
5.3 Thermal Degradation of Polymer
5.3.1 Reaction Kinetics

The kinetic degradation of PP was studied and detailed results presented in Chapter 4.
Thermogravimetric analysis (TGA) is one of the excellent methods to study the kinetics
of thermal degradation of polymer [Aboulkas et al, 2010; Simha et al., 1950]. The
149
TGA profiles of PP at different heating rates were obtained and have been plotted in
Figure 5.9. The TGA curves show almost similar trends, thereby indicating that
degradation of PP occurs in one-step process. No weight loss was observed in the
plateau region (100-280°C) except for small quantities of moisture.
Three heating rates, namely 10°C/min, 15°C/min and 20°C/min were used in this study.
The lowest heating rate showed more rapid degradation compared to 15°C/min and
20°C/min. The degradation of PP begins around 290°C, where about 4% loss has been
detected and the reaction was completed at 486°C. According to Madorsky [1964],
thermal decomposition of PP occurs in the temperature range of 336 to 366°C with
activation energy of 245 kJ mol’1. Yang et al. [2001] reported that the degradation of
PP starts at 250°C, reached a maximum rate at 437°C and was completed at 495°C at
10°C/min heating rate. As heating rates were increased, the TGA curves shifted to
higher temperatures of degradation. The degradation temperatures of PP by 15°C/min
and 20°C/min were around 400 and 415°C, and completed at 590°C and 592°C,
respectively.
These two high heating rates (15°C/min and 20°C/min) have shown almost identical
result as compared to the lower heating rate of 10°C/min, where the plot has shifted
significantly to the left. This phenomenon showed that with increasing heating rates, the
degradation temperature has been narrowed to around 590°C. The poor conductivity of
PP and endothermic nature of polymer decomposition have lead to high energy
consumption [Onu et al., 1999[. Thus, the increasing heating rates did not accelerate
the heat transfer, but the decomposition of PP led to de-polimerization of its chain into
olefin liquid products where all bonds were equally degraded at similar point [Oakes
and Richards, 1949].
This result on the completion of degradation at 500°C is in agreement with previous

studies [Gambiroza-Jukic and Cunko, 1992; Yang et al., 2001]. This lateral shift
presumably occurred due to an increase of heat transfer rate of PP which gets reduced
with increasing heating rates. Therefore, higher temperatures were required for higher
heating rates, in order to provide enough heat to transfer sufficient kinetic energy for PP
polymer bond to decompose.
150
15°C7imn
Sample Temperature (°C)
Figure 5.9: The degradation pattern of PP with different heating rates
Generally, parallel isoconversion lines indicate an independence of activation energy for

conversion in a given temperature range and heating rates; these are the characteristics
of a single reaction mechanism (Gambiroza-Jukic and Cunko, 1992]. The shifting or
displacement of curves due to different heating rates phenomenon has been discussed by
several researchers. The temperature of complete decomposition at a low heating rate is
lower than that obtained at high heating rates. Key factor that may be taken into
consideration is that a change in reaction mechanism can be produced by changing the
heating rate. The inefficient heat transfer from the furnace to sample may also produce
large differences between the furnace and samples, which become significant at high
heating rates. These phenomena can also be explained by a mathematical derivation of
the kinetic laws, which may lead to the displacements of the curves as the heating rate
increases [Aboulkas et al, 2010].
The theory of inefficient heat transfer by Aboulkas et al, [2010] was supported by
Achilias et al. [2007]. His report stated that low thermal conductivity of polymers
together with endothermic reaction consumes large amounts of energy. This suggests
that efficient heat transfer is required in order to allow reactions in polymers. Pyrolysis
151
conditions differ, depending on the methods and equipment used for various research
studies reported in literature; this has resulted in the development of various degradation
mechanisms associated with kinetic results obtained.
Sample j
temperature
Temperature (°C)
Figure 5.10: The difference between programmed temperature and actual measured
temperature of TGA profile at 10°C/min heating rate
Figure 5.10 shows difference between the programmed temperatures and actual
measured temperatures of the samples during thermal degradation analysis of PP at
heating rate of 10°C/min. The highest difference calculated was only about 3%. The
sample temperature therefore, was used for further discussion rather than programmed
temperatures as it will give more reliable results.
The percentage fraction of remaining mass in the platinum crucible was calculated using
the following Equation 1:
Remaining mass (%) = M/M0 * 100 Equation 5.1
where M is the remaining mass at any time and Mo is the initial mass of sample,
including moisture if any.
152
From the data presented in Figure 5.9, it was clearly visible that no char remains in the
crucible irrespective of heating rates. The temperature profiles of PP indicate that the
sample has reached the complete pyrolysis stage at 500°C. The initial degradation
temperatures were strongly influenced by the heating rate, where in the higher heating
rate showed higher degradation temperature than that observed for the lower ones.
Determination of Reaction Order

The calculation of reaction order of the decomposition stage in TGA curves can be done
by considering any particular level of conversion, and determining the instantaneous
rate and temperature at that conversion. The standard rate equation (Equation 2) was
used for nth order decomposition reaction [Milosavljevic and Suuberg, 1995].
dM / dt = -kMpn Equation 5.2
where t is time, k is the reaction rate constant = A exp (-Ea/RT), M is the total sample
mass at any time, and Mp is the remaining pyrolysable mass = M - Mf, and Mf is the
final char mass, here it is treated as zero. The equation can be rewritten by taking
natural logarithm (In) on both sides.
ln[dM / dt ] = -In k + n In Mp Equation 5.3
The advantage of using the logarithm equation is that it converts the exponential
equation into a linear one, thus the curve fitting becomes simpler [Yang et al, 2001].
The plotting of ln[dM/dt] against In Mp in Figure 5.11 gave straight with slope and
intercept values. The results have been summarized in Table 5.4. Results indicate a first
order reaction of degradation kinetics of PP. There is a clear indication that the
assumption of first order reaction is appropriate during most of the polypropylene
degradation, and the yield kinetics were somewhat different than that reported in
literature; reaction order: 0.90 [Yang et al., 2001] and 1.04 (Bockhorn et al, 1999-a].
153
Heating rate at 20 C
y = 1,052x - 3.392
R2 = 0.998
y= l.052x- 3.585
R2 = 0.998
y = 1,069x - 3.680
R2 = 0.998 \
In f M l
Figure 5.11: Calculation of reaction order in PP decomposition with various heating

rates
Table 5.4: Pyrolysis kinetic parameters for polypropylene
Heating rate Reaction order Reaction constant Overall Reaction

(°C min’1) (n) k (min !) constant (k)
10 1.06 3.68 39.64
15 1.05 3.58 35.87
20 1.05 3.39 29.66
Our results indicate that PP has thermally decomposed in a one step process in the
temperature ranges from 290°C to 486°C. Different decomposition products were
generated during the non-isothermal degradation with various activation energies.
According to Gambiroza-Jukic and Cunko [1992], in the process of PP thermal
decomposition, 59.7% olefins were created together with 34.9% paraffin; higher
temperature gives higher amount of olefins.
As noted from the data in Table 5.4, the lower heating rate gave higher activation
energy than higher heating rate. The relationship of heating rate and degradation
mechanism could be understood as follows. Different heating rates result in different
154
level of conversions, where lower heating rate increased the activation energy required
for the decomposition to occur [Gambiroza-Jukic and Cunko, 1992]. With increasing
temperature and heating rate, the products of decomposition and conversion level are
seen to increase. As the conversion level rises, the activation energy rises too because
the number of low molecular decompositions products increase.
For sample heated rapidly enough, the pyrolysis occurs with the “high heating rate
kinetics”, in contrast when the decomposition occurs at lower temperature, the “low
heating rate kinetics pathways” are involved. However, it is not the heating rates per sec
that determines the decomposition pathways, but to a certain extent the temperature at
which the decomposition takes place are considered to be significant. Apparently higher
heating rates favor decomposition at higher temperatures [Yang et al, 2001].
5.3.2 Calculation of Activation Enemy by Arrhenius Equation

The Equation 4 can be rewritten in combination Arrhenius equation:
dM / dt = -kMp Equation 5.4
where, n = 1 due to first order reaction, k is the reaction rate constant = A exp-f-AE/RT)
and A is preexponential constant.
In k = In A - AE/RT Equation 5.5
The plotting of lnfk] against 1/T gave straight with slope and intercept values as shown
in Figure 5.12. The data are presented in Table 5.5.
155
1/1(10 3 K1)
Figure 5.12: Activation energy using Arrhenius equation
Table 5.5: Pyrolysis activation energy using Arrhenius equation
Heating rate R2 Activation

Slope In A A
(°C min ') (J-mol 'K ') energy (kJmol'1)
10 1.04 0.99 5.30 200.33 381.16
15 0.94 0.99 4.89 132.95 313.68
20 0.83 0.99 4.56 95.58 285.03
Apparent kinetic parameters are often used to represent the behavior of plastic
decomposition in general. The variation of activation energy reported is mainly due to
the difference in sample property (the purity and molecular weight), the method of
measuring the weight loss behavior during polymer decomposition, the methodology of
kinetic parameter evolution, heat transfer limitations or complex degradation
mechanisms [Bockhorn et al., 1999-a; Yang et al., 2001J.
Yang et al [2001] in their simulation study on pyrolysis of polymers including

polypropylene have acknowledged that the activation energy of a polymer mainly
depends on the difference between the temperature at which the decomposition begins
156
and the temperature at which the temperature reaches a maximum. In most cases, the
greater the difference between the temperature at which the decomposition begins and
the temperature at which the decomposition reach maximum, the smaller the activation
energy. Similarly, the reaction order mainly depends on the difference between the
temperature at which the rate of decomposition reached maximum and the temperature
at which the decomposition is complete. Generally, the greater the difference, the higher
the reaction order obtained. Milosavljevic and Suuberg [1995] confirmed that it is not
the heating rate per se that is important variable determining the kinetic constant, but
rather the temperature at which the decomposition occurs.
5.4 Degradation Mechanisms for Polypropylene

PP thermal degradation process occurs by the radical mechanism and the final products
of decomposition depend on the type of radicals dominating in the initiation of the
thermal decomposition process [Gambiroza-Jukic and Cunko, 1992]. The
decomposition of PP involved several reactions that occurred simultaneously [Yasushi
and Kaoru, 2003]:
• random scission of the main chain
• (3-scission
• Intra-molecular and inter-molecular radical transfer and
• propagation
• Formation of C=C bonds at the end sites
The choice of temperature and time are critical factors controlling the degradation
product line distribution. The pyrolysis carried out at higher temperature promotes
production of light and volatiles liquids, while longer reaction time enables wider
product ranges. With an increased reaction time, the amount of liquid products
increased and the oil produced became more transparent [Yasushi and Kaoru, 2003].
Williams and Slaney [2007] reported pyrolysis of PP yields with high conversion of
liquid, which amounted up to 95% of liquid, 5% of gas and no solid residue. The
gaseous products had higher amounts of methane, ethane, propane and butane (alkanes)
as compared to alkenes. Depolymerization products that are unable to escape rapidly
enough can undergo secondary reactions and thus contributed to char generation
157
[Milosavljevic and Suuberg, 1995]. Yasushi and Kaoru |2003] suggested a series of
potential plausible degradation reactions of PP as shown in Figure 5.13.
Initiation begins at step (1), where the random scission reaction of C—C bond of the
main chain occurs with heat to produce hydrocarbon radicals. This step is followed by
step (2), where the hydrocarbon radicals decompose to produce small hydrocarbon such
as propylene, then followed by p-scission reaction displayed at step (3) and finally
pulling off H radicals out from another hydrocarbons to produce new radicals in the
propagation stage. The former is treated as intra-molecular radical transfer reaction.
Termination reaction represents the recombination of two free radicals. In proposed
pathways, the reaction steps (1) and (2) reduced molecular weight while step (3) shows
an electron transfer process and final step (4) is a reaction causing an increase of
molecular weight. The recombination may take place for those radicals that do not move
fast [Yasushi and Kaoru, 2003].
Initiation
chjh CHJH
W.
(1) — c-c-c-c — —► —c-c + —c-c
1 1
H H H H H H H H
Chain reaction
H
?H^. i „
(2) -c-c H —C —H
i + c2
H H H
(3)
CHJH
— C~C + —►
W
c-c + —c-c
C=C
H H H H H H H H
Termination
CHJH CHJH
CHJH i 3i i
i Ji.
(4) '"-c-c + —c-c —C-C-C-C-*'
1 1 1 1
1 i
H H H H
H H H H
Figure 5.13: Plausible reaction pathways of PP degradation mechanism as suggested by

Yasushi and Kaoru |2003]
158
5.5 Summary
The interaction between PP and PC was further investigated in this chapter. The effect
of 30 min soaking at 200°C, 250°C 300°C and 350°C on the PP (10, 20 and 30%) and
PC blend, added with binder substrates was investigated. The results show negligible
weight loss after the soaking. This indicates that the blend of PC had helped retain PP
residue, by suppressing the liquid phase produced during PP degradation process. The
soaked samples were then heated up to 600°C and dwelled for 30 min. The weight loss
increased with increasing PP% in the samples. 20% PP substrate sample shows the
more than 20% of loss, which may be caused by loss of binder as well, as it has been
reported that binder starts to devolitilize at temperature range of 200°C to 450°C. The
soaked samples (up to 300°C) containing 30% PP showed almost 2% residue after
600°C heat treatment, but sample soaked at 350°C showed total loss after similar
treatment. This trend is different to that observed for 10% and 20% PP concentrations.
Sample with highest PP composition (30%) was most degraded due to the
devolatilization of PP and that caused the substrate sample to crumble. The FT1R
analysis was carried out on the heat treated substrate sample composed of 10wt% PP
and PC (without binder). There was no chemical bonding form as no new peak was
observed other than peaks originated from PP and PP. The kinetic and degradation
mechanisms of PP was reported for further understanding the cause and reaction of
polymer and the effect of the presence of PC on the PP decomposition.
159
Chapter 6: Results & Discussion - Polyethylene
CHAPTER 6

POLYETHYLENE
160
Chapter 6: Results & Discussion — Polyethylene
Similar to the data reported in Chapter 4 on PP, the experimental results on PE have
been presented in three sections. The first part investigates the interaction and wetting
behavior of PE and PC for temperatures up to 350°C; the second section reports on the
effect of mixtures between PE and PC for temperature up to 1000°C; further studies on
the decomposition reaction of PE to understand the behavior reported in first and second
part is reported in third section.
6.1 Interactions of PE with PC Substrates
Figure 6.1: Raw PE on petroleum coke substrate surface
Figure 6.1 shows an image of raw PE on the petroleum coke substrate prior to carrying
out the experiment. The wetting experiments of PE on petroleum coke substrates were
conducted in temperature range of 150-350°C for 15, 30 and 60 min.
a) Wetting behaviour at 150°C

As seen in Figure 6.2, PE had only melted at 150°C without any penetration into
petroleum coke surface. After 15 min heating, the PE particles were distinct and
separated, and they started to fuse together when time was increased to 30 minutes. As
time was increased to 60 minutes, PE polymer started to liquefy. However, no wetting
or penetration was observed at this stage.
161
15 min 30 min 60 min

Figure 6.2: PE on petroleum coke substrate after 15, 30 and 60 min heat treatment at
150°C

PE melt started to penetrate into the petroleum coke surface after exposure to 200°C.
After 15 min of exposure, PE was still only partially melted on the petroleum coke
substrate (Figure 6.3). Moreover, the colour of PE melt was also similar to that
observed for the raw specimen.
Figure 6.3: PE on petroleum coke substrate and the corresponding SEM micrograph of
An uneven surface was observed from SEM micrographs showing that PE was not fully
molten at this stage. The average contact angle of 82.65° calculated from the SEM
pictures indicated non-wetting of PE with petroleum coke surface. The circled mark on
the SEM picture shows the PE area that did not fully melt, thus did not come into
contact with the petroleum coke surface.
162
Figure 6.4 shows the effect of exposure of PE on petroleum coke to 200°C after 30
min. The particulate form of PE was still visible to some extent, as confirmed by the
SEM image which shows an uneven surface. PE region appeared jagged-edged with
small voids; however the voids were lower in number compared to the 15 minutes
sample. The contact angle calculated was 72.35°.
After 60 min of heat treatment, the colour of PE melt became dense and opaque
(Figure 6.5). The particulate form had disappeared, however spherical granulated raw
PE was still visible from the top of the melt. Small voids noted in the samples heated for
15 and 30 min samples were no longer observed, but one large cavity, which may have
been produced by the cracking of incomplete PE particulate section was observed. The
contact angle was 52.10° indicating the likelihood of PE melting and spreading easily
on the surface.
163
The effect of heating at 250°C for 15 minutes was similar to that seen in the sample
heated for the same duration at 200°C. The PE did not melt completely since the
particulate shape of the raw material could still be seen (Figure 6.6). The solidified PE
could be easily removed off the substrate. SEM image shows an uneven surface of PE
top, with contact angle as high as 131.50°. Large holes were observed showing
unmelted PE granules within.
Similar to the samples reacted for 15 min, the solidified PE sample also could be easily
removed off the petroleum coke substrate even after 30 min. According to Figure 6.7,
the granular form of the raw PE was still recognizable; however, it was more fused
together compared to the sample after 15 minutes of reaction. Longer time has allowed
PE to melt further and leading to the dissolution of initial granular shapes. The lower
contact angle 52.70° shows that PE melt has become more fluid after 30 min of
treatment, leading to better spreading. The contact lengths of the PE melt along the
interface were 10.24 mm and 13.78 mm for the samples exposed to 250°C for 15 and 30
minutes respectively.
164
Figure 6.8 shows the image of the PE sample on coke after exposure to 250°C for 60
minutes. The particulates shape of the original raw PE was not observed anymore. In
addition of that, some small bubbles were detected inside the PE melt. The contraction
of PE on solidification also seems to have caused the lifting up the upper layer of
petroleum coke. The contact angle had also decreased to 38.35° with the increasing time
of the experiment.

The effect of increasing temperature to 300°C is shown in Figure 6.9 to Figure 6.11,
for times of 15, 30 and 60 minutes, respectively.
Reactions at 300°C for 15 min resulted in the incomplete melting of PE with a high
contact angle of 98.80°. The incomplete melting was further confirmed from the SEM
images which revealed PE granules that had not melted and presence of the big cavities
(Figure 6.9).
165
Figure 6.10: PE on petroleum coke substrate and the corresponding SEM micrograph
of the interfacial area after 30 min of reaction at 300°C
Increasing time to 30 min at 300°C heat treatment resulted in a complete melting of PE

with the contact angle also decreasing to 52.10°. Some bubbles were observed inside the
PE melt as seen in Figure 6.10. The melt colour was still opaque, exhibiting the colour
of raw PE. SEM images showed less voids and smoother surface of the PE polymer
phase. Molten PE was seen to penetrate into the petroleum coke surface and tended to
form curved boundaries at the interface.
After 60 min, PE had fully melted and the colour of the melt appeared to be more
translucent and glossy (Figure 6.11). Some bubbles were spotted inside the melt. The
melt had become less viscous and showed a greater spread (13.67 mm along the contact
area length) and a contact angle of 36.30°. The contact angle had significantly reduced
compared to the samples exposed for 15 min and 30 min (which were 98.80° and
52.10°, respectively). Only a few small voids were observed from the SEM
micrographs.
166

When the samples were exposed to 350°C, PE completely melted and changed in colour
to translucent yellow as seen in Figure 6.12. Several small bubbles were clearly seen
inside the polymer melt. The increased melting and wetting tendency of PE melt were
confirmed by low contact angle of 55.50° and greater contact length at the interface
(11.32 mm).
of the interfacial area after 30 min of reaction after at 350°C
The increment in the time to 30 min further reduced the contact angle to 45.50°. More
bubbles were noted on the top of the melt compared to the sample heated for 15 min
(Figure 6.13). However, these small bubbles had disappeared as the time was increased
to 60 min. In addition of having a low contact angle (30.90°), it also has the highest
length of contact between the polymer and petroleum coke (13.66 mm). SEM image in
Figure 6.14 shows that the PE had flattened onto the substrate surface. Moreover, the
polymer region revealed a smooth surface without any cracks or voids.
167
The dimensions of the PE melts were calculated and are summarized in Table 6.1.
These estimated contact angles measurement have also been plotted in Figure 6.15.
Table 6.1: Dimensions of PE melt on PC substrate after reactions at different

temperatures and times

Height Contact
Temperature Time cross-section
(mm) angle (°)
(°C) (min) (mm)
200°C 15 10.78 1.89 82.65
30 9.14 1.24 72.35
60 8.19 1.01 52.10
250°C 15 10.24 1.64 131.50
30 12.37 1.86 52.70
60 13.78 1.39 38.35
300°C 15 10.08 1.16 98.80
30 12.98 1.26 52.10
60 13.67 1.45 36.30
350°C 15 11.32 1.30 55.50
30 11.78 1.02 45.50
60 13.66 0.44 30.90
From Table 6.1, the lengths of contact of samples treated at 200°C decreased with
increasing time. This is probably caused by the partial melting of PE. As the particulate
raw material placed on the substrate were scattered, it may have appeared to have
greater length, but there were some areas where the polymer phase did not contact the
petroleum coke surface. The increasing in time helps in further melting of the liquid PE
phase due to lowered viscosity. The height of the melt also was reduced to 1.01 mm
with increasing the time to 60 minutes, and resulted in a lower contact angle.
168
2.00
Time (min)
200~’C X 250-C A 300°C
Figure 6.15: Effect of time and temperature on estimated contact angle and adhesion of
PE melt on PC substrate
As the temperature was increased to 250°C, the contact lengths increased with increase
in time. This trend was expected since the PE melt was expected to become more fluid
and thus spread further. The contact lengths increased from 10.24 mm at 15 minutes, to
12.37 mm at 30 minutes and to 13.78 mm after 60 minutes. The inconsistent variations
in the PE melt height, at 30 min may be caused by the growth of bubbles inside the melt
at this stage.
This increment in the height becomes more obvious in samples treated at 300°C. This
increase may be due to the bubbles foaming from inside the PE melt. With the
increasing in the temperature to 300°C, PE polymer starts to devolatilize, and starts
releasing gases. The formation of bubbles was observed inside the PE melts from the
SEM images. The length of contact area increased significantly as the time was
increased. These increasing lengths were due to the PE melt becoming less viscous
which resulted in better spreading.
Similar trend was observed in the set of samples exposed to 350°C. The increasing time
has resulted in increased contact length from 11.32 mm (15 min sample), to 11.78 mm
(30 min sample) and 13.66 mm (60 min sample). The height of the samples also
reduced, showing a better spreading of PE melt at 350°C with increase in the time. As
indicated by their respective figures, the PE sample melt has reduced in amount and the
size of bubbles at this temperature.
169
Figure 6.15 shows the effect of time and temperature on the contact angles of PE on
petroleum coke substrate. The highest contact angle (131.50°) was determined at 250°C
after 15 minutes, followed by 80.65° and 72.35°, exhibited by sample heated to 200°C
after 15 minutes and 30 minutes, respectively. These highest values detected from the
observed samples indicate incomplete melting. The low temperature and time was
insuffient for PE to melt and become fluid.
However, increasing time gradually decreased the contact angle as seen in 200°C series.
Time also had a significant role in reducing the contact angle of the PE melt, where the
values decreased to 52.10° after 30 min. Similar trends were observed for samples
heated at different temperatures. Samples heated at 300°C had contact angle values of
98.80°, 52.10° and 36.30° after 15 minutes, 30 minutes and 60 minutes of the
experiment respectively. Samples heated at 350°C showed decreasing contact angle
values, from 55.50° at 15 minutes, to 45.50° after 30 minutes and 30.90° after 60
minutes.
In addition, the increase in the temperature also led to a decrease in the contact angle
values. After 15 minutes of heating, sample at 250°C had contact angle of 131.5°. This
value decreased to 98.80° and 55.50° as the temperature was increased to 300°C and
350°C, respectively. Similar to the 30 minutes heating, the values reduced from 72.35°
at 200°C, 52.70° at 250°C, 52.10° at 300°C and 45.50° at 350°C. The samples after 60
minutes heating also showing gradual decrease with increasing temperature, with the
contact angle values being 52.10°, 38.35°, 36.30° and 30.90° at 200°C, 250°C, 300C
and 350°C, respectively. These decreasing trend has indicated the increasing the
temperature increases the melting of the PE polymer.
170
6.1.1 Interfacial Phenomena Between PE and PC

3000 Point l
2500
2000 H
1500
1000-I,
500
0
1 2 3
keV
3000 Point 2
2500
2000 H
1500
1000
500
0 L
Figure 6.16: EDS analysis of the PE/coke interface after heating at 350°C for 30
minutes
The interfacial properties of PE with petroleum coke were investigated. SEM

micrographs were taken to determine the behaviour of the polymer and petroleum coke
in the interfacial region. The differentiation between the petroleum coke and PE was
done by SEM pictures as these materials appeared to be quite distinct. As shown in
Figure 6.16, EDS analysis showed carbon peaks for both regions (PE and PC). The
gold peaks showed are from to the coating from the coating of Au prior to SEM
analysis.
Figure 6.17 shows the effect on the interfacial region after heating at 200°C for 15
minutes to 60 minutes. Distinct appearance of PE polymer and petroleum coke particles
with apparent boundary lines were observed in 15 min sample. As the area was
magnified, the polymer was seen to be clearly surrounding and penetrating the available
gaps between the petroleum coke particles. The penetration depth was calculated as
only 12.83 pm.
171
After 30 min, the PE polymer interface appears fibrous. In addition, PE was seen to
penetrate deeper into coke. The depth of penetration increased after 30 min to 17.70 pm.
The penetration of the polymer into smaller openings was observed after 60 minutes of
heating. This is expected as the longer times enabled PE melt to become less viscous
(Figure 6.17-c). A number of cracks were observed along the interfacial region between
PE and petroleum coke. The length of the crack measured from the interface was found
to be almost uniform (23.61 pm). SEM image shows that the PE has penetrated most of
the upper layer of the substrate.
The effect of heating at 250°C is presented in Figure 6.18. Like the sample heated at
200°C, PE did not show much penetration during the first 15 minutes. At this time,
incomplete PE melting occurred and thus it was unable to penetrate the gap between the
coke particles (Figure 6.18-a).
172
PE interface also appeared very smooth, with small voids scattered all over it. The PE
interface became much rougher after 30 min (Figure 6.18-b). Short strands were seen
emerging from its surface of voids. Penetration was observed at several places, even
though the penetration depth was quite shallow (average of 22.81 pm). The penetration
depth increased to 30.83 pm as the time increased to 60 minutes. As PE liquid was less
viscous after 60 min, it could penetrate small gaps between the cokes (Figure 6.18-c).
As seen from the SEM micrographs, the fibrous PE had seeped through small openings
between petroleum the coke particles, and got entrapped during its solidification,
creating cracks as seen in the magnified image (2.50 pm size).
173
c) Reactions at 300°C
Figure 6.19 shows the effect of heating at 300°C on the interfacial region of PE on
petroleum coke. Low penetration (17.16 pm) was seen after 15 minutes. However, this
value was higher compared to the penetration of corresponding samples for same time
at lower temperatures, (i.e 12.83° at 200°C and 15.16° at 250°C). 15 minutes was too
short a time for PE to melt even at 300°C and to penetrate deeper as the liquid phase
was still viscous at this stage.
After 30 min, the penetration increased to 25.93 pm. Longer time permitted further
melting and infiltration of PE infiltrated through smaller gaps between the petroleum
coke particle. When the time was increased to 60 minutes, PE clearly penetrated deeper
to 33.24 pm. A schematic of polymer penetration and spreading is demonstrated in
Figure 6.20.
174
Petroleum coke region
Figure 6.20: Penetration and spreading of PE melt into the petroleum coke substrate
The penetration of PE into petroleum coke after heating at 350°C was higher than that
seen for the lower temperatures samples. The penetration was relatively high (24.90
pm), which is comparable to that seen after heating at 250°C for 30 minutes (25.93 pm).
Figure 6.21-a shows the reaction zone, where the PE appears to be wetting petroleum
coke particles.
After 15 min After 30 min After 60 min
12.5 pnm
175
After 30 min, PE penetrated further into petroleum coke substrate. The depth of
penetration significantly increased to 74.88 pm, and this increased only slightly
increased to 75.03 pm when the time was increased to 60 minutes. The penetration and
spreading of PE liquid through pores between petroleum coke particles were similar as
seen in Figure 6.20. The PE was seen sticking to the petroleum coke particles. Figure
6.21 b-c shows the penetration of PE through the cracks between petroleum coke
particles.
The variation in the depths of penetration of PE with temperature and time into
petroleum coke has been summarized in Table 6.2 and has also been plotted in Figure
6.22. Both temperature and time had an impact on the penetration of PE into petroleum
coke.
As seen in Figure 6.22, the lowest penetration depth (12.83 pm) was seen in the sample
heated at 200°C for 15 minutes. The depth increased as time was increased to 30
minutes and 60 minutes, to 17.70 pm and 23.61 pm respectively. Similar trend was also
observed for temperature. At 250°C, samples heated for 15 minutes, 30 minutes and 60
minutes exhibited penetrations of 15.16 pm, 22.81 pm and 30.83 pm, respectively.
Table 6.2: Variations in the depths of penetration depth of PE into petroleum coke with
Experimental conditions
Penetration Depth (pm)
Temperature (°C) Time (min)
200°C 15 12.83
30 17.70
60 23.61
250°C 15 15.16
30 22.81
60 30.83
300°C 15 17.16
30 25.93
60 33.24
350°C 15 24.90
30 74.88
60 75.03
176
l ime (min)
■ 200°C X 250°C A 300°C • 350°C
Figure 6.22: Variation in the penetration depth of PE melt as a function of time and
temperature
The effect of heating time on penetration was more significant at 350°C. The sample
heated for 15 minutes showed 24.90 pm of penetration, but samples heated for 30
minutes and 60 minutes showed significantly high values of increased to 74.88 pm and
75.03 pm, respectively.
6.1.2 Discussion
The heat treatment of PE has shown similar behaviour as PP, both in wettability and the
penetration properties. This was expected as PE and PP were both belong to the same
family of polyolefins. However, PE has showed higher softening and melting
temperature compared to PP. PE begins to soften at a temperature as low as 150°C, but
even after 60 min of heating it does not fully melt. PE has a softening temperature
around 101°C and is expected to melt at 120°C [Collin et al., 1997; Conesa et al.,
1994]. From the observation on the experiment, PE started to fully melt after around 30
to 60 min of exposure to 300°C of heat treatment.
Results show that increasing time and temperature of heat treatment had a significant
impact on the interactions of molten PE with petroleum coke substrate. The SEM
micrographs showed that while the increasing temperature resulted in enhanced melting
177
of PE; time was also seen to play an important role. Longer residence time increased the
extent of melting of PE, which in turn resulted in improved wettability and deeper
penetration into petroleum coke substrate.
However, unlike PP, PE showed low bubble nucleation and foaming activity. This can
be explained as, in spite of having high volatiles content, PE has a very low melt-flow
index. The devolatization at 200°C showed no foaming, which can be attributed to high
viscosity of PE, especially at low shear rates. Therefore, the bubbles grow very slowly
and do not foam due to highlyly viscous PE [Yarin et al., 1999]. This phenomenon was
seen clearly by the observation in the samples, laterally and within the melt (cross-
section view).
The heat treatment in this study led to the thermal cracking of PE polymer into smaller
molecules, consisting of wax, liquid and gaseous products. The mechanism of
polyethylene degradation has been explained in a number of studies [Peterson et al.,
2001; Wall et al., 1954; Wampler, 1989]. The initiation step could involve a random
scission reaction, which is generally followed by two competing reactions, e.g. the
propagation (unzipping) to yield monomers, and free-radicals transfer which involves
hydrogen transfer that results in the formation of unsaturated end, saturated end and new
free radicals [Conesa et al., 1994].
The important aspect that should be considered when determining the end-products of
pyrolysis is the pyrolysis temperature and the residence time of volatiles in the hot zone
of the reactor. This is important as the reaction products are formed from raw material
decomposition (primary reaction) and also through the secondary reactions of the
primary volatiles [Conesa et al., 1994]. The pyrolysis products are very sensitive to
temperature and time, as secondary reactions might take place. Polyethylene is stable up
to 290°C but its molecular weight starts to decrease with increasing temperature. The
product of mild degradation (290°C to 400°C) is a plastic similar to original
polyethylene or hard waxes, while extensive degradation yields semi-solids pastes or
liquids [Oakes and Richards, 1949]. Pyrolysis of PE at 400-450°C yields high liquid
fraction of 69-84wt% and 9-13wt% of gases [Scheirs and Kaminsky, 2006; Feng et
al., 1996].
178
Our experimental results are in agreement with these findings reported in previous
studies. In our experiments, while the gaseous volatiles (fumes) were observed during
the heat treatment at 300°C; the volume of gas produced appeared visually to be lower
in those experiments where the penetrations of molten PE into petroleum coke
substrates were higher. In these cases, penetration could be occurring at a faster rate
than the rate of volatile production by polymer degradation. Also, the presence of coke
could be acting as a physical and thermal barrier to the release to volatile from the
penetrated polymer. However, this postulation cannot be confirmed at this stage.
The viscosity of the liquid product from pyrolysis of PE also depends on the residence
time, as longer time allows for additional carbon-carbon links breaking, as exhibited by
the breaking of thick liquid into a thinner liquid and gases (secondary reaction). The
bubbles formed on top of the molten PE indicate the formation of gaseous products was
caused once again due to secondary reactions. The top of PE tends to heat up faster and
decompose into lower viscosity liquid and gases, whereas due the heat transfer
limitations [Pearson, 1978], PE inside the substrate takes longer time to liquefy and
could explain the increase in penetration with time.
This phenomenon explains to some extent enhanced the penetration of molten PE with
increasing temperature where there is also greater driving force for heat transfer as
temperature is increased. With increasing temperature, the PE becomes more fluid and
is able to flow deeper through petroleum coke particles within the same time period,
e.g., at 350°C after 30 min of treatment, much greater penetration depth was observed
as compared to sample heated at lower temperature. The relative stability and PE
decomposition may be impacted by the length of the chain, where the stability is known
to decrease with decreasing molecular weight and number of weak links present in the
molecules. Oakes and Richards [1949] have reported the initial rate of C-C breaking is
affected by the temperature where numbers of bonds broken are 0.38x10' /g.hour at
305°C and 80xl0'18/g.hour at 360°C.
The re-solidification of the PE liquid after each of the heat treatments caused the surface
of PE to contract. Some petroleum coke particles were also found to be present within
the PE melt. Shrinkage of PE surface with solidification resulted in the detachment of
179
the coke layer close to the interface. Deeper penetration also led to a greater binding of
PE and PC particles from the substrate surface. The schematic diagram of the
phenomena occurring is demonstrated in Figure 4.20 (Chapter 4). Temperature and
time have been found to have a significant influence on the wettability and penetration
behaviour of PE into a petroleum coke substrate. Higher temperature allowed higher
chain breaking rates, while longer residence time allows secondary reaction to take
place. Both conditions may lead to chain breaking, which resulted in more fluid liquid
and therefore better wettability. This phenomenon has enhanced the spreading of molten
PE on the substrate surface and allowed the liquid to flow through between petroleum
coke particles, and increased the depth of penetration into petroleum coke.
The results obtained have shown that when wettability improves due to the lower
interfacial tension, greater depth of penetration is observed. Improved wetting is an
indication that bonding between PE and petroleum coke is developing, and PE could
serve as a binder material and partially substitute for pitch, which is the binder currently
used in the aluminium industry.
6.2 Interactions Between PE and PC for Temperature Up to

1000°C
The mixtures of PE and PC were investigated for 10 to 50% PE in the blend. As the
trend of result was similar to PP and both PE and PP belong to the same polyolefin
family, a study was carried out to primarily seek key differences between them.
6.2.1 Weight Loss Study of PE and PC Mixtures

The effect of heat treatment, according to the baking cycle procedures on the mixtures
of raw PE and PC is presented in Table 6.3. The residues (%) of the mixtures by the
ratio of 10 to 50% PE to PC are shown in Table 6.3 and Figure 6.23 provides
corresponding plots.
The increase of PE ratio in the mixtures has decreased the percentage of residues
obtained (Figure 6.23). The increasing of heat treatment temperature (Cycle 1 to Cycle
3) also lessens the weight of residues measured. This trend is expected as PE is
composed of high volatiles that have been released during the heating cycles. Highest
180
temperature of heating cycles (Cycle 3) results in the lowest residue left for both
mixtures ratios, while the highest residues obtained from Cycle 1, since at this
temperature (150°C) PE had not melted.
Neglecting the moisture loss from Cycle 1, the loss of PE blends was relatively low
compared to PP. As can be seen in Cycle 2, the 20, 30, 40 and 50% PE mixtures loss
about 16%, 21%, 33% and 43%, while with PP mixtures loss was about 17%, 22%,
35% and 47%, respectively. A similar trend was observed in Cycle 3. As discussed
earlier, PE melting and decomposition started later than PP, due to its high viscosity
properties.
Table 6.3: Percentage of solid yield of sample mixtures (PE:PC) obtained after baking
Cycle 1, Cycle 2 and Cycle 3
Heating Initial Weight Final Weight

PE (wt%) Loss (%)
Cycle (g) (g)
10 4.61 4.60 0.29
20 4.19 4.18 0.28
Cycle 1 30 4.52 4.51 0.23
40 4.54 4.55 0.21
50 4.50 4.49 0.11
10 4.59 4.16 9.37
20 4.44 3.74 15.78
Cycle 2 30 4.51 3.56 21.12
40 4.58 3.06 32.99
50 4.53 2.58 43.1
10 4.45 4.03 9.33
20 4.38 3.62 17.43
Cycle 3 30 4.63 3.47 24.97
40 4.57 2.96 35.12
50 4.55 2.33 48.72
181
PE (%)
• Cycle 1 □ Cycle 2 A Cycle 3
Figure 6.23: Residue left (%) of PE and PC mixtures
6.2.2 Carbon Content of PE and PC Mixtures

The percentage of carbon content of heat treated blends of PE and PC was also
investigated for two sets of specimens. The data is presented in Table 6.4 and plotted in
Figure 6.24.
Table 6.4: Carbon content (%) of PE:PC mixtures
PE (%) Heating Treatment Carbon Content (%)

Cycle 1 92.5
40 Cycle 2 94.7
Cycle 3 96.0
Cycle 1 93.9
50 Cycle 2 96.6
Cycle 3 97.1
Higher PE ratio (50:50) seems to yield higher carbon percentage. As the heating level
was increased (temperature was raised), more PE were pyrolyzed and decomposed into
carbon and release of volatiles. The decomposition of PE generated carbon and
increased the carbon to weight ratio, thus raising the percentage of carbon in the
mixtures. At lower temperature (Cycle 1 and 2), low carbon (%) was obtained. This is
due to incomplete devolatilazation of PE at low temperatures. The high volatile content
of PE has impacted the total weight of the mixtures, thus resulting in lower carbon
percentage.
182
Cycle 1 Cycle 2 Cycle 3

Heat treatment
040:60 X 50:50
Figure 6.24: Carbon content (%) of heat treated PE and PC mixtures
6.2.3 Structural Characterizations of PE and PC Mixtures

The XRD profiles of raw PE, raw PC and their mixture (50:50) are displayed in Figure
6.25 - 6.26, while the detailed peak characterization is presented in Table 6.5.
50000
Raw PC
40000
30000
20000
2 10000 ■
'5 150000
Raw PE
100000
50000
10 20 30 40 50
Figure 6.25: XRD profiles of raw PE and PC
183
50000
40000
30000
20000
10000
50000
Cycle 2 -up to 600 °C
40000
z/)
| 30000
0
'__/
&
1 20000
£
10000
50000
g Cycle 3 up to 1000 °C
40000 u
30000
20000
10000
■L.... I------ 1----- 1----- 1----- 1----- 1..... 1......1------L

0
10 20 30 40 50 60
Figure 6.26: XRD profiles of raw PE and PC mixtures at 50:50
184
Table 6.5: XRD peaks characteristics of raw PE, PC and their mixtures at 50:50*
Crystal Size Possible

Peak
Heating d-spacing Compound
Sample Position
Cycle (A) h k 1 (chemical
(20)
formula)
26.03 3.37 0 0 2 C (graphite)
Raw PC
44.64 2.03 1 0 1 C (graphite)
21.63 4.13 1 1 0 (C2H4)n
Raw PE
23.70 3.72 2 0 0 (C2H4)n
21.63 4.13 1 1 0 (C2H4)n
Cycle 1 26.10 3.37 0 0 2 C (graphite)
42.13 2.11 1 0 0 C (graphite)
26.07 3.37 0 0 2 C (graphite)
Mixtures Cycle 2 42.41 2.11 1 0 0 C (graphite)
43.12 2.09 0 1 5 C (graphite)
26.05 3.37 0 0 2 C (graphite)
Cycle 3 42.89 2.11 1 0 0 C (graphite)
43.10 2.09 0 1 5 C (graphite)
Several new peaks were generated whilst some of them were diminished after the
mixtures were heat treated. Raw PC shows high intensity of carbon peak at 26.03° (002)
and 44.64°. The polyethylene peak positions are at 21.63° (110) and 23.70° (200). After
Cycle 1 of heat treatment conducted on the mixing of PE and PC, the only PE peak
visible is at 21.63°, while the peak located at 23.70° was masked by high graphite peak
(26.10°). New weak peak of graphite at 42.13° (100) was observed. The PE peak was
lost after Cycle 2, whilst the XRD profile was dominated by peaks of raw PC and small
peak of graphite originated from Cycle 1 sample, at 42.41° (100) and another new
graphite peak at 43.12° (015). The XRD profile of the mixture sample after Cycle 3
showed similar peaks position. Both of the profiles (Cycle 2 and 3) were seem to be
more similar to raw PC, since at this stage, the PE has totally decomposed due to high
temperature treatment. However, the small graphite peak of raw PC at 44.64, having
(101) structure, has shifted to 42.89° position, which represents (100) structure. This
(100) structure was formerly found after the mixing of PE and PC.
185
6.2.4 FTIR Analysis of PE and PC Blends

The effect of mixing and heat treatment cycles were investigated by FTIR analysis.
Each sample spectrum was stacked and compared in Figure 6.27-6.28 and the
corresponding peak vibration has been characterized in Table 6.6.
Raw PC
Raw PE
Wavenumbers (cm'1)
Figure 6.27: FTIR peaks of raw PE and PC
The CH2 stretching around 2913-2918cm'1 shared by raw PC and raw PE, is visible in
all residues after the heat treatments. In contrast, another vibration of symmetric CH2 at
2850cm'1 in PE is lost after mixing with PC. The double bond stretching absorbed by
raw PC (1685cm1) also has been reduced after the mixing. Another alkene C=C
absorption, which visible in PC and PE around 1645-1655cm'1 was intact throughout
the heating cycles up to 1000°C (Cycle 3). Similar outcomes on the absorption of CH3
bending (1457-1459cm1) and C-CH3 bending (1059-1072cm'1) which originated from
their raw samples also survived in the heat treatment cycles. However, aromatic ring
186
vibration at 893cm"1 from the raw PC, has been significantly reduced after the mixture
and heat treatment. Most of the peaks (raw PE) became significantly broader with
reduced intensities after miximg and heating (Figure 6.28).
Cycle 1 - up to 150 X
6 u
Wavenumbers (cm'1)
Figure 6.28: FT1R peaks of PE and PC mixtures at 50:50
187
Table 6.6: Characterization of FTIR peak profiles of PE, PC and their mixtures at 50:50
Raw PC Raw PE Cycle 1 Cycle 2 Cycle 3 Possible Functional Groups

3484.12 3477.22 3478.10 3489.31 3475.89 Moisture
2913.98 2917.98 2915.61 2918.56 2915.61 CH2 stretching
- 2850.54 - - - CH2 symmetric
H.CS ^CH
✓c =c v
c c
1457.52 1459.20 1457.58 1457.67 1457.61 CH3 bending
- 1379.27 1383.55 - - C-CH3 bending
- 1162.30 1165.86 1168.48 1167.06 C-C stretching
1060.47 1059.33 1072.79 1069.93 1065.25 C-CH3 bending
893.93 - - - - Aromatic ring
- 718.30 711.80 708.86 708.86 Alkynes (triple C bond)
6.2.5 SEM Images of PE and PC Mixtures
J 1 3000 -
c PE
WmMHI
2000 -
(000
1 1 1 1
0 2 4 6 8
keV
Figure 6.29: SEM micrographs and EDS of raw PE
The SEM micrograph of PE surface used in the mixtures and the result of the EDS
scanning are displayed in Figure 6.29. PE has a smooth and glossy plastic surface. EDS
scan showed high intensity of carbon in PE. While PE has a high carbon peak, PC has
been identified with high intensities of carbon and sulphur, and some traces of silica and
aluminium.
Heat treatment of Cycle 1 has no effect on the mixture. The SEM micrograph showed
PE particles scattered on the PC surface (6.30-a). The PE particles are smaller and tend
to accumulate together, with spongy-look properties. These PE particles still can be seen
after Cycle 2, as they were scattered among smaller PC particles on the PC granule
188
surface (5.30-b). However, no PE traces were found after the sample was treated in
Cycle 3. The high temperature of Cycle 3 heat treatment (up to 1000°C and dwell for 30
min) has decomposed the sample and left no traces of the PE polymer.
Figure 6.30: SEM micrographs and EDS of PE and PC (50:50) mixtures samples after
treatment
6.3 Thermal Degradation of PE

The physical changes of polyethylene samples after pyrolysis, in the range 600 - 750°C
are shown in Figure 6.31. The original samples were in the form of white
pellet/powdery as shown in Chapter 3 (Figure 3.1). The physical state of pyrolyzed PE
samples at varying times and temperatures showed significant differences compared to
that of the raw samples. Temperature and residence time can affect activation energy
and reaction kinetics of polymer pyrolysis. For non-isothermal conditions, the heating
rate influenced the degradation pattern. The nature of samples, such as molecular
weight also has an effect on the pyrolysis products. It has been reported that the
189
degradation of polyethylene gives a series of hydrocarbon oligomers, mainly of alkane,

alkene and terminally saturated diene consisting up to 35 carbons [Wampler, 1989].
The lowest temperature of pyrolysis was 600°C and held for 30min, producing
yellowish rich and greasy soft solid resin-like sample, along with thick gas generation.
The solid residues obtained at more than 600°C for 20 min showed dissimilar physical
state and properties. The longer residence times of pyrolysis allowed heat transferred
into the polymer, as polymer is known to be poor a thermal conductor [Raff and
Allison, 1956]. As time increased, from 10 min to 20 until 30 min at temperatures > 650
°C, the predominant yield of pyrolyzed product was a black char. Pyrolyzed
polyethylene generally produces linear hydrocarbons, namely alkanes, alkenes, and
terminally unsaturated diolefins, as in the case of 650 °C samples after 30min
pyrolysis. At higher temperature and longer times, the molecules are degraded to form
smaller products [Wampler, 1989].
These smaller molecules are more volatile and escape as gaseous products, leaving
heavy carbon molecules as solid yield. Similar trends are observed at 700 and 750°C,
where the heat was unable to transfer through PE samples in less than 10 min. However,
after 20 min, the transformation of the end-product of pyrolysis was obvious, where
they were charred and almost no residue was left as indicated in the weight loss study.
This result was in agreement with that reported by Du and Qu [2006] on isothermal
decomposition of PE, which showed almost no residues upon complete degradation at
500°C.
190
r
W *
Polyethyelene
600°C 30min I L/min Ar
PH products alter 600°C pyrolysis
PE products after 650°C pyrolysis
■ -
4
Mi
\
Polyethyelene Polyethyelene
700°C 5min 1 L/min Ar 700°C lOmin 1 L/min Ar
'tb Polyethyelene
m ■
W'
Polyethyelene
700°C 20min 1 L/min Ar 700°C 30min 1 L/min Ar
Polyethyelene
750°C 20min I L/min Ar
Figure 6.31: Physical states of pyrolyzed PE residues with different times and
temperatures
191
6.3.1 Analysis of Solid Residues of PE Pyrolysis

Table 6.7 displays the percentage of solid yield after pyrolysis of PE according to time
and temperature. Data of the yield was plotted according to effect of time and
temperature in Figure 6.32 and Figure 6.33.
Table 6.7: Percentage of residue obtained after pyrolysis of PE
Temperature Initial weight Yield

Polymer Time (min)
(°C) (g) (wt.%)
600 30 10.62 16.98
650 10 11.54 39.30
20 11.45 0.38
30 11.34 0.06
700 5 11.20 19.40
PE 10 11.67 15.46
20 11.50 0.05
30 11.51 0.05
750 5 11.29 18.11
10 11.23 0.92
20 11.35 0.03
Figure 6.32 shows the effect of temperature of the pyrolysis on the percentage of solid
yield of PE samples. The increasing temperature has decreased the solid yield. As for
samples heated at 650°C for 10 min pyrolysis, the yield of samples was about 40%,
decreasing to approximately 20% after the increment of 50 °C and becomes less after
pyrolysis at 750 °C. However, the pyrolysis for 5 min time between 700 and 750 °C did
not show much change where the difference of increasing temperature is just 2%. The
pyrolysis of 30 min time showed a significantly reduced yield after the temperature
increased from 650 to 700°C until 750°C.
The effect of time on pyrolysis of PE is shown in Figure 6.33. The pyrolysis of 650°C
has shown that the increase of residence time from 10 (39.30%) to 20min (0.38%)
shows a huge decrease in its solid yield. However, this effect is not as significant at low
residence time of 5 minutes. The effect only took place at longer residence times.
Similar trend observed for the other temperatures. At 700°C, the yield of solid for 5 min
is 19.40% while at 10 min is 15.46% (approximately 4% reduction). The increment of
pyrolysis residence time from 20 to 30 min leads to huge reduction, giving only 0.05%
of solid yield.
192
Temperature ( C)
A 5 min □ 10 min O 20 min X 30 min
Figure 6.32: Effect of temperature on yield % of products of PE
. . 1H
0 5 10 15 20 25 30 35
Time (min)
□ 650°C A 700°C X 750°C
Figure 6.33: Effect of time on yield % of products of PE
Kim and Kavitha [2006] have explained that thermal degradation of PE depends on the
rates of heating, mass or momentum transfer its thermodynamic property of melt. The
melt thermodynamics, and correlation with the thermal conductivity, is highly
influenced by temperature [Raff and Allison, 1956]. The research on the effect of
temperature on the thermal conductivity of polymers was also done by Fuller and
Fricke [1971]. The results showed significant effect of temperature on polyethylene’s
193
thermal conductivity as compared to polystyrene and polypropylene due to differences

in polymer branching and molecular weight. They also reported that the solid linear
polyethylene (PE) exhibits higher conductivity than solid branched polyethylene (PE),
and melts also exhibit a similar behaviour.
This can be understood on the basis that thermal conductivity decreases as the polymer
chain becomes more complex. Polyethylene (PE) consists of short chain branches, as
opposed to low density polyethylene (LDPE), which contains both short and long chain
branches [Dominik et al., 2009]. Similar results were observed in PP, thus, the
conductivity decreases in the following order: linear polyethylene (PE) > branched
polyethylene (LDPE) > polypropylene (PP). Less branched polymers exhibit higher
segmental mobility, which increased with increasing temperature and consequently
enhanced the thermal conductivity [Fuller and Fricke, 1971].
The comparison between yield% data obtained after pyrolysis of PE alone and PE in the
mixtures has resulted in higher amount of PE residue in the mixtures. It is expected as
PP also has exhibits similar behaviour, where PC had helped contain the polymer
decomposition products, and retaining its residue. This result also indicates that the
presence of PC has retained the PE residue up to 8.88% (30% PE) after the Cycle 2 heat
treatment (closest conditions). Even though the pyrolysis carried out at 600°C for 30
min on PE alone has resulted in 16.98% of residue yield, but in pyrolysis conducted for
20 min at 650°C has only given 0.38% residue. Considering the Cycle 2 involves longer
time of heat exposure, PC has somewhat retained the PE residue from being lost during
the heat treatment.
6.3.2 Chemical Analysis of PE Pyrolysis Residues

The effect of temperature and time on the carbon content in solid residues analysed by
LECO, respectively are displayed in Figure 6.34 and Figure 6.35. The carbon contents
in the range of 45 to 50% were found irrespective of time and temperature used in
pyrolyis. An increase in temperature and time has decreased the percentages of carbon
of the samples, however the differences are about 25%, by comparing the highest
carbon value obtained at 700 °C at 5 min pyrolysis with the lowest carbon value
obtained at 25%.
194
60
50
¥ 40
%O
30
20
550 600 650 700 750 800
Temperature ( C)
□ 10 min O 20 min X 30 min
Figure 6.34: Effect of temperature on carbon content (%) of products of PE
Time (min)
□ 650°C X 700°C A 750°C
Figure 6.35: Effect of time on carbon content (%) of products of PE
The carbon content of the PE residue is lower as compared to the carbon content
measured in the PE:PC mixtures. As PC had helped retaining PE in the mixtures, the
carbon collected also seem to be higher as higher % of PE were still remaining in the
residue, which accounted for the total carbon that has been measured.
Table 6.8 shows the proximate analysis of raw PE. Four different oxides were found
with only 0.5% as ash. Alumina (AI2O3 - 0.280%), iron oxide (Fe2C>3 - 0.089%) and
silicon oxide (SiC>2 - 0.087%) were calculated. These materials were involved in the
manufacture of the polymer.
195
Table 6.8: Ash content (wt.%) of raw PE
Ash (%) Total of ash

Polymer
Fe203 Si02 ai2o3 Ti02 Others (%)
PE 0.089 0.087 0.280 0.026 0.037 0.519
Aluminum oxide (AEO3) is widely used as a catalyst in polymerization of ethylene.

According to Standard Oil of Indiana and Philips Petroleum Company, besides AEO3,
Cr203 and Si02, a group of finely divided metal oxides also can be used, as a carrier in
complex catalyst system to enhance polymerization reaction under mild conditions,
generating high molecular weight polymer and influencing the chain structure of the
polymer. As a carrier, AEO3 has high specific surface development, capable of
absorbing monomer and forming a moderately stable complex with initiator and
monomer. Another type of catalytic process, Ziegler catalyst using Ti and A1 system
was used to enhance the propagation reaction during ethylene polymerization [Raff and
Allison, 1956].
6.3.3 Morphological Study of PE Residues

SEM images were taken to explore the changes occurring in the samples with
temperatures and time. Figure 6.36 (a) shows the overall ground PE samples before
pyrolysing, while Figure 6.36 (b) shows the smooth surface with aligned fibrils of the
samples at higher magnification. The surface of raw PE samples is presented in Figure
6.36, and the effects of time and temperature on the pyrolysis of the polymer are shown
in Figure 6.37.
Figure 6.36: SEM micrographs of raw PE samples (2 mm ground particles) at

(a) lower magnification and (b) higher magnification
196
According to Figure 6.37, pyrolysis of 600°C with 30 min has changed the surface
structure of the polymer. The surface smoothness of the samples has been reduced and
impaired significantly.
The residue after 650°C held for 10 min was still intact. However, after 20 min, the
surface started peeling off and was smudged. After 30 min of pyrolysis, cracks were
spotted and particulates became separated from each other. Similar behaviour was
observed for pyrolysis conducted at 700°C and 750°C. This phenomenon indicates time
has an effect on the residue formed after pyrolysis, by damaging its surface and
microstructures.
The action of higher temperature on damaging the surface and chain of the sample is
more prominent and the specimen appeared as a black powder, indicating the partial
formation of carbonaceous materials. The increase of temperature had also created small
cavities all over the surface. The micro-structure of the solid residue can therefore be
customized with control of influencing parameters such as pyrolysis time and
temperature.
197
600°C Pyrolysis
•
*v
:-'s
■ :W S’k
K;&.v*.
j,
Figure 6.37:
700°C Pyrolysis
Figure 6.37:
6.3.4 Structural Characterization of PE Pyrolysis Residues

The representative XRD peaks of the residues at different temperature with 10 min are
shown in Figure 6.38. These spectra indicate the effect of temperature on the crystalline
structural changes of the residues. The inter-planer distance and the crystal plane along
with associated molecular structure are summarized in Table 6.9. Small traces of Cu
peak were detected positioned at 42-43° in each XRD spectrum. The potential route for
Cu in residue is from the copper crucible used in all pyrolysis experiments.
As noted from the data, raw PE has strong peak at 21.73° and 23.91°. These peaks
correspond to the [110] and [200] diffraction plane of PE crystals with strong
orthorhombic configuration, respectively [Russel et al., 1997]. Weak diffraction peak of
orthorhombic crystal also noted at 36.43° corresponding to [020] plane of orthorhombic
crystal.
These crystal structures have weakened as PE was subjected to 650°C. The

orthorhombic structure has become less crystalline, while the weak peak around 36° has
diminished after being heated. The broadening peak between 18 to 21° indicates the
presence of monoclinic crystal at 18.98°, corresponding to [001] crystal plane. This
pattern was only apparent at 700°C. This shoulder like pattern was expected as this PE
consists of 23.5% monoclinic and 42.7% orthorhombic phase mixture. Determination of
monoclinic component can only be drawn from the observation as its content is
relatively low. This phase becomes observable at this point as the intensity of
orthorhombic phase around 21° has lessened [Russel et al., 1997].
200
150000
Raw PE
100000
50000
150000
650 *C
100000
~ 50000
isity (ci
C 150000
100000
50000
150000
750 °C
100000
50000
0
10 20 30 40 50
Figure 6.38: XRD spectra of samples at 650°C to 750°C for 10 min compared to raw
PE
201
Table 6.9: Effect of temperature on the structural parameters of PE residues XRD peaks
for 10 min
Miller indices Possible

Peak
Temperature d-spacing compound
Samples position
(°C) (A) h k 1 (Chemical
(20°)
Formula)
- 21.73 4.09 1 1 0 (C2H4)n
Raw PE 23.91 3.72 2 0 0 (C2H4)n
36.43 2.47 0 2 0 (C2H4)n
650 21.51 4.13 1 1 0 (C2H4)n
23.85 3.73 2 0 0 (C2H4)n
700 18.98 2.57 0 0 1 (C2H4)n
Residues
21.38 4.15 1 1 0 (C2H4)n
23.89 3.73 2 0 0 (C2H4)n
750 - - - - - -
The orthorhombic weak peak in raw PE appeared around 36° which was then found to
decease in all residues. At 750°C, no crystalline structure was visible, indicating that the
crystalline PE structure had completely diminished and become amorphous.
The effect of time on the structural orientation of PE residues has been presented in
Figure 6.39 and the peaks listed in Table 6.10. The heating of sample for 5 min has
degraded PE peaks at 21.49° and 23.87°, denoted by orthorhombic plane of [110] and
[200], respectively. The small orthorhombic crystal peak at 36.03° was still visible after
5 min. The increasing heating time to 10 and 20 min had completely destroyed the
crystalline structure of the PE.
202
150000
Raw PE
100000
50000
150000
5 min
100000
:ounts)
S'
50000
"" 150000
10 min
100000
50000
150000
20 min
100000
50000
0
10 15 20 25 30 35 40 45 50
Figure 6.39: XRD spectra of samples for 5 min to 20 min at 750°C compared to raw PE
203
These results demonstrate that increasing time and temperature of pyrolysis

significantly reduced the crystalline phase of PE residues. High intensity peak indicate
larger crystallites with slightly tighter packing while smaller peak represents smaller
crystallites with slightly looser packing. The variation of crystallite size depends on the
carbon content and the thermal history of the sample [Russell et al., 1993]. This could
explain the transformation of PE residue’s pattern, where the peaks were reduced by
subjecting sample to elevated temperatures during pyrolysis.
Table 6.10: Effect of time on the crystalline features of PE residues after 750°C
pyrolysis
Possible
Peak Miller indices
Time d-spacing compound
Sample position
(min) (A) (Chemical
(20°) h k 1
Formula)
Raw PE — 21.73 4.09 1 1 0 (C2H4)n
23.91 3.72 2 0 0 (C2H4)n
36.43 2.47 0 2 0 (C2H4)n
Residues 5 21.49 4.13 1 1 0 (C2H4)n
23.87 3.73 2 0 0 (C2H4)n
36.03 2.49 0 2 0 (C2H4)n
10 - - - - - -
20 - - - - - -
6.3.5 FTIR Characterization of PE Residues

The residues of PE pyrolysis were further subjected to FTIR analysis. The results were
compared to raw PE to determine the effect of temperature and time on the structure of
polymer chains.
The FTIR spectrum of raw PE was compared with samples obtained at 700 °C and 750
°C for 10 min pyrolysis in Figure 6.40, while Table 6.11 shows the list of major peaks
from the spectra and the possible functional groups.
The polyethylene is a long molecular chain of CH2 groups with one CH3 group at each
end whose presence does not affect the number of possible chain vibrations in infrared
absorption. The thermal decomposition of polyethylene generates about 30 compounds
consisting of straight alkanes, alkenes, dienes, and cyclic compounds in general.
204
Literature shows that in some cases the monomers evolved before the molecular weight
of residual polymer was appreciably reduced [Raff and Allison, 1956]. This behavior
of polyethylene decomposition is reflected in the results from FTIR.
The vibrations at 2850 cm'1 and 2918 cm'1 which represent CEL stretching and
symmetric, respectively, are small for raw PE samples, whereas the heating at 700°C
has intensified both peaks [Raff and Allison, 1956; Du and Qu, 2006]. Similar trend
was observed for the vibrations around 1642 cm'1 (C=C absorption) and 1463 cm'1 (CH3
bend) [Blazso et al., 1995; Du and Qu, 2006], as small vibration is seen in the raw
data, but that intensified at 700°C and then reduced significantly at 750°C.
The band at 1731cm'1 assigned to C=0 thermo-oxidative products [Du and Qu, 2006].
The vibration of aromatic ring containing oxygen found in the samples may arise due to
the catalytic polymerization of PE [Raff and Allison, 1956]. The other possible reason
is when pyrolysis is done in which air was not completely removed. The residue
containing oxygen may also be present after samples were taken out and the
temperature of the samples was still high which could have caused PE to react with
oxygen in air [Madorsky, 1964; Shlyapnikov and Serenkova, 1983].
The decomposition reaction that takes place during this pyrolysis may have occurred
through random reaction, where the amount of alkane and alkene were seen to increase
after 700°C pyrolysis. This result is also in an agreement with the result obtained by
Marco et al. [2002], where their pyrolysis of LDPE has also generated the paraffin
wax-like semi-solid yield after 500°C (Figure 6.31).
205
120
Raw PE
100
80
r~~MT
60 118 1£ (jh j*11 « 32 “» <jX -b •£>0
40
; SO “IIIO
20
120
650 °C
100
80
£ r~inr1
■ I
S 111! a
60
V .g
40
: 1IJ I1 °
ransmil
. g 0£
20
H
£ • .
120
700 °C
100
80
Pirr^f
11 f15 -5a
1 !i S S
?yjg fi
tj B c O
: I If £ 42 0
u « Lx
0
c/j X
®
120
$
]
100
\
Moisture— f
\\
CHi symmetric
J
CH2 stretching
J
c
80
C=0 stretching---
CHi bending —
C—CHi bending —
CH2 synunetric
60
40
20
4000 3500 3000 2500 2000 1500 1000 500

Wavenumbers (cm'1)
Figure 6.40: FTIR band of samples at 650°C, 700°C and 750°C for 10 min pyrolysis
206
Table 6.11: Effect of temperature on FTIR peak assignment of PE residue for 10 min
Possible Functional
Raw PE After 650°C After 700°C After 750°C
Groups
3477.22 3476.15 3469.78 3479.07 Moisture
2917.92 2918.25 2917.65 2918.67 CH2 stretching
2850.03 2849.20 2849.24 2850.87 CH2 symmetric
- 1730.91 1731.31 1728.62 OO stretch
1637.94 1638.07 1642.04 - C=C stretch
1466.57 1465.04 1463.34 1462.51 CH3 bending
1378.32 1379.62 1377.31 1375.44 C-CH3 bending
1166.19 - - - C-C stretching
1060.97 - - - C-CH3 bending
719.09 720.54 721.64 - Alkynes (C=C) bonds
The effect of time on the FTIR spectrum of the pyrolsysis sample of PE has been
presented in Figure 6.41 and the list of coresponding functional groups recorded in
Table 6.12. FTIR spectrum of PE residues of pyrolysis carried out at 750°C, were
compared to 5 and 20 min pyrolysis.
The intensity of the absorption bands at 2917.92 cm'1 and 2850.03 cm'1, corresponding
to CH2 stretching and CH2 symmetric bonds, respectively have increased after 5 min
pyrolysis as compared to raw PE. The increasing intensity shows the increment of
correspond bands. However, other peaks such at 1378cm'1 (CH3 bending), 1166cm'1
(C-C stretching) and 1060cm'1 (C-CH3 bending) have been lost after 5 min of heating.
The increment of time to 20 min has shown greater affect as the most of the bands have
been reduced. The only remaining peaks found was alkene absorption C=C stretch
(1632.26cm'1). This occurance showed the depolymerizing took place and changed the
chemicals structure greatly after 20 min of pyrolysis.
207
120
Raw PE
too
80
^a 5 1111 &
60 •s £ I k ! * 8
1 *
40
gu u u
20
120
5 min
100
----- *
M o is tu r e — 4^
| 80
§
B
6
C/3
60 8 | 8
J
§
CH-, stretching
c S'
£ 40
£
E
20
5
120
20 min
100
80
60
40
20
0
4000 3500 3000 2500 2000 1500 1000 500
Wavenumbers (cm'1)
Figure 6.41: Effect of pyrolysis time (5 min to 20 min) at 750°C pyrolysis on structure
of solid residues
208
Table 6.12: Effect of time on FTIR absorption bands of residue at 750°C
Raw PE After 5 min After 20 min Possible Functional

Groups
3477.22 3478.21 3479.53 Moisture
2917.92 2919.14 - CH2 stretching
2850.03 2850.25 - CH2 symmetric
1637.94 1642.04 1632.26 C=C stretch
1466.57 1463.53 - CH3 bending
1378.32 - - C-CH3 bending
1166.19 - - C-C stretching
1060.97 - - C-CH3 bending
719.09 718.56 - Alkynes (triple C bonds)
6.4 Degradation Mechanisms for Polyethylene

Thermal degradation mechanism of PE has been compared to that of PP. The
degradation pattern of PE can be interpreted in terms of a chain-reaction mechanism in
which the rupture of weak links occurs to a very low extent. Polyethylene in inert
conditions is stable up to ~ 290°C. The material started to degrade above this
temperature to lower molecular weight and the product resembles hard polyethylene
waxes [Madorsky, 1964). The continuous transition of the samples occurred from the
physical state solid to sequentially viscous, plastic state, intermediary state, into a thick,
oily liquid with partial separated gaseous products [Raff and Allison, 1956]. The
consequences of sample transitions were similar to reactions observed in this study
(Figure 6.31).
Higher temperature pyrolysis resulted in greater fragmentation products [Madorsky,

1964]. Volatilization may start at weak points or defects or near the end of the
macromelecular chain, followed by chain scissions which result in the reduction of
molecular mass [Blazso et al., 1995]. Time factor enables secondary reactions, to
further depolymerise the fragmented products into smaller molecules. Heavier
fragments were predominant at the lower temperatures of pyrolysis, whereas the smaller
ones were significant at higher temperatures. This could be explained on the assumption
that two competing reactions are involved at elevated temperatures (Figure 7.6).
209
Free radicals are unzipped into monomers. At lower temperature, the degradation
procedes by only reaction I, so that no monomers are found among degradation
products. This can be explained as the fragile point of molecular chain was eliminated at
this relatively low temperature process [Blazso et al., 1995]. At higher temperatures
degradations proceeds by the two competing reactions I and II, and as a result the yield
of monomers is considerable [Madorsky, 1964].
H H H H H H H
iiii i i
Reaction I -c-c-c-c
111! ~-c-c-c=c
I t I
+ H-C—C~AW
t t
H H H H H H H H H
H H H H H H H H H
i i i
t «
i i i u
Reaction n *~-C-C-C-C— —C — C -► 2v~C-C-C
lit) i i i i i
H H H H H H H H H
Figure 6.42: Two possible reaction at elevated temperature (400-600°C)
Gao and his co-workers [2003] also stated that as the temperature increases, the
fragment products which can evaporate under prevailing conditions also increase. They
pointed out that the degradation temperature influences the size of volatile products.
Polyethylene degrades via random chain scission which does not depend on the heating
rate, but higher heating rate may lead to degradation at higher temperatures. The
decomposition of polyethylene is easier as it involves the scission reaction of numerous
carbon-carbon bonds, thus producing huge amount of gaseous products which contain
C3 and C4. Blazso et al. [1995] have also done similar work and reported the
decomposition steps of polyethylene.
The cleavage of thermal decomposition of PE may start at the branching point.

Cleavage that occurs near to the end of the macromolecular chain will produce volatiles.
These reactions may have taken place during decomposition of PE. The yield of the
products by Blazso and his co-workers [1995] was also similar to the product yields
observed in this study where as temperature increased around 600°C, more alkanes and
alkenes were produced.
210
Decomposition of PE depends strongly on the rate of heating, mass or momentum

transfer in the polymer and its physico-chemical properties. The applied external heat
melts the PE and thermal decomposition reaction produces volatiles in the molten PE.
At a critical concentration of volatiles in the melt, their supersaturation was attained and
bubbles may begin to nucleate. The nucleated gas bubbles in the PE melt grow rapidly
until the gas pressure in the melt is in the equilibrium with that within the bubbles. The
growing gas bubbles travel to the surface of PE melt and get evaporated [Kim and
Kavitha, 2006].
Cleavage occurs at weak points of defects, such as impurities or near the chain end:
H HH H H H H H H
i i i i i i . I i u
*~~C —C-C-C—C—'» ---------fr A<~'C —C —C “C -C
it i i i till I
H HH H H H H H H H
H HH H H H H H H H
1 i i t i iii# it
•v~~C -*C —C —C —C fr *v~C ~C —C + -C—C ~-v
ill i i III II
H H H H H H H H H H
Termination by disproportionation:
H H H H H
I • I I i i
—C—C + *C —c=c
I I I
~~~*
III I
H H H H H H H
Backbitting (intramolecular radical transfer)

H H
i • i •
**~~C “C =C ”C
till
— c—c
till
—C ■— C
H H H H H H H H
Or, it may also combination of two or more reaction such as disproportionation:

H H H H H H
• • ii i » i
*'~~C ~~C~C—C + ^~C ~C --------fr* C~“C*~C=:C“wV + c =c
I t 1 I II 1 I I I II
HHHH HH HHHH HH
Figure 6.43: Plausible degradation mechanisms suggested by Scheirs and Kaminsky
[2006]
211
6.5 Comparison on the Effect of Heat Treatment on PE and

PP
Table 6.13 shows comparison between PE and PP, effects of mixing with PC,
temperature and time on these polyolefin polymers’s behaviour.
Table 6.13: Comparison between PE and PP behavior
Interactions of Polymer with PC substrate

PP PE
• PP started to soften, melt and • PE started to soften, melt and
agglomerate after 15 min of 200°C agglomerate after 30 min of 150°C
heat treatment of heat treatment.
• PP was totally melted after 60 min at • PE was totally melted after 60 min
200°C of heat treatment. at 250°C of heat treatment.
• PP begins to melt even after 15 min as • PE begins melting after 15 min only
temperature increased to 300°C and at 350°C of heat treatment.
350°C. • PE takes longer time to melt and
• PP devolatizes through bubbling and flow as it has higher viscosity than
blistering activity, which was PP. It did not undergo bubbling and
enhanced with increasing temperature blistering during its devolatization.
and time. • Highest contact angle was measured
• Highest contact angle was measured after 15 min of 250°C of heat
after 15 min of 200°C of heat treatment (131.50° contact angle).
treatment (137.80° contact angle). The The wetting improved with time and
wetting improved with time and temperature, and the lowest value
temperature, and the lowest value was was obtained after 60 min of 350°C
obtained after 60 min of 350°C heat heat treatment (30.90° contact
treatment (12.55° contact angle). angle).
• The adhesion and penetration depth • The adhesion and penetration depth
also improved as time and temperature also improved as time and
were increased. The adhesion and temperature were increased. The
penetration depth showed up to 1.98 adhesion and penetration depth
J/m value and 73 p.m, respectively. showed about 1.85 J/m value and 75
jam, respectively.
212
Polymer and PC Mixtures

• PP mixtures showed relatively higher weight loss as compared to PE mixtures.
• The carbon content of PP mixtures and PE mixtures residue were comparable to
each other, which ranged from 92% to 97% in 40% and 50% polymer mixtures.
• Both PP and PE were still remaining after Cycle 2 of heat treatment (600°C), even
though the TGA result showed decomposition of PP completed around 490°C.
• XRD results on both PP and PE showed that the polymer structures had became
amorphous after Cycle 2 heat treatment.
• FTIR analysis showed that PP functional groups were lost after Cycle 2. Some of
PE peaks were still remaining until Cycle 3.
____________________________Pyrolysis of Polymer_________________________
• PP showed relatively lower wt% yield of residues as compared to PE.
• However, PP has higher carbon % in its residues as compared to PE residues.
• SEM micrographs showed that PE took higher temperature and longer time (30
min at 650°C) to break down into smaller particles, while PP polymer began to
break down after 30 min at 400°C pyrolysis.
6.6 Summary
Unlike PP, PE has a higher viscosity that limits its flow and penetration, where it has
been shown that PE melted at higher temperature, 250°C after 60 min residence time as
compared to PP, where the polymer started to melt during 200°C heating. The contact
angle between the melt and substrate surface showed that the even though the PP has
relatively higher angles during lower temperature heating (200°C), but as temperature
and time were increased, the contact angle between PP melt and PC substrate had
gradually decreased and became smaller as compared to the contact angles measured on
PE melt. As mentioned earlier, PE has high viscosity that is influenced by temperature
and time of the heat treatment. At the highest treatment parameters, of 350°C
temperature and 60 min time, PE melt has flattened on the substrate with contact angle
of 30.90°, while PP of 12.55°. The higher contact angle showed by PE was measured by
its melt, without much disruption, while the bubbling and blistering of PP melt has
caused it to form thin layer on the substrate surface.
213
The study on wettability of PE on PC has shown deep penetration after 30 min heat
treatment at 350°C, where it reached about 75 pm. The heat treatment carried out at
lower temperature (up to 300°C) only resulted in about 12-33 pm range of penetration.
PP on the other hand reached its maximum penetration (73.30pm) only after 60 min of
350°C heating. It is possible that PP had more active decomposition, which is indicated
by the presence of voids, bubbles and blister in the melt, while PE has a higher viscosity
that limits its decomposition on the surface but tended to flow and penetrate deeper as
seen in samples heated at 350°C.
The interaction behaviour between PE and PC was further investigated by the mixing
PE in 10-50% with PC and subjected the mixtures into heating cycle’s heat treatment.
The treatment of Cycle 1 (150°C) has showed no changes in mass loss, while heating of
Cycle 2 (600°C) has showed some traces of PE, where the percentage of loss are about
33% and 43% in 40% and 50% PE mixtures, respectively. However, high temperature
heating of Cycle 3 (up to 1000°C), has totally decomposed PE in the mixtures. This
result is in agreement with SEM images where the EDS analysis detected the presence
of PE untill Cycle 2 (600°C), and was absent after Cycle 3. The carbon content of the
mixtures residues was in the range of 92-97%, with the decreasing value with
increasing time and temperature.
The effect of heat treatment on the structural crystallization of their residues was
investigated by XRD analysis. PE has high crystalline peaks at 21.63° (110) and 23.70°
(200), where both of these peaks were only visible in Cycle 1, but decomposed and not
visible after Cycle 2 and Cycle 3. The (110) peak was topped with C peak (002) at this
stage. New peak was seen emerging around 42.13° giving the crystal plane of (100),
however, after Cycle 2 and Cycle 3, another peak was observed around 43.10° with
(015) plane. As can be seen, PE peak has totally disappeared, leaving the C (002) peak
originated from raw PC and another two new C peak in its profile.
Apart from XRD, the residues of 50% PE blend were also analysed by FTIR, to
investigate the effect of blend, temperature and time on the chemicals bonding of its
residues. PE is composed of many functional groups, some of which were reduced in
intensity and became broad with increasing time and temperature. The CH2 symmetric
214
vibration at 2850cm'1 was gone after the mixture, while C-CH3 vibration around
1380cm'1 was lost after Cycle 2. The reduced intensities and broader peak occurred after
the heat treatment indicates that the temperature has decomposed those functional
groups, and they gradually dissappeared after progressive heating.
Thermal degradation of PE was also investigated to understand the mechanisms of the

PE behaviour in the presence of PC. The pyrolysis of PE was carried out in the
temperature range from 600°C to 750°C, within time ranging from 5 to 30 min. The
residue yields after 30 min of 650°C heating appeared sticky and yellowish in color,
however after 10 min in 650°C heating, the white colour of PE still remains and it
seems hard but brittle as compared to the sample heated longer (30min) at lower
temperature, 600°C. Longer holding time (20 min to 30 min) however, has charred the
PE polymer completely, yielding black residues after the pyrolysis. Similar outcome has
been observed as temperature was increased to 700°C, where residues collected after 5
and 10 min were yellowish and sticky, while the increment time to 20 min and 30 min
has resulted in a black charred powder. As temperature was increased to 750°C, the
sample started to char as early as 10 min holding time, showing the PE has decomposed
at lower holding time at higher temperature. Each of the residues collected was weighed
and plotted.
The yield has indicated that PE lost about 83% of its mass after heating at 600°C for 30
min. Time also played an important role as it was observed that PE lost about 60% of its
mass after 10 min pyrolysis carried out at 650°C. The increase of experiment time to 20
and 30 min had completely decomposed PE and gave 0.38% and 0.06% residues,
respectively. Similar trends were observed in other temperature series (700°C and
750°C). The LECO analysis showed that carbon content in each residue were reduced
corresponding to an increase of time and temperature of pyrolysis, as the residues
collected also reduced with the increasing time and temperature. The analysed result
shows measured carbon from the hydrocarbon that is produced from incomplete PE
decomposition, which resulted in a carbon value range between 25-50% of the total
mass residue.
215
XRD analysis showed that the increasing temperature has major effect on the crystalline
structure of PE. The analysis on the residues of raw PE, 650°C, 700°C and 750°C were
compared. The high crystalline PE peaks were gradually reduced and gone after 750°C.
The increasing of time also showed a similar effect, as can be seen in the comparison of
raw PE with residues of 5, 10 min and 20 min at 750°C pyrolysis. The PE peaks were
completely gone after 10 min of pyrolysis, indicating the decomposition of PE had
completed.
The increasing temperature also has an effect on the pyrolysis residue of PE. The
pyrolysis of PE after 10 min has resulted the loss of C-C and C-H, both vibrated at
11.66 and 1060cm'1, respectively after 650°C, while the vibration of C=C (around
720cm'1) has been reduced after 750°C. However, new C=0 bonding has been detected
(around 1730cm"1) after the pyrolysis, probably occurred as a results of reaction with O
while sample being cooled. The increment of time at 750°C has showed more
significant effect on PE residues. The increasing temperatures also had a significant
impact on PE residues, as can be seen in 750°C pyrolysis. The intensities of C-CH3
bending, C-C stretching and C-H bending vibrations has dramatically reduced even
after 5 min of pyrolysis, and many more others peak vibrations also diminished after 20
min of pyrolysis, indicating that most of the smaller molecular chain generated during
the heating, has been decomposed to amorphous char residue as seen in XRD results.
216
Chapter 7: Results & Discussion - Further Studies on Additional Polymers
CHAPTER 7

INFLUENCE OF POLYMER TYPE
217
In depth investigations carried out on thermoplastic polymers PP and PE have been

presented in Chapters 4 to Chapter 6. Similar experiments were also performed on a
number of additional polymers to compare the effect of specific polymer characteristics.
Additional thermoplastic polymer, polyethylene terephthalate (PET) was chosen for
interactions with PC substrate studies to investigate the effect of oxygen-containing
polymer on the polymer melt wetting properties. The degradation mechanism of PET is
dominated by intramolecular exchange reaction which produces short chain fragments
with carboxylic and unsaturated ester end groups. Furthermore, even though it has
considerably higher melting temperature it has been reported that PET degraded in very
narrow range of temperature [Murty et al., 1996]. These differences made interesting
comparison as PP produces more liquid product and has wide range of degradation
temperature.
The effect of addition of thermoset polymer, Bakelite on the blend with PC was carried
out for heating cycle treatment section. The heat treatment on the mixture of polymer
and petroleum coke studies the effect of petroleum coke and the amount of carbon yield
in their residues. Bakelite was selected as this thermoset does not melt when heated,
thus will not be able to flow into petroleum coke pores like PP.
In the third section, the thermal degradation study on another type of PP (PPB) is
presented. PPB was selected as polypropylene polymer may differ from one to another
according to their molecular weight, tacticity and additives, and PPB is black
pigmented. Therefore, as waste polymer contains various types of additives and
pigments, their effect on the thermal degradation is investigated. Additional studies on
these polymers are expected to help in developing a better understanding of their
behaviour and recycling potential.
7.1 Interactions of PET with PC Substrates

Figure 7.1 shows the sample (PET on petroleum coke) used in the experimental work.
The effect of heat treatment on the wettability and contact angles of PET melt on
petroleum coke has been presented in this section for a range of heating times and
temperatures.
218
Chapter 7: Results & Discussion — Further Studies on Additional Polymers
n111 m 1111 n | m 11111! 11111111111111II m ipm

Cfl V a 3* H* 5
Raw PET on petroleum coke substrate
Figure 7.1: Raw PET on petroleum coke substrate before heat treatment
u) Wetting behaviour at 150°C

Figure 7.2: PET on petroleum coke substrate after 15, 30 and 60min of heating at
150°C
No melting of PET was observed at I50°C. Therefore no further investigations were

carried out at this temperature.

200°C
Once again, PET had not melted at 200°C as well. Therefore no further investigations
were carried out at this temperature as well.
219

250°C
PET has a higher melting point than PP and PE. Figure 7.2 to Figure 7.4 display the
effect of heating at 150°C, 200°C and 250°C, respectively. These figures clearly show
that up to 250°C temperature had no effect on the interfacial behaviour since the raw
samples remained intact on the petroleum coke substrate.
Figure 7.5: PET on petroleum coke substrate and the corresponding SEM micrograph
Significant changes were observed after heating to 300°C. Figure 7.5 shows the effect
of heating for 15 minutes, and the PET is beginning to start melting. The transparent
PET melt appeared to contain gaseous bubbles due to rapid degassing of PET associated
with the release of volatiles from PET. The bubble formation increased the height of
PET melt (0.92 mm) and the contact angle was determined to be 61.35°. The SEM
image showed a smooth surface of PET phase, with few voids and two large cavities in
PET, including a large one at the boundary.
220
The effect of heating at 30 minutes is presented in Figure 7.6. The bubbles formed in
the melt were smaller in size but in higher numbers compared to the samples heated for
15 minutes. Even though the size of the bubbles decreased after heating for 30 minutes
heat treatment, it was foaming rapidly and the accumulation of bubbles made up had a
marginal influence on the contact angle of the molten PET measurements. The contact
angle of the sample heated for 30 minutes was 60.10°, which was almost the same value
as that measured after heating for 15 minutes sample (61.35°). The decrease in the size
of the bubbles from inside of PET may be due to the longer heating time which caused
bigger bubbles to burst, leaving smaller bubbles to develop and grow. SEM image
shows big cracks in the petroleum coke substrate underneath the PET melt. This large
crack was probably caused by the penetration of PET into petroleum coke.
After 60 min of heating, many more small bubbles appeared inside the PET melt
(Figure 7.7). Bubble formation indicated rapid foaming at this temperature and time.
The shape of the PET melt had also become more spherical and smaller. This spherical
shape had a contact length of 9.36 mm, which was higher than the lengths seen for
shorter times (6.10 mm and 30 min at 4.64 mm after 15 minutes and 30 minutes of
heating).
221
SEM image shows the penetration of PET into the petroleum coke along the edges.
Several small voids and large cavities were observed in the polymer phase. In addition,
the top of the PET surface showed traces of the bursting big bubbles formed in the
polymer. The contact angle was about 38.15°, which was significantly lower than the
sample heated for shorter time.
As the temperature was further increased to 350°C, the colour of the PET became
whitish and opaque and the surface appeared to be flaky and blistered (Figure 7.8),
possibly caused by the rapid bubbling in the PET. SEM image showed the presence of
several voids of various sizes. Besides that, at the interface, cracking of the substrate
underneath the PET melt was observed, which is expected to improve the PET melt
penetration into petroleum coke. The length of contact was 7.24 mm with a contact
angle of 59.20°.
After 30 min (Figure 7.9), the colour of the melt turned brownish which was much
darker than the sample heated for 15 min (Figure 7.9). The rapid foaming of bubbles
could be seen inside the melt. This bubbles appeared to be picking up petroleum coke
particles from the substrate. The SEM image showed several large cavities in the PET
phase, as well as a number of smaller voids.
The large cavities represent traces of bursting bubbles. The contact length was higher
(7.74 mm) and the contact angle was lower (54.40°). A further lowering of the contact
angle was expected as more and more volatile matter was released after 30 minutes,
leaving behind a reduced amount of the melt (Figure 7.9).
222
The colour of the melt changed significantly after 60 minutes of heating to an even
more brownish colour (Figure 7.10). Bubbling caused flakiness and blistering in the
melt and thin layers of disintegrating bubbles were seen. As the bubbles were forming
rapidly, the melt became less viscous and spread out across the substrate. The contact
length was up to 10.1 I mm and the contact angle was low (23.85°). SEM image shows
the presence of voids of various sizes on the surface of PET.
f) Wetting behaviour at 400°C

The effect of heating at 400°C is presented in Figure 7.11 - Figure 7.13 for different
times. The colour of the melt significantly changed after this heating stage. The melt
colour appeared darkish brown and was composed of small bubbles packed together in
the melt. The melt had also swelled up (contact length 9.22 mm) on the substrate
surface. Flowever, bubbling produced a relatively high contact angle of 50.80°,
compared to the other samples heated for longer times. This temperature has appeared
to provide sufficient heat for PET to depolymerise and degrade (Figure 7.11). There
were no visible boundaries between the polymer and petroleum coke region, which may
indicate high penetration of PET into the petroleum coke substrate.
223
Rough surface of the melt deposit was observed after 30 min (Figure 7.12). The melt
present across the petroleum coke surface had turned black. The jagged surface was due
to the rapid release of the bubbles. SEM image shows that only a thin layer of polymer
was present, making it hard to distinguish the boundary line.
A black deposit with traces of bubble released from the melt was seen after 60 minutes
(Figure 7.13) and the melt was fused with the substrate. The length of the contact
region was 6.03 mm with a contact angle of 13.87°. Some PET had penetrated into the
petroleum coke region.
224
Results for the PET melt on petroleum coke substrate are summarized in Table 7.1. The
wettability of PET polymer with petroleum coke (measured by contact angle) has been
plotted in Figure 7.14. The characteristics of the melt had a significant influence on the
length spread on the substrate. The forming of bubbles and swelling led to some
inconsistencies in the calculated length of the spread. The same also affected height of
the melt due to the varying sizes of the bubbles formed.
The contact angle of the PET melt with petroleum coke decreased with increasing time
and temperature. The contact angles were relatively high after 30 min. This may due to
the increased of bubbling and blister formation during 30 min, as compared to 15 min
where the bubbles had just begun to form. The contact angles then decreased
significantly after 60 min. During this period, large quantities of gaseous products were
released from the melt.
Table 7.1: Variations in the dimensions of the PET melt on PC substrate with time and
temperature

Height Contact angle
Temperature Time cross-sectioned
(mm) (°)
(°C) (min) (mm)
300°C 15 6.10 0.92 61.35
30 4.64 0.88 60.10
60 9.36 0.95 38.15
350°C 15 7.24 1.46 59.20
30 7.74 1.47 54.40
60 10.11 1.00 23.85
400°C 15 9.22 1.74 50.80
30 7.86 0.14 45.85
60 6.03 0.08 13.87
225
1.5
Time (min)
A 300°C X 35CTC
Figure 7.14: Effect of time and temperature on (a) contact angles and (b) adhesion of
the PET melt on PC substrate
7.1.1 Interfacial Phenomena between PET and PC

Figure 7.15 shows the EDS analysis of different regions in the interface between PET
and petroleum coke. EDS result could not differentiate between these materials. The
interfacial changes in the PET/petroleum coke system after heating at 300°C, 350°C and
400°C have been presented in Figure 7.16 to Figure 7.18, respectively.
2500- Point 1
...
1500-
I......
'.MU
keV
Point 2 Point 3
.... 2000-
1500- 1500-
1000 1000-
500-
Figure 7.15: EDS analysis of points at the PET/PC interface of the samples heated for
15 minutes at 300°C
226
The effect of heating at 300°C on the interfacial region of PET with petroleum coke is
presented in Figure 7.16. PET was seen to penetrate deeply (42.34 pm) after 15 minutes
of heating. As can be seen in Figure 7.16-a, the penetration and spreading of PET into
the petroleum coke substrate was extensive.
As time was increased to 30 min, the penetration of PET increased to 56.79 pm. Figure
7.16-b shows the cracks present in the petroleum coke region. PET melt was seen to
cover petroleum coke particles. The magnified image shows part of petroleum coke
particles that were coated by the PET melt. The penetration of PET had increased to
71.20 pm after 60 minutes. Figure 7.16-c shows the penetration of PET into the gaps
between the petroleum cokes.
Figure 7.16: Interfacial region after heating at 300°C for different time intervals
227
The effect of heating at 350°C on the interfacial behaviour of PET on petroleum coke
samples is presented in Figure 7.17. There were no clear boundaries between PET and
petroleum coke, as the polymer seemed have covered most of the petroleum coke
particles. This is clearly demonstrated in Figure 7.17-c. Interfacial surface of PET
appeared smooth, but a huge cavity was seen in the Figure 7.17-a (top) and fibrous PET
surface was seen in the exposed area.
After 30 min, the penetration of PET was about 235.77 pm. Figure 7.17-b shows the
cracking of petroleum coke and here the PET has filled up the cracks and penetrated
into the petroleum coke. Similar behaviour was seen after 60 minutes heating in Figure
7.17-c. The depth of penetration increased significantly to 493.03 pm. The solidification
of polymer increased the surface tension, and has caused the separation of the attached
outer layer of the petroleum coke substrate.
228
After heat treatment at 400°C, PET sample showed complete wetting and infiltration of
polymer into petroleum coke. It was hard to distinguish the individual petroleum coke
particle edges as these had been covered by PET. Figure 7.18-a to Figure 7.18-c
represent sample after 15 minutes, 30 minutes and 60 minutes, respectively. All these
showed small cracks in the petroleum coke region. PET had penetrated deep through the
substrate and the depth of penetration was measured to be 1650 pm, similar to that seen
in the samples heated for 15 minutes to 60 minutes.
The average penetration depths of PET into petroleum coke are summarized in Table
7.2, and the data is plotted into Figure 7.19.
Depth of penetration increased with increasing time and temperature. The depth of
penetration changed significantly after heat treatment at 350°C for 30 min (235.77 pm).
After 60 min, the value increased to 493.03 pm. The increment in temperature to 400°C
229
also increased the penetration of PET, where samples heated for 15 to 60 minutes
showed a penetration of depth of 1650 pm. This value is estimated to be the maximum
possible penetration.
Table 7.2: Depth of penetration of PET melt into PC substrate
Experimental conditions
Penetration Depth (pm)
Temperature (°C) Time (min)
300°C 15 42.34
30 56.79
60 71.20
350°C 15 70.74
30 235.77
60 493.03
400°C 15 1650.00
30 1650.00
60 1650.00
Time (min)
A 300°C X 350°C • 400°C
Figure 7.19: Variations in penetration depth of PET melt with time and temperature of
heat treatment
230
7.1.2 Discussion
PET has a higher melting point compared to the other two polymers, PP and PE. PE and
PP have melting temperatures of 136°C and 170°C, respectively [National Fire and
Protection Engineers, 1995]. In this study, PP started to soften at 200°C during its 15
min heat treatment, but melted completely only after 30 min. PE on the other hand,
started to soften at temperatures as low as 150°C after 60 min of holding time, but took
up to 300°C and 30 min to melt completely. PET only began to melt after 300°C within
15 min of heat treatment even though it did not show any change at 250°C after 60 min
of exposure. This higher melting temperature is in agreement with the suggestion by
Jabarin and Lofgren [1984], in their study on the PET degradation that the melting
temperature of PET was in the range of 282-320°C. The abrupt change of degradation
of PET is a result of random chain splitting at the ester bond (C-O bonding). This is due
to C-O having a lower strength than the C-C bond [Jabarin and Lofgren, 1984;
Madorsky, 1964].
The rapid softening followed by bubble foaming in the PET samples induced by
temperature accelerated reduction of the melt viscosity. PET, which is a semicrystalline
polymer, has an unstable and continuously evolving microstructure in the temperature
region between equilibrium melting temperature and glass transition point. Under these
conditions, the flow of melt, which is closely related to the viscoelasticity is highly
corresponding to the temperature during heat treatment. However, it also can vary
depending on its thermal history. Moreover, time also plays major role as it is needed
for heat transfer, allowing the PET polymer to melt and start to flow [Buckley and
Salem, 1987].
Literature on the sensitivity of degradation product of PET on temperature, heating rate,

time, sample drying prior degradation and atmosphere condition has been reported by
various researchers [Murty et al., 1996; Dzi^ciol and Trzeszczynski, 2001; Villain et
al., 1995]. These studies have explained the significant change of PET degradation
samples after increasing time and temperature of the heat treatment. PET also showed
rapid formation of bubbles on the lateral surface, as well as the formation of various
voids from the blistering activity within the melts. Due to rapid foaming, the melt had
become less viscous and was able to penetrate and flow deeply through petroleum coke
231
particles into the substrate. Increasing temperature to 350°C had further enhanced the
degradation rates as shown by the increasing contact angle, which is more pronounced
after 30 min of heating. The increasing contact angle, caused by the bursting of bubbles
was reduced after 60 min.
The extensive degradation also caused PET to change its colour, first to cream as
showed by samples after 15 min, and then to brown as showed after 30 of reaction at
350°C. The samples heated for 30 min onwards after 400°C turned black. This colour
change was in agreement with literature and was reported to have a molecular weight of
570±16 with the formula of C34H32O8 [Madorsky, 1964]. This discoloration led by
temperature change is a part of property transformation that includes reduction in
molecular weight and intrinsic viscosity and the emission of volatile compounds. The
studies showed that the main substances evolved from PET are acetaldehyde and
formaldehyde.
Volatiles substances emitted consist of carbon oxide, aldehydes, aliphatic C1-C4

hydrocarbons, aromatic hydrocarbons (benzene, toluene, and styrene), esters,
acethophenone and methyl alcohol, which vary depending on the evolved substances
masses on the degradation temperature. The emission of various volatiles explains the
rapid bubbling activity of heat treated samples and the formation of pale yellow smoke
during the reaction. The degradation products of PET may also vary depending on the
degradation technique used. The short residence time of plastic in a hot zone gives only
primary degradation products, while the longer residence time may allow secondary
degradation reaction [Dzi^ciol and Trzeszczynski, 2000; Dzi^ciol and Trzeszczynski,
2001].
The mechanisms of PET defluidization, from the pyrolysis reaction can be explained by
four stepwise processes, beginning with the softening of the polymer surface, adhesion
to the petroleum coke substrate, followed by volatile release and ending with formation
of aggregates kept together by the carbon residue [Scheirs and Kaminsky, 2006]. PET
had completely decomposed after 30 min of reaction at 400°C, leaving traces of char on
the substrate. This is in the agreement with Marco et al. [2002] and Williams and
232
Williams [1997], which reported that around 18% and 15.6 %, of char collected after
the degradation of PET, respectively.
The solid black residue is an organic carbonized product (char) that formed during
pyrolysis due to the secondary repolymerization reactions between the products derived
from the depolymerisation. The repolymerization tendency increases if there are groups
capable of reacting with hydrogen atoms of the aromatic nuclei, such as -OH or =0. In
the case of PET, the formation of char can be explained by the doubly substituted
aromatic nucleus of the PET structure and by the presence of the =0 of the ester group
of the polymer [Marco et al., 2002].
The SEM micrographs showed highly penetrated substrate after 400°C of heat
treatment. In this case, the PET melt that had flowed in and was trapped inside the
petroleum coke substrate may account for more of solid mass compared to that reported
in the literature due to the limited heat transferred inside the petroleum coke which can
slow down further decomposition that might occur from secondary reactions of PET.
7.2 Interactions between Bakelite and PC for Temperatures

up to 1000°C
The effect of the addition of a thermoset polymer, Bakelite into PC blend was studied
and has been presented in this section. Bakelite was added in 40-50% concentration
range and was subjected to baking cycle’s heat treatment. The baking cycle heat
treatment details are: Cycle 1 - heating up to 150X1 and let dwell for 30 min; Cycle 2 -
heating up to 150 °C and let dwell for 30min, then heating up to 600 °C and let dwell for
30min; Cycle 3 - heating up to 150 °C and let dwell for 30min, then heating up to 600 °C
and let dwell for 30min, and then heating up to 1000 X and let dwell for 30 min. As
thermoset plastics are well known to char directly without having a softening and
melting phase, its thermal properties are expected to be quite different from those of PP
and PE. Bakelite also has a higher decomposition temperature compared to
thermoplastic polymer.
233
7.2.1 Weight Loss Study of Bakelite and PC Mixtures

The weights of residues collected after the heat treatment (Cycle 1, 2 and 3) of the
mixture of bakelite and petroleum coke (40:60 and 50:50) were calculated. The results
are presented in Table 7.3 and plotted in Figure 7.20.
Table 7.3: Percentage of solid yield of each sample mixtures (Bakelite:PC) obtained
after baking cycle
Bakelite Heating Initial Weight Final Weight

Loss (%)
(wt%) Treatment (g) (g)
Cycle 1 7.13 7.05 1.21
40 Cycle 2 7.11 5.93 16.60
Cycle 3 7.11 5.46 23.18
Cycle 1 6.45 6.27 2.72
50 Cycle 2 6.44 5.20 19.27
Cycle 3 6.46 4.90 24.17
The results obtained showed trends similar to other polymers mixed petroleum coke
presented earlier in Chapter 4 and Chapter 5. The increment of heating cycles
(corresponding to the increment of temperature), has reduced the extant of residues after
the treatment. Both mixing ratios (40 and 50% of Bakelite) showed a similar behavior.
The ratio of polymer used in the mixture had a great impact on the residues collected,
where the lower percentage of Bakelite (higher PC content) has resulted in higher
residues. This is due to the fact that Bakelite was decomposed under the heat treatment,
while the corresponding temperature has no effect on the PC.
However, the data on the residues left showed that a fraction of Bakelite was still
present after various heat treatment cycles. Residues of 83.40% and 76.82% were
retained respectively for the 40:60 mixtures after Cycle 2 and 3 of heat treatments. This
result reveals that even after 1000°C (Cycle 3), the remainder from Bakelite (initially
40%) was around 16%. Higher percent of Bakelite was seen after Cycle 2 (23%) due to
lower temperatures. As for 50:50 mixing ratio, the Bakelite fraction was still remaining
about 30% after Cycle 2 and 25% of its initial mass after Cycle 3. This behavior is
attributed to the basic characteristics of the thermoset polymer, which tends to directly
go through a charring process without melting and depolymerisation steps. This
charring process generates burnt char, which is collected as solid residues at the end of
heat treatment.
234

Heat treatment
O 40/60 X 50/50
Figure 7.20: Residue (%) of Bakelite:PC mixtures after heat treatment
7.2.2 Carbon Content of Bakelite and PC Mixtures

The carbon content (%) of the mixture after the heat treatment was determined and
analysed. These results plotted in Figure 7.21 show a trend similar to that observed for
thermoplastic polymer mixtures (PP and PE), where the higher ratio of polymer
(Bakelite) yielded higher carbon content of the mixtures. This is due to the
devolatilazation of Bakelite by the heat treatment. The devolatilazation produced char,
thus the higher Bakelite ratio, resulted in more char being produced. The char amount
has increased the carbon to weight ratio, thus increased the carbon percentages
determined.
The increasing number of heating cycles increased temperature applied in the analysis
[Cycle 1 (150°C) < Cycle 2 (600°C) < Cycle 3 (1000°C)]. Higher temperatures
permitted rapid devolatization, which resulted in higher char residues. The higher char
residues thus increased the percentages of carbon to weight ratio in the sample.
Table 7.4: Carbon content (%) of Bakelite:PC mixtures

Bakelite (wt%) Heating Treatment Carbon Content (%)
Cycle 1 69.1
40 Cycle 2 76.7
Cycle 3 82.0
Cycle 1 71.0
50 Cycle 2 78.4
Cycle 3 84.0
235

Heat treatment
040:60 X 50:50
Figure 7.21: Carbon content (%) of Bakelite:PC (50:50) residues after heat treatment
7.2.3 Structural Characterizations of Bakelite and PC Mixtures

The residues of samples after heating cycles were studied by XRD to reveal any
structural change in them or their components. The XRD spectrums of raw PC and
Bakelite and their mixture after the heat treatment are presented in Figure 7.22 and
Figure 7.23, respectively and characterizations of peaks detailed in Table 7.5.
40000
Raw PC
30000
20000
^ 10000
£>40000
Raw Bakelite
- 30000
20000 *v /*“*\
o co
10000 o O Ci
10 20 30 40 50 60
Figure 7.22: XRD profiles of raw PC and Bakelite

236
As can be seen in Figure 7.22, raw PC profile included a high intensity C (002) peak at
26.03°. Another weak peak of graphite (101) is also seen at 44.64°. As for raw Bakelite
used in this study, the profile reveals a high CaCC>3 (104) peak at 29.44°(used as filler).
Besides, it has many other small peaks corresponding to CaC03: (012) at 23.09°; (110)
at 36.01°; (113) at 39.44°; (018) at 47.53°; (116) at 48.52°; and (122) at 57.44°. In
addition, the raw Bakelite also contained another small peak of SiC>2 (011) at 26.66° and
(200) at 43.19°.
40000
30000
gv
o
r—^
20000 C?
O, Q gN oo
o 'w'
V—✓ •— <N
r—
B 1
o
3\l
10000 O' O' o'
003 o<53 CJ
CO oCO
o u 33 o u r)
A -A- ,.AA
40000
Cycle 2 -up to 600°C
f 30000
O3
.£ 20000
C/5
/----> /'—s
OO VO
§ *—•
r7
cS* I r**>
oo
ro
" 10000 oCO u uu

o § co
ou
a
Q
■ A—A ■ . AA.
40000
30000
20000
10000
0
10 20 30 40 50 60
Figure 7.23: XRD profiles of Bakelite and PC mixtures sample after heat treatment
237
The effect of mixing and heat treatments (Cycle 1-Cycle 3) on the crystallization
structure of the mixtures has been presented in Figure 7.23. The mixing and heating of
PC and Bakelite caused some shifting and overlapping of the peaks, e.g. the CaCC>3
peak (012) of Bakelite located at 23.09° had appeared on the graphite shoulder peak of
(002) peak at 25.96°. Moreover, the SiC>2 peak (Oil) located at 26.66° in raw Bakelite
profile also was not apparent as it has been masked by high C (002) peak from raw PC
sample. Other peaks of original component remained at their original locations. Similar
pattern were observed after Cycle 1 and Cycle 2 heat treatment.
After Cycle 3, all of the CaC03 peaks from raw Bakelite components had decreased.
However, the Si02 (200) remains stable at this stage. Similar phenomenon has been
detected for the C (002) originated from raw PC, as temperature as high as 1000°C did
not has any impact on PC structure. Several new peaks of CaO (111), (200) and (220)
have been detected after the high temperature heating of Cycle 3. The new CaO peak
may have resulted from the reaction of CaC03 releasing CO2 and leaving CaO as
residues.
238
Table 7.5: XRD peaks characteristics of raw PC, Bakelite and their mixtures at 50:50*
Sample Heating Peak d-spacing Mil er indices Possible
Cycle Position (A) h k 1 Compound
(20) (chemical
formula)
Raw PC 26.03 3.37 0 0 2 C (graphite)
44.64 2.03 1 0 1 C (graphite)
Raw 23.09 3.86 0 1 2 CaCOs
Bakelite 26.66 3.26 0 1 1 Si02
29.44 3.04 1 0 4 CaC03
36.01 2.50 1 1 0 CaC03
39.44 2.29 1 1 3 CaC03
43.19 2.06 2 0 0 Si02
47.53 1.91 0 1 8 CaC03
48.52 1.88 1 1 6 CaC03
57.44 1.61 1 2 2 CaC03
Mixtures Cycle 1 23.09 3.86 0 1 2 CaC03
50:50 25.96 3.37 0 0 2 C (graphite)
29.44 3.04 1 0 4 CaC03
36.03 2.50 1 1 0 CaC03
39.47 2.29 1 1 3 CaC03
43.19 2.06 2 0 0 Si02
47.50 1.91 0 1 8 CaC03
48.56 1.88 1 1 6 CaC03
57.44 1.61 1 2 2 CaC03
Cycle 2 23.04 3.86 0 1 2 CaC03
25.84 3.37 0 0 2 C (graphite)
29.40 3.04 1 0 4 CaC03
36.00 2.50 1 1 0 CaC03
39.43 2.29 1 1 3 CaC03
43.18 2.06 2 0 0 Si02
47.49 1.91 0 1 8 CaC03
48.50 1.88 1 1 6 CaC03
57.46 1.61 1 2 2 CaC03
Cycle 3 26.09 3.37 0 0 2 C (graphite)
32.19 2.78 1 1 1 CaO
37.32 2.41 2 0 0 CaO
42.94 2.06 2 0 0 Si02
53.82 1.70 2 2 0 CaO
*Additional peaks have been bolded
239
7.2.4 FTIR Analysis of Bakelite and PC Blends
Raw Bakelite
o 0 .
Wavenumbers (cm1)
Figure 7.24: FTIR peaks of raw Bakelite and PC
The FTIR spectrum of raw samples has been plotted in Figure 7.24 and the possible
chemical reactions of mixing due to the heating investigated has been presented in
Figure 7.25. Various peaks and their corresponding possible functional groups are
listed in Table 7.6. There are several missing peaks as well as new ones coming up due
to the mixing and heat treatment of the samples.
Raw Bakelite showed a couple of vibrations of CH2 asymmetric and symmetric

stretching at 2917cm'1 and 2847cm'1 and the mixing and heat treatments had reduced
their intensities with increasing temperature. However, the C-C in aromatic ring
(1511cm1) peak has been diminished after the mixture. On the other hand, the C-H2
identified around 1235cm'1 has been reduced after heating of Cycle 2. Some peaks
originally identified in raw PC showed a similar behaviour. Couple of C=C stretching
peaks occurring at 1685cm'1 and 1653cm'1, had decreased after mixing with Bakelite.
240
Wavenumbers (cm'1)
Figure 7.25: FTIR peaks as observed in the 50:50 mixtures after various cycles of heat
treatment
241
Table 7.6: Characterization of FTIR peak profiles of PC, Bakelite and their mixtures at
50:50
Raw PC Raw Cycle 1 Cycle 2 Cycle 3 Possible Functional Groups

Bakelite
3484.12 3485.73 3479.03 3467.22 3480.51 Moisture
2913.98 2917.94 2917.13 2925.88 2914.57 CH2 stretching
- 2847.85 2845.33 2849.13 2847.18 CH2 symmetric
H , CH
=C ^
c c
1653.83 - - - - C=C (alkene absorption)
- 1610.48 1617.57 1617.92 1622.11 C=C stretching
- 1511.03 - - - C-C stretch in aromatic ring
1457.52 1432.13 1425.14 1419.84 1458.10 CH3 bending
- 1235.75 1232.68 - - c-h2
- 1 168.85 1166.19 1 166.19 1166.19 C-C stretching
1060.47 - - - - C-CH3 bending
893.93 874.11 874.84 872.32 893.91 Aromatic ring
7.2.5 SEM Images of Bakelite and PC Mixtures
Figure 7.26: SEM micrographs and EDS of raw Bakelite
Figure 7.26 shows the SEM image of raw Bakelite as well as the EDS results on an
overall scan of the surface. EDS showed high intensities of carbon and calcium (Ca) and
a small peak of oxygen (O). The Ca peak occurs due to the presence of CaCC>3 filler in
Bakelite.
The effect of mixing of raw Bakelite and PC (50:50) and heating cycle treatments on the
mixture samples is illustrated in Figure 7.27. The sharp layering structure of PC makes
the identification of each raw component easier after the mixing process Figure 7.27-a.
242
The low temperatures of Cycle 1 had no impact on the sample, thus the image is just
showing each of the original components of the mixture. The SEM image shows some
PC chunks on the grainy Bakelite particles.
After Cycle 2, the size of Bakelite has been reduced. The SEM image shows small
Bakelite residues on the PC (Figure 7.27-b). The heat treatment of Cycle 3 has further
reduced the size of Bakelite particles. The SEM image shows traces of Bakelite on PC
surface as confirmed by EDS result that revealed the presents of Ca peaks.
(c)
Figure 7.27: SEM micrographs and EDS of Bakelite:PC (50:50) mixtures after heat
treatment
243
7.3 Thermal Degradation of Polypropylene Black (PPB)

Another polypropylene, which is black in colour (PPB) was obtained from Sigma-
Aldrich and was also investigated to study how degradation depends on polymer
property. The difference of two types of polypropylene PP and PPB investigated are
their colour and physical states. The PPB contains carbon black as a pigment and its
implications on the degradation are discussed later. Further details along with the high
resolution pictures were given in Chapter 3.
7.3.1 Analysis of Solid Residues of PPB Pyrolysis

The pyrolysis yields of solid residues of PPB are presented in Table 7.7, and their
dependence on temperature has been plotted in Figure 7.28. Pyrolysis of PP at low
temperature range from 300 to 400°C up to 30 min showed no weight changes, which
remained as high as 92% of weight residue. However, after 60 min (at 400°C), the
weight residue has showed a reduction of more than 25%.
Significant changes started to take place at 420°C. A similar trend was also noted at
440°C, where each increment of time has nearly reduced half of the wt.% of its
residues. Significant decrements were observed at higher temperatures, 480°C and
500°C, where pyrolysis held for 10 min showed yields of only 47% and 18% of solid
residues, respectively.
Table 7.7: Solid residues (wt.%) of PPB
Solid Yield (%)

Temperature
Pyrolysis Time (min)
(°C)
10 20 30 60
300 99.87 99.73 99.56 99.27
320 99.86 99.73 99.53 99.22
340 99.84 99.63 99.18 98.53
360 99.64 99.11 98.59 97.15
380 99.13 98.05 96.58 92.44
400 99.50 96.02 91.61 75.89
420 97.83 85.34 68.50 13.50
440 95.12 59.25 23.52 3.08
460 84.02 5.97 3.04 2.99
480 47.63 2.76 3.04 3.00
500 18.66 3.18 2.85 2.89
244
Temperature ( C)
O 10 min ■ 20 min O 30 min X 60 min
Figure 7.28: Effect of temperature on solid yield (%) of products of PPB
7.3.2 Degradation of PPB

A TGA study was carried out to investigate the step degradation reaction of PPB. The
TGA profile showed a one step reaction of degradation in the range of 290°C to 550°C.
This result was similar to that observed for PP degradation range discussed previously
in the section 4.2.2. There was however differences between these two materials as have
been presented in Figure 7.29, where effect of time appears to be quite important.
0 100 200 300 400 500 600 0 100 200 300 400 500 600
Temperature (°C)
Figure 7.29: TGA profiles of PP and PPB
245
Both of PP and PPB started to loose their weight at same point, 290°C. However, PPB
has a longer decomposition range compared to PP, where it took up to 580°C for the
completion, while the PP completed its decomposition around 490°C. This behaviour
can be presumably caused by the more complex polymeric chains possessed by PPB
that required more time to decompose.
Further studies on the behaviour of PPB need to be conducted as it may have better
characteristics for wetting properties, as it can seen that it has longer resilience and may
be able to provide better interactions with PC.
7.3.3 Morphology Study of PPB Residues
(a) (b)
Figure 7.30: SEM micrographs of raw PPB samples (2mm particles) at (a) low
magnification and (b) high magnification
The surface morphology of raw PPB sample is displayed in Figure 7.30 and the effect
of temperature and time presented in Figure 7.31 and Figure 7.32, respectively. The
residues at 300, 400 and 500 °C for 30 min pyrolysis were compared to determine the
effect of temperature on structures. The surface residue has started to split into smaller
pieces at 400°C, and the increase of temperature to 500°C had damaged the smooth
plasticity of PPB into a rough surface.
This behaviour is similar to that reported in PP pyrolysis where the increasing of
temperature has significantly degraded PP into smaller fragments and reduced the
polymer into smaller size particles.
246
300 °C 400 °C 500 °C

Figure 7.31: SEM micrographs of PPB residues - effect of temperature for 30 min
Figure 7.32 shows the change of the morphology of PP and PPB samples with pyrolysis
time at 500°C. The basic structural morphology of PPB apparently appeared to be quite
similar at various times. PPB was able to retain agglomeration of its particles up to 60
min of pyrolysis even though the surface roughness indicates that degradation had
occurred due to release of volatiles. This phenomenon has proven the long degradation
durability of TGA result presented in Figure 7.29.
The long decomposition time possessed by PPB may be due to the pigment employed in
the polymer. As PPB is black in color, the pigment normally employed is carbon black,
which also can act as a UV stabilizer, anti-oxidants, thermal insulator, weathering
resistance, reinforcing filler [Horrocks et al., 1999; Hawkins et al., 1971; Newland et
al., 1969|. These thermal stabilizer properties have somewhat prolonged the thermal
decomposition of PPB, as Hawkins et al. 11971J have reported that black pigment is the
most stable at lower temperatures (80-140°C). Generally pigments are stable at melting
temperature of polymer [Zeisberger, 1973|. It is also reported that carbon black with
low volatile content, small particle size and high structure improved the thermal
stability of polymer significantly. Moreover, literature also suggested that carbon black
particles nucleate additional crystallization sites, resulting in smaller crystalline domains
with narrower size distribution, as observed in Figure 7.32 where PPB has more
homogeneous particle size distributions as compared to PP (Miraftab et al., 2002;
Horrocks et al., 1999],
In addition of that, low particle size of carbon black provides high specific area for
interaction with polymer, which results in higher secondary structures and increased
agglomerate size. This phenomenon leads to immobilized molecular polymer segments
as they are trapped in this secondary structure. In this state, the degradation rate of PP is
247
slowed down since the mechanism is driven by chain scission. These secondary
structures are broken down only at high temperatures [Miraftab et al., 2002; Medalia,
1970].
After 20 min
Figure 7.32: SEM micrographs of yield products of PP compared to PPB - effect of

time on morphology of pyrolysis residues after 500°C
248
7.4 Summary
1. Unlike PP and PE, PET started melting only at 300°C. PET sample also showed a
strong tendency to bubble during the heating. This impacted the sample shape and
size, as these tended to blister and swell up. The swelling also affected the contact
angles. The contact angles calculated at 300°C were 61.35°, 60.10° and 38.15° after
15, 30 and 60 minutes of heat treatment respectively. The bubbling increased rapidly
after 300°C and further intensified at 350°C heat treatment. The production of
bubbles at 350°C extensively changed the colour and properties of the PET melt.
The blistering effect caused the peeling of the sample surfaces. The contact angle
seen after heating for 15, 30 and 60 minutes were 59.20°, 54.40° and 23.85°. The
effect of bubbling slowly decreased and ceased at 400°C. The presence of lesser
number of bubbles resulted in lower contact angles of 50.80°, 45.85° and 13.87° after
heating for 15, 30 and 60 minutes.
Both temperature and time clearly impacted the wettability of polymers with
petroleum coke. PET sample displayed a much deeper penetration compared to PP
and PE. Even though PET started to melt at higher temperatures than PP and PE, the
penetration depths for the corresponding temperature were very high. This is due to
the rapid bubbling of volatiles released has reduced significantly its melt viscosity,
thus enabling it to go through the petroleum coke particles deeper, get trapped inside
the substrate and stop further decomposition that might occur from secondary
degradation reaction.
2. The effect of the addition of Bakelite on the polymer-PC blend was investigated in
this study. As Bakelite has higher decomposition temperature, there were significant
amount of Bakelite residues left even after the Cycle 3 (up to 1000°C) heat treatment.
In Cycle 2, 40% and 50% Bakelite mixtures had loss about 16.60%, and about 23%
of its initial weight, respectively. After Cycle 3, results show there were almost 27%
of Bakelite residue still remained in 40% mixture, and about 26% of Bakelite residue
still remained in the 50% mixtures with PC.
Higher ratio of Bakelite also yielded higher carbon content (%) of the mixtures.
Increasing the heat treatment cycle promotes higher decomposition of Bakelite.
249
Decomposition of Bakelite produced chars and thus increased the carbon percentages
of the mixtures. The range of carbon percentages of the mixtures varies from 69 to
84%, which simultaneously showed the high ash content. This result has been
confirmed in XRD analysis, where the profile revealed the presence of CaCC>3 in the
raw Bakelite sample. The high temperature of Cycle 3 was found to generate CaO
from the decomposition of CaCOs.
The FTIR analysis shows the increasing temperature (heating cycle 1 to 3) has
reduced the intensities most of the peaks present in raw Bakelite sample. However,
there were traces of Bakelite observed after Cycle 3, as indicated by the C=C
stretching at 1622cm'1. SEM images have also showed the presence of Bakelite in
the mixtures after up to Cycle 3 heat treatment.
3. The thermal degradation of PP was compared to PPB. PPB was charged into the
furnace under temperature range from 300 to 500°C, in 10, 20, 30 up to 60 min
residing times. PPB began to loose about 25% of its mass after 400°C in 60 min
pyrolysis, while the increment of temperature up to 420°C by the same holding time
has decreased about 86% of its mass. PPB started to loose significantly about 50% its
mass after 480°C in 10 min time. Further time and temperature increase beyond this
point had completely decomposed PPB. The TGA profile showed one step
decomposition process, similar to trend showed by PP. The decomposition started to
occur after 290°C and was completed around 580°C.
The increasing time and temperature has increased the breaking and separation of
their polymeric surface, causing the smaller molecules clustered together and formed
agglomerates. The size of the particles became smaller as time and temperature of
pyrolysis were increased. However, PPB exhibits longer decomposition time as
compared to PP. This may be attributed by the presence of carbon black as a pigment
in PPB. Carbon black may delay the degradation of PPB as its interactions with PPB
polymer generated the secondary structures that slowdowns the rate of degradation.
The homogeneous particle size distribution structures of PPB also may be conveyed
by carbon black as it was suggested that it carbon black provides additional
nucleation sites which has smaller crystalline domains.
250
Table 7.8 summarizes the comparison between polymers used in this study along with
the effect of properties of each polymer on the process and residues characteristics
during the heat treatment.
Table 7.8: Comparison of polymers properties during the heat treatment
Interactions of Polymer with PC substrate

PP PE PET
• PP starts to soften, • PE started to soften, • Nothing happened to
melt and agglomerate melt and agglomerate PET up till 250°C of
after 15 min of 200°C after 30 min of 150°C heat treatment.
heat treatment. of heat treatment. • PET had totally melted
• PP had totally melted • PE had totally melted after 15 min at 300°C
after 60 min at 200°C after 60 min at 250°C of heat treatment.
of heat treatment. of heat treatment
• Highest contact angle • Highest contact angle • Highest contact angle
(137.80°) measured at (131.50°) measured at (61.35°) measured at
15 min of 200°C heat 15 min of 250°C heat 15 min of 300°C heat
treatment. treatment treatment
• Lowest contact angle • Lowest contact angle • Lowest contact angle
(12.55°) measured at (30.90°) measured at (13.87°) measured at
60 min of 350°C of 60 min of 350°C of 60 min of 400°C of
heat treatment. heat treatment. heat treatment.
• Highest adhesion • Highest adhesion • Highest adhesion
(1.98 J/m3) and (1.85 J/m3) and (1.97 J/m3) and
penetration depth (73 penetration depth (75 penetration depth
pm) measured at 60 pm) measured at 60 (1650 pm) measured at
min of 350°C heat min of 350°C heat 15 min of 400°C heat
treatment. treatment. treatment.
Polymer and PC Mixtures

40% PP - Cycle 2 (5.41% 40% PE - Cycle 2 (7.01% 40% Bakelite - Cycle 2
residue) Cycle 3 (1.99% residue) Cycle 3 (4.88% (23.40% residue) Cycle 3
residue) residue) (26.82% residue)
251
50% PP - Cycle 2 (2.51% 50% PE - Cycle 2 (6.90% 50% Bakelite - Cycle 2

residue) Cycle 3 (0.38% residue) Cycle 3 (1.28% (30.73% residue) Cycle 3
residue) residue) (25.83% residue)
Bakelite showed highest yield % residues after the heat treatment of the polymers
and PC mixtures. This is to be expected as Bakelite is a thermoset polymer that

decomposes as a char without the depolymerisation process. However, Bakelite
also contained up to 10% ash that may account for the high residues yield after the
heating.
• Carbon content of the • Carbon content of the • Carbon content of the
residue ranged from residue ranged from residue ranged from
92% to 98% of the 92% to 97% of the 69% to 84% of the
mixtures mass (10% to mixtures mass (40% mixtures mass (40%
90% of PP). and 50% of PE). and 50% of Bakelite).
• Bakelite had the
lowest carbon content
% and high ash
content.
• Both PP and PE were still remaining after Cycle 2 • Bakelite still
of heat treatment (600°C), even though the TGA remained up till Cycle
result showed decomposition of PP was completed 3.
around 490°C.
• XRD results on both PE and PC showed that the • XRD result showed a
polymer structures became amorphous after Cycle significant level of ash
2 heat treatment. (CaCC>3) present in
Bakelite.
Pyrolysis of Polymer
PP PE PPB
• Relativity of residue yield % of three polymers : PPB < PP < PE
• TGA profile of PPB showed a one step reaction of degradation in the range of
290°C to 550°C, while PP also showed a one step reaction of degradation that
started at 290°C and completed at 490°C.
252
• The presence of carbon black as a pigment in PPB was seen to delay the
degradation process of PPB, while PE exhibited highly viscosity behaviour
that slowed down its degradation process.
• SEM micrographs showed that PE required higher temperature and longer time
(30 min at 650°C) to break down into smaller particles, while PP and PPB
polymers began to break down after 30 min at 400°C pyrolysis.
253
Chapter 8: Summary and Conclusions
CHAPTER 8
SUMMARY AND CONCLUSIONS
254
8.1 Summary and Conclusions

In depth studies have been carried out on the wettability and interactions of polymers
with petroleum coke to understand the effect of temperature and heating time. These
have been supplemented with studies on polymer degradation kinetics and mechanisms.
This research was carried out over a wide temperature range in order to investigate and
determine optimum characteristics for recycling waste polymers in carbon anodes.
Key conclusions from this study are summarized below:
1. Significant differences were observed in the wetting characteristics of three

thermoplastic polymers. Figure 8.1 shows the comparison of the wettability of three
different types of polymers (PP, PE and PET).
150
O 100
a>
*3)
a
| 50
g
O
200°C X 250°C □ 400°C
Figure 8.1: Variations in contact angles for three different polymers (PP, PE and PET)
with petroleum coke as a function of temperature and time
These polymers generally showed a good wetting with petroleum coke. Increasing time
and temperature resulted in melting, flowing and subsequent spreading of polymer on
the substrate and penetrating into the substrate. Wettability was seen to improve as
temperature was increased to 400°C. As seen from Figure 8.1, the relative reduction in
contact angles for PP and PE as a function of temperature was highest at 15 min and this
decrease became relatively smaller with increasing times. An opposite trend was
observed in case of PET. Increasing time and temperature had a significant influence on
the viscosity of these polymers.
255
PET
Time (min)
□ 200°C O 250°C A 300°C X 350°C • 400°C
Figure 8.2: Average penetration depths of three different polymers (PP, PE and PET)
into petroleum coke as a function of temperature and time
Figure 8.2 compares the penetration depth of PP, PE and PET as a function of time. It
can clearly be seen that time and temperature played an important role in penetration
depth of polymers into PC substrate. While PE had relatively lower penetration than PP
during lower temperature treatment, but their magnitudes became similar after 350°C.
For PET, the maximum temperature was reach at 400°C.
2. PP was observed to soften and melt gradually with increasing temperature and
the wettability improved at longer holding time and started to melt after 60 minutes of
heating at 200°C (M.P of PP: 164°C). The bubbling and blistering of PP was seen to
enhance the decomposing reaction which also impacted its spreading onto the surface
and penetration into the substrate of PC. The degradation of PP produced high
percentage of liquid and gases that varied according to the time and temperature, as the
longer residence time aided secondary degradation reactions of PP degradation.
3. The soaking of PP blended with PC has showed negligible weight loss up to

350C heating. This result shows that PC had helped PP retaining its residue, as PC
contains liquid phase produced by PP during its degradation process. The further heat
treatment on the soaked samples showed more than polymer mass percentage that may
be attributed by loss of binder. The soaked sample at 300°C showed almost 2% of
residue after 600°C heat treatment but sample soaked at 350°C showed total loss after
the similar treatment.
256
4. PE has slightly higher boiling point than PP. PE still retained its raw particulate
form after 200°C experiments, even though it had started to soften in as low as 150°C
heating. Compared to PP, PE has higher viscosity that reduced its ability to spread
across the PC surface. PE also showed relatively higher contact angle, which indicates
poorer wetting of PC, but it was found to improve with increasing temperature and time
of the heating. The penetration of PE also improved significantly only after 350°C
temperatures, while the depth of penetration was low at lower temperature. The
behaviour shown by PE somewhat indicated that PE required more energy for its
decomposition as compared to PP.
5. PET on the other hand was observed to start melting only after 300°C. However,
the presence of oxygen in the polymer has facilitated the rapid thermal decomposition
of PET. Even though PET started to melt at higher temperatures than PP and PE, the
penetration depths for the corresponding temperature were very high. The 400°C
heating has showed optimum penetration of PET into PC substrate. Rapid bubbling of
volatiles released has reduced significantly its melt viscosity, thus enabling it to go
through the petroleum coke particles deeper, get trapped inside the substrate and stop
further decomposition that might occur from secondary degradation reaction.
6. Therefore, it can be concluded that each of these polymers showed differences in

their wettability and interfacial phenomena with petroleum coke substrates. PP tended to
melt and wet PC at low temperature, but PE in the other hand required higher energy
and therefore higher temperatures before it could start to flow efficiently. PET has been
recorded as highest melting polymer, but the once it started to flow, it decomposed
rapidly and produced a runny liquid/solid chars aided by the presence of oxygen in the
polymer.
7. Co-pyrolysis of petroleum coke with three different polymers namely, PP, PE

and Bakelite was also carried out. Various heating cycles determined the effect of
isothermal heating with low heating rates in inert conditions for plausible reactions
between polymers and petroleum coke. These results are summarized below.
257
8. Cycle 1 (up to 150°C) heat treatment had no impact on any of the polymers, as
the temperature was too low to cause softening and melting of polymer. This yield of
residue has also been confirmed by SEM images as the polymeric component (PP, PE,
Bakelite) were seen scattered among PC particles after Cycle 1. After Cycle 2 (up to
600°C), the polymers started to melt and devolatize. Thermoset Bakelite has the highest
yield of residues after heating. Thermoplastics PP and PE behaved similarly and their
thermal degradation products consisted of liquid, gas and solid. Bakelite produced solid
char on heating. Only Bakelite was still remaining after heating Cycle 3 (up to 1000°C).
9. Thermal degradation behaviour and decomposition of PP is expected to be

complete around 450°C, but interestingly, some polymer component was still present in
the PP:PC mixtures for temperature up to 600°C during Cycle 2. This behavior indicates
that the addition of petroleum coke particles has inhibited the decomposition of
polymer. The presence of polymers (PP, PE and Bakelite) in their mixtures were
showed by SEM images and also confirmed by XRD analysis. The heat treatment has
significantly reduced the polymeric components of the mixture samples, remarkably
after Cycle 3. Lower polymer ratio has yielded higher residues, but higher ratio gave
slightly higher carbon percent of the overall residues.
10. The yield of residues decreased with increasing temperature and time for all
thermoplastic polymers studied. Additional investigations on pyrolysis of polymer PP
was carried out for a better understanding of the degradation. The PP residue color
changed from white to brown and became relatively dark with increasing time and
temperature, until eventually it became a black powder after 850°C and 5 min of
pyrolysis, thereby indicating PP had completely decomposed and charred. PP started to
lose around 30% of its mass after 20 min in 450°C pyrolysis, and then its loss increased
to 70% after 30min. The increasing time and temperature speeded up the decomposition
rate of PP, where PP lost about 95% of its mass after 30min of 500°C pyrolysis.
11. The pyrolysis of thermoplastic polymers (PP and PE) at lower temperatures
yielded wax and heavy liquid product that solidified at room temperature. The
increasing time caused further decomposition of polymer chains that yielded lighter
liquids and gaseous products. On the other hand, pyrolysis at higher temperature range
yielded higher proportions of gas and solid, indicating high conversion rates. The XRD
258
results showed a decreasing crystalline peak with increasing time and temperature. The
FTIR analysis showed a relative increase in the intensity of several functional group
peaks as a function of pyrolysis time and temperature. This was attributed to the
generation of several smaller molecules containing associated groups. However, as the
temperature was increased, most of the vibration peaks were seen to decrease and
reduced in intensity, indicating the completing of polymer decomposition. Four main
ash impurities found in the polymers residues were Fe2C>3, Si02, AI2O3 and MgO.
12. The morphology of PP was also affected by the pyrolysis conditions, where the
PP particles were broken down into smaller sizes, and then agglomerated as PP has a
tendency of sticking together. The charred residues after 850°C at 5 min pyrolysis had
generated 1.53pm powder particles.
13. The kinetics of degradation reactions in TGA analysis on PP showed a single

step degradation process that started at 290°C and was completed at 486°C, leaving
behind no solid residue. The TGA profile of PPB, another form of PP, also showed a
one step decomposition process. The decomposition started to occur after 290°C and
complete around 550°C. The heating rates shifted the degradation temperature profile as
the product of degradation varied according to time and temperature factors. The
heating rates also influenced the activation energy of PP; higher activation energy was
needed at lower heating rate (10°C/min) then that observed for higher heating rates (15
and 20°C/min). This trend was explained as lower heating rate generating heavier
(longer) polymer molecules that needed more energy for degradation process.
14. It can be concluded from this study that among PP, PE, PET and Bakelite, PP
had lowest level of thermal stability. It started to melt and decompose much faster than
PE and PET. It’s rapid bubbling and blister formation also contributed to the loss of
polymer, as compared to PE that had higher viscosity and took longer time to melt and
spread on the PC surface. This observed phenomenon also is in agreement with heat
treatment on mixtures studies and thermal decomposition studies, where PE showed a
higher melting point compared to PP. The incorporation of PET however indicated a
deep penetration into the PC particles. Bakelite can act as carbon resource, but ones
need to carefully select Bakelite polymer as it generally contains several fillers that may
affect the final product qualities.
259
8.2 Future Studies

The research reported in this thesis provides the basic framework required for recycling
waste plastics. Further investigations need to be carried out to understand and to
enhance the potential of polymers as carbon sources and binder.
• Thermoset polymer may provide carbon but is not expected to act as a binder as it
showed only charring with increasing temperature. Therefore, it may be possible to
utilize a blend of thermoset and thermoplastics together as thermoplastics are known
to decompose at lower temperatures to form lower chain hydrocarbons, which can
act as a binder in carbon anodes.
• The combination of more than one type of polymer has showed higher residues
yield based on the literature reported. Therefore, PP, PE, PET and many other types
of polymer could be combined, for enhancing properties such as durability during
heating, binding properties and compatibility with PC and binder.
• The effect of coal-tar pitch needs to be further investigated as it is the primary

binder used in anode baking process nowadays. Further investigations need to be
conducted on coal-tar pitch-polymer interactions in order to understand and to
compare the potential of polymers to partially substitute as a binder. The
compatibility between these three main materials namely petroleum coke, polymer
and coal-tar pitch needs to be fully determined.
• The kinetics of reaction influences the polymer degradation behavior. This study has
clearly shown that waste polymers can be successfully used as a carbon and binder
resource. Future studies will be conducted to optimize the time and temperature
correlation during the interaction with petroleum coke.
260
Chapter 9: References
CHAPTER 9
REFERENCES
261
Aboulkas A., El harfi K. and El Bouadili A. (2010) "Thermal degradation behaviors

ofpolyethylene and polypropylene. Part I: Pyrolysis kinetics and mechanisms." Energy
Conversion and Management 51(7) pp. 1363-1369
Achillas D. S., Roupakias C., Megalokonomos P., Lappas A. A. and Antonakou V.

(2007) "Chemical recycling of plastic wastes made from polyethylene (LDPE and
HOPE) and polypropylene (PP)." Journal of Hazardous Materials 149(3) pp. 536-542
Aguado J., Serrano D. P., Miguel G. S., Escola J. M. and Rodriguez J. M. (2007)
"Catalytic activity of zeolitic and mesostructured catalysts in the cracking of pure and
waste polyolefins." Journal of Analytical and Applied Pyrolysis 78(1) pp. 153-161
Albalak R. J., Tadmor Z. and Talmon Y. (1990) "Polymer melt devolatilization

mechanisms." AIChE Journal 36(9) pp. 1313-1320
Angyal A., Miskolczi N. Bartha L. (2007) "Petrochemical feedstock by thermal

cracking of plastic waste." Journal of Analytical and Applied Pyrolysis 79(1-2)
pp. 409-414
Asanuma M., Ariyama T., Sato M., Murai R. and Sumigama T. (1997)
"Combustion and gasification behaviors of plastics injected into raceway of blast
furnace." Tetsu-to-hagane 83( 10) pp. 617-622
Ashraf G. S. “Environmental Waste Management and Plastics Recycling-An

Overview ” Brunei University, Londres
Attar A. (1978) "Bubble Nucleation In Viscous Material Due To Gas Formation By A

Chemical Reaction: Application To Coal Pyrolysis." AIChE Journal 24(1) pp. 106-115
Bhaskar T., Uddin M. A., Murai K., Kaneko J., Hamano K., Kusaba T., Muto A.
and Sakata Y. (2003) "Comparison of thermal degradation products from real
municipal waste plastic and model mixed plastics." Journal of Analytical and Applied
Pyrolysis 70(2) pp. 579-587
262
Bhaskar T., Kaneko J., Muto A., Sakata Y., Jakab E., Matsui T. and Uddin M. A.
(2004) "Pyrolysis studies of PP/PE/PS/PVC/HIPS-Br plastics mixed with PET and
dehalogenation (Br, Cl) of the liquid products." Journal of Analytical and Applied
Pyrolysis 72(1) pp. 27-33
Bhaskar T., Tanabe M., Muto A., Sakata Y., Liu C., Chen M., and Chao C. C.
(2005) "Analysis of chlorine distribution in the pyrolysis products of poly(vinylidene
chloride) mixed with polyethylene, polypropylene or polystyrene." Polymer Degradation
and Stability 89(1) pp. 38-42
Blazso M., Zelei B. and Jakab E. (1995) "Thermal decomposition of low-density

polyethylene in the presence of chlorine-containing polymers." Journal of Analytical
and Applied Pyrolysis 35(2) pp. 221-235
Bockhorn H., Hornung A., Hornung U. and Schawaller D. (1999-a) "Kinetic study
on the thermal degradation of polypropylene and polyethylene. " Journal of Analytical
Bockhorn H., Hornung A., Hornung U. and Jakobstroer P. (1999-b) "Modelling of

isothermal and dynamic pyrolysis of plastics considering non-homogeneous
temperature distribution and detailed degradation mechanism." Journal of Analytical
and Applied Pyrolysis 49(1-2) pp. 53-74
BP (2001) “Technical Services: How Calcined Petroleum Coke is Produced. ” BP

Coke: A Global Carbon Company Website 2001 Access on 19 July 2010 from
http://coke.bp.com/tech/tech.cfm
Brzozowska T., Zielinski J. and Machnikowski J. (1998,) "Effect of polymeric

additives to coal tar pitch on carbonization behaviour and optical texture of resultant
cokes." Journal of Analytical and Applied Pyrolysis 48(1) pp. 45-58
Buckley C. P. and Salem D. R. (1987) "High-temperature viscoelasticity and heat

setting ofpolyethylene terephthalate)." Polymer 28(1) pp. 69-85
263
Buekens, A. G. and Huang H. (1998) "Catalytic plastics cracking for recovery of

gasoline-range hydrocarbons from municipal plastic wastes." Resources, Conservation
and Recycling 23(3) pp. 163-181
Busscher H. J., van Pelt A. W. J., de Boer P., de Jong H. P. and Arends J. (1984)
"The effect of surface roughening ofpolymers on measured contact angles of liquids."
Colloids and Surfaces 9(4) pp. 319-331
Cao J., Buckley A. and Tomsett A. (2002) "Re-examining the pitch/coke wetting and
penetration test." JOM Journal of the Minerals, Metals and Materials Society 54(2) pp.
30-33
Ciesinska W., Zielinski J. and Brzozowska T. (2009). "Thermal treatment of pitch-

polymer blends." Journal of Thermal Analysis and Calorimetry 95(1) pp. 193-196
Collin G., Bujnowska B. and Polaczek, J. (1997) "Co-coking of coal with pitches and
waste plastics." Fuel Processing Technology 50(2-3) pp. 179-184
Conesa J. A., Font R, Marcilla A. and Garcia A. N. (1994) "Pyrolysis of

Polyethylene in a Fluidized Bed Reactor." Energy & Fuels 8(6) pp. 1238-1246
Couderc P., Hyvernat P. and Lemarchand J. L. (1986) "Correlations between ability

of pitch to penetrate coke and the physical characteristics of prebaked anodes for the
aluminium industry." Fuel 65(2) pp. 281-287
Coutinho F. M. B., Costa T. H. S. and Daisy L. C. (1997) "Polypropylene-wood fiber

composites: Effect of treatment and mixing conditions on mechanical properties."
Journal of Applied Polymer Science 65(6) pp. 1227-1235
Cranmer J. H., Plotzker I. G., Peebles Jr, L. H. and Uhlmann D. R. (1983) "Carbon
mesophase-substrate interactions." Carbon 21(3) pp. 201-207
264
Curlee R. T. (1986) "Plastics recycling: Economic and institutional issues."

Conservation & Recycling 9(4) pp. 335-350
Czernik S., Elam C. C., Evans R. J., Meglen R. R., Moens L. and Tatsumoto K.
(1998) "Catalytic pyrolysis of nylon-6 to recover caprolactam." Journal of Analytical
Delattre C., Forissier M., Forissier, M., Pitault I., Schweich D. and Bernard J. R.
(2001) "Improvement of the microactivity test for kinetic and deactivation studies
involved in catalytic cracking." Chemical Engineering Science 56(4) pp. 1337-1345
Di Blasi C. (1997) "Linear pyrolysis of cellulosic and plastic waste." Journal of

Analytical and Applied Pyrolysis 40-41pp. 463-479
Di Blasi C. (1999) "Transition between regimes in the degradation of thermoplastic

polymers." Polymer Degradation and Stability 64(3) pp. 359-367
Dominik, A., Chapman, Walter G., Swindoll R. D., Eversdyk D., Jog P. K. and
Srivastava R. (2009) "Compositional Polydispersity in Linear Low Density
Polyethylene." Industrial & Engineering Chemistry Research 48(8) pp. 4127-4135
Du L. and Qu B. (2006) "Structural characterization and thermal oxidation properties

of LLDPE/MgAl-LDH nanocomposites " Journal of Materials Chemistry 16 pp. 1549-
1554
Dzi^ciol M. and Trzeszczynski J. (2001) "Temperature and atmosphere influences on

smoke composition during thermal degradation of poly (ethylene terephthalate).”
Journal of Applied Polymer Science 81(12) pp. 3064-3068
Dzi^ciol M. and Trzeszczynski J. (2000) "Volatile products of poly (ethylene

terephthalate) thermal degradation in nitrogen atmosphere." Journal of Applied
Polymer Science 77(9) pp. 1894-1901
265
Ehrburger P. (1994) “Glassy properties of coal tar pitch materials” Energeia

University of Kentucky Center for Applied Energy Research, University of Kentucky 5
pp. 1-3
Ehrburger P. and Lahaye J. (1984) "Capillary flow of liquid pitch." Fuel 63(12) pp.
1677-1680.
Felix J. M. and Gatenholm P. (1991) "The nature of adhesion in composites of

modified cellulose fibers and polypropylene." Journal of Applied Polymer Science 42(3)
pp. 609-620
Feng Z., Zhao J., Rockwell J., Bailey D. and Huffman G. (1996) "Direct liquefaction
of waste plastics and coliquefaction of coal-plastic mixtures." Fuel Processing
Technology 49(1-3) pp. 17-30
Fuller T. R. and Fricke A. L. (1971) "Thermal conductivity ofpolymer melts." Journal

of Applied Polymer Science 15(7) pp. 1729-1736
Gambiroza-Jukic M. and Cunko R. (1992) "Kinetics of the thermal degradation of

polypropylene fibres." Acta Polymerica 43(5) pp. 258-260
Gao Z., Amasaki I. and Nakada M. (2003) "A thermogravimetric study on thermal
degradation ofpolyethylene." Journal of Analytical and Applied Pyrolysis 67(1) pp 1 -9
Garcia-Morales M., Partal P., Navarro F. J. and Gallegos C. (2006) "Effect of waste
polymer addition on the rheology of modified bitumen." Fuel 85(7-8) pp. 936-943
Hawkins W. L., Chan M. G. and Link G. L. (1971) "Factors influencing the thermal
oxidation ofpolyethylene." Polymer Engineering & Science 11(5) pp. 377-380
Hegberg B. A. (1992) Mixed Plastics Recycling Technology Noyes Publications
266
Hornsby P. R., Hinrichsen E. and Tarverdi, K. (1997) "Preparation and properties

of polypropylene composites reinforced with wheat and flax straw fibres: Part I Fibre
characterization. "Journal of Materials Science 32(2) pp. 443-449
Horrocks A. R., Mwila J., Miraftab M., Liu M. and Chohan S. S. (1999) "The
influence of carbon black on properties of orientated polypropylene 2. Thermal and
photodegradation." Polymer Degradation and Stability 65(1) pp. 25-36
Hotta H. (2003) "Recycling Technologies for Promoting Recycling-oriented Society."

NKK Technical Review(Japan) 88(88) pp. 160-166
Hsu C.-K. (2002) "Thermal decomposition properties of polymer fibers."

Thermochimica Acta 392-393 pp. 163-167
Huang Y., Bird R. N. and Heidrich O. (2007) "A review of the use of recycled solid
waste materials in asphalt pavements." Resources, Conservation and Recycling 52(1)
pp. 58-73
Iwaya T., Sasaki M. and Goto M. (2006) "Kinetic analysis for hydrothermal
depolymerization of nylon 6" Polymer Degradation and Stability 91 pp. 1989-1995
Jabarin S. A. and Lofgren E. A. (1984) "Thermal stability of polyethylene

terephthalate." Polymer Engineering & Science 24( 13) pp. 1056-1063
Kakuta M., Kohriki M. and Sanada, Y. (1980) "Relationships between the

characteristics ofpetroleum feedstocks and the graphitizability of the petroleum cokes."
Journal of Materials Science 15 pp. 1671-1679
Kaminsky W., Schlesselmann B. and Simon C. (1995) "Olefins from polyolefins and
mixed plastics by pyrolysis." Journal of Analytical and Applied Pyrolysis 32 pp. 19-27
267
Kato K., Fukuda K., Kondoh H., Nomura S. and Uematsu H. (2006) "Development
of Waste Plastics Recycling Process Using Coke Oven." Nippon Steel Technical Report
pp 75-79
Kim S. and Kavitha D. (2006) "Identification of Pyrolysis Reaction Model of Linear

Low Density Polyethylene (LLDPE)." Chemistry Letters 35(4) pp. 446-447
Kim S., Kavitha D., Yu T. U., Jung J. S., Song J. H, Lee S. W. and Kong S. H
(2008) "Using isothermal kinetic results to estimate kinetic triplet ofpyrolysis reaction
of polypropylene." Journal of Analytical and Applied Pyrolysis 81(1) pp. 100-105
Kim S., Kim Y.C., Kim Y.M. and Eom, Y. (2005). “Identification of Pyrolysis
Reaction Model of Polypropylene. ” Chemistry Letters 34(9) pp. 1268-1269.
Kiran C. N., Ekinci E., Snape C. E. (2004) "Pyrolysis of virgin and waste
polypropylene and its mixtures with waste polyethylene and polystyrene." Waste
Management 24(2) pp. 173-181
Lee K.-H. and D.-H. Shin (2007) "Characteristics of liquid product from the pyrolysis
of waste plastic mixture at low and high temperatures: Influence of lapse time of
reaction." Waste Management 27(2) pp. 168-176
Levitskiy S. P., Khusid B. M. and Shulman Z. P. (1996) "Growth of vapour bubbles

in boiling polymer solutions—II. Nucleate boiling heat transfer." International Journal of
Heat and Mass Transfer 39(3) pp. 639-644
Lewis I. C. (1982) "Chemistry of carbonization." Carbon 20(6) pp. 519-529
Luechinger N. A., Booth N., Greg H., Sri B., Robert N. G. and Wendelin J. S.
(2008) "Surfactant-free, melt-processable metal-polymer hydrid materials: use of
graphene as a dispersing agent." Advances Materials 20 pp. 3044-3049
268
Machado G., Kinast E. J., Scholten J. D., Thompson A., Vargas T. D., Teixeira S.
R. and Samios D. (2009) "Morphological and crystalline studies of isotactic
polypropylene plastically deformed and evaluated by small-angle X-ray scattering,
scanning electron microscopy and X-ray diffraction." European Polymer Journal 45(3)
pp. 700-713
Machnikowski J., Machnikowska H., Brzozowska T. and Zielinski J. (2002)

"Mesophase development in coal-tar pitch modified with various polymers." Journal of
Analytical and Applied Pyrolysis 65(2) pp. 147-160
Madorsky S. L. (1964) "Thermal Degradation of Organic Polymers” John Wiley &

Sons Inc.
Maier C. and Calafut T. (1998) “Polypropylene - The Definitive User's Guide and
Databook” William Andrew Publishing/Plastics Design Library
Marco I. d., Caballero B., Torres A., Laresgoiti M. F., Chomon M. J. and Cabrero
M. A. (2002) "Recycling polymeric wastes by means ofpyrolysis." Journal of Chemical
Technology & Biotechnology 77(7) pp. 817-824
Marin N., Collura S., Sharypov V. I., Beregovtsova N. G., Baryshnikov S. V.,
Kutnetzov B. N., Cebolla V. and Weber J. V. (2002) "Copyrolysis of wood biomass
and synthetic polymers mixtures. Part II: characterisation of the liquid phases." Journal
of Analytical and Applied Pyrolysis 65(1) pp. 41-55
Marsh H., Martinez-Escandell M. and Rodriguez-Reinoso F. (1999) "Semicokes

from pitch pyrolysis: mechanisms and kinetics." Carbon 37(3) pp. 363-390
Mcllveen-Wright D. R., Pinto F., Armesto L., Caballero M. A., Aznar M. P.,
Cabanillas A., Huang Y., Franco C., Gulyurtlu I. and McMullan J. T. (2006) "A
comparison of circulating fluidised bed combustion and gasification power plant
technologies for processing mixtures of coal, biomass and plastic waste." Fuel
Processing Technology 87(9) pp. 793-801
269
Medalia A. I. (1970) "Morphology of aggregates, : VI. Effective volume of aggregates

of carbon black from electron microscopy; Application to vehicle absorption and to die
swell offdled rubber." Journal of Colloid and Interface Science 32(1) pp. 115-131
Meille S. V., Bruckner S. and Porzio W. (1990) "y-Isotactic polypropylene. A

structure with nonparallel chain axes." Macromolecules 23(18) pp. 4114-4121
Milosavljevic I. and Suuberg E. M. (1995) "Cellulose Thermal Decomposition

Kinetics: Global Mass Loss Kinetics." Industrial & Engineering Chemistry Research
34(4) pp. 1081-1091
Min Z., Cai-Hong L., Lei M., Xiao-Mei W. and Zhi-lei Zheng (2009) "Estimate
research on co-carbonization of blend coal with waste plastics." Procedia Earth and
Planetary Science 1(1) pp. 807-813
Miraftab M., Horrocks A. R. and Mwila J. (2002) "The influence of carbon black on
properties of oriented polypropylene 3. Thermal degradation under applied stress."
Polymer Degradation and Stability 78(2) pp. 225-235
Miskolczi N., Bartha L. and Angyal A. (2006). "High Energy Containing Fractions
from Plastic Wastes by Their Chemical Recycling." Macromolecular Symposia 245-
246(1) pp. 599-606
Mitera J. and Michal J. (1985) "The combustion products ofpolymeric materials. Ill:
GC-MS Analysis of the combustion products of polyethylene, polypropylene,
polystyrene and polyamide." Fire and Materials 9(3) pp. 111-116
Mochida I., Yoon S.-H. and Qiao W. (2006) "Catalysts in syntheses of carbon and
carbon precursors." Journal of the Brazilian Chemical Society 17 pp. 1059-1073
Murty M. V. S., Rangarajan P., Grulke E. A. and Bhattacharyya D. (1996)

"Thermal degradation/hydrogenation of commodity plastics and characterization of
their liquefaction products." Fuel Processing Technology 49(1-3) pp. 75-90
270
National Fire, P. A. and S. O. F. Protection Engineers (1995) Thermal

decomposition ofpolymers SFPE Handbook of fire protection engineering. C. L. Beyler
and M. M. Hirschler. Boston, National Fire Protection Association: pp.1200
Newland G.C., George R. G. and John W. T. (1969). “Polyolefin Composition and

Degradable Films Made Therefrom ” E. K. Company. United States. 3454510: 4.
Oakes W. G. and Richards R. B. (1949) ”619. The thermal degradation of ethylene

polymers." Journal of the Chemical Society (Resumed) pp. 2929-2935
Onu P., Vasile C., Ciocilteu S., Iojoiu E. and Darie H. (1999). Thermal and catalytic
decomposition of polyethylene and polypropylene. Journal of Analytical and Applied
Pyrolysis 49(1-2) pp. 145-153.
Owens D. K. (1970) "Some thermodynamic aspects of polymer adhesion." Journal of

Applied Polymer Science 14(7) pp. 1725-1730
PACIA (2009). 2009 National Plastics Recycling Survey. Access on 5 February 2011
from http://www.pacia.org.au/DisplayFile.aspx?FileID=92
Panda A. K., Singh R. K. and Mishra, D. K. (2010) "Thermolysis of waste plastics to

liquid fuel: A suitable method for plastic waste management and manufacture of value
added products—A world prospective." Renewable and Sustainable Energy Reviews
14(1) pp. 233-248
Pattanakul C., Selke S., Lai C. and Miltz J. (1991) "Properties of recycled high
density polyethylene from milk bottles." Journal of Applied Polymer Science 43 pp.
2147-2150
Pearson J. R. A. (1978) "Polymer flows dominated by high heat generation and low
heat transfer." Polymer Engineering & Science 18(3) pp. 222-229
271
Perez M., Granda M., Garcia R., Santamaria R., Romero E. and Menendez R.
(2002) "Pyrolysis behaviour of petroleum pitches prepared at different conditions.”
Journal of Analytical and Applied Pyrolysis 63(2) pp. 223-239
Peterlin A. (1971) "Molecular model of drawing polyethylene and polypropylene.”

Journal of Materials Science 6(6) pp. 490-508
Peterson J. D., Sergey V., and Charles A. W. (2001) "Kinetics of the Thermal and
Thermo-Oxidative Degradation of Polystyrene, Polyethylene and Poly (propylene)."
Macromolecular Chemistry and Physics 202(6) pp. 775-784
Pinto F., Costa P., Gulyurtlu I. and Cabrita I. (1999-a) "Pyrolysis ofplastic wastes.
1. Effect of plastic waste composition on product yield." Journal of Analytical and
Applied Pyrolysis 51(1-2) pp. 39-55
Pinto F., Costa P., Gulyurtlu I. and Cabrita I. (1999-b) "Pyrolysis ofplastic wastes:
2. Effect of catalyst on product yield." Journal of Analytical and Applied Pyrolysis
51(1-2) pp. 57-71
Plastic Waste Mangement Institute (2004) "An Introduction to Plastic

Recycling ’’Report
Raff R. A. V. and Allison J. B. (1956) “Polyethylene ” Interscience Publishers
Ragan S. and Goodarzi F. (1984) "Disclinations in the optical texture of a high-rank

vitrinite coke and a petroleum needle-coke." Fuel 63(10) pp. 1382-1384
Riga A., Collins R. and Mlachak G. (1998) "Oxidative behavior of polymers by

thermogravimetric analysis, differential thermal analysis and pressure differential
scanning calorimetry." Thermochimica Acta 324(1-2) pp. 135-149
Rocha V. G., Blanco C., Santamaria R., Diestre E. I., Menendez R. and Granda M.
(2005) "Pitch/coke wetting behaviour." Fuel 84(12-13) pp. 1550-1556
272
Roskill (2007) The Economics of Petroleum Coke, 5th edition 2007 Industrial Mineral
Reports. London Roskill 247 pages
Russell K. E., Hunter B. K. and Heyding R. D. (1993) "Polyethylenes revisited: Waxs

and the phase structure." European Polymer Journal 29(2-3) pp. 211-217
Russell K. E., Hunter B. K. and Heyding R. D. (1997) "Monoclinic polyethylene

revisited." Polymer 38(6) pp. 1409-1414
Sadler B. A. and Welch B. J. (2001) “Anode Consumption Mechanisms — A Practical

Review of the Theory and Anode Property Considerations. ” In Proceedings of the 7th
Australasian Aluminium Smelting Technology Conference and Workshops; UNSW
Press: Sydney Australia pp. 294
Sahajwalla V, Rahman M., Khanna R, Saha-Chaudhury N., Knights D. and

O’Kane P. (2006) “Recycling of Waste Plastics for Slag Foaming in EAF
Steelmaking. ” AISTech 2006 Proceedings - Vol L Indianapolis, USA pp. 547-553
Sai'hi D., El-Achari A., Ghenaim A. and Claude C. (2002) "Wettability of grafted
polyethylene terephthalate) fibers." Polymer Testing 21 (6) pp. 615-618
Sakata Y., Uddin M. A., Koizumi K. and Murata K. (1996) "Catalytic Degradation
of Polypropylene into Liquid Hydrocarbons Using Silica-Alumina Catalyst." Chemistry
Letters 25(3) pp. 245-246
Scheirs and Kaminsky (2006) “Feedstock Recycling and Pyrolysis of Waste Plastics:
Converting Waste Plastics into Diesel and Other Fuels” John Wiley & Sons, Ltd.
Shah V. (2007) “Handbook of plastics testing and failure analysis - 3rd Edition”
Wiley-Interscience
Shlyapnikov Y. A. and Serenkova I. A. (1983) "High-temperature oxidation of

polymers." Die Angewandte Makromolekulare Chemie 114(1) pp. 1-11
273
Shulman Z. P. and Levitskiy S. P. (1996) "Growth of vapour bubbles in boiling

polymer solutions—I. Rheological and diffusional effects." International Journal of Heat
and Mass Transfer 39(3) pp. 631-638
Siddiqui M. N. and Redhwi H. H. (2009) "Pyrolysis of mixed plastics for the recovery
of useful products." Fuel Processing Technology 90(4) pp. 545-552
Simha R., Wall L. A. and Blatz P. J. (1950) "Depolymerization as a chain reaction."

Journal of Polymer Science 5(5) pp. 615-632
Strong A. B. (2006) “Plastics Materials and Processing ” Prentice Hall
Syunyaev Z. I. and Voloshin N. D. (1966) "Mechanism of coking and the calcination of

petroleum coke." Chemistry and Technology of Fuels & Oils 2 pp. 473-476
Ven S. V. D. (1990) “Polypropylene and other polyolefins : polymerization and

characterization ” Amsterdam, Elsevier Science Publishers
Villain F., Coudane J. and Vert Michel (1995) "Thermal degradation ofpolyethylene
terephthalate: study ofpolymer stabilization." Polymer Degradation and Stability 49(3)
pp. 393-397
Wall L. A. and Straus S. (1960) "Pyrolysis ofpolyolefins." Journal of Polymer Science

44(144) pp. 313-323
Wall L. A., Madorsky S. L., Brown D. W., Straus S. and Simha R. (1954) "The
Depolymerization of Polymethylene and Polyethylene." Journal of the American
Chemical Society 76(13) pp. 3430-3437
Wampler T. P. (1989) "Thermometric behavior of polyolefins." Journal of Analytical

and Applied Pyrolysis 15 pp. 187-195
274
Williams E. A. and Williams P. T. (1997) "The pyrolysis of individual plastics and a

plastic mixture in a fixed bed reactor." Journal of Chemical Technology &
Biotechnology 70(1) pp. 9-20
Williams P. T. and Slaney E. (2007) "Analysis of products from the pyrolysis and
liquefaction of single plastics and waste plastic mixtures." Resources, Conservation and
Recycling 51(4) pp. 754-769
Williams P. T. and Williams E. A. (1999) "Interaction of plastics in mixed-plastics

pyrolysis." Energy & Fuels 13(1) pp. 188-196
Wombles R. H. and Sadler B. (2004) “The Effect of Binder Pitch Quinoline Insolubles
Content on Aluminium Anode Physical Properties. ” 8th Australasian Aluminium
Smelting Technology Conference, Queensland, Australia
Xie X. L., Li R.K.Y, Tjong S.C and Mai Y.W (2002) "Structural properties and
mechanical behaviour of injection molded composites ofpolypropylene and sisal fiber."
Polymer Composites 23(3) pp. 319-328
Yaman S. (2004) "Pyrolysis of biomass to produce fuels and chemical feedstocks."

Energy Conversion and Management 45(5) pp. 651 -671
Yang J., Miranda R. and Roy C. (2001) "Using the DTG curve fitting method to
determine the apparent kinetic parameters of thermal decomposition of polymers."
Polymer Degradation and Stability 73(3) pp. 455-461
Yarin A. L., D. Lastochkin, Talmon Y. and Tadmor Z. (1999) "Bubble nucleation

during devolatilization ofpolymer melts." AIChE Journal 45(12) pp. 2590-2605
Yasushi S. and Kaoru F. (2003) "Catalytic degradation of PP with an Fe/activated

carbon catalyst. ''Journal of Material Cycles and Waste Management 5(2) pp. 107-112
275
Yu-Hu C. and Jou S. (2005) "Carbon nanotubes from catalytic pyrolysis of

polypropylene." Materials Chemistry and Physics 92 pp. 256-259
Zeisberger R. (1973) “Pigment Composition for Coloring Polypropylene ’’ Germany

Patent No. 3767444
Zhao W., Hasegawa S., Fujita J., Yoshii F., Sasaki T., Makuuchi K., Sun J. and
Nishimoto S. (1996) "Effects of zeolites on the pyrolysis of polypropylene." Polymer
Degradation and Stability 53(1) pp. 129-135.
276

S Baker, 2011. Fundamental Investigations On Petroleum Coke and Polymer Interactions For Recycling Waste Polymers in Aluminium Production

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

S Baker, 2011. Fundamental Investigations On Petroleum Coke and Polymer Interactions For Recycling Waste Polymers in Aluminium Production

Uploaded by

Copyright:

Available Formats

Fundamental investigations on petroleum coke and polymer

interactions for recycling waste polymers in aluminium

Downloaded from http://hdl.handle.net/1959.4/55284 in https://

FUNDAMENTAL INVESTIGATIONS ON PETROLEUM COKE

Submitted as Partial Fulfilment

First and foremost I offer my sincerest gratitude to my supervisors, Professor Veena

I am grateful to Professor Dato' Dr. Kamarudin Hussin, my supervisor in Malaysia, who

CHAPTER 2 LITERATURE REVIEW 6

CHAPTER 4 RESULTS & DISCUSSION: 78

CHAPTER 6 RESULT & DISCUSSION: 160

CHAPTER 7 RESULT & DISCUSSION: INFLUENCE 217

CHAPTER 9 REFERENCES 261

Figure 2.2 : Breakdown of mechanically recycled waste by plastic types 11

Figure 2.3 : Reaction mechanism of polymer (Scheirs and Kaminsky, 26

Figure 2.4 : Random chain scission in polyethylene 27

Figure 2.5 : Degradation mechanisms of polyethylene as suggested by 28

Figure 2.6 : Polymerization of propylene monomer with metal catalyst 30

Figure 2.7 : Stereochemical configurations of polypropylene 31

Figure 2.8 : Oxidative radicals formation of PP 32

Figure 2.9 : The major products of PP by their pathways 37

Figure 2.10 : TGA and DSC curves of thermal degradation of PP in 37

Figure 2.11 : Product of PP pyrolysis as a function of temperature 38

Figure 2.12 : General reaction scheme of carbonization |Lewis,1982] 41

Figure 2.13 : Carbonization and graphitization of organic materials 43

Figure 2.14 : Schematic illustration of the solid-liquid wetting principle 47

Figure 2.15 : Effect of temperature on the droplet on the solid surface in 48

Figure 2.16 : Contact angle of liquid droplet and solid surface 49

Figure 3.1 : Raw polypropylene samples 58

Figure 3.2 : Raw polyethylene samples 59

Figure 3.3 : Raw polypropylene black samples 59

Figure 3.4 : PET glass and raw samples after ground 60

Figure 3.5 : Raw bakelite samples 60

Figure 3.7 : RockLab ring milling machine 64

Figure 3.8 : Enerpac 10 Tonne Hydraulic Press 64

Figure 3.9 : Petroleum coke substrate 64

Figure 3.10 : Horizontal tube furnace used for pyrolysis 66

Figure 3.11 : Polymer on PC substrate assemblyon alumina powder 66

Figure 3.13 : Petroleum coke was screened by Retsch sieving machine to 69

Figure 3.14 : Fritsch knife grinder 69

Figure 3.15 : Schematic diagram of horizontal tube furnace set up for 71

Figure 3.16 : Perkin Elmer thermogravimetric analyser 73

Figure 3.17 : LECO CS 444 carbon and sulphur analyzer 74

Figure 3.18 : Philips Xpert Multipurpose X-ray Diffraction System 75

Figure 3.19 : Sample preparation material and instruments (a) KBr 76

Figure 3.20 : Nicolet Avatar 320 FT-IR instrument 76

Figure 3.21 : Hitachi S3400-X scanning electron microscope 77

Figure 4.1 : Raw PP on petroleum coke substrate before heat treatment 79

Figure 4.2 : PP on petroleum coke substrate after exposure to 150°C for 80

Figure 4.3 : PP on petroleum coke substrate and the corresponding 80

Figure 4.5 : PP on petroleum coke substrate and the corresponding 81

Figure 4.6 : PP on petroleum coke substrate and the corresponding 82

Figure 4.7 : PP on petroleum coke substrate and the corresponding 82

Figure 4.8 : PP on petroleum coke substrate and the corresponding 82

Figure 4.9 : PP on petroleum coke substrate and the corresponding 83

Figure 4.10 : PP on petroleum coke substrate and the corresponding 83

Figure 4.11 : PP on petroleum coke substrate and the corresponding 84

Figure 4.12 : PP on petroleum coke substrate and the corresponding 84

Figure 4.13 : PP on petroleum coke substrate and the corresponding 85

Figure 4.14 : PP on petroleum coke substrate and the corresponding 85

Figure 4.16 : EDS analysis of points in the interface of PP-PC system 88

Figure 4.18 : Interfacial changes due to heating at 250°C for different 90

Figure 4.19 : Interfacial changes due to heating at 300°C for different 91

Figure 4.20 : Schematic diagram of PP melt penetration into PC 92